lwig R. Struik Internet-Draft Struik Security Consultancy Intended status: Informational~~March 9,~~April 23,2020 Expires:~~September 10,~~October 25,2020 Alternative Elliptic Curve Representations~~draft-ietf-lwig-curve-representations-09~~draft-ietf-lwig-curve-representations-10Abstract This document specifies how to represent Montgomery curves and (twisted) Edwards curves as curves in short-Weierstrass form and illustrates how this can be used to carry out elliptic curve computations using existing implementations of, e.g., ECDSA and ECDH using NIST prime curves. We also provide extensive background material that may be useful for implementers of elliptic curve cryptography. Requirements Language The key words "MUST", "MUST NOT", "REQUIRED", "SHALL", "SHALL NOT", "SHOULD", "SHOULD NOT", "RECOMMENDED", "NOT RECOMMENDED", "MAY", and "OPTIONAL" in this document are to be interpreted as described in BCP 14 [RFC2119] [RFC8174] when, and only when, they appear in all capitals, as shown here. Status of This Memo This Internet-Draft is submitted in full conformance with the provisions of BCP 78 and BCP 79. Internet-Drafts are working documents of the Internet Engineering Task Force (IETF). Note that other groups may also distribute working documents as Internet-Drafts. The list of current Internet- Drafts is at https://datatracker.ietf.org/drafts/current/. Internet-Drafts are draft documents valid for a maximum of six months and may be updated, replaced, or obsoleted by other documents at any time. It is inappropriate to use Internet-Drafts as reference material or to cite them other than as "work in progress." This Internet-Draft will expire on~~September 10,~~October 25,2020. Copyright Notice Copyright (c) 2020 IETF Trust and the persons identified as the document authors. All rights reserved. This document is subject to BCP 78 and the IETF Trust's Legal Provisions Relating to IETF Documents (https://trustee.ietf.org/license-info) in effect on the date of publication of this document. Please review these documents carefully, as they describe your rights and restrictions with respect to this document. Code Components extracted from this document must include Simplified BSD License text as described in Section 4.e of the Trust Legal Provisions and are provided without warranty as described in the Simplified BSD License. Table of Contents 1. Fostering Code Reuse with New Elliptic Curves . . . . . . . . 5 2. Specification of Wei25519 . . . . . . . . . . . . . . . . . . 5 3. Use of Representation Switches . . . . . . . . . . . . . . .~~5~~64. Examples . . . . . . . . . . . . . . . . . . . . . . . . . .~~6~~74.1. Implementation of X25519 . . . . . . . . . . . . . . . .~~6~~74.2. Implementation of Ed25519 . . . . . . . . . . . . . . . . 7 4.3. Specification of ECDSA25519 . . . . . . . . . . . . . . .~~7~~84.4. Other Uses (Wei448, ECDH448, ECDSA448, and Others) . . . 8 5. Caveats . . . . . . . . . . . . . . . . . . . . . . . . . . . 9 5.1. Wire Format . . . . . . . . . . . . . . . . . . . . . . . 9 5.2. Representation Conventions . . . . . . . . . . . . . . . 9 5.3. Domain Parameters . . . . . . . . . . . . . . . . . . . .~~9~~106. Implementation Considerations . . . . . . . . . . . . . . . . 10 7. Implementation Status . . . . . . . . . . . . . . . . . . . .~~11~~128. Security Considerations . . . . . . . . . . . . . . . . . . . 12 9. Privacy Considerations . . . . . . . . . . . . . . . . . . . 13 10.Using Wei25519 and Wei448 with COSE and JOSE . . . . . . . . 14 10.1. Using Wei25519 and Wei448 Keys with COSE and JOSE . . . 14 10.1.1. Encoding of Short-Weierstrass Curves with COSE . . . 14 10.1.2. Encoding of Short-Weierstrass Curves with JOSE . . . 15 10.2. Using ECDSA25519 and ECDSA448 with COSE and JOSE . . . . 16 10.2.1. Encoding of ECDSA Instantiations with COSE . . . . . 17 10.2.2. Encoding of ECDSA Instantiations with JOSE . . . . . 17 10.3. Using ECDH25519 and ECDH448 with COSE and JOSE . . . . . 18 10.3.1. Encoding of co-factor ECDH with COSE . . . . . . . . 19 10.3.2. Encoding of co-factor ECDH with JOSE . . . . . . . . 19 11.IANA Considerations . . . . . . . . . . . . . . . . . . . . .~~13 10.1.~~20 11.1.IANA Considerations for Wei25519 . . . . . . . . . . . .~~13 10.1.1.~~20 11.1.1.COSE Elliptic Curves Registration . . . . . . . . .~~14 10.1.2.~~20 11.1.2.COSE Algorithms Registration (1/2) . . . . . . . . .~~14 10.1.3.~~21 11.1.3.COSE Algorithms Registration (2/2) . . . . . . . . .~~14 10.1.4.~~21 11.1.4.JOSE Elliptic Curves Registration . . . . . . . . .~~15 10.1.5.~~21 11.1.5.JOSE Algorithms Registration (1/2) . . . . . . . . .~~15 10.1.6.~~22 11.1.6.JOSE Algorithms Registration (2/2) . . . . . . . . .~~16 10.2.~~22 11.2.IANA Considerations for Wei448 . . . . . . . . . . . . .~~16 10.2.1.~~23 11.2.1.COSE Elliptic Curves Registration . . . . . . . . .~~16 10.2.2.~~23 11.2.2.COSE Algorithms Registration (1/2) . . . . . . . . .~~17 10.2.3.~~23 11.2.3.COSE Algorithms Registration (2/2) . . . . . . . . .~~17 10.2.4.~~23 11.2.4.JOSE Elliptic Curves Registration . . . . . . . . .~~17 10.2.5.~~24 11.2.5.JOSE Algorithms Registration (1/2) . . . . . . . . .~~18 10.2.6.~~24 11.2.6.JOSE Algorithms Registration (2/2) . . . . . . . . .~~18 11.~~25 12.Acknowledgements . . . . . . . . . . . . . . . . . . . . . .~~18 12.~~25 13.References . . . . . . . . . . . . . . . . . . . . . . . . .~~19 12.1.~~25 13.1.Normative References . . . . . . . . . . . . . . . . . .~~19 12.2.~~25 13.2.Informative References . . . . . . . . . . . . . . . . .~~20~~28Appendix A. Some (Non-Binary) Elliptic Curves . . . . . . . . .~~22~~29A.1. Curves in Short-Weierstrass Form . . . . . . . . . . . .~~22~~29A.2. Montgomery Curves . . . . . . . . . . . . . . . . . . . .~~22~~30A.3. Twisted Edwards Curves . . . . . . . . . . . . . . . . .~~23~~30Appendix B. Elliptic Curve Nomenclature and Finite Fields . . .~~23~~30B.1. Elliptic Curve Nomenclature . . . . . . . . . . . . . . .~~23~~30B.2. Finite Fields . . . . . . . . . . . . . . . . . . . . . .~~25~~32Appendix C. Elliptic Curve Group Operations . . . . . . . . . .~~26~~33C.1. Group Laws for Weierstrass Curves . . . . . . . . . . . .~~26~~33C.2. Group Laws for Montgomery Curves . . . . . . . . . . . .~~27~~34C.3. Group Laws for Twisted Edwards Curves . . . . . . . . . .~~28~~35Appendix D. Relationship Between Curve Models . . . . . . . . .~~29~~36D.1. Mapping between Twisted Edwards Curves and Montgomery Curves . . . . . . . . . . . . . . . . . . . . . . . . .~~29~~36D.2. Mapping between Montgomery Curves and Weierstrass Curves~~30~~37D.3. Mapping between Twisted Edwards Curves and Weierstrass Curves . . . . . . . . . . . . . . . . . . . . . . . . .~~31~~38Appendix E. Curve25519 and Cousins . . . . . . . . . . . . . . .~~31~~38E.1. Curve Definition and Alternative Representations . . . .~~31~~38E.2. Switching between Alternative Representations . . . . . .~~31~~38E.3. Domain Parameters . . . . . . . . . . . . . . . . . . . .~~33~~40Appendix F. Further Mappings . . . . . . . . . . . . . . . . . .~~35~~42F.1. Isomorphic Mapping between Twisted Edwards Curves . . . .~~35~~42F.2. Isomorphic Mapping between Montgomery Curves . . . . . .~~36~~43F.3. Isomorphic Mapping between Weierstrass Curves . . . . . .~~36~~43F.4. Isogenous Mapping between Weierstrass Curves . . . . . .~~37~~44Appendix G. Further Cousins of Curve25519 . . . . . . . . . . .~~39~~46G.1. Further Alternative Representations . . . . . . . . . . .~~39~~46G.2. Further Switching . . . . . . . . . . . . . . . . . . . .~~39~~46G.3. Further Domain Parameters . . . . . . . . . . . . . . . .~~40~~47G.4. Isogeny Details . . . . . . . . . . . . . . . . . . . . .~~41~~48G.4.1. Isogeny Parameters . . . . . . . . . . . . . . . . .~~42~~49G.4.2. Dual Isogeny Parameters . . . . . . . . . . . . . . .~~48~~55Appendix H. Point Compression . . . . . . . . . . . . . . . . .~~54~~61H.1. Point Compression for Weierstrass Curves . . . . . . . .~~54~~61H.2. Point Compression for Montgomery Curves . . . . . . . . .~~55~~62H.3. Point Compression for Twisted Edwards Curves . . . . . .~~56~~63Appendix I. Data Conversions . . . . . . . . . . . . . . . . . .~~57~~64I.1.Strings and String Operations . . . . . . . . . . . . . . 64 I.2.Conversion between Bit Strings and Integers (BS2I, I2BS)~~57 I.2.~~64 I.3.Conversion between Octet Strings and Integers (OS2I, I2OS) . . . . . . . . . . . . . . . . . . . . . . . . . .~~58 I.3.~~65 I.4.Conversion between Octet Strings and Bit Strings (OS2BS, BS2OS) . . . . . . . . . . . . . . . . . . . . . . . . .~~58 I.4.~~65 I.5.Conversion between Field Elements and Octet Strings (FE2OS, OS2FE) . . . . . . . . . . . . . . . . . . . . .~~58 I.5.~~66 I.6.Conversion between Elements of Z mod n and Octet Strings (ZnE2OS, OS2ZnE) . . . . . . . . . . . . . . . . . . . .~~59 I.6.~~66 I.7.Ordering Conventions . . . . . . . . . . . . . . . . . .~~59~~67 I.8. Conversion Between Curve Points and Octet Strings . . . . 68Appendix J. Representation Examples Curve25519 Family Members .~~61~~70J.1. Example with Curve25519 . . . . . . . . . . . . . . . . .~~61~~71J.2. Example with Edwards25519 . . . . . . . . . . . . . . . .~~63~~73J.3. Example with Wei25519 . . . . . . . . . . . . . . . . . .~~65~~74J.4. Example with Wei25519.2 . . . . . . . . . . . . . . . . .~~67~~77J.5. Example with Wei25519.-3 . . . . . . . . . . . . . . . .~~68~~78Appendix K. Auxiliary Functions . . . . . . . . . . . . . . . .~~70~~80K.1. Square Roots in GF(q) . . . . . . . . . . . . . . . . . .~~70~~80K.1.1. Square Roots in GF(q), where q = 3 (mod 4) . . . . .~~70~~80K.1.2. Square Roots in GF(q), where q = 5 (mod 8) . . . . .~~70~~80K.2. Inversion . . . . . . . . . . . . . . . . . . . . . . . .~~71~~81K.3. Mappings to Curve Points . . . . . . . . . . . . . . . .~~71~~81K.3.1. Mapping to Points of Weierstrass Curve . . . . . . .~~71~~81K.3.2. Mapping to Points of Montgomery Curve . . . . . . . .~~72~~82K.3.3. Mapping to Points of Twisted Edwards Curve . . . . .~~74~~83K.4. Mappings to High-Order Curve Points . . . . . . . . . . .~~74~~84K.4.1. Mapping to High-Order Points of Weierstrass Curve . .~~74~~84K.4.2. Mapping to High-Order Points of Montgomery Curve . .~~75~~85K.4.3. Mapping to High-Order Points of Twisted Edwards Curve~~76~~86K.5. Randomized Representation of Curve Points . . . . . . . .~~77~~87Appendix L. Curve secp256k1 and Friend . . . . . . . . . . . . .~~78~~88L.1. Curve Definition and Alternative Representation . . . . .~~78~~88L.2. Switching Between Representations . . . . . . . . . . . .~~79~~89L.3. Domain Parameters . . . . . . . . . . . . . . . . . . . .~~79~~89L.4. Isogeny Details . . . . . . . . . . . . . . . . . . . . .~~81~~91L.4.1. Isogeny Parameters . . . . . . . . . . . . . . . . .~~81~~91L.4.2. Dual Isogeny Parameters . . . . . . . . . . . . . . .~~81~~91Appendix M. Curve448 and Cousins . . . . . . . . . . . . . . . .~~82~~92M.1. Curve Definition and Alternative Representations . . . .~~82~~92M.2. Switching between Alternative Representations . . . . . .~~83~~93M.3. Domain Parameters . . . . . . . . . . . . . . . . . . . .~~84~~94Appendix N. Further Cousins of Curve448 . . . . . . . . . . . .~~87~~97N.1. Further Alternative Representations . . . . . . . . . . .~~87~~97N.2. Further Switching . . . . . . . . . . . . . . . . . . . .~~87~~97N.3. Further Domain Parameters . . . . . . . . . . . . . . . .~~89~~99N.4. Isogeny Details . . . . . . . . . . . . . . . . . . . . .~~91~~101N.4.1. Isogeny Parameters . . . . . . . . . . . . . . . . .~~92~~102N.4.2. Dual Isogeny Parameters . . . . . . . . . . . . . . .~~92~~102Appendix O. Representation Examples Curve448 Family Members . .~~93~~103O.1. Example with Curve448 . . . . . . . . . . . . . . . . . .~~93~~104O.2. Example with Ed448 . . . . . . . . . . . . . . . . . . .~~96~~106O.3. Example with Wei448 . . . . . . . . . . . . . . . . . . .~~98~~108O.4. Example with Wei448.-3 . . . . . . . . . . . . . . . . .~~101~~111O.5. Example with Edwards448 . . . . . . . . . . . . . . . . .~~103~~113Author's Address . . . . . . . . . . . . . . . . . . . . . . . .~~105~~1151. Fostering Code Reuse with New Elliptic Curves Elliptic curves can be represented using different curve models. Recently, IETF standardized elliptic curves that are claimed to have better performance and improved robustness against "real world" attacks than curves represented in the traditional "short" Weierstrass model. This document specifies an alternative representation of points of Curve25519, a so-called Montgomery curve, and of points of Edwards25519, a so-called twisted Edwards curve, which are both specified in [RFC7748], as points of a specific so- called "short" Weierstrass curve, called Wei25519. We also define how to efficiently switch between these different representations. Use of Wei25519 allows easy definition of new instantiations of signature schemes and key agreement schemes already specified for traditional NIST prime curves, thereby allowing easy integration with existing specifications, such as NIST SP 800-56a [SP-800-56a], FIPS Pub 186-4 [FIPS-186-4], and ANSI X9.62-2005 [ANSI-X9.62], and fostering code reuse on platforms that already implement some of these schemes using elliptic curve arithmetic for curves in "short" Weierstrass form (see Appendix C.1). 2. Specification of Wei25519 For the specification of Wei25519 and its relationship to Curve25519 and Edwards25519, see Appendix E. For further details and background information on elliptic curves, we refer to the other appendices. The use of Wei25519 allows reuse of existing generic code that implements short-Weierstrass curves, such as the NIST curve P-256, to also implement the CFRG curves Curve25519 and Edwards25519. (Here, generic code refers to an implementation that does not depend on hardcoded domain parameters (see also Section 6).) We also cater to reusing of existing code where some domain parameters may have been hardcoded, thereby widening the scope of applicability. To this end, we specify the short-Weierstrass curves Wei25519.2 and Wei25519.-3, with hardcoded domain parameter a=2 and a=-3 (mod p), respectively; see Appendix G. (Here, p is the characteristic of the field over which these curves are defined.) 3. Use of Representation Switches The curves Curve25519, Edwards25519, and Wei25519, as specified in Appendix E.3, are all isomorphic, with the transformations of Appendix E.2. These transformations map the specified base point of each of these curves to the specified base point of each of the other curves. Consequently, a public-private key pair (k,R:=k*G) for any one of these curves corresponds, via these isomorphic mappings, to the public-private key pair (k, R':=k*G') for each of these other curves (where G and G' are the corresponding base points of these curves). This observation extends to the case where one also considers curve Wei25519.2 (which has hardcoded domain parameter a=2), as specified in Appendix G.3, since it is isomorphic to Wei25519, with the transformation of Appendix G.2, and, thereby, also isomorphic to Curve25519 and Edwards25519. The curve Wei25519.-3 (which has hardcoded domain parameter a=-3 (mod p)) is not isomorphic to the curve Wei25519, but is related in a slightly weaker sense: the curve Wei25519 is isogenous to the curve Wei25519.-3, where the mapping of Appendix G.2 is an isogeny of degree l=47 that maps the specified base point G of Wei25519 to the specified base point G' of Wei25519.-3 and where the so-called dual isogeny (which maps Wei25519.-3 to Wei25519) has the same degree l=47, but does not map G' to G, but to a fixed multiple hereof, where this multiple is l=47. Consequently, a public-private key pair (k,R:=k*G) for Wei25519 corresponds to the public-private key pair (k, R':= k*G') for Wei25519.-3 (via the l-isogeny), whereas the public-private key pair (k, R':=k*G') corresponds to the public- private key pair (l*k, l*R=l*k*G) of Wei25519 (via the dual isogeny). (Note the extra scalar l=47 here.) Alternative curve representations can, therefore, be used in any cryptographic scheme that involves computations on public-private key pairs, where implementations may carry out computations on the corresponding object for the isomorphic or isogenous curve and convert the results back to the original curve (where, in case this involves an l-isogeny, one has to take into account the factor l). This includes use with elliptic-curve based signature schemes and key agreement and key transport schemes. For some examples of curve computations on each of the curves specified in Appendix E.3 and Appendix G.3, see Appendix J. 4. Examples 4.1. Implementation of X25519 RFC 7748 [RFC7748] specifies the use of X25519, a co-factor Diffie- Hellman key agreement scheme, with instantiation by the Montgomery curve Curve25519. This key agreement scheme was already specified in Section 6.1.2.2 of NIST SP 800-56a [SP-800-56a] for elliptic curves in short Weierstrass form. Hence, one can implement X25519 using existing NIST routines by (1) representing a point of the Montgomery curve Curve25519 as a point of the Weierstrass curve Wei25519; (2) instantiating the co-factor Diffie-Hellman key agreement scheme of the NIST specification with the resulting point and Wei25519 domain parameters; (3) representing the key resulting from this scheme (which is a point of the curve Wei25519 in Weierstrass form) as a point of the Montgomery curve Curve25519. The representation change can be implemented via a simple wrapper and involves a single modular addition (see Appendix E.2). Using this method has the additional advantage that one can reuse the public-private key pair routines, domain parameter validation, and other checks that are already part of the NIST specifications. A NIST-compliant version of co-factor Diffie-Hellman key agreement (denoted by ECDH25519) results if one keeps inputs (key contributions) and outputs (shared key) in the short-Weierstrass format (and, hence, does not perform~~Step~~Steps (1) and(3) above). NOTE: At this point, it is unclear whether this implies that a FIPS- accredited module implementing co-factor Diffie-Hellman for, e.g., P-256 would also extend this accreditation to X25519. 4.2. Implementation of Ed25519 RFC 8032 [RFC8032] specifies Ed25519, a "full" Schnorr signature scheme, with instantiation by the twisted Edwards curve Edwards25519. One can implement the computation of the ephemeral key pair for Ed25519 using an existing Montgomery curve implementation by (1) generating a public-private key pair (k, R':=k*G') for Curve25519; (2) representing this public-private key as the pair (k, R:=k*G) for Ed25519. As before, the representation change can be implemented via a simple wrapper. Note that the Montgomery ladder specified in Section 5 of RFC7748 [RFC7748] does not provide sufficient information to reconstruct R':=(u, v) (since it does not compute the v-coordinate of R'). However, this deficiency can be remedied by using a slightly modified version of the Montgomery ladder that includes reconstruction of the v-coordinate of R':=k*G' at the end of the Montgomery ladder (which uses the v-coordinate of the base point of Curve25519 as well). For details, see Appendix C.1. 4.3. Specification of ECDSA25519 FIPS Pub 186-4 [FIPS-186-4] specifies the signature scheme ECDSA and can be instantiated not just with the NIST prime curves, but also with other Weierstrass curves (that satisfy additional cryptographic criteria). In particular, one can instantiate this scheme with the Weierstrass curve Wei25519 and the hash function SHA-256 [FIPS-180-4], where an implementation may generate an ephemeral public-private key pair for Wei25519 by (1) internally carrying out these computations on the Montgomery curve Curve25519, the twisted Edwards curve Edwards25519, or even the Weierstrass curve Wei25519.-3 (with hardcoded a=-3 domain parameter); (2) representing the result as a key pair for the curve Wei25519. Note that, in either case, one can implement these schemes with the same representation conventions as used with existing NIST specifications, including bit/byte- ordering, compression functions, and the-like. This allows generic implementations of ECDSA with the hash function SHA-256 and with the NIST curve P-256 or with the curve Wei25519 specified in this specification to reuse the same implementation (instantiated with, respectively, the NIST P-256 elliptic curve domain parameters or with the domain parameters of curve Wei25519 specified in Appendix E). We denote by ECDSA25519 the instantiation of ECDSA with SHA-256 and with curve Wei25519, where the signature (r,s) is represented as the right-concatenation of the integers r and s, each represented as fixed-size strings with tight MSB/msb ordering (see Appendix I). 4.4. Other Uses (Wei448, ECDH448, ECDSA448, and Others) Any existing specification of cryptographic schemes using elliptic curves in Weierstrass form and that allows introduction of a new elliptic curve (here: Wei25519) is amenable to similar constructs, thus spawning "offspring" protocols, simply by instantiating these using the new curve in "short" Weierstrass form, thereby allowing code and/or specifications reuse and, for implementations that so desire, carrying out curve computations "under the hood" on Montgomery curve and twisted Edwards curve cousins hereof (where these exist). This would simply require definition of a new object identifier for any such envisioned "offspring" protocol. This could significantly simplify standardization of schemes and help keeping the resource and maintenance cost of implementations supporting algorithm agility [RFC7696] at bay. We illustrate the construction of such offspring protocols for Curve448, another Montgomery curve recently standardized by IETF (see [RFC7748]). Similar to the case with Curve25519, one can represent points of this curve via different curve models, viz. as points of an Edwards curve (Ed448) or as points of a short-Weierstrass curve (Wei448). For the specification of Wei448 and its relationship to Curve448 and Ed448, see Appendix M. As with ECDH25519, one can now easily define a NIST-compliant version of co-factor Diffie-Hellman key agreement (denoted by ECDH448), by simply reusing the example of Section 4.1, but now using the short-Weierstrass curve Wei448, rather than Wei25519. Similarly, one can easily specify ECDSA with Wei448 and a suitable hash function, by simply reusing the example of Section 4.3, but now using the short-Weierstrass curve Wei448, rather than Wei25519, and picking as hash function SHAKE256 [FIPS-202] with output size of d=446 bits. We denote by ECDSA448 the resulting signature scheme (with the same bit/byte-ordering conventions). 5. Caveats The examples above illustrate how specifying the Weierstrass curve Wei25519 (or any curve in short-Weierstrass format, for that matter) may facilitate reuse of existing code and may simplify standards development. However, the following caveats apply: 5.1. Wire Format The transformations between alternative curve representations can be implemented at negligible relative incremental cost if the curve points are represented as affine points. If a point is represented in compressed format, conversion usually requires a costly point decompression step. This is the case in [RFC7748], where the inputs to the co-factor Diffie-Hellman scheme X25519, as well as its output, are represented in u-coordinate-only format. This is also the case in [RFC8032], where the EdDSA signature includes the ephemeral signing key represented in compressed format (see Appendix H for details). Note that in the latter case compression is lossless, whereas it is lossy in the former~~case;~~case.5.2. Representation Conventions While elliptic curve computations are carried-out in a field GF(q) and, thereby, involve large integer arithmetic, these integers are represented as bit- and byte-strings. Here, [RFC8032] uses least- significant-byte (LSB)/least-significant-bit (lsb) conventions, whereas [RFC7748] uses LSB/most-significant-bit (msb) conventions, and where most other cryptographic specifications, including NIST SP800-56a [SP-800-56a], FIPS Pub 186-4 [FIPS-186-4], and ANSI X9.62-2005 [ANSI-X9.62] use~~MSB/msb~~most-significant-byte (MSB)/msbconventions. Since each pair of conventions is different (see Appendix I for details and Appendix J for examples), this does necessitate bit/byte representation~~conversions;~~conversions.5.3. Domain Parameters All traditional NIST curves are Weierstrass curves with domain parameter a=-3, while all Brainpool curves [RFC5639] are isomorphic to a Weierstrass curve of this form. Thus, one can expect there to be existing Weierstrass implementations with a hardcoded a=-3 domain parameter ("Jacobian-friendly"). For those implementations, including the curve Wei25519 as a potential vehicle for offering support for the CFRG curves Curve25519 and Edwards25519 is not possible, since not of the required form. Instead, one has to implement Wei25519.-3 and include code that implements the isogeny and dual isogeny from and to Wei25519. The lowest odd-degree isogeny has degree l=47 and requires roughly 9kB of storage for isogeny and dual-isogeny computations (see the tables in Appendix G.4). Note that storage would have reduced to a single 64-byte table if only theCurve25519curve would have been generated so as to be isomorphic to a Weierstrass curve with hardcoded a=-3 parameter (this corresponds to l=1). NOTE 1: An example of a Montgomery curve defined over the same field as Curve25519 that is isomorphic to a Weierstrass curve with hardcoded a=-3 parameter is the Montgomery curve M_{A,B} with B=1 and A=-1410290 (or, if one wants the base point to still have u-coordinate u=9, with B=1 and A=-3960846). In either case, the resulting curve has the same cryptographic properties as Curve25519 and the same performance (which relies on A being a 3-byte integer, as is the case with the domain parameter A=486662 of Curve25519, and using the same special prime p=2^255-19), while at the same time being "Jacobian-friendly" by design. NOTE 2: While an implementation of Curve25519 via an isogenous Weierstrass curve with domain parameter a=-3 requires a relatively large table (of size roughly 9kB), for the quadratic twist of Curve25519 (i.e., the Montgomery curve M_{A,B'} with A=486662 and B'=2) this implementation approach only requires a table of size less than 0.5kB (over 20x smaller), solely due to the fact that it is l-isogenous to a Weierstrass curve with a=-3 parameter with relatively small parameter l=2 (compared to l=47, as is the case with Curve25519 itself). 6. Implementation Considerations The efficiency of elliptic curve arithmetic is primarily determined by the efficiency of its group operations (see Appendix C). Numerous optimized formulae exist, such as the use of so-called Montgomery ladders with Montgomery curves [Mont-Ladder] or with Weierstrass curves [Wei-Ladder], the use of hardcoded a=-3 domain parameter for Weierstrass curves [ECC-Isogeny], and the use of hardcoded a=-1 domain parameters for twisted Edwards curves [tEd-Formulas]. These all target reduction of the number of finite field operations (primarily, finite field multiplications and squarings). Other optimizations target more efficient modular reductions underlying these finite field operations, by specifying curves defined over a field GF(q), where the field size q has a special form or a specific bit-size (typically, close to a multiple of a machine word). Depending on the implementation strategy, the bit-size of q may also facilitate reduced so-called "carry-effects" of integer arithmetic. Most curves use a combination of these design philosophies. All NIST curves [FIPS-186-4] and Brainpool curves [RFC5639] are Weierstrass curves with a=-3 domain parameter, thus facilitating more efficient elliptic curve group operations (via so-called Jacobian coordinates). The NIST curves and the Montgomery curve Curve25519 are defined over prime fields, where the prime number has a special form, whereas the Brainpool curves - by design - use a generic prime number. None of the NIST prime curves, nor the Brainpool curves, can be expressed as Montgomery or twisted Edwards curves, whereas - conversely - Montgomery curves and twisted curves can be expressed as Weierstrass curves. While use of Wei25519 allows reuse of existing generic code that implements short Weierstrass curves, such as the NIST curve P-256, to also implement the CFRG curves Curve25519 or Edwards25519, this obviously does not result in an implementation of these CFRG curves that exploits the specific structure of the underlying field or other specific domain parameters (since generic). Reuse of generic code, therefore, may result in a less computationally efficient curve implementation than would have been possible if the implementation had specifically targeted Curve25519 or Edwards25519 alone (with the overall cost differential estimated to be somewhere in the interval [1.00-1.25]). If existing generic code offers hardware support, however, the overall speed may still be larger, since less efficient formulae for curve arithmetic using Wei25519 curves compared to a direct implementation of Curve25519 or Edwards25519 arithmetic may be more than compensated for by faster implementations of the finite field arithmetic itself. Overall, one should consider not just code reuse and computational efficiency, but also development and maintenance cost, and, e.g, the cost of providing effective implementation attack countermeasures (see also Section 8). 7. Implementation Status [Note to the RFC Editor] Please remove this entire section before publication, as well as the reference to [RFC7942]. This section records the status of known implementations of the protocol defined by this specification at the time of posting of this Internet-Draft, and is based on a proposal described in [RFC7942]. The description of implementations in this section is intended to assist the IETF in its decision processes in progressing drafts to RFCs. Please note that the listing of any individual implementation here does not imply endorsement by the IETF. Furthermore, no effort has been spent to verify the information presented here that was supplied by IETF contributors. This is not intended as, and must not be construed to be, a catalog of available implementations or their features. Readers are advised to note that other implementations may exist. According to [RFC7942], "this will allow reviewers and working groups to assign due consideration to documents that have the benefit of running code, which may serve as evidence of valuable experimentation and feedback that have made the implemented protocols more mature. It is up to the individual working groups to use this information as they see fit. Nikolas Rosener evaluated the performance of switching between different curve models in his Master's thesis [Rosener]. For an implementation of Wei25519, see <https://github.com/ncme/c25519>. For support of this curve in tinydtls, see <https://github.com/ncme/ tinydtls>. According to <https://community.nxp.com/docs/DOC-330199>, an implementation of Wei25519 on the Kinets LTC ECC HW platform improves the performance by over a factor ten compared to a stand-alone implementation of Curve25519 without hardware support. The signature scheme ECDSA25519 (see Section 4.3) is supported in <https://datatracker.ietf.org/doc/draft-ietf-6lo-ap-nd/>. 8. Security Considerations The different representations of elliptic curve points discussed in this document are all obtained using a publicly known transformation, which is either an isomorphism or a low-degree isogeny. It is well- known that an isomorphism maps elliptic curve points to equivalent mathematical objects and that the complexity of cryptographic problems (such as the discrete logarithm problem) of curves related via a low-degree isogeny are tightly related. Thus, the use of these techniques does not negatively impact cryptographic security of elliptic curve operations. As to implementation security, reusing existing high-quality code or generic implementations that have been carefully designed to withstand implementation attacks for one curve model may allow a more economical way of development and maintenance than providing this same functionality for each curve model separately (if multiple curve models need to be supported) and, otherwise, may allow a more gradual migration path, where one may initially use existing and accredited chipsets that cater to the pre-dominant curve model used in practice for over 15 years. Elliptic curves are generally used as objects in a broader cryptographic scheme that may include processing steps that depend on the representation conventions used (such as with, e.g., key derivation following key establishment). These schemes should (obviously) unambiguously specify fixed representations of each input and output (e.g., representing each elliptic curve point always in short-Weierstrass form and in uncompressed tight MSB/msb format). To prevent cross-protocol attacks, private keys SHOULD only be used with one cryptographic scheme. Private keys MUST NOT be reused between Ed25519 (as specified in [RFC8032]) and ECDSA25519 (as specified in Section 4.3). To prevent intra-protocol cross-instantiation attacks, ephemeral private keys MUST NOT be reused between instantiations of~~ECDSA25519.~~ECDSA25519 or ECDSA448.9. Privacy Considerations The transformations between different curve models described in this document are publicly known and, therefore, do not affect privacy provisions. Use of a public key in any protocol for which successful execution evidences knowledge of the corresponding private key implicitly indicates the entity holding this private key. Reuse of this public key with more than one protocol or more than one protocol instantiation may, therefore, allow traceability of this entity. It may also allow correlation of meta-data communicated with this common data element (e.g., different addressing information), even if an observer cannot technically verify the binding of this meta-data. The randomized representation described in Appendix K.5 allows random curve points to be represented as random pairs of field elements, thereby assisting in obfuscating the presence of these curve points in some applications. 10.~~IANA Considerations Code points are requested for curve~~UsingWei25519 and Wei448with COSEand~~its~~JOSE This section defines algorithm encodings and representations enabling the use of the curves Wei25519 and Wei448 and theiruse withECDH andECDSAwith JOSE [RFC7518]and~~co-factor ECDH, using~~COSE [RFC8152] messages. All octet string encodings below usethe~~representation~~MSB/msb-orderingconventions~~of this document. New code points would be required~~as definedin~~case one wishes~~Appendix I.7. For CBOR representation details, we referto~~specify one or more other "offspring" protocols beyond those exemplified in Section 4.4. Specification hereof is, however, outside scope of the current document. 10.1. IANA Considerations~~[RFC7049];forbase64url encodings, we refer to [RFC4648]. 10.1. UsingWei25519~~10.1.1.~~and Wei448 Keys withCOSE~~Elliptic Curves Registration This section registers~~and JOSE For Weierstrass curves,the~~following value in~~representation ofthe~~IANA "COSE Elliptic Curves" registry [IANA.COSE.Curves]. Name: Wei25519; Value: TBD (Requested value: -1); Key Type: EC2 or OKP (where OKP uses~~point at infinity O is curve-specific (see Appendix H.1). Forthe~~squeezed MSB/msb representation of this specification); Description:~~short-Weierstrass curve~~Wei25519; Change Controller: IESG; Reference: Appendix E.3 of this specification; Recommended: Yes. (Note that The "kty" value~~Wei25519, we define O:=(-1,0), whereasfor~~Wei25519 may be "OKP" or "EC2".) 10.1.2. COSE Algorithms Registration (1/2) This section registers the following value in~~Wei448, we define O:=(1,0). The encodings below specifythe~~IANA "COSE Algorithms" registry [IANA.COSE.Algorithms]. Name: ECDSA25519; Value: TBD (Requested value: -1); Description: ECDSA~~use of short-Weierstrass curveswith~~SHA-256~~COSE (see Section 10.1.1)and~~curve Wei25519; Change Controller: IESG; Reference:~~JOSE (seeSection~~4.3~~10.1.2), where the encoding for a specific curve results by setting the "crv" parameter to the unique nameof~~this specification; Recommended: Yes. 10.1.3. COSE Algorithms Registration (2/2) This section registers~~the~~following value~~curveinquestion (e.g., "Wei25519" forthe~~IANA "COSE Algorithms" registry [IANA.COSE.Algorithms]. Name: ECDH25519; Value: TBD (Requested value: -2); Description: NIST-compliant co-factor Diffie-Hellman w/~~curve Wei25519 and"Wei448" for the curve Wei448). 10.1.1. Encoding of Short-Weierstrass Curves with COSE With COSE, points of short-Weierstrass curves are encoded using the "EC2"key~~derivation function HKDF SHA256; Change Controller: IESG; Reference: Section 4.1~~type (Section 13.1.1of~~this specification (for~~[RFC8152]) or the "OKP"key~~derivation, see Section 11.1~~type (Section 7.2of~~[RFC8152]); Recommended: Yes. 10.1.4. JOSE Elliptic Curves Registration This section registers~~[I-D.ietf-cose-rfc8152bis-algs]), which are instantiated by settingthe~~following value in~~"crv" parameter tothe~~IANA "JSON Web Key Elliptic Curve" registry [IANA.JOSE.Curves]. Curve Name: Wei25519; Curve Description: short-Weierstrass curve Wei25519; JOSE Implementation Requirements: Optional; Change Controller: IESG; Reference: Appendix E.3~~(unique) nameof~~this specification. 10.1.5. JOSE Algorithms Registration (1/2) This section registers~~the~~following value~~curvein~~the IANA "JSON Web Signature and Encryption Algorithms" registry [IANA.JOSE.Algorithms]. Algorithm Name: ECDSA25519; Algorithm Description: ECDSA using SHA-256~~questionand~~curve Wei25519; Algorithm Usage Locations: alg; JOSE Implementation Requirements: Optional; Change Controller: IESG; Reference: Section 4.3 of this specification; Algorithm Analysis Document(s): Section 4.3~~the "kty" parameter to "EC2" or "OKP", respectively, where key type-specific settings are as follows: a. With the "EC2" type, each affine point (X, Y) is encoded by setting the parameters "x" and "y" to the octet string representationsof~~this specification. 10.1.6. JOSE Algorithms Registration (2/2) This section registers~~the~~following value~~elements X and Y, respectively,intight MSB/msb-order, and converting each to a CBOR byte string. Each compressed point (X, t) is encoded by settingthe~~IANA "JSON Web Signature~~parameter "x" to the octet representation of the element X, in tight MSB/msb- order, converted to a CBOR byte string,and~~Encryption Algorithms" registry [IANA.JOSE.Algorithms]. Algorithm Name: ECDH25519; Algorithm Description: NIST-compliant co-factor Diffie-Hellman w/ curve Wei25519~~by setting the parameter "y" to the CBOR False or CBOR True value, depending on whether, respectively, t=0 or t=1. For representation detailsand~~key derivation function HKDF SHA256; Algorithm Usage Locations: alg; JOSE Implementation Requirements: Optional; Change Controller: IESG; Reference: Section 4.1 of this specification (for key derivation,~~for details on the reverse mappings,see~~Section 5 of [SP-800-56c]); Algorithm Analysis Document(s): Section 4.1 of~~Appendix I.8. (Note that for affine pointsthis~~specification (for key derivation, see~~representation is consistent with the "EC2" representation inSection~~5~~13.1.1of~~[SP-800-56c]). 10.2. IANA Considerations for Wei448 10.2.1. COSE Elliptic Curves Registration This section registers~~[RFC8152].) b. Withthe~~following value in~~"OKP" type, each point is encoded by settingthe~~IANA "COSE Elliptic Curves" registry [IANA.COSE.Curves]. Name: Wei448; Value: TBD (Requested value: -2); Key Type: EC2 or OKP (where OKP uses~~parameter "x" tothe~~squeezed MSB/msb~~"squeezed" pointrepresentation of this~~specification); Description: short-Weierstrass curve Wei448; Change Controller: IESG; Reference: Appendix M.3 of~~point, in MSB/msb-order, and convertingthis~~specification; Recommended: Yes. (Note that The "kty" value~~to a CBOR byte string. For representation details andfor~~Wei448 may be "OKP" or "EC2".) 10.2.2. COSE Algorithms Registration (1/2) This section registers the following value in~~details onthe~~IANA "COSE Algorithms" registry [IANA.COSE.Algorithms]. Name: ECDSA448; Value: TBD (Requested value: -21); Description: ECDSA~~reverse mappings, see Appendix I.8. (Note that for affine points this representation is consistentwith~~SHAKE256 and curve Wei448; Change Controller: IESG; Reference:~~the "OKP" representation inSection~~4.4~~7.2of[I-D.ietf-cose-rfc8152bis-algs], which affords a curve-specific octet string encoding.) In either case, if the point is a public key, the parameter "d" encodes the corresponding private key, using the octet string representation, in tight MSB/msb-order, and convertingthis~~specification; Recommended: Yes. 10.2.3. COSE Algorithms Registration (2/2) This section registers~~to a CBOR byte string (see Appendix I.6). For curve points,the~~following value~~"crv" parameter and the parameters referenced with the applicable key type-specific settings above MUST be presentin the~~IANA "COSE Algorithms" registry [IANA.COSE.Algorithms]. Name: ECDH448; Value: TBD (Requested value: -22); Description: NIST-compliant co-factor Diffie-Hellman w/ curve Wei25519~~structure, whereas the parameter "d" MUST NOT be present, while for private keys, the parameters "crv" and "d" MUST be presentandthe applicablekey~~derivation function HKDF SHA512; Change Controller: IESG; Reference: Section 4.4~~type-specific parametersof~~this specification (for key derivation, see Section 11.1~~the corresponding public-key are RECOMMENDED to be present. 10.1.2. Encodingof~~[RFC8152]); Recommended: Yes. 10.2.4. JOSE Elliptic~~Short-WeierstrassCurves~~Registration This section registers the following value in the IANA "JSON Web Key Elliptic Curve" registry [IANA.JOSE.Curves]. Curve Name: Wei448; Curve Description: short-Weierstrass curve Wei448;~~withJOSE~~Implementation Requirements: Optional; Change Controller: IESG; Reference: Appendix M.3~~With JOSE, pointsof~~this specification. 10.2.5. JOSE Algorithms Registration (1/2) This section registers~~short-Weierstrass curves are encoded usingthe~~following value in~~"EC" key type (Section 6.2 of [RFC7518]) orthe~~IANA "JSON Web Signature and Encryption Algorithms" registry [IANA.JOSE.Algorithms]. Algorithm Name: ECDSA448; Algorithm Description: ECDSA using SHAKE256 and curve Wei448; Algorithm Usage Locations: alg; JOSE Implementation Requirements: Optional; Change Controller: IESG; Reference: Section 4.4~~"OKP" key type (Section 2of~~this specification; Algorithm Analysis Document(s): Section 4.4~~[RFC8037]), which are instantiated by setting the "crv" parameter to the (unique) nameof~~this specification. 10.2.6. JOSE Algorithms Registration (2/2) This section registers~~the~~following value~~curvein~~the IANA "JSON Web Signature~~questionand~~Encryption Algorithms" registry [IANA.JOSE.Algorithms]. Algorithm Name: ECDH448; Algorithm Description: NIST-compliant co-factor Diffie-Hellman w/~~the "kty" parameter to "EC" or "OKP", respectively, where key type-specific settings are as follows: a. With the "EC" type, each affinecurve~~Wei25519~~point (X, Y) is encoded by setting the parameters "x"and~~key derivation function HKDF SHA512; Algorithm Usage Locations: alg; JOSE Implementation Requirements: Optional; Change Controller: IESG; Reference: Section 4.4~~"y" to the octet string representationsofthe elements X and Y, respectively, in tight MSB/msb-order, and converting each using the base64url encoding. The point at infinity O is encoded as ifthis~~specification (for key derivation,~~were an affine point. For representation details and details on the reverse mappings,see~~Section 5 of [SP-800-56c]); Algorithm Analysis Document(s): Section 4.4 of~~Appendix I.8. (Note that for affine pointsthis~~specification (for key derivation, see~~representation is consistent with the "EC" representation inSection~~5~~6.2of~~[SP-800-56c]). 11. Acknowledgements Thanks~~[RFC7518]).) b. With the "OKP" type, each curve point is encoded by setting the parameter "x"to~~Nikolas Rosener for discussions surrounding implementation details of~~the~~techniques described~~"squeezed" point representation of this point,inMSB/msb-order, and convertingthis~~document~~using the base64url encoding. For representation detailsand~~to Phillip Hallam-Baker~~for~~triggering inclusion of verbiage~~detailson the~~use of Montgomery ladders~~reverse mappings, see Appendix I.8. (Note that for affine points this representation is consistentwith~~recovery of~~the~~y-coordinate. Thanks to Stanislav Smyshlyaev~~"OKP" representation in Section 2 of [RFC8037], which affords a curve-specific octet string encoding.) In either case, if the point is a public key (i.e., the private key is well-defined), the parameter "d" encodes the corresponding private key, using the octet string representation, in tight MSB/msb-order,and~~Vasily Nikolaev for their careful reviews. 12. References 12.1. Normative References [ANSI-X9.62] ANSI X9.62-2005, "Public Key Cryptography for~~converting this usingthe~~Financial Services Industry: The Elliptic Curve Digital Signature Algorithm (ECDSA)", American National Standard~~base64url encoding (see Appendix I.6). For curve points, the "crv" parameter and the parameters referenced with the applicable key type-specific settings above MUST be present in the structure, whereas the parameter "d" MUST NOT be present, whilefor~~Financial Services, Accredited Standards Committee X9, Inc, Anapolis, MD, 2005. [FIPS-180-4] FIPS 180-4, "Secure Hash Standard (SHS), Federal Information Processing Standards Publication 180-4", US Department of Commerce/National Institute of Standards~~private keys, the parameters "crv"and~~Technology, Gaithersburg, MD, August 2015. [FIPS-186-4] FIPS 186-4, "Digital Signature Standard (DSS), Federal Information Processing Standards Publication 186-4", US Department of Commerce/National Institute~~"d" MUST be present and the applicable key type-specific parametersof~~Standards~~the corresponding public-key are RECOMMENDED to be present. 10.2. Using ECDSA25519and~~Technology, Gaithersburg, MD, July 2013. [FIPS-202]~~ECDSA448 with COSE and JOSEFIPS~~202, "SHA-3 Standard: Permutation-Based Hash~~Pub 186-4 [FIPS-186-4] specifies the signature scheme ECDSAand~~Extendable-Output Functions, Federal Information Processing Standards Publication 202", US Department of Commerce/National Institute~~can be instantiated with suitable combinationsof~~Standards and Technology, Gaithersburg, MD, August 2015. [RFC2119] Bradner, S., "Key words for use~~elliptic curvesin~~RFCs to Indicate Requirement Levels", BCP 14, RFC 2119, DOI 10.17487/RFC2119, March 1997, <https://www.rfc-editor.org/info/rfc2119>. [RFC5639] Lochter, M.~~short-Weierstrass formand~~J. Merkle, "Elliptic Curve Cryptography (ECC) Brainpool Standard Curves and Curve Generation", RFC 5639, DOI 10.17487/RFC5639, March 2010, <https://www.rfc-editor.org/info/rfc5639>. [RFC7696] Housley, R., "Guidelines for Cryptographic Algorithm Agility~~hash functions (that satisfy particular cryptographic criteria). While this completely specifies the internal workings of the signingand~~Selecting Mandatory-to-Implement Algorithms", BCP 201, RFC 7696, DOI 10.17487/RFC7696, November 2015, <https://www.rfc-editor.org/info/rfc7696>. [RFC7748] Langley, A., Hamburg, M.,~~signature verification operations, this does not uniquely specify the input/output formats: a. The signing operation takes as inputs a message m (represented as a bit string)and~~S. Turner, "Elliptic Curves for Security", RFC 7748, DOI 10.17487/RFC7748, January 2016, <https://www.rfc-editor.org/info/rfc7748>. [RFC7942] Sheffer, Y.~~a private key d in the interval [1,n-1]and~~A. Farrel, "Improving Awareness~~produces as output a signature, which is an ordered pair (r, s)of~~Running Code:~~integers in the interval [1,n-1], where n is the order of the base point of the curve in question; b.The~~Implementation Status Section", BCP 205, RFC 7942, DOI 10.17487/RFC7942, July 2016, <https://www.rfc-editor.org/info/rfc7942>. [RFC8032] Josefsson, S.~~signature verification operation takes as inputs a message m, a public key Q,and~~I. Liusvaara, "Edwards-Curve Digital Signature Algorithm (EdDSA)", RFC 8032, DOI 10.17487/RFC8032, January 2017, <https://www.rfc-editor.org/info/rfc8032>. [RFC8152] Schaad, J., "CBOR Object Signing~~a signature (r,s)and~~Encryption (COSE)", RFC 8152, DOI 10.17487/RFC8152, July 2017, <https://www.rfc-editor.org/info/rfc8152>. [RFC8174] Leiba, B., "Ambiguity~~produces as output the value "valid" or "invalid", depending upon whether the message was purportedly signed by a holderof~~Uppercase vs Lowercase in RFC 2119 Key Words", BCP 14, RFC 8174, DOI 10.17487/RFC8174, May 2017, <https://www.rfc-editor.org/info/rfc8174>. [SEC1] SEC1, "SEC 1: Elliptic Curve Cryptography, Version 2.0", Standards for Efficient Cryptography, , June 2009. [SEC2] SEC2, "SEC 2: Elliptic Curve Cryptography, Version 2.0", Standards~~the private key of the public-private key pair (d, Q)for~~Efficient Cryptography, , January 2010. [SP-800-56a] NIST SP 800-56a, "Recommendation~~the curve used with the signature scheme in question. All inputs and outputs are uniquely determined by specifying the encodings of the message m, the private key d, the public key Q, the signature, and the values "valid" and "invalid". The encodings below specify the use of instantiations of ECDSA with COSE (see Section 10.2.1) and JOSE (see Section 10.2.2), where the encodingfor~~Pair-Wise Key Establishment Schemes Using Discrete Log Cryptography, Revision 3", US Department~~a specific ECDSA instantiation (i.e., with a specific short-Weierstrass curve and specific hash function) results by setting the "crv" parameter to the unique nameof~~Commerce/National Institute~~the underlying curve in question and the "alg" parameter to the unique nameof~~Standards~~the specific signature scheme scheme instantiation (e.g., "ECDSA25519" for the ECDSA scheme defined in Section 4.3and~~Technology, Gaithersburg, MD, April 2018. [SP-800-56c] NIST SP 800-56c, "Recommendation~~"ECDSA448"for~~Key-Derivation Methods~~the scheme definedin~~Key-Establishment Schemes, Revision 1", US Department~~Section 4.4). Note that, in this case, the "alg" name uniquely defines the curve (and, thereby, implicitly the underlying "crv" parameter) and the underlying hash function. 10.2.1. Encodingof~~Commerce/National Institute~~ECDSA Instantiations with COSE Instantiationsof~~Standards~~ECDSA used with COSE use the following encodings of inputsand~~Technology, Gaithersburg, MD, April 2018. 12.2. Informative References [ECC] I.F. Blake, G. Seroussi, N.P. Smart, "Elliptic Curves in Cryptography", Cambridge University Press, Lecture Notes Series 265, July 1999. [ECC-Isogeny] E. Brier, M. Joye, "Fast Point Multiplication on Elliptic Curves through Isogenies", AAECC, Lecture Notes~~outputs: a. The message m is the COSE_Sign structure as specifiedin~~Computer Science, Vol. 2643, New York: Springer-Verlag, 2003. [GECC] D. Hankerson, A.J. Menezes, S.A. Vanstone, "Guide~~Section 4.1 of [RFC8152], convertedto~~Elliptic Curve Cryptography", New York: Springer-Verlag, 2004. [IANA.COSE.Algorithms] IANA, "COSE Algorithms", IANA, https://www.iana.org/assignments/cose/ cose.xhtml#algorithms. [IANA.COSE.Curves] IANA, "COSE Elliptic Curves", IANA, https://www.iana.org/assignments/cose/cose.xhtml#elliptic- curves. [IANA.JOSE.Algorithms] IANA, "JSON Web Signature and Encryption Algorithms", IANA, https://www.iana.org/assignments/jose/jose.xhtml#web- signature-encryption-algorithms. [IANA.JOSE.Curves] IANA, "JSON Web Key Elliptic Curve", IANA, https://www.iana.org/assignments/jose/jose.xhtml#web-key- elliptic-curve. [Mont-Ladder] P.L. Montgomery, "Speeding~~a bit-string, usingthe~~Pollard~~OS2BS mapping of Appendix I.4; b. The public key Qand~~Elliptic Curve Methods~~the private key d are encoded as specified in Section 10.1.1, where the "crv" parameter is set to the unique nameof~~Factorization", Mathematics~~the curve used with this particular instantiationof~~Computation, Vol. 48, 1987. [Rosener] N. Rosener, "Evaluating~~ECDSA; c. The Cose signature is encoded asthe~~Performance~~right-concatenationof~~Transformations Between Curve Representations in Elliptic Curve Cryptography for Constrained Device Security", M.Sc. Universitat Bremen, August 2018. [SWUmap] E. Brier, J-S. Coron, Th. Icart, D. Madore, H. Randriam, M. Tibouchi, "Efficient Indifferentiable Hashing into Ordinary Elliptic Curves", CRYPTO 2010, Lecture Notes in Computer Science, Vol. 6223, New York: Springer-Verlag, 2010. [tEd] D.J. Bernstein, P. Birkner, M. Joye, T. Lange, C. Peters, "Twisted Edwards Curves", Africacrypt 2008, Lecture Notes in Computer Science, Vol. 5023, New York: Springer-Verlag, 2008. [tEd-Formulas] H. Hisil, K.K.H. Wong, G. Carter, E. Dawson, "Twisted Edwards Curves Revisited", ASIACRYPT 2008, Lecture Notes in Computer Science, Vol. 5350, New York: Springer-Verlag, 2008. [Tibouchi] M. Tibouchi, "Elligator Squared -- Uniform Points on Elliptic Curves~~the octet string representationsof~~Prime Order~~the coordinates of the signature pair (r, s), in left-to-right order, where r and s are each representedas~~Uniform Random Strings", Financial Cryptography 2014, Lecture Notes~~octet stringsin~~Computer Science, Vol. 8437, New York: Springer-Verlag, 2014. [Wei-Ladder] T. Izu, Ts. Takagi,, "A Fast Parallel Elliptic Curve Multiplication Resistant Against Side Channel Attacks", Centre for Applied Cryptographic Research, Corr 2002-03, 2002. Appendix A. Some (Non-Binary) Elliptic Curves This section defines~~tight MSB/msb-order usingthe~~three different curve models~~ZnE2OS mapping of Appendix I.6, converted to a CBOR byte string. Note that, sincewe~~consider, viz. short-Weierstrass curves, Montgomery curves, and twisted Edwards curves. A.1. Curves in Short-Weierstrass Form Let GF(q) denote the finite field with q elements,~~use a tight representation, this right- concatenated octet string has fixed size 2*l,where~~q~~the parameter lis~~an odd prime power~~uniquely defined by the set Z_n in question. The inverse mapping results by checking that the purported encoded signature (after CBOR decoding) has indeed size 2*l,and~~where q is not divisible~~by~~three. Let W_{a,b} be~~convertingthe~~Weierstrass curve with defining equation Y^2 = X^3 + a*X + b, where a~~left-sideand~~b are elements~~right-side halvesof~~GF(q) and where 4*a^3 + 27*b^2 is nonzero. The points~~this octet string (eachof~~W_{a,b} are~~length l) to, respectively,the~~ordered pairs (X, Y) whose coordinates are elements of GF(q)~~integers rand~~that satisfy~~s in Z_n, viathe~~defining equation (the so-called affine points), together with~~strict OS2ZnE mapping of Appendix I.6. When using a COSE key for this algorithm, ifthe~~special point O (the so-called "point at infinity"). This~~"alg" field is present, it MUST beset~~forms a group under addition, via~~tothe~~so-called "chord-and-tangent" rule, where~~(unique) name of this particular instantiation of ECDSA andthe~~point at infinity serves as~~"crv" parameter MUST be set tothe~~identity element. See Appendix C.1 for details~~(unique) nameof the~~group operation. A.2. Montgomery Curves Let GF(q) denote~~corresponding curve; ifthe~~finite~~"key_ops"field~~with q elements, where q~~ispresent, it MUST include "sign" when creatingan~~odd prime power. Let M_{A,B} be the Montgomery curve with defining equation B*v^2 = u^3 + A*u^2 + u, where A~~ECDSA signatureand~~B are elements~~it MUST include "verify" when verifying an ECDSA signature. 10.2.2. Encodingof~~GF(q) and where A is unequal to (+/-)2 and where B is nonzero. The points~~ECDSA Instantiations with JOSE Instantiationsof~~M_{A,B} are~~ECDSA used with JOSE usethe~~ordered pairs (u, v) whose coordinates are elements~~following encodingsof~~GF(q)~~inputsand~~that satisfy the defining equation (the so- called affine points), together with the special point O (the so- called "point at infinity"). This set forms a group under addition, via the so-called "chord-and-tangent" rule, where~~outputs: a. The message m isthe~~point at infinity serves~~JWS Signing Inputasspecified in [RFC7515], converted to a bit-string, usingthe~~identity element. See Appendix C.2 for details~~OS2BS mappingofAppendix I.4; b. The public key andthe~~group operation. A.3. Twisted Edwards Curves Let GF(q) denote the finite field with q elements,~~private key are encoded as specified in Section 10.1.2,where~~q~~the "crv" parameteris~~an odd prime power. Let E_{a,d} be~~set to the unique name ofthe~~twisted Edwards~~curveusedwith~~defining equation a*x^2 + y^2 = 1+ d*x^2*y^2, where a and d are distinct nonzero elements~~this particular instantiationof~~GF(q).~~ECDSA; c.The~~points~~JWS signature is encoded as the right-concatenation of the octet string representationsof~~E_{a,d} are~~the~~ordered pairs (x, y) whose~~coordinates~~are elements~~of~~GF(q) and that satisfy~~the~~defining equation (the so-called affine points). It can be shown that this set forms a group under addition if a is a square~~signature pair (r, s),in~~GF(q), whereas d is not,~~left-to-right order,where~~the point O:=(0, 1) serves as the identity element. (Note that the identity element satisfies the defining equation.) See~~r and s are each represented in tight MSB/msb-order (seeAppendix~~C.3 for details of~~I.7), converted usingthe~~group operation. An Edwards curve is~~base64url encoding. Note that, since we usea~~twisted Edwards curve with a=1. Appendix B. Elliptic Curve Nomenclature and Finite Fields This section provides brief background information on elliptic curves and finite fields that should be sufficient to understand constructions and examples in~~tight representation,this~~document. B.1. Elliptic Curve Nomenclature Each curve~~right-concatenated octet string has fixed size 2*l, where the parameter l is uniquelydefined~~in Appendix A forms a commutative group under addition (denoted~~by~~'+'). In Appendix C we specify the group laws, which depend on~~the~~curve model~~set Z_nin question.~~For completeness, we here include some common elliptic curve nomenclature~~The inverse mapping results by checking that the purported encoded signature (after base64url decoding) has indeed size 2*l,and~~basic properties (primarily so as to keep~~by converting the left-side and right-side halves ofthis~~document self-contained). These notions are mainly used in Appendix E~~octet string (each of length l) to, respectively, the integers rands in Z_n, via the strict OS2ZnE mapping ofAppendix~~G and not essential~~I.6. When using a JOSE keyfor~~our exposition. This section can~~this algorithm, if the "alg" field is present, it MUSTbe~~skipped at first reading. Any point P~~set to the (unique) nameof~~a curve E is a generator~~this particular instantiationofECDSA andthe~~cyclic subgroup (P):={k*P | k = 0, 1, 2,...}~~"crv" parameter MUST be set to the (unique) nameof the~~curve. (Here, k*P denotes~~corresponding curve; ifthe~~sum of k copies of P, where 0*P~~"key_ops" fieldispresent, it MUST include "sign" when creating an ECDSA signature and it MUST include "verify" when verifying an ECDSA signature; ifthe~~identity element O of the curve; k*P is commonly referred to as scalar multiplication of P by k.) If (P) has cardinality l, then l~~JWK _use_ fieldis~~called~~present, its value MUST be "sig". 10.3. Using ECDH25519 and ECDH448 with COSE and JOSE NIST SP 800-56a [SP-800-56a] specifiesthe~~order of P. The order of~~co-factor elliptic-curve Diffie-Hellman key agreement scheme (co-factor ECDH) and can be instantiated with a suitable ellipticcurve~~E is~~in short-Weierstrass form (that satisfies particular cryptographic criteria). While this completely specifiesthe~~cardinality~~internal workingsof the~~set of its points, commonly denoted by |E|. A curve is cyclic if it is generated by some point of~~key agreement scheme in question,this~~curve. All curves of prime order are cyclic, while all curves of order h*n, where n~~does not uniquely specify the input/output formats: a. The co-factor ECDH schemeis a~~large prime number~~two-party key agreement scheme that takes as inputs a private key d from one of the partiesand~~where h is~~a~~small number (the so-called co-factor), have~~point Q' obtained from the other party and producesa~~large cyclic subgroup of prime~~shared key K:=h*(d*Q'), where h and n are, respectively, the co-factor and theorder~~n. In this case, a generator~~of~~order n is called a~~thebase~~point, commonly denoted by G. A~~point of~~order dividing h is said to be in~~the~~small subgroup. For curves~~curve in question and where Q' is a pointof~~prime order,~~this~~small subgroup~~curve. If this shared key Kis the~~singleton set, consisting~~point at infinity Oof~~only~~the~~identity element O. A point that is not in~~curve,the~~small subgroup~~outputis~~said to be a high-order point (since it has order at least n).~~an error message; b.If~~R~~the shared key Kis~~a~~an affinepoint of the~~curve that is also contained in (P), there is a unique integer k in~~curve,the~~interval [0, l-1] so that R=k*P, where l~~outputis the~~order of P. This number~~(raw) shared secret Z, whichis~~called~~the~~discrete logarithm~~octet representationof~~R to the base P. The discrete logarithm problem is~~the~~problem~~x-coordinateof~~finding~~K, usingthe~~discrete logarithm~~FE2OS mappingof~~R to~~Appendix I.5, represented in tight-MSB/msb-order (see Appendix I.7). (NOTE: A subsequent key derivation function (kdf) takes as inputsthe~~base P for any two points P~~shared secret Zand~~R~~side information OtherInfo and produces as output an octet stringofDerivedKeyingMaterial, where details depend onthe~~curve, if such a number exists. If P~~used kdf in question. This stepis~~a fixed base point G~~outofscope.) The inputs and outputs are uniquely determined by specifyingthe~~curve,~~encodings of private keys, curve points, andthe~~pair (k, R:=k*G) is commonly called a public-private~~error message for thiskey~~pair, the integer k~~agreement scheme. The encodings below specifythe~~private key,~~use of instantiations of ECDH with COSE (see Section 10.3.1)andJOSE (see Section 10.3.2), wherethe~~point R~~encoding for a specific co-factor ECDH instantiation (i.e., with a specific short-Weierstrass curve) results by settingthe~~corresponding public key. The private key k can be represented as an integer in~~"crv" parameter tothe~~interval [0,n-1], where G has order n. If this representation is nonzero, R has order n; otherwise, it has order one~~unique name of the underlying curve in questionand~~is~~the~~identity element O~~"alg" parameter to the unique nameof the~~curve. In this document, a quadratic twist of a curve E~~specific key agreement scheme instantiation (e.g., "ECDH25519" for the co-factor ECDH schemedefined~~over a field GF(q) is a~~in Section 4.1 and "ECDH448" for the scheme defined in Section 4.4). Note that, in this case, the "alg" name uniquely defines thecurve~~E' related to E,~~(and, thereby, implicitly the underlying "crv" parameter). 10.3.1. Encoding of co-factor ECDHwith~~cardinality |E'|, where |E|+|E'|=2*(q+1). If E is a curve in one~~COSE Instantiationsofco-factor ECDH used with COSE usethe~~curve models~~following encodings of inputs and outputs: a. Curve points and private keys are encoded asspecified in~~this document, a quadratic twist~~Section 10.1.1, where the "crv" parameter is set to the unique nameof~~this curve can be expressed using~~the~~same~~curve~~model, although (naturally)~~usedwith~~its own curve parameters. Two curves E and E' defined over~~this particular instantiation of ECDH. When usinga~~field GF(q) are said to be isogenous~~COSE key for this algorithm,if~~these have~~the~~same order and are said to~~"alg" field is present, it MUSTbe~~isomorphic if these have the same group structure. Note that isomorphic curves have necessarily~~set tothe~~same order and are, thus, a special type~~(unique) nameof~~isogenous curves. Further details are out~~this particular instantiationof~~scope. Weierstrass curves can have prime order, whereas Montgomery curves~~co-factor ECDHand~~twisted Edwards curves always have an order that is a multiple of four (and, thereby, a small subgroup of cardinality four). An ordered pair (x, y) whose coordinates are elements of GF(q) can~~the "crv" parameter MUSTbe~~associated with any ordered triple~~set to the (unique) nameof the~~form [x*z: y*z: z], where z~~corresponding curve; if the "key_ops" fieldis~~a nonzero element of GF(q), and can be uniquely recovered from such a representation. The latter representation is commonly called a representation in projective coordinates. Sometimes, yet other representations are useful (e.g., representation in Jacobian coordinates). Further details are out of scope. The group laws in Appendix C are mostly expressed in terms~~present, it MUST include "key agreement" when creating an ECDH shared secret. 10.3.2. Encodingof~~affine points, but can also be expressed in terms~~co-factor ECDH with JOSE Instantiationsofco-factor ECDH used with JOSE usethe~~representation of these points in projective coordinates, thereby allowing clearing~~following encodingsof~~denominators. The group laws may also involve non-affine~~inputs and outputs: a. Curvepoints~~(such~~and private keys are encodedas~~the point at infinity O of a Weierstrass curve or of a Montgomery curve). Those can also be represented~~specifiedin~~projective coordinates. Further details are out of scope. B.2. Finite Fields The field GF(q),~~Section 10.1.2,where~~q is a prime power, is defined as follows. If q:=p~~the "crv" parameteris~~a prime number,~~set tothe~~field GF(p) consists~~unique nameof the~~integers in the interval [0,p-1] and two binary operations on~~curve used withthis~~set: addition and multiplication modulo p. This field is commonly called a prime field. If q:=p^m, where p is~~particular instantiation of ECDH. When usinga~~prime number and where m>0,~~JOSE key for this algorith, ifthe"alg"field~~GF(q)~~is~~defined in terms of an irreducible polynomial f(z) in z of degree m with coefficients in GF(p) (i.e., f(z) cannot~~present, it MUSTbe~~written as~~set tothe~~product of two polynomials in z~~(unique) nameof~~lower degree with coefficients in GF(p)): in~~this~~case, GF(q) consists~~particular instantiationof~~the polynomials in z of degree smaller than m with coefficients in GF(p) and two binary operations on this set: polynomial addition~~co-factor ECDHand~~polynomial multiplication modulo~~the~~irreducible polynomial f(z). By definition, each element x of GF(q) is a polynomial in z of degree smaller than m and can, therefore,~~"crv" parameter MUSTbe~~uniquely represented as a vector (x_{m-1}, x_{m-2}, ..., x_1, x_0) of length m with coefficients in GF(p), where x_i is~~set tothe~~coefficient of z^i~~(unique) nameof~~polynomial x. Note that this representation depends on~~the~~irreducible polynomial f(z) of~~corresponding curve; ifthe"key_ops"field~~GF(p^m) in question (which~~is~~often fixed in practice). Note that GF(q) contains the prime field GF(p) as a subset. If m=1, the definitions of GF(p)~~present, it MUST include "key agreement" when creating an ECDH shared secret. 11. IANA Considerations Code points are requested for curves Wei25519and~~GF(p^1) above coincide, since each nonzero element~~Wei448 and their use with ECDSA and co-factor ECDH, using the representation conventionsof~~GF(p) can~~this document. New code points wouldbe~~viewed as a polynomial~~requiredin~~z of degree zero. If m>1, then GF(q) is called a (nontrivial) extension field~~case one wishes to specify one or more other "offspring" protocols beyond those exemplified in Section 4.4. Specification hereof is, however, outside scopeof~~GF(p). The number p is called~~the~~characteristic of GF(q). A field element y is called a square in GF(q) if it can be expressed as y:=x^2~~current document. 11.1. IANA Considerationsfor~~some x in GF(q); it is called a non-square~~Wei25519 11.1.1. COSE Elliptic Curves Registration This section registers the following valuein~~GF(q) otherwise. If y is a square~~the IANA "COSE Elliptic Curves" registry [IANA.COSE.Curves]. Name: Wei25519; Value: TBD (Requested value: -1); Key Type: EC2 or OKP; Description: short-Weierstrass curve Wei25519; Change Controller: IESG; Reference: specifiedin~~GF(q), we denote by sqrt(y) one~~Appendix E.3of~~its square roots (the other one being -sqrt(y)). For methods~~this specification;for~~computing square roots and inverses in GF(q) - if these exist -~~encodings,see~~Appendix K.1 and Appendix K.2, respectively. For methods for mapping a nonzero field element~~Section 10.1; Recommended: Yes. (Notethat~~is not a square in GF(q) to a point of a curve, see Appendix K.3. NOTE:~~The~~curves in Appendix E and Appendix G are all defined over a prime field GF(p), thereby reducing all operations to simple modular integer arithmetic. Strictly speaking we could, therefore, have refrained from introducing extension fields. Nevertheless, we included~~"kty" value for Wei25519 may be "EC2" or "OKP".) 11.1.2. COSE Algorithms Registration (1/2) This section registersthe~~more general exposition, so as to accommodate potential introduction of new curves that are defined over a (nontrivial) extension field at some point~~following valuein the~~future. This includes curves proposed for post-quantum isogeny-based schemes, which are defined over a quadratic extension field (i.e., where q:=p^2), and elliptic curves used~~IANA "COSE Algorithms" registry [IANA.COSE.Algorithms]. Name: ECDSA25519; Value: TBD (Requested value: -1); Description: ECDSAwith~~pairing-based cryptography. The exposition in either case is almost the same~~SHA-256and~~now automatically yields, e.g., data conversion routines for any finite field object (see Appendix I). Readers not interested~~curve Wei25519; Change Controller: IESG; Reference: specifiedin~~this, could simply view all fields as prime fields. Appendix C. Elliptic Curve Group Operations~~Section 4.3 of this specification; for encodings, see Section 10.2; Recommended: Yes. 11.1.3. COSE Algorithms Registration (2/2)This section~~specifies group operations for elliptic curves~~registers the following valuein~~short- Weierstrass form, for Montgomery curves,~~the IANA "COSE Algorithms" registry [IANA.COSE.Algorithms]. Name: ECDH25519; Value: TBD (Requested value: -2); Description: NIST-compliant co-factor Diffie-Hellman w/ curve Wei25519andkey derivation function HKDF SHA256; Change Controller: IESG; Reference: defined in Section 4.1 of this specification;for~~twisted Edwards curves. C.1. Group Laws for Weierstrass~~encodings, see Section 10.3; Recommended: Yes. 11.1.4. JOSE EllipticCurves~~For each point P of the Weierstrass curve W_{a,b},~~Registration This section registersthe~~point at infinity O serves as identity element, i.e., P + O = O + P = P. For each affine point P:=(X, Y) of~~following value inthe~~Weierstrass~~IANA "JSON Web Key Elliptic Curve" registry [IANA.JOSE.Curves]. Curve Name: Wei25519; Curve Description: short-Weierstrasscurve~~W_{a,b}, the point -P is the point (X, -Y) and one has P + (-P) = O (i.e., -P is the inverse~~Wei25519; JOSE Implementation Requirements: Optional; Change Controller: IESG; Reference: specified in Appendix E.3of~~P). For the point at infinity O, one has -O:=O. Let P1:=(X1, Y1) and P2:=(X2, Y2)~~this specification; for encodings, see Section 10.1. (Note that The "kty" value for Wei25519 maybe~~distinct affine points of~~"EC" or "OKP".) 11.1.5. JOSE Algorithms Registration (1/2) This section registersthe~~Weierstrass curve W_{a,b} and let Q:=P1 + P2, where Q is not~~following value inthe~~identity element. Then Q:=(X, Y), where X + X1 + X2 = lambda^2~~IANA "JSON Web Signatureand~~Y + Y1 = lambda*(X1 - X), where lambda:= (Y2 - Y1)/(X2 - X1). Let P:=(X1, Y1) be an affine point of the Weierstrass curve W_{a,b} and let Q:=2*P, where Q is not the identity element. Then Q:=(X, Y), where X + 2*X1 = lambda^2 and Y + Y1 = lambda*(X1 - X), where lambda:=(3*X1^2 + a)/(2*Y1). From the group laws above it follows that if P=(X, Y), P1=(X1, Y1),~~Encryption Algorithms" registry [IANA.JOSE.Algorithms]. Algorithm Name: ECDSA25519; Algorithm Description: ECDSA using SHA-256and~~P2=(X2, Y2) are distinct affine points of the Weierstrass~~curve~~W_{a,b} with P2:=P+P1 and if Y is nonzero, then the Y-coordinate of P1 can be expressed~~Wei25519; Algorithm Usage Locations: alg; JOSE Implementation Requirements: Optional; Change Controller: IESG; Reference: specifiedin~~terms of the X-coordinates~~Section 4.3of~~P, P1, and P2, and the Y-coordinate~~this specification; for encodings, see Section 10.2; Algorithm Analysis Document(s): Section 4.3of~~P, since 2*Y*Y1=(X*X1+a)*(X+X1)+2*b-X2*(X-X1)^2.~~this specification. 11.1.6. JOSE Algorithms Registration (2/2)This~~property allows recovery of the Y-coordinate of a point P1=k*P that is computed via the so-called Montgomery ladder, where P is an affine point with nonzero Y-coordinate (i.e., it does not have order two). For future reference, note that the expression above uniquely determines~~section registersthe~~X-coordinate of P2~~following valuein~~terms of~~the~~X-coordinates of P~~IANA "JSON Web Signatureand~~P1~~Encryption Algorithms" registry [IANA.JOSE.Algorithms]. Algorithm Name: ECDH25519; Algorithm Description: NIST-compliant co-factor Diffie-Hellman w/ curve Wei25519and~~the product of their Y-coordinates. Further details are out~~key derivation function HKDF SHA256; Algorithm Usage Locations: alg; JOSE Implementation Requirements: Optional; Change Controller: IESG; Reference: Section 4.1of~~scope. C.2. Group Laws~~this specification;for~~Montgomery Curves For each point P~~encodings, see Section 10.1; Algorithm Analysis Document(s): Section 4.1ofthis specification. 11.2. IANA Considerations for Wei448 11.2.1. COSE Elliptic Curves Registration This section registersthe~~Montgomery curve M_{A,B}, the point at infinity O serves as identity element, i.e., P + O = O + P = P. For each affine point P:=(u, v) of~~following value inthe~~Montgomery~~IANA "COSE Elliptic Curves" registry [IANA.COSE.Curves]. Name: Wei448; Value: TBD (Requested value: -2); Key Type: EC2 or OKP; Description: short-Weierstrasscurve~~M_{A,B},~~Wei448; Change Controller: IESG; Reference: specified in Appendix M.3 of this specification; for encodings, see Section 10.1; Recommended: Yes. (Note that The "kty" value for Wei448 may be "EC2" or "OKP".) 11.2.2. COSE Algorithms Registration (1/2) This section registersthe~~point -P is~~following value inthe~~point (u, -v)~~IANA "COSE Algorithms" registry [IANA.COSE.Algorithms]. Name: ECDSA448; Value: TBD (Requested value: -47); Description: ECDSA with SHAKE256and~~one has P + (-P) = O (i.e., -P is the inverse~~curve Wei448; Change Controller: IESG; Reference: Section 4.4of~~P). For~~this specification; Recommended: Yes. 11.2.3. COSE Algorithms Registration (2/2) This section registersthe~~point at infinity O, one has -O:=O. Let P1:=(u1, v1) and P2:=(u2, v2) be distinct affine points of~~following value inthe~~Montgomery~~IANA "COSE Algorithms" registry [IANA.COSE.Algorithms]. Name: ECDH448; Value: TBD (Requested value: -48); Description: NIST-compliant co-factor Diffie-Hellman w/curve~~M_{A,B}~~Wei448and~~let Q:=P1 + P2, where Q is not~~key derivation function HKDF SHA512; Change Controller: IESG; Reference: Section 4.4 of this specification (for key derivation, see Section 11.1 of [RFC8152]); Recommended: Yes. 11.2.4. JOSE Elliptic Curves Registration This section registersthe~~identity element. Then Q:=(u, v), where u + u1 + u2 = B*lambda^2 - A and v + v1 = lambda*(u1 - u), where lambda:=(v2 - v1)/(u2 - u1). Let P:=(u1, v1) be an affine point of the Montgomery curve M_{A,B} and let Q:=2*P, where Q is not the identity element. Then Q:=(u, v), where u + 2*u1 = B*lambda^2 - A and v + v1 = lambda*(u1 - u), where lambda:=(3*u1^2 + 2*A*u1+1)/(2*B*v1). From the group laws above it follows that if P=(u, v), P1=(u1, v1), and P2=(u2, v2) are distinct affine points of~~following value inthe~~Montgomery~~IANA "JSON Web Key Elliptic Curve" registry [IANA.JOSE.Curves]. Curve Name: Wei448; Curve Description: short-Weierstrasscurve~~M_{A,B} with P2:=P+P1 and if v is nonzero, then the v-coordinate~~Wei448; JOSE Implementation Requirements: Optional; Change Controller: IESG; Reference: specified in Appendix M.3of~~P1 can~~this specification; for encodings, see Section 10.1. (Note that The "kty" value for Wei448 maybe~~expressed~~"EC" or "OKP".) 11.2.5. JOSE Algorithms Registration (1/2) This section registers the following valuein~~terms of~~the~~u-coordinates of P, P1,~~IANA "JSON Web Signatureand~~P2,~~Encryption Algorithms" registry [IANA.JOSE.Algorithms]. Algorithm Name: ECDSA448; Algorithm Description: ECDSA using SHAKE256and~~the v-coordinate~~curve Wei448; Algorithm Usage Locations: alg; JOSE Implementation Requirements: Optional; Change Controller: IESG; Reference: specified in Section 4.4of~~P, since 2*B*v*v1=(u*u1+1)*(u+u1+2*A)-2*A-u2*(u-u1)^2. This property allows recovery~~this specification; for encodings, see Section 10.2; Algorithm Analysis Document(s): Section 4.4ofthis specification. 11.2.6. JOSE Algorithms Registration (2/2) This section registersthe~~v-coordinate of a point P1=k*P that is computed via~~following value inthe~~so-called Montgomery ladder, where P is an affine point with nonzero v-coordinate (i.e., it does not have order two). For future reference, note that the expression above uniquely determines the u-coordinate of P2 in terms of the u-coordinates of P and P1~~IANA "JSON Web Signatureand~~the product of their v-coordinates. Further details are out of scope. C.3. Group Laws for Twisted Edwards Curves Note: The group laws below hold for twisted Edwards curves E_{a,d} where a is a square~~Encryption Algorithms" registry [IANA.JOSE.Algorithms]. Algorithm Name: ECDH448; Algorithm Description: NIST-compliant co-factor Diffie-Hellman w/ curve Wei448; Algorithm Usage Locations: alg; JOSE Implementation Requirements: Optional; Change Controller: IESG; Reference: specifiedin~~GF(q), whereas d is not. In~~Section 4.4 ofthis~~case, the addition formulae below are defined~~specification;for~~each pair of points, without exceptions. Generalizations~~encodings, see Section 10.3 Algorithm Analysis Document(s): Section 4.4of this~~group law~~specification. 12. Acknowledgements Thanksto~~other twisted Edwards curves are out of scope. For each point P of the twisted Edwards curve E_{a,d}, the point O:=(0,1) serves as identity element, i.e., P + O = O + P = P. For each point P:=(x, y)~~Nikolas Rosener for discussions surrounding implementation detailsof the~~twisted Edwards curve E_{a,d}, the point -P is the point (-x, y)~~techniques described in this documentand~~one has P + (-P) = O (i.e., -P is~~to Phillip Hallam-Baker for triggering inclusion of verbiage onthe~~inverse~~useof~~P). Let P1:=(x1, y1) and P2:=(x2, y2) be points~~Montgomery ladders with recoveryof the~~twisted Edwards curve E_{a,d} and let Q:=P1 + P2. Then Q:=(x, y), where x = (x1*y2 + x2*y1)/(1 + d*x1*x2*y1*y2)~~y-coordinate. Thanks to Stanislav Smyshlyaevand~~y = (y1*y2 - a*x1*x2)/(1 - d*x1*x2*y1*y2). Let P:=(x1, y1) be a point of the twisted Edwards curve E_{a,d} and let Q:=2*P. Then Q:=(x, y), where x = (2*x1*y1)/(1 + d*x1^2*y1^2) and y = (y1^2 - a*x1^2)/(1 - d*x1^2*y1^2). Note that one can use the formulae~~Vasily Nikolaevfor~~point addition~~their careful reviews. 13. References 13.1. Normative References [ANSI-X9.62] ANSI X9.62-2005, "Public Key Cryptographyfor~~point doubling, taking inverses, and adding the identity element as well (i.e., the point addition formulae are uniform and complete (subject to our Note above)). From~~the~~group laws above (subject to our Note above) it follows that if P=(x, y), P1=(x1, y1), and P2=P=(x2, y2) are points~~Financial Services Industry: The Elliptic Curve Digital Signature Algorithm (ECDSA)", American National Standard for Financial Services, Accredited Standards Committee X9, Inc, Anapolis, MD, 2005. [FIPS-180-4] FIPS 180-4, "Secure Hash Standard (SHS), Federal Information Processing Standards Publication 180-4", US Departmentof~~the twisted Edwards curve E_{a,d} with P2:=P+P1 and if x is nonzero, then the x-coordinate~~Commerce/National Instituteof~~P1 can be expressed in terms~~Standards and Technology, Gaithersburg, MD, August 2015. [FIPS-186-4] FIPS 186-4, "Digital Signature Standard (DSS), Federal Information Processing Standards Publication 186-4", US Departmentof~~the y-coordinates~~Commerce/National Instituteof~~P, P1,~~Standardsand~~P2,~~Technology, Gaithersburg, MD, July 2013. [FIPS-202] FIPS 202, "SHA-3 Standard: Permutation-Based Hashand~~the x-coordinate of P, since x*x1*(a-d*y*y1*y2)=y*y1-y2. (Here, observe that a-d*y*y1*y2 is nonzero per our Note above.) This property allows recovery~~Extendable-Output Functions, Federal Information Processing Standards Publication 202", US Departmentof~~the x-coordinate~~Commerce/National Instituteof~~a point P1=k*P that is computed via the so-called Montgomery ladder, where P is an affine point with nonzero x-coordinate (i.e., it does not have order one or two). For future reference, note that the group law (subject to our Note above) uniquely determines the y-coordinate of P2 in terms of the y-coordinates of P~~Standardsand~~P1~~Technology, Gaithersburg, MD, August 2015. [I-D.ietf-cose-rfc8152bis-algs] Schaad, J., "CBOR Object Signingand~~the product of their x-coordinates. Further details are out of scope. Appendix D. Relationship Between Curve Models The non-binary curves specified in Appendix A are expressed~~Encryption (COSE): Initial Algorithms", draft-ietf-cose-rfc8152bis-algs-07 (workin~~different curve models, viz. as curves~~progress), March 2020. [RFC2119] Bradner, S., "Key words for usein~~short-Weierstrass form, as Montgomery curves, or as twisted Edwards curves. These curve models are related, as follows. D.1. Mapping between Twisted Edwards Curves~~RFCs to Indicate Requirement Levels", BCP 14, RFC 2119, DOI 10.17487/RFC2119, March 1997, <https://www.rfc-editor.org/info/rfc2119>. [RFC4648] Josefsson, S., "The Base16, Base32,and~~Montgomery~~Base64 Data Encodings", RFC 4648, DOI 10.17487/RFC4648, October 2006, <https://www.rfc-editor.org/info/rfc4648>. [RFC5639] Lochter, M. and J. Merkle, "Elliptic Curve Cryptography (ECC) Brainpool StandardCurves~~One can map points of the Montgomery curve M_{A,B} to points of the twisted Edwards curve E_{a,d}, where a:=(A+2)/B~~and~~d:=(A-2)/B and, conversely, map points of the twisted Edwards curve E_{a,d} to points of the Montgomery curve M_{A,B}, where A:=2*(a+d)/(a-d)~~Curve Generation", RFC 5639, DOI 10.17487/RFC5639, March 2010, <https://www.rfc-editor.org/info/rfc5639>. [RFC7049] Bormann, C.and~~where B:=4/(a-d). For twisted Edwards curves we consider (i.e., those where a is a square in GF(q), whereas d is not), this defines a one- to-one correspondence, which - in fact - is an isomorphism between M_{A,B}~~P. Hoffman, "Concise Binary Object Representation (CBOR)", RFC 7049, DOI 10.17487/RFC7049, October 2013, <https://www.rfc-editor.org/info/rfc7049>. [RFC7515] Jones, M., Bradley, J.,and~~E_{a,d}, thereby showing that, e.g., the discrete logarithm problem in either curve model is equally hard. For the Montgomery curves~~N. Sakimura, "JSON Web Signature (JWS)", RFC 7515, DOI 10.17487/RFC7515, May 2015, <https://www.rfc-editor.org/info/rfc7515>. [RFC7518] Jones, M., "JSON Web Algorithms (JWA)", RFC 7518, DOI 10.17487/RFC7518, May 2015, <https://www.rfc-editor.org/info/rfc7518>. [RFC7696] Housley, R., "Guidelines for Cryptographic Algorithm Agilityand~~twisted Edwards curves we consider, the mapping from M_{A,B} to E_{a,d} is defined by mapping the point at infinity O~~Selecting Mandatory-to-Implement Algorithms", BCP 201, RFC 7696, DOI 10.17487/RFC7696, November 2015, <https://www.rfc-editor.org/info/rfc7696>. [RFC7748] Langley, A., Hamburg, M.,and~~the point (0, 0) of order two of M_{A,B} to, respectively, the point (0, 1)~~S. Turner, "Elliptic Curves for Security", RFC 7748, DOI 10.17487/RFC7748, January 2016, <https://www.rfc-editor.org/info/rfc7748>. [RFC7942] Sheffer, Y.and~~the point (0, -1) of order two of E_{a,d}, while mapping each other point (u, v) of M_{A,B} to the point (x,y):=(u/v,(u-1)/(u+1))~~A. Farrel, "Improving Awarenessof~~E_{a,d}.~~Running Code:The~~inverse mapping from E_{a,d} to M_{A,B} is defined by mapping the point (0, 1)~~Implementation Status Section", BCP 205, RFC 7942, DOI 10.17487/RFC7942, July 2016, <https://www.rfc-editor.org/info/rfc7942>. [RFC8032] Josefsson, S.and~~the point (0, -1) of order two of E_{a,d} to, respectively, the point at infinity O~~I. Liusvaara, "Edwards-Curve Digital Signature Algorithm (EdDSA)", RFC 8032, DOI 10.17487/RFC8032, January 2017, <https://www.rfc-editor.org/info/rfc8032>. [RFC8037] Liusvaara, I., "CFRG Elliptic Curve Diffie-Hellman (ECDH)and~~the point (0, 0) of order two of M_{A,B}, while each other point (x, y) of E_{a,d} is mapped to the point (u,v):=((1+y)/(1-y),(1+y)/((1-y)*x)) of M_{A,B}. Implementations may take advantage of this mapping to carry out elliptic curve group operations originally defined for a twisted Edwards curve on the corresponding Montgomery curve, or vice-versa,~~Signatures in JSON Object Signingand~~translating the result back to the original curve, thereby potentially allowing code reuse. D.2. Mapping between Montgomery Curves~~Encryption (JOSE)", RFC 8037, DOI 10.17487/RFC8037, January 2017, <https://www.rfc-editor.org/info/rfc8037>. [RFC8152] Schaad, J., "CBOR Object Signingand~~Weierstrass Curves One can map points of the Montgomery curve M_{A,B} to points~~Encryption (COSE)", RFC 8152, DOI 10.17487/RFC8152, July 2017, <https://www.rfc-editor.org/info/rfc8152>. [RFC8174] Leiba, B., "Ambiguityof~~the Weierstrass curve W_{a,b}, where a:=(3-A^2)/(3*B^2) and b:=(2*A^3-9*A)/(27*B^3). This defines a one-to-one correspondence, which - in fact - is an isomorphism between M_{A,B} and W_{a,b}, thereby showing that, e.g., the discrete logarithm problem~~Uppercase vs Lowercasein~~either curve model is equally hard. The mapping from M_{A,B} to W_{a,b} is defined by mapping the point at infinity O of M_{A,B} to the point at infinity O~~RFC 2119 Key Words", BCP 14, RFC 8174, DOI 10.17487/RFC8174, May 2017, <https://www.rfc-editor.org/info/rfc8174>. [SEC1] SEC1, "SEC 1: Elliptic Curve Cryptography, Version 2.0", Standards for Efficient Cryptography, , June 2009. [SEC2] SEC2, "SEC 2: Elliptic Curve Cryptography, Version 2.0", Standards for Efficient Cryptography, , January 2010. [SP-800-56a] NIST SP 800-56a, "Recommendation for Pair-Wise Key Establishment Schemes Using Discrete Log Cryptography, Revision 3", US Departmentof~~W_{a,b}, while mapping each other point (u,v)~~Commerce/National Instituteof~~M_{A,B} to the point (X,Y):=((u+A/3)/B,v/B)~~Standards and Technology, Gaithersburg, MD, April 2018. [SP-800-56c] NIST SP 800-56c, "Recommendation for Key-Derivation Methods in Key-Establishment Schemes, Revision 1", US Departmentof~~W_{a,b}. Note that not all Weierstrass curves can be injectively mapped to Montgomery curves, since the latter have a point~~Commerce/National Instituteof~~order two~~Standardsand~~the former may not. In particular, if a Weierstrass curve has prime order, such as is the case with the so-called "NIST prime curves", this inverse mapping is not defined. If the Weierstrass curve W_{a,b} has a point (alpha,0) of order two and c:=a+3*(alpha)^2 is a square~~Technology, Gaithersburg, MD, April 2018. 13.2. Informative References [ECC] I.F. Blake, G. Seroussi, N.P. Smart, "Elliptic Curvesin~~GF(q), one can map points of this curve~~Cryptography", Cambridge University Press, Lecture Notes Series 265, July 1999. [ECC-Isogeny] E. Brier, M. Joye, "Fast Point Multiplication on Elliptic Curves through Isogenies", AAECC, Lecture Notes in Computer Science, Vol. 2643, New York: Springer-Verlag, 2003. [GECC] D. Hankerson, A.J. Menezes, S.A. Vanstone, "Guideto~~points of the Montgomery curve M_{A,B}, where A:=3*alpha/ gamma and B:=1/gamma~~Elliptic Curve Cryptography", New York: Springer-Verlag, 2004. [IANA.COSE.Algorithms] IANA, "COSE Algorithms", IANA, https://www.iana.org/assignments/cose/ cose.xhtml#algorithms. [IANA.COSE.Curves] IANA, "COSE Elliptic Curves", IANA, https://www.iana.org/assignments/cose/cose.xhtml#elliptic- curves. [IANA.JOSE.Algorithms] IANA, "JSON Web Signatureand~~where gamma is any square root of c. In this case, the mapping from W_{a,b} to M_{A,B} is defined by mapping the point at infinity O of W_{a,b} to~~Encryption Algorithms", IANA, https://www.iana.org/assignments/jose/jose.xhtml#web- signature-encryption-algorithms. [IANA.JOSE.Curves] IANA, "JSON Web Key Elliptic Curve", IANA, https://www.iana.org/assignments/jose/jose.xhtml#web-key- elliptic-curve. [Mont-Ladder] P.L. Montgomery, "Speedingthe~~point at infinity O~~Pollard and Elliptic Curve Methodsof~~M_{A,B}, while mapping each other point (X,Y)~~Factorization", Mathematicsof~~W_{a,b} to~~Computation, Vol. 48, 1987. [Rosener] N. Rosener, "Evaluatingthe~~point (u,v):=((X-alpha)/gamma,Y/gamma)~~Performanceof~~M_{A,B}. As before, this defines a one-to-one correspondence, which -~~Transformations Between Curve Representationsin~~fact - is an isomorphism between W_{a,b} and M_{A,B}. It is easy to see that the mapping from W_{a,b} to M_{A,B} and that from M_{A,B} to W_{a,b} (if defined) are each other's inverse. This mapping can be used to implement elliptic curve group operations originally defined~~Elliptic Curve Cryptographyfor~~a twisted~~Constrained Device Security", M.Sc. Universitat Bremen, August 2018. [SWUmap] E. Brier, J-S. Coron, Th. Icart, D. Madore, H. Randriam, M. Tibouchi, "Efficient Indifferentiable Hashing into Ordinary Elliptic Curves", CRYPTO 2010, Lecture Notes in Computer Science, Vol. 6223, New York: Springer-Verlag, 2010. [tEd] D.J. Bernstein, P. Birkner, M. Joye, T. Lange, C. Peters, "TwistedEdwards~~curve or for a Montgomery curve using group operations~~Curves", Africacrypt 2008, Lecture Notes in Computer Science, Vol. 5023, New York: Springer-Verlag, 2008. [tEd-Formulas] H. Hisil, K.K.H. Wong, G. Carter, E. Dawson, "Twisted Edwards Curves Revisited", ASIACRYPT 2008, Lecture Notes in Computer Science, Vol. 5350, New York: Springer-Verlag, 2008. [Tibouchi] M. Tibouchi, "Elligator Squared -- Uniform Points on Elliptic Curves of Prime Order as Uniform Random Strings", Financial Cryptography 2014, Lecture Notes in Computer Science, Vol. 8437, New York: Springer-Verlag, 2014. [Wei-Ladder] T. Izu, Ts. Takagi,, "A Fast Parallel Elliptic Curve Multiplication Resistant Against Side Channel Attacks", CentreforApplied Cryptographic Research, Corr 2002-03, 2002. Appendix A. Some (Non-Binary) Elliptic Curves This section definesthe~~corresponding elliptic~~three differentcurve~~in~~models we consider, viz.short-Weierstrass~~form~~curves, Montgomery curves,and~~translating the result back to~~twisted Edwards curves. A.1. Curves in Short-Weierstrass Form Let GF(q) denotethe~~original curve, thereby potentially allowing code reuse. Note that implementations for elliptic curves~~finite fieldwith~~short-Weierstrass form that hard-code the domain parameter a to a= -3 (which value~~q elements, where qis~~known to allow more efficient implementations) cannot always be used this way, since the curve W_{a,b} resulting from~~an~~isomorphic mapping cannot always~~odd prime power and where q is not divisible by three. Let W_{a,b}be~~expressed as a~~theWeierstrass curve with~~a=-3 via~~defining equation Y^2 = X^3 + a*X + b, wherea~~coordinate transformation. For more details, see Appendix F. D.3. Mapping between Twisted Edwards Curves~~and~~Weierstrass Curves One can map~~b are elements of GF(q) and where 4*a^3 + 27*b^2 is nonzero. Thepoints ofW_{a,b} arethe~~twisted Edwards curve E_{a,d} to points~~ordered pairs (X, Y) whose coordinates are elementsofGF(q) and that satisfythe~~Weierstrass curve W_{a,b},~~defining equation (the so-called affine points), together with the special point O (the so-called "point at infinity"). This set forms a group under addition,via~~function composition,~~the so-called "chord-and-tangent" rule,where~~one uses~~the~~isomorphic mapping between twisted Edwards curves and Montgomery curves of Appendix D.1 and~~point at infinity serves asthe~~one between Montgomery and Weierstrass curves of~~identity element. SeeAppendix~~D.2. Obviously, one can use function composition (now using the respective inverses - if these exist) to realize the inverse~~C.1 for detailsof~~this mapping. Appendix E. Curve25519 and Cousins This section introduces curves related to Curve25519 and explains their relationships. E.1. Curve Definition and Alternative Representations The elliptic curve Curve25519 is~~thegroup operation. A.2.Montgomery~~curve M_{A,B} defined over~~Curves Let GF(q) denotethe~~prime~~finitefield~~GF(p),~~with~~p:=2^{255}-19, where A:=486662 and B:=1. This curve has order h*n, where h=8 and~~q elements,where~~n~~qis~~a~~an oddprime~~number. For this curve, A^2-4 is not a square in GF(p), whereas A+2 is. The quadratic twist of this~~power. Let M_{A,B} be the Montgomerycurve~~has order h1*n1,~~with defining equation B*v^2 = u^3 + A*u^2 + u,where~~h1=4~~A and B are elements of GF(q)and where~~n1~~Ais~~a prime number. For this curve, the base point~~unequal to (+/-)2 and where Bisnonzero. The points of M_{A,B} are the ordered pairs (u, v) whose coordinates are elements of GF(q) and that satisfy the defining equation (the so- called affine points), together withthespecialpoint~~(Gu, Gv),~~O (the so- called "point at infinity"). This set forms a group under addition, via the so-called "chord-and-tangent" rule,where~~Gu=9 and~~the point at infinity serves as the identity element. See Appendix C.2 for details of the group operation. A.3. Twisted Edwards Curves Let GF(q) denote the finite field with q elements,where~~Gv~~qis an odd~~integer in the interval [0, p-1]. This curve has the same group structure as (is "isomorphic" to)~~prime power. Let E_{a,d} bethe twisted Edwards curve~~E_{a,d} defined over GF(p),~~with~~as base point the point (Gx, Gy),~~defining equation a*x^2 + y^2 = 1+ d*x^2*y^2,where~~parameters~~a and d are distinct nonzero elements of GF(q). The points of E_{a,d}are~~as specified in Appendix E.3. This curve is denoted as Edwards25519. For this curve,~~the~~parameter~~ordered pairs (x, y) whose coordinates are elements of GF(q) and that satisfy the defining equation (the so-called affine points). It can be shown that this set forms a group under addition ifa is a square in~~GF(p),~~GF(q),whereas d is not,~~so~~wherethe~~group laws of Appendix C.3 apply. The~~point O:=(0, 1) serves as the identity element. (Note that the identity element satisfies the defining equation.) See Appendix C.3 for details of the group operation. An Edwardscurve is~~also isomorphic~~a twisted Edwards curve with a=1. Appendix B. Elliptic Curve Nomenclature and Finite Fields This section provides brief background information on elliptic curves and finite fields that should be sufficienttounderstand constructions and examples in this document. B.1. Elliptic Curve Nomenclature Each curve defined in Appendix A forms a commutative group under addition (denoted by '+'). In Appendix C we specify the group laws, which depend onthe~~elliptic~~curve~~W_{a,b}~~modelin~~short- Weierstrass form defined over GF(p), with~~question. For completeness, we here include some common elliptic curve nomenclature and basic properties (primarily soas~~base point the point (GX, GY), where parameters~~to keep this document self-contained). These notionsare~~as specified~~mainly usedin Appendix~~E.3.~~E and Appendix G and not essential for our exposition.This~~curve is denoted as Wei25519. E.2. Switching between Alternative Representations Each affine~~section can be skipped at first reading. Anypoint~~(u, v)~~Pof~~Curve25519 corresponds to the point (X, Y):=(u + A/3, v)~~a curve E is a generatorof~~Wei25519, while~~the~~point at infinity~~cyclic subgroup (P):={k*P | k = 0, 1, 2,...}of~~Curve25519 corresponds to~~the~~point at infinity of Wei25519.~~curve.(Here,~~we used~~k*P denotesthe~~mappings~~sumof~~Appendix D.2 and that B=1.) Under this mapping, the base point (Gu, Gv)~~k copiesof~~Curve25519 corresponds to~~P, where 0*P isthe~~base point (GX, GY)~~identity element Oof~~Wei25519. The inverse mapping maps~~the~~affine point (X, Y) of Wei25519~~curve; k*P is commonly referredto~~(u, v):=(X - A/3, Y)~~as scalar multiplicationof~~Curve25519, while mapping~~P by k.) If (P) has cardinality l, then l is calledthe~~point at infinity~~order of P. The order of curve E is the cardinalityof~~Wei25519 to~~theset of its points, commonly denoted by |E|. A curve is cyclic if it is generated by somepoint~~at infinity~~of~~Curve25519. Note that~~this~~mapping involves a simple shift~~curve. All curvesof~~the first coordinate~~prime order are cyclic, while all curves of order h*n, where n is a large prime numberand~~can be implemented via integer-only arithmetic as~~where h isa~~shift~~small number (the so-called co-factor), have a large cyclic subgroupof~~(p+A)/3 for the isomorphic mapping and~~prime order n. In this case,a~~shift~~generatorof~~-(p+A)/3 for its inverse, where delta=(p+A)/3~~order nis~~the element of GF(p) defined~~called a base point, commonly denotedby~~delta 19298681539552699237261830834781317975544997444273427339909597 334652188435537 (=0x2aaaaaaa aaaaaaaa aaaaaaaa aaaaaaaa aaaaaaaa aaaaaaaa aaaaaaaa aaad2451). (Note that, depending on the implementation details~~G. A pointof~~the field arithmetic, one may have~~order dividing h is saidto~~shift~~be inthe~~result by +p or -p if~~small subgroup. For curves of prime order,this~~integer~~small subgroup is the singleton set, consisting of only the identity element O. A point thatis not in the~~interval [0,p-1].) The curve Edwards25519~~small subgroupis~~isomorphic~~saidto~~the curve Curve25519, where the base~~be a high-order point (since it has order at least n). If R is apoint~~(Gu, Gv)~~of~~Curve25519 corresponds to~~the~~base point (Gx,Gy) of Edwards25519 and where~~curve that is also contained in (P), there is a unique integer k inthe~~point at infinity and~~interval [0, l-1] so that R=k*P, where l isthe~~point (0,0) of~~order~~two~~of~~Curve25519 correspond to, respectively, the point (0, 1) and~~P. This number is calledthe~~point (0, -1) of order two of Edwards25519 and where each other point (u, v)~~discrete logarithmof~~Curve25519 corresponds~~Rto the~~point (c*u/v, (u-1)/(u+1)) of Edwards25519, where c~~base P. The discrete logarithm problemis the~~element of GF(p) defined by c sqrt(-(A+2)/B) 51042569399160536130206135233146329284152202253034631822681833788 666877215207 (=0x70d9120b 9f5ff944 2d84f723 fc03b081 3a5e2c2e b482e57d 3391fb55 00ba81e7). (Here, we used the mapping of Appendix D.1 and normalized this using the mapping~~problemof~~Appendix F.1 (where~~findingthe~~element s~~discrete logarithmof~~that appendix is set to c above).) The inverse mapping from Edwards25519~~Rto~~Curve25519 is defined by mapping the point (0, 1) and the point (0, -1) of order two of Edwards25519 to, respectively, the point at infinity and~~the~~point (0,0) of order~~base P for anytwo~~of Curve25519~~points Pand~~having each other point (x, y)~~Rof~~Edwards25519 correspond to~~the~~point ((1 + y)/(1 - y), c*(1 + y)/((1-y)*x)) of Curve25519. The curve Edwards25519~~curve, if such a number exists. If Pis~~isomorphic to the Weierstrass curve Wei25519, where the~~a fixedbase point~~(Gx, Gy)~~Gof~~Edwards25519 corresponds to~~the~~base point (GX,GY) of Wei25519 and where~~curve,the~~identity element (0,1)~~pair (k, R:=k*G) is commonly called a public-private key pair, the integer k the private key,and the point~~(0,-1) of order two of Edwards25519 correspond to, respectively, the point at infinity O and~~Rthe~~point (A/3, 0) of order two of Wei25519 and where each other point (x, y) of Edwards25519 corresponds to~~corresponding public key. The private key k can be represented as an integer inthe~~point (X, Y):=((1+y)/(1-y)+A/3, c*(1+y)/((1-y)*x)) of Wei25519,~~interval [0,n-1],where~~c was defined before. (Here, we used the mapping of Appendix D.3.) The inverse mapping from Wei25519 to Edwards25519~~G has order n. If this representationis~~defined by mapping the point at infinity O and the point (A/3, 0) of~~nonzero, R hasorder~~two of Wei25519 to, respectively,~~n; otherwise, it has order one and isthe identity element~~(0,1) and the point (0,-1) of order two of Edwards25519 and having each other point (X, Y)~~Oof~~Wei25519 correspond to~~the~~point (c*(X-A/3)/Y, (X-A/3-1)/(X-A/3+1))~~curve. In this document, a quadratic twistof~~Edwards25519. Note that these mappings can be easily realized if points are represented in projective coordinates, using~~a~~few~~curve E defined over afield~~multiplications only, thus allowing switching between alternative~~GF(q) is acurve~~representations~~E' related to E,with~~negligible relative incremental cost. E.3. Domain Parameters The parameters of the Montgomery~~cardinality |E'|, where |E|+|E'|=2*(q+1). If E is acurve~~and the corresponding isomorphic curves~~in~~twisted Edwards curve and short-Weierstrass form are as indicated below. Here, the domain parameters of the Montgomery curve Curve25519 and~~oneof the~~twisted Edwards~~curve~~Edwards25519 are as~~modelsspecified in~~[RFC7748]; the domain parameters~~this document, a quadratic twistof~~Wei25519 are "new". General parameters (for all~~thiscurve~~models): p 2^{255}-19 (=0x7fffffff ffffffff ffffffff ffffffff ffffffff ffffffff ffffffff ffffffed) h 8 n 72370055773322622139731865630429942408571163593799076060019509382 85454250989 (=2^{252} + 0x14def9de a2f79cd6 5812631a 5cf5d3ed) h1 4 n1 14474011154664524427946373126085988481603263447650325797860494125 407373907997 (=2^{253} - 0x29bdf3bd 45ef39ac b024c634 b9eba7e3) Montgomery curve-specific parameters (for Curve25519): A 486662 (=0x076d06) B 1 (=0x01) Gu 9 (=0x09) Gv 14781619447589544791020593568409986887264606134616475288964881837 755586237401 (=0x20ae19a1 b8a086b4 e01edd2c 7748d14c 923d4d7e 6d7c61b2 29e9c5a2 7eced3d9) Twisted Edwards curve-specific parameters (for Edwards25519): a -1 (-0x01) d -121665/121666 = - (A-2)/(A+2) (=370957059346694393431380835087545651895421138798432190163887855 33085940283555) (=0x52036cee 2b6ffe73 8cc74079 7779e898 00700a4d 4141d8ab 75eb4dca 135978a3) Gx 15112221349535400772501151409588531511454012693041857206046113283 949847762202 (=0x216936d3 cd6e53fe c0a4e231 fdd6dc5c 692cc760 9525a7b2 c9562d60 8f25d51a) Gy 4/5 (=463168356949264781694283940034751631413079938662562256157830336 03165251855960) (=0x66666666 66666666 66666666 66666666 66666666 66666666 66666666 66666658) Weierstrass curve-specific parameters (for Wei25519): a 19298681539552699237261830834781317975544997444273427339909597334 573241639236 (=0x2aaaaaaa aaaaaaaa aaaaaaaa aaaaaaaa aaaaaaaa aaaaaaaa aaaaaa98 4914a144) b 55751746669818908907645289078257140818241103727901012315294400837 956729358436 (=0x7b425ed0 97b425ed 097b425e d097b425 ed097b42 5ed097b4 260b5e9c 7710c864) GX 19298681539552699237261830834781317975544997444273427339909597334 652188435546 (=0x2aaaaaaa aaaaaaaa aaaaaaaa aaaaaaaa aaaaaaaa aaaaaaaa aaaaaaaa aaad245a) GY 14781619447589544791020593568409986887264606134616475288964881837 755586237401 (=0x20ae19a1 b8a086b4 e01edd2c 7748d14c 923d4d7e 6d7c61b2 29e9c5a2 7eced3d9) Appendix F. Further Mappings The non-binary curves specified in Appendix A are~~can beexpressed~~in different~~using the samecurve~~models, viz. as curves in short-Weierstrass form, as Montgomery curves, or as twisted Edwards curves. In Appendix D we already described relationships between these various~~model, although (naturally) with its owncurve~~models. Further mappings exist between elliptic~~parameters. Twocurves~~within~~E and E' defined over a field GF(q) are said to be isogenous if these havethe same~~curve model. These can be exploited~~order and are saidto~~force some of~~be isomorphic if these havethe~~domain parameters to specific values~~same group structure. Notethat~~allow for a more efficient implementation of~~isomorphic curves have necessarilythe~~addition formulae. F.1. Isomorphic Mapping between Twisted Edwards Curves One can map points~~same order and are, thus, a special typeof~~the~~isogenous curves. Further details are out of scope. Weierstrass curves can have prime order, whereas Montgomery curves andtwisted Edwards~~curve E_{a,d} to points~~curves always have an order that is a multiple of four (and, thereby, a small subgroup of cardinality four). An ordered pair (x, y) whose coordinates are elements of GF(q) can be associated with any ordered tripleof the~~twisted Edwards curve E_{a',d'},~~form [x*z: y*z: z],where~~a:=a'*s^2 and d:=d'*s^2 for some~~z is anonzero element~~s~~of~~GF(q). This defines a one-to-one correspondence, which - in fact - is an isomorphism between E_{a,d}~~GF(q),and~~E_{a',d'}. The mapping~~can be uniquely recoveredfrom~~E_{a,d} to E_{a',d'}~~such a representation. The latter representationis~~defined by mapping the point (x,y) of E_{a,d} to the point (x', y'):=(s*x, y)~~commonly called a representation in projective coordinates. Sometimes, yet other representations are useful (e.g., representation in Jacobian coordinates). Further details are outof~~E_{a',d'}.~~scope.The~~inverse mapping from E_{a',d'} to E_{a,d} is defined by mapping the point (x', y')~~group laws in Appendix C are mostly expressed in terms of affine points, but can also be expressed in termsof~~E_{a',d'} to~~the~~point (x, y):=(x'/s, y')~~representationof~~E_{a,d}. Implementations may take advantage~~these points in projective coordinates, thereby allowing clearingof~~this mapping to carry out elliptic curve~~denominators. Thegroup~~operations originally defined for~~laws may also involve non-affine points (such as the point at infinity O ofa~~twisted Edwards~~Weierstrasscurve~~with generic domain parameters a and d on a corresponding isomorphic twisted Edwards curve with domain parameters a' and d' that have a more special form and that are known to allow for more efficient implementations~~orof~~addition laws and translating the result back to the original curve. In particular, it is known that such efficiency improvements exist if a':=(+/-)1 (see [tEd-Formulas]). F.2. Isomorphic Mapping between~~aMontgomery~~Curves One~~curve). Thosecan~~map points~~also be represented in projective coordinates. Further details are outofscope. B.2. Finite Fields The field GF(q), where q is a prime power, is defined as follows. If q:=p is a prime number,the~~Montgomery curve M_{A,B} to points~~field GF(p) consistsof the~~Montgomery curve M_{A',B'}, where A:=A'~~integers in the interval [0,p-1]and~~B:=B'*s^2 for some nonzero element s of GF(q).~~two binary operations on this set: addition and multiplication modulo p.This~~defines~~field is commonly calleda~~one-to-one correspondence, which - in fact -~~prime field. If q:=p^m, where pis~~an isomorphism between M_{A,B}~~a prime numberand~~M_{A',B'}. The mapping from M_{A,B} to M_{A',B'}~~where m>0, the field GF(q)is defined~~by mapping the point at infinity O~~in termsof~~M_{A,B} to~~an irreducible polynomial f(z) in z of degree m with coefficients in GF(p) (i.e., f(z) cannot be written asthe~~point at infinity O~~productof~~M_{A',B'}, while mapping each other point (u,v)~~two polynomials in zof~~M_{A,B} to the point (u', v'):=(u, s*v)~~lower degree with coefficients in GF(p)): in this case, GF(q) consistsof~~M_{A',B'}. The inverse mapping from M_{A',B'} to M_{A,B} is defined by mapping~~the~~point at infinity O~~polynomials in zof~~M_{A',B'} to~~degree smaller than m with coefficients in GF(p) and two binary operations on this set: polynomial addition and polynomial multiplication modulothe~~point at infinity O of M_{A,B}, while mapping~~irreducible polynomial f(z). By definition,each~~other point (u',v') of M_{A',B'} to the point (u,v):=(u',v'/s) of M_{A,B}. One can also map points~~element xof~~the Montgomery curve M_{A,B} to points~~GF(q) is a polynomial in zof~~the Montgomery curve M_{A',B'}, where A':=-A~~degree smaller than mand~~B':=-B. This defines~~can, therefore, be uniquely represented asa~~one-to-one correspondence, which -~~vector (x_{m-1}, x_{m-2}, ..., x_1, x_0) of length m with coefficientsin~~fact - is an isomorphism between M_{A,B} and M_{A',B'}. In this case, the mapping from M_{A,B} to M_{A',B'}~~GF(p), where x_iis~~defined by mapping the point at infinity O of M_{A,B} to~~the~~point at infinity O~~coefficientof~~M_{A',B'}, while mapping each other point (u,v)~~z^iof~~M_{A,B} to~~polynomial x. Note that this representation depends onthe~~point (u',v'):=(-u,v)~~irreducible polynomial f(z)of~~M_{A',B'}. The inverse mapping from M_{A',B'} to M_{A,B}~~the field GF(p^m) in question (whichis~~defined by mapping~~often fixed in practice). Note that GF(q) containsthe~~point at infinity O of M_{A',B'} to~~prime field GF(p) as a subset. If m=1,the~~point at infinity O~~definitionsof~~M_{A,B}, while mapping~~GF(p) and GF(p^1) above coincide, sinceeach~~other point (u',v')~~nonzero elementof~~M_{A',B'} to the point (u,v):=(-u',v')~~GF(p) can be viewed as a polynomial in zof~~M_{A,B}. Implementations may take advantage~~degree zero. If m>1, then GF(q) is called a (nontrivial) extension fieldof~~these mappings to carry out elliptic curve groups operations originally defined for a Montgomery curve with generic domain parameters~~GF(p). The number p is called the characteristic of GF(q).A~~and B on a corresponding isomorphic Montgomery curve with domain parameters A' and B' that have~~field element y is calleda~~more special form and that are known to allow~~square in GF(q) if it can be expressed as y:=x^2for~~more efficient implementations of addition laws and translating the result back to the original curve. In particular,~~some x in GF(q);it is~~known that such efficiency improvements exist if B' assumes~~calleda~~small absolute value, such as B':=(+/-)1. (see [Mont-Ladder]). F.3. Isomorphic Mapping between Weierstrass Curves One can map points of the Weierstrass curve W_{a,b} to points~~non-square in GF(q) otherwise. If y is a square in GF(q), we denote by sqrt(y) oneof~~the Weierstrass curve W_{a',b'}, where a':=a*s^4 and b':=b*s^6~~its square roots (the other one being -sqrt(y)). For methodsfor~~some nonzero element s of GF(q). This defines a one-to-one correspondence, which -~~computing square roots and inversesin~~fact~~GF(q)-~~is an isomorphism between W_{a,b}~~if these exist - see Appendix K.1and~~W_{a',b'}. The~~Appendix K.2, respectively. For methods formapping~~from W_{a,b} to W_{a',b'}~~a nonzero field element thatis~~defined by mapping the point at infinity O of W_{a,b} to the point at infinity O of W_{a',b'}, while mapping each other point (X,Y) of W_{a,b}~~not a square in GF(q)to~~the~~apoint~~(X',Y'):=(X*s^2, Y*s^3)~~of~~W_{a',b'}.~~a curve, see Appendix K.3. NOTE:The~~inverse mapping from W_{a',b'} to W_{a,b} is defined by mapping the point at infinity O of W_{a',b'} to the point at infinity O of W_{a,b}, while mapping each other point (X', Y') of W_{a',b'} to the point (X,Y):=(X'/s^2,Y'/s^3) of W_{a,b}. Implementations may take advantage of this mapping to carry out elliptic curve group operations originally defined for a Weierstrass curve with generic domain parameters a~~curves in Appendix Eand~~b on~~Appendix G are all defined overa~~corresponding isomorphic Weierstrass curve with domain parameter a' and b' that~~prime field GF(p), thereby reducing all operations to simple modular integer arithmetic. Strictly speaking we could, therefore,have~~a~~refrained from introducing extension fields. Nevertheless, we included themore~~special form and that are known~~general exposition, so asto~~allow for more efficient implementations~~accommodate potential introductionof~~addition laws and translating the result back to the original curve. In particular, it is known~~new curvesthat~~such efficiency improvements exist if a'=-3 (mod p), where p is the characteristic of GF(q), and one uses so-called Jacobian coordinates with~~are defined overa~~particular projective version of~~(nontrivial) extension field at some point inthe~~addition laws of Appendix C.1. While not all Weierstrass curves can be put into this form, all traditional NIST curves have domain parameter a=-3, while all Brainpool~~future. This includescurves~~[RFC5639]~~proposed for post-quantum isogeny-based schemes, whichare~~isomorphic to~~defined overa~~Weierstrass curve of this form via the above mapping. Note that implementations for~~quadratic extension field (i.e., where q:=p^2), andelliptic curvesusedwith~~short-Weierstrass form that hard-code~~pairing-based cryptography. The exposition in either case is almostthe~~domain parameter a to a= -3 cannot always be used this way, since the curve W_{a,b} cannot always be expressed in terms of a Weierstrass curve with a'=-3 via a coordinate transformation: this only holds if a'/a is a fourth power in GF(q)~~same and now automatically yields, e.g., data conversion routines for any finite field object(see~~Section 3.1.5 of [GECC]). However, even~~Appendix I). Readers not interestedin~~this case, one can still express the curve W_{a,b}~~this, could simply view all fieldsas~~a~~prime fields. Appendix C. Elliptic Curve Group Operations This section specifies group operations for elliptic curves in short-Weierstrass~~curve with a small domain parameter value a', thereby still allowing a more efficient implementation than with a general domain parameter value a. F.4. Isogenous Mapping between~~form, for Montgomery curves, and for twisted Edwards curves. C.1. Group Laws forWeierstrass Curves~~One can still map points~~For each point Pof the Weierstrass curve~~W_{a,b} to points~~W_{a,b}, the point at infinity O serves as identity element, i.e., P + O = O + P = P. For each affine point P:=(X, Y)of the Weierstrass curve~~W_{a',b'}, where a':=-3 (mod p) and where p is~~W_{a,b},the~~characteristic of GF(q), even if a'/a~~point -Pis~~not a fourth power in GF(q). In that case, this mappping cannot be an isomorphism (see Appendix F.3). Instead,~~the~~mapping~~point (X, -Y) and one has P + (-P) = O (i.e., -Pis~~a so-called isogeny (or homomorphism). Since most elliptic curve operations process points~~the inverseof~~prime order or use so-called "co-factor multiplication", in practice~~P). Forthe~~resulting mapping has similar properties as an isomorphism. In particular,~~point at infinity O,one~~can still take advantage~~has -O:=O. Let P1:=(X1, Y1) and P2:=(X2, Y2) be distinct affine pointsof~~this mapping to carry out elliptic curve group operations originally defined for a~~theWeierstrass curve~~with domain parameter a unequal to -3 (mod p) on a corresponding isogenous~~W_{a,b} and let Q:=P1 + P2, where Q is not the identity element. Then Q:=(X, Y), where X + X1 + X2 = lambda^2 and Y + Y1 = lambda*(X1 - X), where lambda:= (Y2 - Y1)/(X2 - X1). Let P:=(X1, Y1) be an affine point of theWeierstrass curve~~with domain parameter a'=-3 (mod p)~~W_{a,b}and~~translating~~let Q:=2*P, where Q is notthe~~result back to~~identity element. Then Q:=(X, Y), where X + 2*X1 = lambda^2 and Y + Y1 = lambda*(X1 - X), where lambda:=(3*X1^2 + a)/(2*Y1). Fromthe~~original curve. In this case,~~group laws above it follows that if P=(X, Y), P1=(X1, Y1), and P2=(X2, Y2) are distinct affine points ofthe~~mapping from~~Weierstrass curveW_{a,b}~~to W_{a',b'}~~with P2:=P+P1 and if Yis~~defined by mapping~~nonzero, thenthe~~point at infinity O~~Y-coordinate of P1 can be expressed in termsof~~W_{a,b} to~~the~~point at infinity O~~X-coordinatesof~~W_{a',b'}, while mapping each other point (X,Y)~~P, P1, and P2, and the Y-coordinate of P, since 2*Y*Y1=(X*X1+a)*(X+X1)+2*b-X2*(X-X1)^2. This property allows recoveryof~~W_{a,b} to~~the~~point (X',Y'):=(u(X)/w(X)^2,Y*v(X)/w(X)^3)~~Y-coordinateof~~W_{a',b'}. Here, u(X), v(X), and w(X) are polynomials in X~~a point P1=k*Pthat~~depend on~~is computed viathe~~isogeny in question, as do domain parameters a' and b'. The inverse mapping from W_{a',b'} to W_{a,b}~~so-called Montgomery ladder, where Pis~~again~~an~~isogeny (called~~affine point with nonzero Y-coordinate (i.e., it does not have order two). For future reference, note thatthe~~dual isogeny)~~expression above uniquely determines the X-coordinate of P2 in terms of the X-coordinates of P and P1and~~defined by mapping~~theproduct of their Y-coordinates. Further details are out of scope. C.2. Group Laws for Montgomery Curves For eachpoint~~at infinity O~~Pof~~W_{a',b'} to~~the Montgomery curve M_{A,B},the point at infinity O~~of W_{a,b}, while mapping~~serves as identity element, i.e., P + O = O + P = P. Foreach~~other point (X', Y') of W_{a',b'} to the~~affinepoint~~(X,Y):=(u'(X')/w'(X')^2,Y'*v'(X')/w'(X')^3)~~P:=(u, v)of~~W_{a,b}, where -- again -- u'(X'), v'(X'), and w'(X') are polynomials in X' that depend on~~the~~isogeny in question. These mappings have~~Montgomery curve M_{A,B},the~~property that their composition~~point -Pis~~not the identity mapping (as was the case with~~the~~isomorphic mappings discussed in Appendix F.3), but rather a fixed multiple hereof: if this multiple~~point (u, -v) and one has P + (-P) = O (i.e., -Pis~~l then~~the~~isogeny is called an isogeny~~inverseof~~degree l (or l-isogeny) and u, v, and w (and, similarly, u', v',~~P). For the point at infinity O, one has -O:=O. Let P1:=(u1, v1)and~~w') are polynomials~~P2:=(u2, v2) be distinct affine pointsof~~degrees l, 3*(l-1)/2,~~the Montgomery curve M_{A,B}and~~(l-1)/2, respectively. Note that an isomorphism~~let Q:=P1 + P2, where Qis~~simply an isogeny of degree l=1. Details of how to determine isogenies are out of scope~~not the identity element. Then Q:=(u, v), where u + u1 + u2 = B*lambda^2 - A and v + v1 = lambda*(u1 - u), where lambda:=(v2 - v1)/(u2 - u1). Let P:=(u1, v1) be an affine pointof~~this document. The above formulas assume that~~the~~isogeny has odd degree (i.e., l~~Montgomery curve M_{A,B} and let Q:=2*P, where Qis~~odd); detailed formulas for even- degree isogenies~~not the identity element. Then Q:=(u, v), where u + 2*u1 = B*lambda^2 - A and v + v1 = lambda*(u1 - u), where lambda:=(3*u1^2 + 2*A*u1+1)/(2*B*v1). From the group laws above it follows that if P=(u, v), P1=(u1, v1), and P2=(u2, v2)are~~similar, but out of scope. Implementations may take advantage~~distinct affine pointsof~~this mapping to carry out elliptic curve group operations originally defined for a Weierstrass curve with a generic domain parameter a on a corresponding isogenous Weierstrass~~the MontgomerycurveM_{A,B}with~~domain parameter a'=-3 (mod p), where one~~P2:=P+P1 and if v is nonzero, then the v-coordinate of P1can~~use so-called Jacobian coordinates with a particular projective version~~be expressed in termsof the~~addition laws~~u-coordinatesof~~Appendix C.1. Since all traditional NIST curves have domain parameter a=-3, while all Brainpool curves [RFC5639] are isomorphic to a Weierstrass curve~~P, P1, and P2, and the v-coordinateof~~this form, this~~P, since 2*B*v*v1=(u*u1+1)*(u+u1+2*A)-2*A-u2*(u-u1)^2. This propertyallows~~taking advantage~~recoveryof~~existing implementations for these curves that may have a hardcoded a=-3 (mod p) domain parameter, provided one switches back and forth to this curve form using~~the~~isogenous mapping in question. Note that isogenous mappings can be easily realized using representations in projective coordinates and involves roughly 3*l finite field multiplications, thus allowing switching between alternative representations at relatively low incremental cost compared to that~~v-coordinateof~~elliptic curve scalar multiplications (provided the isogeny has low degree l). Note, however,~~a point P1=k*Pthat~~this~~is computed via the so-called Montgomery ladder, where P is an affine point with nonzero v-coordinate (i.e., itdoes~~require storage of~~not have order two). For future reference, note thatthe~~polynomial coefficients of~~expression above uniquely determinesthe~~isogeny and dual isogeny involved. This illustrates that low-degree isogenies are to be preferred, since an l-isogeny (usually) requires storing roughly 6*l elements~~u-coordinate of P2 in termsof~~GF(q). While there are many isogenies, we therefore only consider those with~~the~~desired property with lowest possible degree. Appendix G. Further Cousins~~u-coordinatesof~~Curve25519 This section introduces some further curves related to Curve25519~~Pand~~explains~~P1 and the product oftheir~~relationships. G.1.~~v-coordinates.Further~~Alternative Representations~~details are out of scope. C.3. Group Laws for Twisted Edwards Curves Note:The~~Weierstrass curve Wei25519~~group laws below hold for twisted Edwards curves E_{a,d} where ais~~isomorphic to~~a square in GF(q), whereas d is not. In this case,the~~Weierstrass curve Wei25519.2~~addition formulae below aredefined~~over GF(p), with as base point the~~for eachpair~~(G2X,G2Y), and isogenous~~of points, without exceptions. Generalizations of this group lawto~~the Weierstrass curve Wei25519.-3 defined over GF(p), with as base~~other twisted Edwards curves are out of scope. For eachpointP ofthe~~pair (G3X, G3Y), where parameters are as specified in Appendix G.3 and where~~twisted Edwards curve E_{a,d},the~~related mappings are~~point O:=(0,1) servesas~~specified in Appendix G.2. G.2. Further Switching Each affine~~identity element, i.e., P + O = O + P = P. For eachpoint~~(X, Y)~~P:=(x, y)of~~Wei25519 corresponds to~~the twisted Edwards curve E_{a,d},the point~~(X', Y'):=(X*s^2,Y*s^3) of Wei25519.2, where s~~-Pis the~~element of GF(p) defined by s 20343593038935618591794247374137143598394058341193943326473831977 39407761440 (=0x047f6814 6d568b44 7e4552ea a5ed633d 02d62964 a2b0a120 5e7941e9 375de020), while the~~point~~at infinity of Wei25519 corresponds to~~(-x, y) and one has P + (-P) = O (i.e., -P isthe~~point at infinity~~inverseof~~Wei25519.2. (Here, we used the mapping~~P). Let P1:=(x1, y1) and P2:=(x2, y2) be pointsof~~Appendix F.3.) Under this mapping,~~the~~base~~twisted Edwards curve E_{a,d} and let Q:=P1 + P2. Then Q:=(x, y), where x = (x1*y2 + x2*y1)/(1 + d*x1*x2*y1*y2) and y = (y1*y2 - a*x1*x2)/(1 - d*x1*x2*y1*y2). Let P:=(x1, y1) be apoint~~(GX, GY)~~of~~Wei25519 corresponds to~~the~~base point (G2X,G2Y) of Wei25519.2. The inverse mapping maps~~twisted Edwards curve E_{a,d} and let Q:=2*P. Then Q:=(x, y), where x = (2*x1*y1)/(1 + d*x1^2*y1^2) and y = (y1^2 - a*x1^2)/(1 - d*x1^2*y1^2). Note that one can usethe~~affine~~formulae forpoint~~(X', Y') of Wei25519.2 to (X,Y):=(X'/s^2,Y'/s^3) of Wei25519, while mapping~~addition for point doubling, taking inverses, and adding the identity element as well (i.e.,the point~~at infinity O of Wei25519.2~~addition formulae are uniform and complete (subjecttoour Note above)). Fromthe~~point at infinity O of Wei25519.~~group laws above (subject to ourNoteabove) it followsthat~~this mapping (and its inverse) involves a modular multiplication of both coordinates with fixed constants s^2 and s^3 (respectively, 1/s^2 and 1/s^3), which can be precomputed. Each affine point (X,Y) of Wei25519 corresponds to the point (X',Y'):=(X1*t^2,Y1*t^3) of Wei25519.-3, where (X1,Y1)=(u(X)/w(X)^2,Y*v(X)/w(X)^3), where u, v,~~if P=(x, y), P1=(x1, y1),and~~w~~P2=P=(x2, y2)arepoints ofthe~~polynomials~~twisted Edwards curve E_{a,d}with~~coefficients in GF(p) as defined in Appendix G.4.1~~P2:=P+P1and~~where t~~if xisnonzero, thenthe~~element~~x-coordinate of P1 can be expressed in termsof~~GF(p) defined by t 35728133398289175649586938605660542688691615699169662967154525084 644181596229 (=0x4efd6829 88ff8526 e189f712 5999550c e9ef729b ed1a7015 73b1bab8 8bfcd845), while~~the~~point at infinity~~y-coordinatesof~~Wei25519 corresponds to~~P, P1, and P2, andthe~~point at infinity~~x-coordinateof~~Wei25519.-3.~~P, since x*x1*(a-d*y*y1*y2)=y*y1-y2.(Here,~~we used the isogenous mapping~~observe that a-d*y*y1*y2 is nonzero per our Note above.) This property allows recoveryof~~Appendix F.4.) Under this isogenous mapping,~~the~~base point (GX,GY)~~x-coordinateof~~Wei25519 corresponds to the base~~apoint~~(G3X,G3Y) of Wei25519.-3. The dual isogeny maps~~P1=k*P that is computed viatheso-called Montgomery ladder, where P is anaffine point~~(X',Y') of Wei25519.-3~~with nonzero x-coordinate (i.e., it does not have order one or two). For future reference, note that the group law (subjecttoour Note above) uniquely determinesthe~~affine point (X,Y):=(u'(X1)/w'(X1)^2,Y1*v'(X1)/w'(X1)^3)~~y-coordinateof~~Wei25519, where (X1,Y1)=(X'/t^2,Y'/t^3) and where u', v', and w' are the polynomials with coefficients in GF(p) as defined~~P2in~~Appendix G.4.2, while mapping the point at infinity O~~termsof~~Wei25519.-3 to~~the~~point at infinity O~~y-coordinatesof~~Wei25519. Under this dual isogenous mapping,~~P and P1 andthe~~base point (G3X, G3Y) of Wei25519.-3 corresponds to a multiple of the base point (GX, GY)~~productof~~Wei25519, where this multiple is l=47 (the degree~~their x-coordinates. Further details are outof~~the isogeny; see the description~~scope. Appendix D. Relationship Between Curve Models The non-binary curves specifiedin Appendix~~F.4). Note that this isogenous~~A are expressed in different curve models, viz. as curves in short-Weierstrass form, as Montgomery curves, or as twisted Edwards curves. These curve models are related, as follows. D.1. Mapping between Twisted Edwards Curves and Montgomery Curves One canmap~~(and its dual) primarily involves the evaluation~~pointsof~~three fixed polynomials involving~~the~~x-coordinate, which takes roughly 140 modular multiplications (or less than 5-10% relative incremental cost compared~~Montgomery curve M_{A,B}topoints ofthe~~cost~~twisted Edwards curve E_{a,d}, where a:=(A+2)/B and d:=(A-2)/B and, conversely, map pointsof~~an elliptic~~the twisted Edwardscurve~~scalar multiplication). G.3. Further Domain Parameters The parameters~~E_{a,d} to pointsof the~~Weierstrass~~Montgomerycurve~~with a=2 that~~M_{A,B}, where A:=2*(a+d)/(a-d) and where B:=4/(a-d). For twisted Edwards curves we consider (i.e., those where ais~~isomorphic with Wei25519~~a square in GF(q), whereas d is not), this defines a one- to-one correspondence, which - in fact - is an isomorphism between M_{A,B}andE_{a,d}, thereby showing that, e.g.,the~~parameters of the Weierstrass~~discrete logarithm problem in eithercurve~~with a=-3 that~~modelis~~isogenous with Wei25519 are as indicated below. Both domain parameter sets can be exploited directly~~equally hard. For the Montgomery curves and twisted Edwards curves we consider, the mapping from M_{A,B}to~~derive more efficient~~E_{a,d} is defined by mapping thepoint~~addition formulae, should an implementation facilitate this. General parameters: same as for Wei25519 (see Appendix E.3) Weierstrass curve-specific parameters (for Wei25519.2, i.e., with a=2): a 2 (=0x02) b 12102640281269758552371076649779977768474709596484288167752775713 178787220689 (=0x1ac1da05 b55bc146 33bd39e4 7f94302e f19843dc f669916f 6a5dfd01 65538cd1) G2X 10770553138368400518417020196796161136792368198326337823149502681 097436401658 (=0x17cfeac3 78aed661 318e8634 582275b6 d9ad4def 072ea193 5ee3c4e8 7a940ffa) G2Y 54430575861508405653098668984457528616807103332502577521161439773 88639873869 (=0x0c08a952 c55dfad6 2c4f13f1 a8f68dca dc5c331d 297a37b6 f0d7fdcc 51e16b4d) Weierstrass curve-specific parameters (for Wei25519.-3, i.e., with a=-3): a -3 (=0x7fffffff ffffffff ffffffff ffffffff ffffffff ffffffff ffffffff ffffffea) b 29689592517550930188872794512874050362622433571298029721775200646 451501277098 (=0x41a3b6bf c668778e be2954a4 b1df36d1 485ecef1 ea614295 796e1022 40891faa) G3X 53837179229940872434942723257480777370451127212339198133697207846 219400243292 (=0x7706c37b 5a84128a 3884a5d7 1811f1b5 5da3230f fb17a8ab 0b32e48d 31a6685c) G3Y 69548073091100184414402055529279970392514867422855141773070804184 60388229929 (=0x0f60480c 7a5c0e11 40340adc 79d6a2bf 0cb57ad0 49d025dc 38d80c77 985f0329) G.4. Isogeny Details The isogeny~~at infinity Oand~~dual isogeny are both isogenies with degree l=47. Both are specified by a triple~~the point (0, 0)of~~polynomials u, v, and w (resp. u', v', and w')~~order twoof~~degree 47, 69, and 23,~~M_{A,B} to,respectively,~~with coefficients in GF(p). The coeffients~~the point (0, 1) and the point (0, -1) of order twoofE_{a,d}, while mappingeachother point (u, v)of~~these polynomials are specified in Appendix G.4.1 (for~~M_{A,B} tothe~~isogeny)~~point (x,y):=(u/v,(u-1)/(u+1)) of E_{a,d}. The inverse mapping from E_{a,d} to M_{A,B} is defined by mapping the point (0, 1)and~~in Appendix G.4.2 (for~~the~~dual isogeny). For each polynomial in variable x,~~point (0, -1) of order two of E_{a,d} to, respectively,the~~coefficients are tabulated as sequence~~point at infinity O and the point (0, 0)of~~coefficients~~order twoof~~x^0, x^1, x^2, ..., in hexadecimal format. G.4.1. Isogeny Parameters G.4.1.1. Coefficients~~M_{A,B}, while each other point (x, y)of~~u(x) 0 0x670ed14828b6f1791ceb3a9cc0edfe127dee8729c5a72ddf77bb1abaebbba1e8 1 0x1135ca8bd5383cb3545402c8bce2ced14b45c29b241e4751b035f27524a9f932 2 0x3223806ff5f669c430efd74df8389f058d180e2fcffa5cdef3eacecdd2c34771 3 0x31b8fecf3f17a819c228517f6cd9814466c8c8bea2efccc47a29bfc14c364266 4 0x2541305c958c5a326f44efad2bec284e7abee840fadb08f2d994cd382fd8ce42 5 0x6e6f9c5792f3ff497f860f44a9c469cec42bd711526b733e10915be5b2dbd8c6 6 0x3e9ad2e5f594b9ce6b06d4565891d28a1be8790000b396ef0bf59215d6cabfde 7 0x278448895d236403bbc161347d19c913e7df5f372732a823ed807ee1d30206be 8 0x42f9d171ea8dc2f4a14ea46cc0ee54967175ecfe83a975137b753cb127c35060 9 0x128e40efa2d3ccb51567e73bae91e7c31eac45700fa13ce5781cbe5ddc985648 10 0x450e5086c065430b496d88952dd2d5f2c5102bc27074d4d1e98bfa47413e0645 11 0x487ef93da70dfd44a4db8cb41542e33d1aa32237bdca3a59b3ce1c59585f253d 12 0x33d209270026b1d2db96efb36cc2fa0a49be1307f49689022eab1892b010b785 13 0x4732b5996a20ebc4d5c5e2375d3b6c4b700c681bd9904343a14a0555ef0ecd48 14 0x64dc9e8272b9f5c6ad3470db543238386f42b18cb1c592cc6caf7893141b2107 15 0x52bbacd1f85c61ef7eafd8da27260fa2821f7a961867ed449b283036508ac5c5 16 0x320447ed91210985e2c401cfe1a93db1379424cf748f92fd61ab5cc356bc89a2 17 0x23d23a49bbcdf8cf4c4ce8a4ff7dd87d1ad1970317686254d5b4d2ec050d019f 18 0x1601fca063f0bbbf15f198b3c20e474c2170294fa981f73365732d2372b40cd4 19 0x7bf3f93840035e9688cfff402cee204a17c0de9779fc33503537dd78021bf4c4 20 0x311998ce59fb7e1cd6af591ece3e84dfcb1c330cbcf28c0349e37b9581452853 21 0x7ae5e41acfd28a9add2216dfed34756575a19b16984c1f3847b694326dad7f99 22 0x704957e279244a5b107a6c57bd0ab9afe5227b7c0be2052cd3513772a40efee7 23 0x56b918b5a0c583cb763550f8f71481e57c13bdcef2e5cfc8091d0821266f233b 24 0x677073fed43ab291e496f798fbcf217bac3f014e35d0c2fa07f041ae746a04d7 25 0x22225388e76f9688c7d4053b50ba41d0d8b71a2f21da8353d98472243ef50170 26 0x66930b3dffdd3995a2502cef790d78b091c875192d8074bb5d5639f736400555 27 0x79eb677c5e36971e8d64d56ebc0dedb4e9b7dd2d7b01343ebbd4d358d376e490 28 0x48a204c2ca6d8636e9994842605bd648b91b637844e38d6c7dd707edce8256e2 29 0x0fb3529b0d4b9ce2d70760f33e8ce997a58999718e9277caf48623d27ae6a788 30 0x4352604bffd0c7d7a9ed898a2c6e7cf2512ffb89407271ba1f2c2d0ead8cc5aa 31 0x6667697b29785fb6f0bd5e04d828991a5fe525370216f347ec767a26e7aac936 32 0x09fc950b083c56dbd989badf9887255e203c879f123a7cb28901e50aea6d64dc 33 0x41e51b51b5caadd1c15436bbf37596a1d7288a5f495d6b5b1ae66f8b2942b31d 34 0x073b59fec709aa1cabd429e981c6284822a8b7b07620c831ab41fd31d5cf7430 35 0x67e9b88e9a1bfbc2554107d67d814986f1b09c3107a060cba21c019a2d5dc848 36 0x6881494a1066ca176c5e174713786040affb4268b19d2abf28ef4293429f89c1 37 0x5f4d30502ff1e1ccd624e6f506569454ab771869d7483e26afc09dea0c5ccd3d 38 0x02a814cfc5859bca51e539c159955cbe729a58978b52329575d09bc6c3bf97ad 39 0x1313c8aaae20d6f4397f0d8b19e52cfcdf8d8e10fba144aec1778fd10ddf4e9c 40 0x7008d38f434b98953a996d4cc79fcbef9502411dcdf92005f725cea7ce82ad47 41 0x5a74d1296aaaa245ffb848f434531fa3ba9e5cb9098a7091d36c2777d4cf5a13 42 0x4bd3b700606397083f8038177bdaa1ac6edbba0447537582723cae0fd29341a9 43 0x573453fb2b093016f3368356c786519d54ed05f5372c01723b4da520597ec217 44 0x77f5c605bdb3a30d7d9c8840fce38650910d4418eed707a212c8927f41c2c812 45 0x16d6b9f7ff57ca32350057de1204cc6d69d4ef1b255dfef8080118e2fef6ace3 46 0x34e8595832a4021f8b5744014c6b4f7da7df0d0329e8b6b4d44c8fadad6513b7 47 0x01 G.4.1.2. Coefficients~~E_{a,d} is mapped to the point (u,v):=((1+y)/(1-y),(1+y)/((1-y)*x))of~~v(x) 0 0x0f9f5eb7134e6f8dafa30c45afa58d7bfc6d4e3ccbb5de87b562fd77403972b2 1 0x36c2dcd9e88f0d2d517a15fc453a098bbbb5a05eb6e8da906fae418a4e1a13f7 2 0x0b40078302c24fa394a834880d5bf46732ca1b4894172fb7f775821276f558b3 3 0x53dd8e2234573f7f3f7df11e90a7bdd7b75d807f9712f521d4fb18af59aa5f26 4 0x6d4d7bb08de9061988a8cf6ff3beb10e933d4d2fbb8872d256a38c74c8c2ceda 5 0x71bfe5831b30e28cd0fbe1e9916ab2291c6beacc5af08e2c9165c632e61dd2f5 6 0x7c524f4d17ff2ee88463da012fc12a5b67d7fb5bd0ab59f4bbf162d76be1c89c 7 0x758183d5e07878d3364e3fd4c863a5dc1fe723f48c4ab4273fc034f5454d59a4 8 0x1eb41ef2479444ecdccbc200f64bde53f434a02b6c3f485d32f14da6aa7700e1 9 0x1490f3851f016cc3cf8a1e3c16a53317253d232ed425297531b560d70770315c 10 0x09bc43131964e46d905c3489c9d465c3abbd26eab9371c10e429b36d4b86469c 11 0x5f27c173d94c7a413a288348d3fc88daa0bcf5af8f436a47262050f240e9be3b 12 0x1d20010ec741aaa393cd19f0133b35f067adab0d105babe75fe45c8ba2732ceb 13 0x01b3c669ae49b86be2f0c946a9ff6c48e44740d7d9804146915747c3c025996a 14 0x24c6090f79ec13e3ae454d8f0f98e0c30a8938180595f79602f2ba013b3c10db 15 0x4650c5b5648c6c43ac75a2042048c699e44437929268661726e7182a31b1532f 16 0x0957a835fb8bac3360b5008790e4c1f3389589ba74c8e8bf648b856ba7f22ba5 17 0x1cd1300bc534880f95c7885d8df04a82bd54ed3e904b0749e0e3f8cb3240c7c7 18 0x760b486e0d3c6ee0833b34b64b7ebc846055d4d1e0beeb6aedd5132399ada0ea 19 0x1c666846c63965ef7edf519d6ada738f2b676ae38ff1f4621533373931b3220e 20 0x365055118b38d4bc0df86648044affea2ef33e9a392ad336444e7d15e45585d1 21 0x736487bde4b555abfccd3ea7ddcda98eda0d7c879664117dee906a88bc551194 22 0x70de05ab9520222a37c7a84c61eedff71cb50c5f6647fc2a5d6e0ff2305cea37 23 0x59053f6cdf6517ab3fe4bd9c9271d1892f8cf353d8041b98409e1e341a01f8b5 24 0x375db54ed12fe8df9a198ea40200e812c2660b7022681d7932d89fafe7c6e88d 25 0x2a070c31d1c1a064daf56c79a044bd1cd6d13f1ddb0ff039b03a6469aaa9ed77 26 0x41482351e7f69a756a5a2c0b3fa0681c03c550341d0ca0f76c5b394db9d2de8d 27 0x747ac1109c9e9368d94a302cb5a1d23fcc7f0fd8a574efb7ddcaa738297c407a 28 0x45682f1f2aab6358247e364834e2181ad0448bb815c587675fb2fee5a2119064 29 0x148c5bf44870dfd307317f0a0e4a8c163940bee1d2f01455a2e658aa92c13620 30 0x6add1361e56ffa2d2fbbddba284b35be5845aec8069fc28af009d53290a705ce 31 0x6631614c617400dc00f2c55357f67a94268e7b5369b02e55d5db46c935be3af5 32 0x17cffb496c64bb89d91c8c082f4c288c3c87feabd6b08591fe5a92216c094637 33 0x648ff88155969f54c955a1834ad227b93062bb191170dd8c4d759f79ad5da250 34 0x73e50900b89e5f295052b97f9d0c9edb0fc7d97b7fa5e3cfeefe33dd6a9cb223 35 0x6afcb2f2ffe6c08508477aa4956cbd3dc864257f5059685adf2c68d4f2338f00 36 0x372fd49701954c1b8f00926a8cb4b157d4165b75d53fa0476716554bf101b74c 37 0x0334ed41325f3724ff8becbf2b3443fea6d30fa543d1ca13188aceb2bdaf5f4e 38 0x70e629c95a94e8e1b3974acb25e18ba42f8d5991786f0931f650c283adfe82fd 39 0x738a625f4c62d3d645f1274e09ab344e72d441f3c0e82989d3e21e19212f23f3 40 0x7093737294b29f21522f5664a9941c9b476f75d443b647bd2c777040bcd12a6a 41 0x0a996bad5863d821ccb8b89fa329ddbe5317a46bcb32552db396bea933765436 42 0x2da237e3741b75dd0264836e7ef634fc0bc36ab187ebc790591a77c257b06f53 43 0x1902f3daa86fa4f430b57212924fdc9e40f09e809f3991a0b3a10ab186c50ee5 44 0x12baffec1bf20c921afd3cdf67a7f1d87c00d5326a3e5c83841593c214dadcb1 45 0x6460f5a68123cb9e7bc1289cd5023c0c9ccd2d98eea24484fb3825b59dcd09aa 46 0x2c7d63a868ffc9f0fd034f821d84736c5bc33325ce98aba5f0d95fef6f230ec8 47 0x756e0063349a702db7406984c285a9b6bfba48177950d4361d8efa77408dc860 48 0x037f3e30032b21e0279738e0a2b689625447831a2ccf15c638672da9aa7255ae 49 0x1107c0dbe15d6ca9e790768317a40bcf23c80f1841f03ca79dd3e3ef4ea1ae30 50 0x61ff7f25721d6206041c59a788316b09e05135a2aad94d539c65daa68b302cc2 51 0x5dbfe346cbd0d61b9a3b5c42ec0518d3ae81cabcc32245060d7b0cd982b8d071 52 0x4b6595e8501e9ec3e75f46107d2fd76511764efca179f69196eb45c0aa6fade3 53 0x72d17a5aa7bd8a2540aa9b02d9605f2a714f44abfb4c35d518b7abc39b477870 54 0x658d8c134bac37729ec40d27d50b637201abbf1ab4157316358953548c49cf22 55 0x36ac53b9118581ace574d5a08f9647e6a916f92dda684a4dbc405e2646b0243f 56 0x1917a98f387d1e323e84a0f02d53307b1dd949e1a27b0de14514f89d9c0ef4b6 57 0x21573434fde7ce56e8777c79539479441942dba535ade8ecb77763f7eb05d797 58 0x0e0bf482dc40884719bea5503422b603f3a8edb582f52838caa6eaab6eeac7ef 59 0x3b0471eb53bd83e14fbc13928fe1691820349a963be8f7e9815848a53d03f5eb 60 0x1e92cb067b24a729c42d3abb7a1179c577970f0ab3e6b0ce8d66c5b8f7001262 61 0x74ea885c1ebed6f74964262402432ef184c42884fceb2f8dba3a9d67a1344dd7 62 0x433ebce2ce9b0dc314425cfc2b234614d3c34f2c9da9fff4fdddd1ce242d035b 63 0x33ac69e6be858dde7b83a9ff6f11de443128b39cec6e410e8d3b570e405ff896 64 0x0dab71e2ae94e6530a501ed8cf3df26731dd1d41cd81578341e12dca3cb71aa3 65 0x537f58d52d18ce5b1d5a6bd3a420e796e64173491ad43dd4d1083a7dcc7dd201 66 0x49c2f6afa93fdcc4e0f8128a8b06da4c75049be14edf3e103821ab604c60f8ae 67 0x10a333eabd6135aeaa3f5f5f7e73d102e4fd7e4bf0902fc55b00da235fa1ad08 68 0x0f5c86044bf6032f5102e601f2a0f73c7bce9384bedd120f3e72d78484179d9c 69 0x01 G.4.1.3. Coefficients~~M_{A,B}. Implementations may take advantageof~~w(x) 0 0x3da24d42421264f30939ff00203880f2b017eb3fecf8933ae61e18df8c8ba116 1 0x0457f20bc393cdc9a66848ce174e2fa41d77e6dbae05a317a1fb6e3ae78760f8 2 0x7f608a2285c480d5c9592c435431fae94695beef79d770bb6d029c1d10a53295 3 0x3832accc520a485100a0a1695792465142a5572bed1b2e50e1f8f662ac7289bb 4 0x2df1b0559e31b328eb34beedd5e537c3f4d7b9befb0749f75d6d0d866d26fbaa 5 0x25396820381d04015a9f655ddd41c74303ded05d54a7750e2f58006659adda28 6 0x6fa070a70ca2bc6d4d0795fb28d4990b2cc80cd72d48b603a8ac8c8268bef6a6 7 0x27f488578357388b20fbc7503328e1d10de602b082b3c7b8ceb33c29fea7a0d2 8 0x15776851a7cabcfe84c632118306915c0c15c75068a47021968c7438d46076e6 9 0x101565b08a9af015c172fb194b940a4df25c4fb1d85f72d153efc79131d45e8f 10 0x196b0ffbf92f3229fea1dac0d74591b905ccaab6b83f905ee813ee8449f8a62c 11 0x01f55784691719f765f04ee9051ec95d5deb42ae45405a9d87833855a6d95a94 12 0x628858f79cca86305739d084d365d5a9e56e51a4485d253ae3f2e4a379fa8aff 13 0x4a842dcd943a80d1e6e1dab3622a8c4d390da1592d1e56d1c14c4d3f72dd01a5 14 0x0f3bfc9cb17a1125f94766a4097d0f1018963bc11cb7bc0c7a1d94d65e282477 15 0x1c4bd70488c4882846500691fa7543b7ef694446d9c3e3b4707ea2c99383e53c 16 0x2d7017e47b24b89b0528932c4ade43f09091b91db0072e6ebdc5e777cb215e35 17 0x781d69243b6c86f59416f91f7decaca93eab9cdc36a184191810c56ed85e0fdc 18 0x5f20526f4177357da40a18da054731d442ad2a5a4727322ba8ed10d32eca24fb 19 0x33e4cab64ed8a00d8012104fe8f928e6173c428eff95bbbe569ea46126a4f3cd 20 0x050555b6f07e308d33776922b6566829d122e19b25b7bbacbb0a4b1a7dc40192 21 0x533fa4bf1e2a2aae2f979065fdbb5b667ede2f85543fddbba146aa3a4ef2d281 22 0x5a742cac1952010fc5aba200a635a7bed3ef868194f45b5a6a2647d6d6b289d2 23 0x01 G.4.2. Dual Isogeny Parameters G.4.2.1. Coefficients~~this mapping to carry out elliptic curve group operations originally defined for a twisted Edwards curve on the corresponding Montgomery curve, or vice-versa, and translating the result back to the original curve, thereby potentially allowing code reuse. D.2. Mapping between Montgomery Curves and Weierstrass Curves One can map pointsof~~u'(x) 0 0x0f0eddb584a20aaac8f1419efdd02a5cca77b21e4cfae78c49b5127d98bc5882 1 0x7115e60d44a58630417df33dd45b8a546fa00b79fea3b2bdc449694bade87c0a 2 0x0b3f3a6f3c445c7dc1f91121275414e88c32ff3f367ba0edad4d75b7e7b94b65 3 0x1eb31bb333d7048b87f2b3d4ec76d69035927b41c30274368649c87c52e1ab30 4 0x552c886c2044153e280832264066cce2a7da1127dc9720e2a380e9d37049ac64 5 0x4504f27908db2e1f5840b74ae42445298755d9493141f5417c02f04d47797dda 6 0x082c242cce1eb19698a4fa30b5affe64e5051c04ae8b52cb68d89ee85222e628 7 0x480473406add76cf1d77661b3ff506c038d9cdd5ad6e1ea41969430bb876d223 8 0x25f47bb506fba80c79d1763365fa9076d4c4cb6644f73ed37918074397e88588 9 0x10f13ed36eab593fa20817f6bb70cac292e18d300498f6642e35cbdf772f0855 10 0x7d28329d695fb3305620f83a58df1531e89a43c7b3151d16f3b60a8246c36ade 11 0x02c5ec8c42b16dc6409bdd2c7b4ffe9d65d7209e886badbd5f865dec35e4ab4a 12 0x7f4f33cd50255537e6cde15a4a327a5790c37e081802654b56c956434354e133 13 0x7d30431a121d9240c761998cf83d228237e80c3ef5c7191ec9617208e0ab8cec 14 0x4d2a7d6609610c1deed56425a4615b92f70a507e1079b2681d96a2b874cf0630 15 0x74676df60a9906901d1dc316c639ff6ae0fcdb02b5571d4b83fc2eedcd2936a8 16 0x22f8212219aca01410f06eb234ed53bd5b8fbe7c08652b8002bcd1ea3cdae387 17 0x7edb04449565d7c566b934a87fadade5515f23bda1ce25daa19fff0c6a5ccc2f 18 0x106ef71aa3aa34e8ecf4c07a67d03f0949d7d015ef2c1e32eb698dd3bec5a18c 19 0x0017913eb705db126ac3172447bcd811a62744d505ad0eea94cfcfdde5ca7428 20 0x2cc793e6d3b592dcf5472057a991ff1a5ab43b4680bb34c0f5faffc5307827c1 21 0x6dafcc0b16f98300cddb5e0a7d7ff04a0e73ca558c54461781d5a5ccb1ea0122 22 0x7e418891cf222c021b0ae5f5232b9c0dc8270d4925a13174a0f0ac5e7a4c8045 23 0x76553bd26fecb019ead31142684789fea7754c2dc9ab9197c623f45d60749058 24 0x693efb3f81086043656d81840902b6f3a9a4b0e8f2a5a5edf5ce1c7f50a3898e 25 0x46c630eac2b86d36f18a061882b756917718a359f44752a5caf41be506788921 26 0x01dcfa01773628753bc6f448ac11be8a3bffa0011b9284967629b827e064f614 27 0x08430b5b97d49b0938d1f66ecb9d2043025c6eec624f8f02042b9621b2b5cb19 28 0x66f66a6669272d47d3ec1efea36ee01d4a54ed50e9ec84475f668a5a9850f9be 29 0x539128823b5ef3e87e901ab22f06d518a9bad15f5d375b49fe1e893ab38b1345 30 0x2bd01c49d6fff22c213a8688924c10bf29269388a69a08d7f326695b3c213931 31 0x3f7bea1baeccea3980201dc40d67c26db0e3b15b5a19b6cdac6de477aa717ac1 32 0x6e0a72d94867807f7150fcb1233062f911b46e2ad11a3eac3c6c4c91e0f4a3fa 33 0x5963f3cc262253f56fc103e50217e7e5b823ae8e1617f9e11f4c9c595fbb5bf6 34 0x41440b6fe787777bc7b63afac9f4a38ddadcebc3d72f8fc73835247ba05f3a1d 35 0x66d185401c1d2d0b84fcf6758a6a985bf9695651271c08f4b69ce89175fb7b34 36 0x2673fb8c65bc4fe41905381093429a2601c46a309c03077ca229bac7d6ccf239 37 0x1ce4d895ee601918a080de353633c82b75a3f61e8247763767d146554dd2f862 38 0x18efa6c72fa908347547a89028a44f79f22542baa588601f2b3ed25a5e56d27c 39 0x53de362e2f8ff220f8921620a71e8faa1aa57f8886fcbb6808fa3a5560570543 40 0x0dc29a73b97f08aa8774911474e651130ed364e8d8cffd4a80dee633aacecc47 41 0x4e7eb8584ae4de525389d1e9300fc4480b3d9c8a5a45ecfbe33311029d8f6b99 42 0x6c3cba4aa9229550fa82e1cfaee4b02f2c0cb86f79e0d412b8e32b00b7959d80 43 0x5a9d104ae585b94af68eeb16b1349776b601f97b7ce716701645b1a75b68dcf3 44 0x754e014b5e87af035b3d5fe6fb49f4631e32549f6341c6693c5172a6388e273e 45 0x6710d8265118e22eaceba09566c86f642ab42da58c435083a353eaa12d866c39 46 0x6e88ac659ce146c369f8b24c3a49f8dca547827250cf7963a455851cfc4f8d22 47 0x0971eb5f253356cd1fde9fb21f4a4902aa5b8d804a2b57ba775dc130181ae2e8 G.4.2.2. Coefficients~~the Montgomery curve M_{A,B} to pointsof~~v'(x) 0 0x043c9b67cc5b16e167b55f190db61e44d48d813a7112910f10e3fd8da85d61d3 1 0x72046db07e0e7882ff3f0f38b54b45ca84153be47a7fd1dd8f6402e17c47966f 2 0x1593d97b65a070b6b3f879fe3dc4d1ef03c0e781c997111d5c1748f956f1ffc0 3 0x54e5fec076b8779338432bdc5a449e36823a0a7c905fd37f232330b026a143a0 4 0x46328dd9bc336e0873abd453db472468393333fbf2010c6ac283933216e98038 5 0x25d0c64de1dfe1c6d5f5f2d98ab637d8b39bcf0d886a23dabac18c80d7eb03ce 6 0x3a175c46b2cd8e2b313dde2d5f3097b78114a6295f283cf58a33844b0c8d8b34 7 0x5cf4e6f745bdd61181a7d1b4db31dc4c30c84957f63cdf163bee5e466a7a8d38 8 0x639071c39b723eea51cfd870478331d60396b31f39a593ebdd9b1eb543875283 9 0x7ea8f895dcd85fc6cb2b58793789bd9246e62fa7a8c7116936876f4d8dff869b 10 0x503818acb535bcaacf8ad44a83c213a9ce83af7c937dc9b3e5b6efedc0a7428c 11 0x0e815373920ec3cbf3f8cae20d4389d367dc4398e01691244af90edc3e6d42b8 12 0x7e4b23e1e0b739087f77910cc635a92a3dc184a791400cbceae056c19c853815 13 0x145322201db4b5ec0a643229e07c0ab7c36e4274745689be2c19cfa8a702129d 14 0x0fde79514935d9b40f52e33429621a200acc092f6e5dec14b49e73f2f59c780d 15 0x37517ac5c04dc48145a9d6e14803b8ce9cb6a5d01c6f0ad1b04ff3353d02d815 16 0x58ae96b8eefe9e80f24d3b886932fe3c27aaea810fa189c702f93987c8c97854 17 0x6f6402c90fa379096d5f436035bebc9d29302126e9b117887abfa7d4b3c5709a~~the Weierstrass curve W_{a,b}, where a:=(3-A^2)/(3*B^2) and b:=(2*A^3-9*A)/(27*B^3). This defines a one-to-one correspondence, which - in fact - is an isomorphism between M_{A,B} and W_{a,b}, thereby showing that, e.g., the discrete logarithm problem in either curve model is equally hard. The mapping from M_{A,B} to W_{a,b} is defined by mapping the point at infinity O of M_{A,B} to the point at infinity O of W_{a,b}, while mapping each other point (u,v) of M_{A,B} to the point (X,Y):=((u+A/3)/B,v/B) of W_{a,b}. Note that not all Weierstrass curves can be injectively mapped to Montgomery curves, since the latter have a point of order two and the former may not. In particular, if a Weierstrass curve has prime order, such as is the case with the so-called "NIST prime curves", this inverse mapping is not defined. If the Weierstrass curve W_{a,b} has a point (alpha,0) of order two and c:=a+3*(alpha)^2 is a square in GF(q), one can map points of this curve to points of the Montgomery curve M_{A,B}, where A:=3*alpha/ gamma and B:=1/gamma and where gamma is any square root of c. In this case, the mapping from W_{a,b} to M_{A,B} is defined by mapping the point at infinity O of W_{a,b} to the point at infinity O of M_{A,B}, while mapping each other point (X,Y) of W_{a,b} to the point (u,v):=((X-alpha)/gamma,Y/gamma) of M_{A,B}. As before, this defines a one-to-one correspondence, which - in fact - is an isomorphism between W_{a,b} and M_{A,B}. It is easy to see that the mapping from W_{a,b} to M_{A,B} and that from M_{A,B} to W_{a,b} (if defined) are each other's inverse. This mapping can be used to implement elliptic curve group operations originally defined for a twisted Edwards curve or for a Montgomery curve using group operations for the corresponding elliptic curve in short-Weierstrass form and translating the result back to the original curve, thereby potentially allowing code reuse. Note that implementations for elliptic curves with short-Weierstrass form that hard-code the domain parameter a to a= -3 (which value is known to allow more efficient implementations) cannot always be used this way, since the curve W_{a,b} resulting from an isomorphic mapping cannot always be expressed as a Weierstrass curve with a=-3 via a coordinate transformation. For more details, see Appendix F. D.3. Mapping between Twisted Edwards Curves and Weierstrass Curves One can map points of the twisted Edwards curve E_{a,d} to points of the Weierstrass curve W_{a,b}, via function composition, where one uses the isomorphic mapping between twisted Edwards curves and Montgomery curves of Appendix D.1 and the one between Montgomery and Weierstrass curves of Appendix D.2. Obviously, one can use function composition (now using the respective inverses - if these exist) to realize the inverse of this mapping. Appendix E. Curve25519 and Cousins This section introduces curves related to Curve25519 and explains their relationships. E.1. Curve Definition and Alternative Representations The elliptic curve Curve25519 is the Montgomery curve M_{A,B} defined over the prime field GF(p), with p:=2^{255}-19, where A:=486662 and B:=1. This curve has order h*n, where h=8 and where n is a prime number. For this curve, A^2-4 is not a square in GF(p), whereas A+2 is. The quadratic twist of this curve has order h1*n1, where h1=4 and where n1 is a prime number. For this curve, the base point is the point (Gu, Gv), where Gu=9 and where Gv is an odd integer in the interval [0, p-1]. This curve has the same group structure as (is "isomorphic" to) the twisted Edwards curve E_{a,d} defined over GF(p), with as base point the point (Gx, Gy), where parameters are as specified in Appendix E.3. This curve is denoted as Edwards25519. For this curve, the parameter a is a square in GF(p), whereas d is not, so the group laws of Appendix C.3 apply. The curve is also isomorphic to the elliptic curve W_{a,b} in short- Weierstrass form defined over GF(p), with as base point the point (GX, GY), where parameters are as specified in Appendix E.3. This curve is denoted as Wei25519. For this curve, the parameter b is a square in GF(p). (For future reference, we note that this curve has no affine points with x-coordinate -1.) E.2. Switching between Alternative Representations Each affine point (u, v) of Curve25519 corresponds to the point (X, Y):=(u + A/3, v) of Wei25519, while the point at infinity of Curve25519 corresponds to the point at infinity of Wei25519. (Here, we used the mappings of Appendix D.2 and that B=1.) Under this mapping, the base point (Gu, Gv) of Curve25519 corresponds to the base point (GX, GY) of Wei25519. The inverse mapping maps the affine point (X, Y) of Wei25519 to (u, v):=(X - A/3, Y) of Curve25519, while mapping the point at infinity of Wei25519 to the point at infinity of Curve25519. Note that this mapping involves a simple shift of the first coordinate and can be implemented via integer-only arithmetic as a shift of (p+A)/3 for the isomorphic mapping and a shift of -(p+A)/3 for its inverse, where delta=(p+A)/3 is the element of GF(p) defined by delta 19298681539552699237261830834781317975544997444273427339909597 334652188435537 (=0x2aaaaaaa aaaaaaaa aaaaaaaa aaaaaaaa aaaaaaaa aaaaaaaa aaaaaaaa aaad2451). (Note that, depending on the implementation details of the field arithmetic, one may have to shift the result by +p or -p if this integer is not in the interval [0,p-1].) The curve Edwards25519 is isomorphic to the curve Curve25519, where the base point (Gu, Gv) of Curve25519 corresponds to the base point (Gx,Gy) of Edwards25519 and where the point at infinity and the point (0,0) of order two of Curve25519 correspond to, respectively, the point (0, 1) and the point (0, -1) of order two of Edwards25519 and where each other point (u, v) of Curve25519 corresponds to the point (c*u/v, (u-1)/(u+1)) of Edwards25519, where c is the element of GF(p) defined by c sqrt(-(A+2)/B) 51042569399160536130206135233146329284152202253034631822681833788 666877215207 (=0x70d9120b 9f5ff944 2d84f723 fc03b081 3a5e2c2e b482e57d 3391fb55 00ba81e7). (Here, we used the mapping of Appendix D.1 and normalized this using the mapping of Appendix F.1 (where the element s of that appendix is set to c above).) The inverse mapping from Edwards25519 to Curve25519 is defined by mapping the point (0, 1) and the point (0, -1) of order two of Edwards25519 to, respectively, the point at infinity and the point (0,0) of order two of Curve25519 and having each other point (x, y) of Edwards25519 correspond to the point ((1 + y)/(1 - y), c*(1 + y)/((1-y)*x)) of Curve25519. The curve Edwards25519 is isomorphic to the Weierstrass curve Wei25519, where the base point (Gx, Gy) of Edwards25519 corresponds to the base point (GX,GY) of Wei25519 and where the identity element (0,1) and the point (0,-1) of order two of Edwards25519 correspond to, respectively, the point at infinity O and the point (A/3, 0) of order two of Wei25519 and where each other point (x, y) of Edwards25519 corresponds to the point (X, Y):=((1+y)/(1-y)+A/3, c*(1+y)/((1-y)*x)) of Wei25519, where c was defined before. (Here, we used the mapping of Appendix D.3.) The inverse mapping from Wei25519 to Edwards25519 is defined by mapping the point at infinity O and the point (A/3, 0) of order two of Wei25519 to, respectively, the identity element (0,1) and the point (0,-1) of order two of Edwards25519 and having each other point (X, Y) of Wei25519 correspond to the point (c*(X-A/3)/Y, (X-A/3-1)/(X-A/3+1)) of Edwards25519. Note that these mappings can be easily realized if points are represented in projective coordinates, using a few field multiplications only, thus allowing switching between alternative curve representations with negligible relative incremental cost. E.3. Domain Parameters The parameters of the Montgomery curve and the corresponding isomorphic curves in twisted Edwards curve and short-Weierstrass form are as indicated below. Here, the domain parameters of the Montgomery curve Curve25519 and of the twisted Edwards curve Edwards25519 are as specified in [RFC7748]; the domain parameters of Wei25519 are "new". General parameters (for all curve models): p 2^{255}-19 (=0x7fffffff ffffffff ffffffff ffffffff ffffffff ffffffff ffffffff ffffffed) h 8 n 72370055773322622139731865630429942408571163593799076060019509382 85454250989 (=2^{252} + 0x14def9de a2f79cd6 5812631a 5cf5d3ed) h1 4 n1 14474011154664524427946373126085988481603263447650325797860494125 407373907997 (=2^{253} - 0x29bdf3bd 45ef39ac b024c634 b9eba7e3) Montgomery curve-specific parameters (for Curve25519): A 486662 (=0x076d06) B 1 (=0x01) Gu 9 (=0x09) Gv 14781619447589544791020593568409986887264606134616475288964881837 755586237401 (=0x20ae19a1 b8a086b4 e01edd2c 7748d14c 923d4d7e 6d7c61b2 29e9c5a2 7eced3d9) Twisted Edwards curve-specific parameters (for Edwards25519): a -1 (-0x01) d -121665/121666 = - (A-2)/(A+2) (=370957059346694393431380835087545651895421138798432190163887855 33085940283555) (=0x52036cee 2b6ffe73 8cc74079 7779e898 00700a4d 4141d8ab 75eb4dca 135978a3) Gx 15112221349535400772501151409588531511454012693041857206046113283 949847762202 (=0x216936d3 cd6e53fe c0a4e231 fdd6dc5c 692cc760 9525a7b2 c9562d60 8f25d51a) Gy 4/5 (=463168356949264781694283940034751631413079938662562256157830336 03165251855960) (=0x66666666 66666666 66666666 66666666 66666666 66666666 66666666 66666658) Weierstrass curve-specific parameters (for Wei25519): a 19298681539552699237261830834781317975544997444273427339909597334 573241639236 (=0x2aaaaaaa aaaaaaaa aaaaaaaa aaaaaaaa aaaaaaaa aaaaaaaa aaaaaa98 4914a144) b 55751746669818908907645289078257140818241103727901012315294400837 956729358436 (=0x7b425ed0 97b425ed 097b425e d097b425 ed097b42 5ed097b4 260b5e9c 7710c864) GX 19298681539552699237261830834781317975544997444273427339909597334 652188435546 (=0x2aaaaaaa aaaaaaaa aaaaaaaa aaaaaaaa aaaaaaaa aaaaaaaa aaaaaaaa aaad245a) GY 14781619447589544791020593568409986887264606134616475288964881837 755586237401 (=0x20ae19a1 b8a086b4 e01edd2c 7748d14c 923d4d7e 6d7c61b2 29e9c5a2 7eced3d9) Appendix F. Further Mappings The non-binary curves specified in Appendix A are expressed in different curve models, viz. as curves in short-Weierstrass form, as Montgomery curves, or as twisted Edwards curves. In Appendix D we already described relationships between these various curve models. Further mappings exist between elliptic curves within the same curve model. These can be exploited to force some of the domain parameters to specific values that allow for a more efficient implementation of the addition formulae. F.1. Isomorphic Mapping between Twisted Edwards Curves One can map points of the twisted Edwards curve E_{a,d} to points of the twisted Edwards curve E_{a',d'}, where a:=a'*s^2 and d:=d'*s^2 for some nonzero element s of GF(q). This defines a one-to-one correspondence, which - in fact - is an isomorphism between E_{a,d} and E_{a',d'}. The mapping from E_{a,d} to E_{a',d'} is defined by mapping the point (x,y) of E_{a,d} to the point (x', y'):=(s*x, y) of E_{a',d'}. The inverse mapping from E_{a',d'} to E_{a,d} is defined by mapping the point (x', y') of E_{a',d'} to the point (x, y):=(x'/s, y') of E_{a,d}. Implementations may take advantage of this mapping to carry out elliptic curve group operations originally defined for a twisted Edwards curve with generic domain parameters a and d on a corresponding isomorphic twisted Edwards curve with domain parameters a' and d' that have a more special form and that are known to allow for more efficient implementations of addition laws and translating the result back to the original curve. In particular, it is known that such efficiency improvements exist if a':=(+/-)1 (see [tEd-Formulas]). F.2. Isomorphic Mapping between Montgomery Curves One can map points of the Montgomery curve M_{A,B} to points of the Montgomery curve M_{A',B'}, where A:=A' and B:=B'*s^2 for some nonzero element s of GF(q). This defines a one-to-one correspondence, which - in fact - is an isomorphism between M_{A,B} and M_{A',B'}. The mapping from M_{A,B} to M_{A',B'} is defined by mapping the point at infinity O of M_{A,B} to the point at infinity O of M_{A',B'}, while mapping each other point (u,v) of M_{A,B} to the point (u', v'):=(u, s*v) of M_{A',B'}. The inverse mapping from M_{A',B'} to M_{A,B} is defined by mapping the point at infinity O of M_{A',B'} to the point at infinity O of M_{A,B}, while mapping each other point (u',v') of M_{A',B'} to the point (u,v):=(u',v'/s) of M_{A,B}. One can also map points of the Montgomery curve M_{A,B} to points of the Montgomery curve M_{A',B'}, where A':=-A and B':=-B. This defines a one-to-one correspondence, which - in fact - is an isomorphism between M_{A,B} and M_{A',B'}. In this case, the mapping from M_{A,B} to M_{A',B'} is defined by mapping the point at infinity O of M_{A,B} to the point at infinity O of M_{A',B'}, while mapping each other point (u,v) of M_{A,B} to the point (u',v'):=(-u,v) of M_{A',B'}. The inverse mapping from M_{A',B'} to M_{A,B} is defined by mapping the point at infinity O of M_{A',B'} to the point at infinity O of M_{A,B}, while mapping each other point (u',v') of M_{A',B'} to the point (u,v):=(-u',v') of M_{A,B}. Implementations may take advantage of these mappings to carry out elliptic curve groups operations originally defined for a Montgomery curve with generic domain parameters A and B on a corresponding isomorphic Montgomery curve with domain parameters A' and B' that have a more special form and that are known to allow for more efficient implementations of addition laws and translating the result back to the original curve. In particular, it is known that such efficiency improvements exist if B' assumes a small absolute value, such as B':=(+/-)1. (see [Mont-Ladder]). F.3. Isomorphic Mapping between Weierstrass Curves One can map points of the Weierstrass curve W_{a,b} to points of the Weierstrass curve W_{a',b'}, where a':=a*s^4 and b':=b*s^6 for some nonzero element s of GF(q). This defines a one-to-one correspondence, which - in fact - is an isomorphism between W_{a,b} and W_{a',b'}. The mapping from W_{a,b} to W_{a',b'} is defined by mapping the point at infinity O of W_{a,b} to the point at infinity O of W_{a',b'}, while mapping each other point (X,Y) of W_{a,b} to the point (X',Y'):=(X*s^2, Y*s^3) of W_{a',b'}. The inverse mapping from W_{a',b'} to W_{a,b} is defined by mapping the point at infinity O of W_{a',b'} to the point at infinity O of W_{a,b}, while mapping each other point (X', Y') of W_{a',b'} to the point (X,Y):=(X'/s^2,Y'/s^3) of W_{a,b}. Implementations may take advantage of this mapping to carry out elliptic curve group operations originally defined for a Weierstrass curve with generic domain parameters a and b on a corresponding isomorphic Weierstrass curve with domain parameter a' and b' that have a more special form and that are known to allow for more efficient implementations of addition laws and translating the result back to the original curve. In particular, it is known that such efficiency improvements exist if a'=-3 (mod p), where p is the characteristic of GF(q), and one uses so-called Jacobian coordinates with a particular projective version of the addition laws of Appendix C.1. While not all Weierstrass curves can be put into this form, all traditional NIST curves have domain parameter a=-3, while all Brainpool curves [RFC5639] are isomorphic to a Weierstrass curve of this form via the above mapping. Note that implementations for elliptic curves with short-Weierstrass form that hard-code the domain parameter a to a= -3 cannot always be used this way, since the curve W_{a,b} cannot always be expressed in terms of a Weierstrass curve with a'=-3 via a coordinate transformation: this only holds if a'/a is a fourth power in GF(q) (see Section 3.1.5 of [GECC]). However, even in this case, one can still express the curve W_{a,b} as a Weierstrass curve with a small domain parameter value a', thereby still allowing a more efficient implementation than with a general domain parameter value a. F.4. Isogenous Mapping between Weierstrass Curves One can still map points of the Weierstrass curve W_{a,b} to points of the Weierstrass curve W_{a',b'}, where a':=-3 (mod p) and where p is the characteristic of GF(q), even if a'/a is not a fourth power in GF(q). In that case, this mappping cannot be an isomorphism (see Appendix F.3). Instead, the mapping is a so-called isogeny (or homomorphism). Since most elliptic curve operations process points of prime order or use so-called "co-factor multiplication", in practice the resulting mapping has similar properties as an isomorphism. In particular, one can still take advantage of this mapping to carry out elliptic curve group operations originally defined for a Weierstrass curve with domain parameter a unequal to -3 (mod p) on a corresponding isogenous Weierstrass curve with domain parameter a'=-3 (mod p) and translating the result back to the original curve. In this case, the mapping from W_{a,b} to W_{a',b'} is defined by mapping the point at infinity O of W_{a,b} to the point at infinity O of W_{a',b'}, while mapping each other point (X,Y) of W_{a,b} to the point (X',Y'):=(u(X)/w(X)^2,Y*v(X)/w(X)^3) of W_{a',b'}. Here, u(X), v(X), and w(X) are polynomials in X that depend on the isogeny in question, as do domain parameters a' and b'. The inverse mapping from W_{a',b'} to W_{a,b} is again an isogeny (called the dual isogeny) and defined by mapping the point at infinity O of W_{a',b'} to the point at infinity O of W_{a,b}, while mapping each other point (X', Y') of W_{a',b'} to the point (X,Y):=(u'(X')/w'(X')^2,Y'*v'(X')/w'(X')^3) of W_{a,b}, where -- again -- u'(X'), v'(X'), and w'(X') are polynomials in X' that depend on the isogeny in question. These mappings have the property that their composition is not the identity mapping (as was the case with the isomorphic mappings discussed in Appendix F.3), but rather a fixed multiple hereof: if this multiple is l then the isogeny is called an isogeny of degree l (or l-isogeny) and u, v, and w (and, similarly, u', v', and w') are polynomials of degrees l, 3*(l-1)/2, and (l-1)/2, respectively. Note that an isomorphism is simply an isogeny of degree l=1. Details of how to determine isogenies are out of scope of this document. The above formulas assume that the isogeny has odd degree (i.e., l is odd); detailed formulas for even- degree isogenies are similar, but out of scope. Implementations may take advantage of this mapping to carry out elliptic curve group operations originally defined for a Weierstrass curve with a generic domain parameter a on a corresponding isogenous Weierstrass curve with domain parameter a'=-3 (mod p), where one can use so-called Jacobian coordinates with a particular projective version of the addition laws of Appendix C.1. Since all traditional NIST curves have domain parameter a=-3, while all Brainpool curves [RFC5639] are isomorphic to a Weierstrass curve of this form, this allows taking advantage of existing implementations for these curves that may have a hardcoded a=-3 (mod p) domain parameter, provided one switches back and forth to this curve form using the isogenous mapping in question. Note that isogenous mappings can be easily realized using representations in projective coordinates and involves roughly 3*l finite field multiplications, thus allowing switching between alternative representations at relatively low incremental cost compared to that of elliptic curve scalar multiplications (provided the isogeny has low degree l). Note, however, that this does require storage of the polynomial coefficients of the isogeny and dual isogeny involved. This illustrates that low-degree isogenies are to be preferred, since an l-isogeny (usually) requires storing roughly 6*l elements of GF(q). While there are many isogenies, we therefore only consider those with the desired property with lowest possible degree. Appendix G. Further Cousins of Curve25519 This section introduces some further curves related to Curve25519 and explains their relationships. G.1. Further Alternative Representations The Weierstrass curve Wei25519 is isomorphic to the Weierstrass curve Wei25519.2 defined over GF(p), with as base point the pair (G2X,G2Y), and isogenous to the Weierstrass curve Wei25519.-3 defined over GF(p), with as base point the pair (G3X, G3Y), where parameters are as specified in Appendix G.3 and where the related mappings are as specified in Appendix G.2. G.2. Further Switching Each affine point (X, Y) of Wei25519 corresponds to the point (X', Y'):=(X*s^2,Y*s^3) of Wei25519.2, where s is the element of GF(p) defined by s 20343593038935618591794247374137143598394058341193943326473831977 39407761440 (=0x047f6814 6d568b44 7e4552ea a5ed633d 02d62964 a2b0a120 5e7941e9 375de020), while the point at infinity of Wei25519 corresponds to the point at infinity of Wei25519.2. (Here, we used the mapping of Appendix F.3.) Under this mapping, the base point (GX, GY) of Wei25519 corresponds to the base point (G2X,G2Y) of Wei25519.2. The inverse mapping maps the affine point (X', Y') of Wei25519.2 to (X,Y):=(X'/s^2,Y'/s^3) of Wei25519, while mapping the point at infinity O of Wei25519.2 to the point at infinity O of Wei25519. Note that this mapping (and its inverse) involves a modular multiplication of both coordinates with fixed constants s^2 and s^3 (respectively, 1/s^2 and 1/s^3), which can be precomputed. Each affine point (X,Y) of Wei25519 corresponds to the point (X',Y'):=(X1*t^2,Y1*t^3) of Wei25519.-3, where (X1,Y1)=(u(X)/w(X)^2,Y*v(X)/w(X)^3), where u, v, and w are the polynomials with coefficients in GF(p) as defined in Appendix G.4.1 and where t is the element of GF(p) defined by t 35728133398289175649586938605660542688691615699169662967154525084 644181596229 (=0x4efd6829 88ff8526 e189f712 5999550c e9ef729b ed1a7015 73b1bab8 8bfcd845), while the point at infinity of Wei25519 corresponds to the point at infinity of Wei25519.-3. (Here, we used the isogenous mapping of Appendix F.4.) Under this isogenous mapping, the base point (GX,GY) of Wei25519 corresponds to the base point (G3X,G3Y) of Wei25519.-3. The dual isogeny maps the affine point (X',Y') of Wei25519.-3 to the affine point (X,Y):=(u'(X1)/w'(X1)^2,Y1*v'(X1)/w'(X1)^3) of Wei25519, where (X1,Y1)=(X'/t^2,Y'/t^3) and where u', v', and w' are the polynomials with coefficients in GF(p) as defined in Appendix G.4.2, while mapping the point at infinity O of Wei25519.-3 to the point at infinity O of Wei25519. Under this dual isogenous mapping, the base point (G3X, G3Y) of Wei25519.-3 corresponds to a multiple of the base point (GX, GY) of Wei25519, where this multiple is l=47 (the degree of the isogeny; see the description in Appendix F.4). Note that this isogenous map (and its dual) primarily involves the evaluation of three fixed polynomials involving the x-coordinate, which takes roughly 140 modular multiplications (or less than 5-10% relative incremental cost compared to the cost of an elliptic curve scalar multiplication). G.3. Further Domain Parameters The parameters of the Weierstrass curve with a=2 that is isomorphic with Wei25519 and the parameters of the Weierstrass curve with a=-3 that is isogenous with Wei25519 are as indicated below. Both domain parameter sets can be exploited directly to derive more efficient point addition formulae, should an implementation facilitate this. General parameters: same as for Wei25519 (see Appendix E.3) Weierstrass curve-specific parameters (for Wei25519.2, i.e., with a=2): a 2 (=0x02) b 12102640281269758552371076649779977768474709596484288167752775713 178787220689 (=0x1ac1da05 b55bc146 33bd39e4 7f94302e f19843dc f669916f 6a5dfd01 65538cd1) G2X 10770553138368400518417020196796161136792368198326337823149502681 097436401658 (=0x17cfeac3 78aed661 318e8634 582275b6 d9ad4def 072ea193 5ee3c4e8 7a940ffa) G2Y 54430575861508405653098668984457528616807103332502577521161439773 88639873869 (=0x0c08a952 c55dfad6 2c4f13f1 a8f68dca dc5c331d 297a37b6 f0d7fdcc 51e16b4d) Weierstrass curve-specific parameters (for Wei25519.-3, i.e., with a=-3): a -3 (=0x7fffffff ffffffff ffffffff ffffffff ffffffff ffffffff ffffffff ffffffea) b 29689592517550930188872794512874050362622433571298029721775200646 451501277098 (=0x41a3b6bf c668778e be2954a4 b1df36d1 485ecef1 ea614295 796e1022 40891faa) G3X 53837179229940872434942723257480777370451127212339198133697207846 219400243292 (=0x7706c37b 5a84128a 3884a5d7 1811f1b5 5da3230f fb17a8ab 0b32e48d 31a6685c) G3Y 69548073091100184414402055529279970392514867422855141773070804184 60388229929 (=0x0f60480c 7a5c0e11 40340adc 79d6a2bf 0cb57ad0 49d025dc 38d80c77 985f0329) G.4. Isogeny Details The isogeny and dual isogeny are both isogenies with degree l=47. Both are specified by a triple of polynomials u, v, and w (resp. u', v', and w') of degree 47, 69, and 23, respectively, with coefficients in GF(p). The coeffients of each of these polynomials are specified in Appendix G.4.1 (for the isogeny) and in Appendix G.4.2 (for the dual isogeny). For each polynomial in variable x, the coefficients are tabulated as sequence of coefficients of x^0, x^1, x^2, ..., in hexadecimal format. G.4.1. Isogeny Parameters G.4.1.1. Coefficients of u(x) 0 0x670ed14828b6f1791ceb3a9cc0edfe127dee8729c5a72ddf77bb1abaebbba1e8 1 0x1135ca8bd5383cb3545402c8bce2ced14b45c29b241e4751b035f27524a9f932 2 0x3223806ff5f669c430efd74df8389f058d180e2fcffa5cdef3eacecdd2c34771 3 0x31b8fecf3f17a819c228517f6cd9814466c8c8bea2efccc47a29bfc14c364266 4 0x2541305c958c5a326f44efad2bec284e7abee840fadb08f2d994cd382fd8ce42 5 0x6e6f9c5792f3ff497f860f44a9c469cec42bd711526b733e10915be5b2dbd8c6 6 0x3e9ad2e5f594b9ce6b06d4565891d28a1be8790000b396ef0bf59215d6cabfde 7 0x278448895d236403bbc161347d19c913e7df5f372732a823ed807ee1d30206be 8 0x42f9d171ea8dc2f4a14ea46cc0ee54967175ecfe83a975137b753cb127c35060 9 0x128e40efa2d3ccb51567e73bae91e7c31eac45700fa13ce5781cbe5ddc985648 10 0x450e5086c065430b496d88952dd2d5f2c5102bc27074d4d1e98bfa47413e0645 11 0x487ef93da70dfd44a4db8cb41542e33d1aa32237bdca3a59b3ce1c59585f253d 12 0x33d209270026b1d2db96efb36cc2fa0a49be1307f49689022eab1892b010b785 13 0x4732b5996a20ebc4d5c5e2375d3b6c4b700c681bd9904343a14a0555ef0ecd48 14 0x64dc9e8272b9f5c6ad3470db543238386f42b18cb1c592cc6caf7893141b2107 15 0x52bbacd1f85c61ef7eafd8da27260fa2821f7a961867ed449b283036508ac5c5 16 0x320447ed91210985e2c401cfe1a93db1379424cf748f92fd61ab5cc356bc89a2 17 0x23d23a49bbcdf8cf4c4ce8a4ff7dd87d1ad1970317686254d5b4d2ec050d019f 18 0x1601fca063f0bbbf15f198b3c20e474c2170294fa981f73365732d2372b40cd4 19 0x7bf3f93840035e9688cfff402cee204a17c0de9779fc33503537dd78021bf4c4 20 0x311998ce59fb7e1cd6af591ece3e84dfcb1c330cbcf28c0349e37b9581452853 21 0x7ae5e41acfd28a9add2216dfed34756575a19b16984c1f3847b694326dad7f99 22 0x704957e279244a5b107a6c57bd0ab9afe5227b7c0be2052cd3513772a40efee7 23 0x56b918b5a0c583cb763550f8f71481e57c13bdcef2e5cfc8091d0821266f233b 24 0x677073fed43ab291e496f798fbcf217bac3f014e35d0c2fa07f041ae746a04d7 25 0x22225388e76f9688c7d4053b50ba41d0d8b71a2f21da8353d98472243ef50170 26 0x66930b3dffdd3995a2502cef790d78b091c875192d8074bb5d5639f736400555 27 0x79eb677c5e36971e8d64d56ebc0dedb4e9b7dd2d7b01343ebbd4d358d376e490 28 0x48a204c2ca6d8636e9994842605bd648b91b637844e38d6c7dd707edce8256e2 29 0x0fb3529b0d4b9ce2d70760f33e8ce997a58999718e9277caf48623d27ae6a788 30 0x4352604bffd0c7d7a9ed898a2c6e7cf2512ffb89407271ba1f2c2d0ead8cc5aa 31 0x6667697b29785fb6f0bd5e04d828991a5fe525370216f347ec767a26e7aac936 32 0x09fc950b083c56dbd989badf9887255e203c879f123a7cb28901e50aea6d64dc 33 0x41e51b51b5caadd1c15436bbf37596a1d7288a5f495d6b5b1ae66f8b2942b31d 34 0x073b59fec709aa1cabd429e981c6284822a8b7b07620c831ab41fd31d5cf7430 35 0x67e9b88e9a1bfbc2554107d67d814986f1b09c3107a060cba21c019a2d5dc848 36 0x6881494a1066ca176c5e174713786040affb4268b19d2abf28ef4293429f89c1 37 0x5f4d30502ff1e1ccd624e6f506569454ab771869d7483e26afc09dea0c5ccd3d 38 0x02a814cfc5859bca51e539c159955cbe729a58978b52329575d09bc6c3bf97ad 39 0x1313c8aaae20d6f4397f0d8b19e52cfcdf8d8e10fba144aec1778fd10ddf4e9c 40 0x7008d38f434b98953a996d4cc79fcbef9502411dcdf92005f725cea7ce82ad47 41 0x5a74d1296aaaa245ffb848f434531fa3ba9e5cb9098a7091d36c2777d4cf5a13 42 0x4bd3b700606397083f8038177bdaa1ac6edbba0447537582723cae0fd29341a9 43 0x573453fb2b093016f3368356c786519d54ed05f5372c01723b4da520597ec217 44 0x77f5c605bdb3a30d7d9c8840fce38650910d4418eed707a212c8927f41c2c812 45 0x16d6b9f7ff57ca32350057de1204cc6d69d4ef1b255dfef8080118e2fef6ace3 46 0x34e8595832a4021f8b5744014c6b4f7da7df0d0329e8b6b4d44c8fadad6513b7 47 0x01 G.4.1.2. Coefficients of v(x) 0 0x0f9f5eb7134e6f8dafa30c45afa58d7bfc6d4e3ccbb5de87b562fd77403972b2 1 0x36c2dcd9e88f0d2d517a15fc453a098bbbb5a05eb6e8da906fae418a4e1a13f7 2 0x0b40078302c24fa394a834880d5bf46732ca1b4894172fb7f775821276f558b3 3 0x53dd8e2234573f7f3f7df11e90a7bdd7b75d807f9712f521d4fb18af59aa5f26 4 0x6d4d7bb08de9061988a8cf6ff3beb10e933d4d2fbb8872d256a38c74c8c2ceda 5 0x71bfe5831b30e28cd0fbe1e9916ab2291c6beacc5af08e2c9165c632e61dd2f5 6 0x7c524f4d17ff2ee88463da012fc12a5b67d7fb5bd0ab59f4bbf162d76be1c89c 7 0x758183d5e07878d3364e3fd4c863a5dc1fe723f48c4ab4273fc034f5454d59a4 8 0x1eb41ef2479444ecdccbc200f64bde53f434a02b6c3f485d32f14da6aa7700e1 9 0x1490f3851f016cc3cf8a1e3c16a53317253d232ed425297531b560d70770315c 10 0x09bc43131964e46d905c3489c9d465c3abbd26eab9371c10e429b36d4b86469c 11 0x5f27c173d94c7a413a288348d3fc88daa0bcf5af8f436a47262050f240e9be3b 12 0x1d20010ec741aaa393cd19f0133b35f067adab0d105babe75fe45c8ba2732ceb 13 0x01b3c669ae49b86be2f0c946a9ff6c48e44740d7d9804146915747c3c025996a 14 0x24c6090f79ec13e3ae454d8f0f98e0c30a8938180595f79602f2ba013b3c10db 15 0x4650c5b5648c6c43ac75a2042048c699e44437929268661726e7182a31b1532f 16 0x0957a835fb8bac3360b5008790e4c1f3389589ba74c8e8bf648b856ba7f22ba5 17 0x1cd1300bc534880f95c7885d8df04a82bd54ed3e904b0749e0e3f8cb3240c7c7 18 0x760b486e0d3c6ee0833b34b64b7ebc846055d4d1e0beeb6aedd5132399ada0ea 19 0x1c666846c63965ef7edf519d6ada738f2b676ae38ff1f4621533373931b3220e 20 0x365055118b38d4bc0df86648044affea2ef33e9a392ad336444e7d15e45585d1 21 0x736487bde4b555abfccd3ea7ddcda98eda0d7c879664117dee906a88bc551194 22 0x70de05ab9520222a37c7a84c61eedff71cb50c5f6647fc2a5d6e0ff2305cea37 23 0x59053f6cdf6517ab3fe4bd9c9271d1892f8cf353d8041b98409e1e341a01f8b5 24 0x375db54ed12fe8df9a198ea40200e812c2660b7022681d7932d89fafe7c6e88d 25 0x2a070c31d1c1a064daf56c79a044bd1cd6d13f1ddb0ff039b03a6469aaa9ed77 26 0x41482351e7f69a756a5a2c0b3fa0681c03c550341d0ca0f76c5b394db9d2de8d 27 0x747ac1109c9e9368d94a302cb5a1d23fcc7f0fd8a574efb7ddcaa738297c407a 28 0x45682f1f2aab6358247e364834e2181ad0448bb815c587675fb2fee5a2119064 29 0x148c5bf44870dfd307317f0a0e4a8c163940bee1d2f01455a2e658aa92c13620 30 0x6add1361e56ffa2d2fbbddba284b35be5845aec8069fc28af009d53290a705ce 31 0x6631614c617400dc00f2c55357f67a94268e7b5369b02e55d5db46c935be3af5 32 0x17cffb496c64bb89d91c8c082f4c288c3c87feabd6b08591fe5a92216c094637 33 0x648ff88155969f54c955a1834ad227b93062bb191170dd8c4d759f79ad5da250 34 0x73e50900b89e5f295052b97f9d0c9edb0fc7d97b7fa5e3cfeefe33dd6a9cb223 35 0x6afcb2f2ffe6c08508477aa4956cbd3dc864257f5059685adf2c68d4f2338f00 36 0x372fd49701954c1b8f00926a8cb4b157d4165b75d53fa0476716554bf101b74c 37 0x0334ed41325f3724ff8becbf2b3443fea6d30fa543d1ca13188aceb2bdaf5f4e 38 0x70e629c95a94e8e1b3974acb25e18ba42f8d5991786f0931f650c283adfe82fd 39 0x738a625f4c62d3d645f1274e09ab344e72d441f3c0e82989d3e21e19212f23f3 40 0x7093737294b29f21522f5664a9941c9b476f75d443b647bd2c777040bcd12a6a 41 0x0a996bad5863d821ccb8b89fa329ddbe5317a46bcb32552db396bea933765436 42 0x2da237e3741b75dd0264836e7ef634fc0bc36ab187ebc790591a77c257b06f53 43 0x1902f3daa86fa4f430b57212924fdc9e40f09e809f3991a0b3a10ab186c50ee5 44 0x12baffec1bf20c921afd3cdf67a7f1d87c00d5326a3e5c83841593c214dadcb1 45 0x6460f5a68123cb9e7bc1289cd5023c0c9ccd2d98eea24484fb3825b59dcd09aa 46 0x2c7d63a868ffc9f0fd034f821d84736c5bc33325ce98aba5f0d95fef6f230ec8 47 0x756e0063349a702db7406984c285a9b6bfba48177950d4361d8efa77408dc860 48 0x037f3e30032b21e0279738e0a2b689625447831a2ccf15c638672da9aa7255ae 49 0x1107c0dbe15d6ca9e790768317a40bcf23c80f1841f03ca79dd3e3ef4ea1ae30 50 0x61ff7f25721d6206041c59a788316b09e05135a2aad94d539c65daa68b302cc2 51 0x5dbfe346cbd0d61b9a3b5c42ec0518d3ae81cabcc32245060d7b0cd982b8d071 52 0x4b6595e8501e9ec3e75f46107d2fd76511764efca179f69196eb45c0aa6fade3 53 0x72d17a5aa7bd8a2540aa9b02d9605f2a714f44abfb4c35d518b7abc39b477870 54 0x658d8c134bac37729ec40d27d50b637201abbf1ab4157316358953548c49cf22 55 0x36ac53b9118581ace574d5a08f9647e6a916f92dda684a4dbc405e2646b0243f 56 0x1917a98f387d1e323e84a0f02d53307b1dd949e1a27b0de14514f89d9c0ef4b6 57 0x21573434fde7ce56e8777c79539479441942dba535ade8ecb77763f7eb05d797 58 0x0e0bf482dc40884719bea5503422b603f3a8edb582f52838caa6eaab6eeac7ef 59 0x3b0471eb53bd83e14fbc13928fe1691820349a963be8f7e9815848a53d03f5eb 60 0x1e92cb067b24a729c42d3abb7a1179c577970f0ab3e6b0ce8d66c5b8f7001262 61 0x74ea885c1ebed6f74964262402432ef184c42884fceb2f8dba3a9d67a1344dd7 62 0x433ebce2ce9b0dc314425cfc2b234614d3c34f2c9da9fff4fdddd1ce242d035b 63 0x33ac69e6be858dde7b83a9ff6f11de443128b39cec6e410e8d3b570e405ff896 64 0x0dab71e2ae94e6530a501ed8cf3df26731dd1d41cd81578341e12dca3cb71aa3 65 0x537f58d52d18ce5b1d5a6bd3a420e796e64173491ad43dd4d1083a7dcc7dd201 66 0x49c2f6afa93fdcc4e0f8128a8b06da4c75049be14edf3e103821ab604c60f8ae 67 0x10a333eabd6135aeaa3f5f5f7e73d102e4fd7e4bf0902fc55b00da235fa1ad08 68 0x0f5c86044bf6032f5102e601f2a0f73c7bce9384bedd120f3e72d78484179d9c 69 0x01 G.4.1.3. Coefficients of w(x) 0 0x3da24d42421264f30939ff00203880f2b017eb3fecf8933ae61e18df8c8ba116 1 0x0457f20bc393cdc9a66848ce174e2fa41d77e6dbae05a317a1fb6e3ae78760f8 2 0x7f608a2285c480d5c9592c435431fae94695beef79d770bb6d029c1d10a53295 3 0x3832accc520a485100a0a1695792465142a5572bed1b2e50e1f8f662ac7289bb 4 0x2df1b0559e31b328eb34beedd5e537c3f4d7b9befb0749f75d6d0d866d26fbaa 5 0x25396820381d04015a9f655ddd41c74303ded05d54a7750e2f58006659adda28 6 0x6fa070a70ca2bc6d4d0795fb28d4990b2cc80cd72d48b603a8ac8c8268bef6a6 7 0x27f488578357388b20fbc7503328e1d10de602b082b3c7b8ceb33c29fea7a0d2 8 0x15776851a7cabcfe84c632118306915c0c15c75068a47021968c7438d46076e6 9 0x101565b08a9af015c172fb194b940a4df25c4fb1d85f72d153efc79131d45e8f 10 0x196b0ffbf92f3229fea1dac0d74591b905ccaab6b83f905ee813ee8449f8a62c 11 0x01f55784691719f765f04ee9051ec95d5deb42ae45405a9d87833855a6d95a94 12 0x628858f79cca86305739d084d365d5a9e56e51a4485d253ae3f2e4a379fa8aff 13 0x4a842dcd943a80d1e6e1dab3622a8c4d390da1592d1e56d1c14c4d3f72dd01a5 14 0x0f3bfc9cb17a1125f94766a4097d0f1018963bc11cb7bc0c7a1d94d65e282477 15 0x1c4bd70488c4882846500691fa7543b7ef694446d9c3e3b4707ea2c99383e53c 16 0x2d7017e47b24b89b0528932c4ade43f09091b91db0072e6ebdc5e777cb215e35 17 0x781d69243b6c86f59416f91f7decaca93eab9cdc36a184191810c56ed85e0fdc 18 0x5f20526f4177357da40a18da054731d442ad2a5a4727322ba8ed10d32eca24fb 19 0x33e4cab64ed8a00d8012104fe8f928e6173c428eff95bbbe569ea46126a4f3cd 20 0x050555b6f07e308d33776922b6566829d122e19b25b7bbacbb0a4b1a7dc40192 21 0x533fa4bf1e2a2aae2f979065fdbb5b667ede2f85543fddbba146aa3a4ef2d281 22 0x5a742cac1952010fc5aba200a635a7bed3ef868194f45b5a6a2647d6d6b289d2 23 0x01 G.4.2. Dual Isogeny Parameters G.4.2.1. Coefficients of u'(x) 0 0x0f0eddb584a20aaac8f1419efdd02a5cca77b21e4cfae78c49b5127d98bc5882 1 0x7115e60d44a58630417df33dd45b8a546fa00b79fea3b2bdc449694bade87c0a 2 0x0b3f3a6f3c445c7dc1f91121275414e88c32ff3f367ba0edad4d75b7e7b94b65 3 0x1eb31bb333d7048b87f2b3d4ec76d69035927b41c30274368649c87c52e1ab30 4 0x552c886c2044153e280832264066cce2a7da1127dc9720e2a380e9d37049ac64 5 0x4504f27908db2e1f5840b74ae42445298755d9493141f5417c02f04d47797dda 6 0x082c242cce1eb19698a4fa30b5affe64e5051c04ae8b52cb68d89ee85222e628 7 0x480473406add76cf1d77661b3ff506c038d9cdd5ad6e1ea41969430bb876d223 8 0x25f47bb506fba80c79d1763365fa9076d4c4cb6644f73ed37918074397e88588 9 0x10f13ed36eab593fa20817f6bb70cac292e18d300498f6642e35cbdf772f0855 10 0x7d28329d695fb3305620f83a58df1531e89a43c7b3151d16f3b60a8246c36ade 11 0x02c5ec8c42b16dc6409bdd2c7b4ffe9d65d7209e886badbd5f865dec35e4ab4a 12 0x7f4f33cd50255537e6cde15a4a327a5790c37e081802654b56c956434354e133 13 0x7d30431a121d9240c761998cf83d228237e80c3ef5c7191ec9617208e0ab8cec 14 0x4d2a7d6609610c1deed56425a4615b92f70a507e1079b2681d96a2b874cf0630 15 0x74676df60a9906901d1dc316c639ff6ae0fcdb02b5571d4b83fc2eedcd2936a8 16 0x22f8212219aca01410f06eb234ed53bd5b8fbe7c08652b8002bcd1ea3cdae387 17 0x7edb04449565d7c566b934a87fadade5515f23bda1ce25daa19fff0c6a5ccc2f 18 0x106ef71aa3aa34e8ecf4c07a67d03f0949d7d015ef2c1e32eb698dd3bec5a18c 19 0x0017913eb705db126ac3172447bcd811a62744d505ad0eea94cfcfdde5ca7428 20 0x2cc793e6d3b592dcf5472057a991ff1a5ab43b4680bb34c0f5faffc5307827c1 21 0x6dafcc0b16f98300cddb5e0a7d7ff04a0e73ca558c54461781d5a5ccb1ea0122 22 0x7e418891cf222c021b0ae5f5232b9c0dc8270d4925a13174a0f0ac5e7a4c8045 23 0x76553bd26fecb019ead31142684789fea7754c2dc9ab9197c623f45d60749058 24 0x693efb3f81086043656d81840902b6f3a9a4b0e8f2a5a5edf5ce1c7f50a3898e 25 0x46c630eac2b86d36f18a061882b756917718a359f44752a5caf41be506788921 26 0x01dcfa01773628753bc6f448ac11be8a3bffa0011b9284967629b827e064f614 27 0x08430b5b97d49b0938d1f66ecb9d2043025c6eec624f8f02042b9621b2b5cb19 28 0x66f66a6669272d47d3ec1efea36ee01d4a54ed50e9ec84475f668a5a9850f9be 29 0x539128823b5ef3e87e901ab22f06d518a9bad15f5d375b49fe1e893ab38b1345 30 0x2bd01c49d6fff22c213a8688924c10bf29269388a69a08d7f326695b3c213931 31 0x3f7bea1baeccea3980201dc40d67c26db0e3b15b5a19b6cdac6de477aa717ac1 32 0x6e0a72d94867807f7150fcb1233062f911b46e2ad11a3eac3c6c4c91e0f4a3fa 33 0x5963f3cc262253f56fc103e50217e7e5b823ae8e1617f9e11f4c9c595fbb5bf6 34 0x41440b6fe787777bc7b63afac9f4a38ddadcebc3d72f8fc73835247ba05f3a1d 35 0x66d185401c1d2d0b84fcf6758a6a985bf9695651271c08f4b69ce89175fb7b34 36 0x2673fb8c65bc4fe41905381093429a2601c46a309c03077ca229bac7d6ccf239 37 0x1ce4d895ee601918a080de353633c82b75a3f61e8247763767d146554dd2f862 38 0x18efa6c72fa908347547a89028a44f79f22542baa588601f2b3ed25a5e56d27c 39 0x53de362e2f8ff220f8921620a71e8faa1aa57f8886fcbb6808fa3a5560570543 40 0x0dc29a73b97f08aa8774911474e651130ed364e8d8cffd4a80dee633aacecc47 41 0x4e7eb8584ae4de525389d1e9300fc4480b3d9c8a5a45ecfbe33311029d8f6b99 42 0x6c3cba4aa9229550fa82e1cfaee4b02f2c0cb86f79e0d412b8e32b00b7959d80 43 0x5a9d104ae585b94af68eeb16b1349776b601f97b7ce716701645b1a75b68dcf3 44 0x754e014b5e87af035b3d5fe6fb49f4631e32549f6341c6693c5172a6388e273e 45 0x6710d8265118e22eaceba09566c86f642ab42da58c435083a353eaa12d866c39 46 0x6e88ac659ce146c369f8b24c3a49f8dca547827250cf7963a455851cfc4f8d22 47 0x0971eb5f253356cd1fde9fb21f4a4902aa5b8d804a2b57ba775dc130181ae2e8 G.4.2.2. Coefficients of v'(x) 0 0x043c9b67cc5b16e167b55f190db61e44d48d813a7112910f10e3fd8da85d61d3 1 0x72046db07e0e7882ff3f0f38b54b45ca84153be47a7fd1dd8f6402e17c47966f 2 0x1593d97b65a070b6b3f879fe3dc4d1ef03c0e781c997111d5c1748f956f1ffc0 3 0x54e5fec076b8779338432bdc5a449e36823a0a7c905fd37f232330b026a143a0 4 0x46328dd9bc336e0873abd453db472468393333fbf2010c6ac283933216e98038 5 0x25d0c64de1dfe1c6d5f5f2d98ab637d8b39bcf0d886a23dabac18c80d7eb03ce 6 0x3a175c46b2cd8e2b313dde2d5f3097b78114a6295f283cf58a33844b0c8d8b34 7 0x5cf4e6f745bdd61181a7d1b4db31dc4c30c84957f63cdf163bee5e466a7a8d38 8 0x639071c39b723eea51cfd870478331d60396b31f39a593ebdd9b1eb543875283 9 0x7ea8f895dcd85fc6cb2b58793789bd9246e62fa7a8c7116936876f4d8dff869b 10 0x503818acb535bcaacf8ad44a83c213a9ce83af7c937dc9b3e5b6efedc0a7428c 11 0x0e815373920ec3cbf3f8cae20d4389d367dc4398e01691244af90edc3e6d42b8 12 0x7e4b23e1e0b739087f77910cc635a92a3dc184a791400cbceae056c19c853815 13 0x145322201db4b5ec0a643229e07c0ab7c36e4274745689be2c19cfa8a702129d 14 0x0fde79514935d9b40f52e33429621a200acc092f6e5dec14b49e73f2f59c780d 15 0x37517ac5c04dc48145a9d6e14803b8ce9cb6a5d01c6f0ad1b04ff3353d02d815 16 0x58ae96b8eefe9e80f24d3b886932fe3c27aaea810fa189c702f93987c8c97854 17 0x6f6402c90fa379096d5f436035bebc9d29302126e9b117887abfa7d4b3c5709a18 0x01dbdf2b9ec09a8defeb485cc16ea98d0d45c5b9877ff16bd04c0110d2f64961 19 0x53c51706af523ab5b32291de6c6b1ee7c5cbd0a5b317218f917b12ff38421452 20 0x1b1051c7aec7d37a349208e3950b679d14e39f979db4fcd7b50d7d27dc918650 21 0x1547e8d36262d5434cfb029cdd29385353124c3c35b1423c6cca1f87910b305b 22 0x198efe984efc817835e28f704d41e4583a1e2398f7ce14045c4575d0445c6ce7 23 0x492276dfe9588ee5cd9f553d990f377935d721822ecd0333ce2eb1d4324d539c 24 0x77bad5319bacd5ed99e1905ce2ae89294efa7ee1f74314e4095c618a4e580c9b 25 0x2cb3d532b8eac41c61b683f7b02feb9c2761f8b4286a54c3c4b60dd8081a312e 26 0x37d189ea60443e2fee9b7ba8a34ed79ff3883dcefc06592836d2a9dd2ee3656e 27 0x79a80f9a0e6b8ded17a3d6ccf71eb565e3704c3543b77d70bca854345e880aba 28 0x47718530ef8e8c75f069acb2d9925c5537908e220b28c8a2859b856f46d5f8db 29 0x7dc518f82b55a36b4fa084b05bf21e3efce481d278a9f5c6a49701e56dac01ec 30 0x340a318dad4b8d348a0838659672792a0f00b7105881e6080a340f708a9c7f94 31 0x55f04d9d8891636d4e9c808a1fa95ad0dae7a8492257b20448023aad3203278e 32 0x39dc465d58259f9f70bb430d27e2f0ab384a550e1259655443e14bdecba85530 33 0x757385464cff265379a1adfadfd6f6a03fa8a2278761d4889ab097eff4d1ac28 34 0x4d575654dbe39778857f4e688cc657416ce524d54864ebe8995ba766efa7ca2b 35 0x47adb6aecc1949f2dc9f01206cc23eb4a0c29585d475dd24dc463c5087809298 36 0x30d39e8b0c451a8fcf3d2abab4b86ffa374265abbe77c5903db4c1be8cec7672 37 0x28cf47b39112297f0daeaa621f8e777875adc26f35dec0ba475c2ee148562b41 38 0x36199723cc59867e2e309fe9941cd33722c807bb2d0a06eeb41de93f1b93f2f5 39 0x5cdeb1f2ee1c7d694bdd884cb1c5c22de206684e1cafb8d3adb9a33cb85e19a2 40 0x0f6e6b3fc54c2d25871011b1499bb0ef015c6d0da802ae7eccf1d8c3fb73856c 41 0x0c1422c98b672414344a9c05492b926f473f05033b9f85b8788b4bb9a080053c 42 0x19a8527de35d4faacb00184e0423962247319703a815eecf355f143c2c18f17f 43 0x7812dc3313e6cf093da4617f06062e8e8969d648dfe6b5c331bccd58eb428383 44 0x61e537180c84c79e1fd2d4f9d386e1c4f0442247605b8d8904d122ee7ef9f7be 45 0x544d8621d05540576cfc9b58a3dab19145332b88eb0b86f4c15567c37205adf9 46 0x11be3ef96e6e07556356b51e2479436d9966b7b083892b390caec22a117aa48e 47 0x205cda31289cf75ab0759c14c43cb30f7287969ea3dc0d5286a3853a4d403187 48 0x048d8fc6934f4f0a99f0f2cc59010389e2a0b20d6909bfcf8d7d0249f360acdc 49 0x42cecc6d9bdca6d382e97fcea46a79c3eda2853091a8f399a2252115bf9a1454 50 0x0117d41b24f2f69cb3270b359c181607931f62c56d070bbd14dc9e3f9ab1432e 51 0x7c51564c66f68e2ad4ce6ea0d68f920fafa375376709c606c88a0ed44207aa1e 52 0x48f25191fc8ac7d9f21adf6df23b76ccbca9cb02b815acdbebfa3f4eddc71b34 53 0x4fc21a62c4688de70e28ad3d5956633fc9833bc7be09dc7bc500b7fae1e1c9a8 54 0x1f23f25be0912173c3ef98e1c9990205a69d0bf2303d201d27a5499247f06789 55 0x3131495618a0ac4cb11a702f3f8bab66c4fa1066d0a741af3c92d5c246edd579 56 0x0d93fe40faa53913638e497328a1b47603cb062c7afc9e96278603f29fd11fd4 57 0x6b348bc59e984c91d696d1e3c3cfae44021f06f74798c787c355437fb696093d 58 0x65af00e73043edcb479620c8b48098b89809d577a4071c8e33e8678829138b8a 59 0x5e62ffb032b2ddb06591f86a46a18effd5d6ecf3f129bb2bacfd51a3739a98b6 60 0x62c974ef3593fc86f7d78883b8727a2f7359a282cbc0196948e7a793e60ce1a1 61 0x204d708e3f500aad64283f753e7d9bab976aa42a4ca1ce5e9d2264639e8b1110 62 0x0a90f0059da81a012e9d0a756809fab2ce61cb45965d4d1513a06227783ee4ea 63 0x39fa55971c9e833f61139c39e243d40869fd7e8a1417ee4e7719dd2dd242766f 64 0x22677c1e659caa324f0c74a013921facf62d0d78f273563145cc1ddccfcc4421 65 0x3468cf6df7e93f7ff1fe1dd7e180a89dec3ed4f72843b4ea8a8d780011a245b2 66 0x68f75a0e2210f52a90704ed5f511918d1f6bcfcd26b462cc4975252369db6e9d 67 0x6220c0699696e9bcab0fe3a80d437519bd2bdf3caef665e106b2dd47585ddd9f 68 0x553ad47b129fb347992b576479b0a89f8d71f1196f83e5eaab5f533a1dd6f6d7 69 0x239aef387e116ec8730fa15af053485ca707650d9f8917a75f22acf6213197df G.4.2.3. Coefficients of~~w'(x) 0 0x6bd7f1fc5dd51b7d832848c180f019bcbdb101d4b3435230a79cc4f95c35e15e 1 0x17413bb3ee505184a504e14419b8d7c8517a0d268f65b0d7f5b0ba68d6166dd0 2 0x47f4471beed06e5e2b6d5569c20e30346bdba2921d9676603c58e55431572f90 3 0x2af7eaafd04f6910a5b01cdb0c27dca09487f1cd1116b38db34563e7b0b414eb 4 0x57f0a593459732eef11d2e2f7085bf9adf534879ba56f7afd17c4a40d3d3477b 5 0x4da04e912f145c8d1e5957e0a9e44cca83e74345b38583b70840bdfdbd0288ed 6 0x7cc9c3a51a3767d9d37c6652c349adc09bfe477d99f249a2a7bc803c1c5f39ed 7 0x425d7e58b8adf87eebf445b424ba308ee7880228921651995a7eab548180ad49 8 0x48156db5c99248234c09f43fedf509005943d3d5f5d7422621617467b06d314f 9 0x0d837dbbd1af32d04e2699cb026399c1928472aa1a7f0a1d3afd24bc9923456a 10 0x5b8806e0f924e67c1f207464a9d025758c078b43ddc0ea9afe9993641e5650be 11 0x29c91284e5d14939a6c9bc848908bd9df1f8346c259bbd40f3ed65182f3a2f39 12 0x25550b0f3bceef18a6bf4a46c45bf1b92f22a76d456bfdf19d07398c80b0f946 13 0x495d289b1db16229d7d4630cb65d52500256547401f121a9b09fb8e82cf01953 14 0x718c8c610ea7048a370eabfd9888c633ee31dd70f8bcc58361962bb08619963e 15 0x55d8a5ceef588ab52a07fa6047d6045550a5c52c91cc8b6b82eeb033c8ca557d 16 0x620b5a4974cc3395f96b2a0fa9e6454202ef2c00d82b0e6c534b3b1d20f9a572 17 0x4991b763929b00241a1a9a68e00e90c5df087f90b3352c0f4d8094a51429524e 18 0x18b6b49c5650fb82e36e25fd4eb6decfdd40b46c37425e6597c7444a1b6afb4e 19 0x6868305b4f40654460aad63af3cb9151ab67c775eaac5e5df90d3aea58dee141 20 0x16bc90219a36063a22889db810730a8b719c267d538cd28fa7c0d04f124c8580 21 0x3628f9cf1fbe3eb559854e3b1c06a4cd6a26906b4e2d2e70616a493bba2dc574 22 0x64abcc6759f1ce1ab57d41e17c2633f717064e35a7233a6682f8cf8e9538afec 23 0x01 Appendix H. Point Compression Point~~w'(x) 0 0x6bd7f1fc5dd51b7d832848c180f019bcbdb101d4b3435230a79cc4f95c35e15e 1 0x17413bb3ee505184a504e14419b8d7c8517a0d268f65b0d7f5b0ba68d6166dd0 2 0x47f4471beed06e5e2b6d5569c20e30346bdba2921d9676603c58e55431572f90 3 0x2af7eaafd04f6910a5b01cdb0c27dca09487f1cd1116b38db34563e7b0b414eb 4 0x57f0a593459732eef11d2e2f7085bf9adf534879ba56f7afd17c4a40d3d3477b 5 0x4da04e912f145c8d1e5957e0a9e44cca83e74345b38583b70840bdfdbd0288ed 6 0x7cc9c3a51a3767d9d37c6652c349adc09bfe477d99f249a2a7bc803c1c5f39ed 7 0x425d7e58b8adf87eebf445b424ba308ee7880228921651995a7eab548180ad49 8 0x48156db5c99248234c09f43fedf509005943d3d5f5d7422621617467b06d314f 9 0x0d837dbbd1af32d04e2699cb026399c1928472aa1a7f0a1d3afd24bc9923456a 10 0x5b8806e0f924e67c1f207464a9d025758c078b43ddc0ea9afe9993641e5650be 11 0x29c91284e5d14939a6c9bc848908bd9df1f8346c259bbd40f3ed65182f3a2f39 12 0x25550b0f3bceef18a6bf4a46c45bf1b92f22a76d456bfdf19d07398c80b0f946 13 0x495d289b1db16229d7d4630cb65d52500256547401f121a9b09fb8e82cf01953 14 0x718c8c610ea7048a370eabfd9888c633ee31dd70f8bcc58361962bb08619963e 15 0x55d8a5ceef588ab52a07fa6047d6045550a5c52c91cc8b6b82eeb033c8ca557d 16 0x620b5a4974cc3395f96b2a0fa9e6454202ef2c00d82b0e6c534b3b1d20f9a572 17 0x4991b763929b00241a1a9a68e00e90c5df087f90b3352c0f4d8094a51429524e 18 0x18b6b49c5650fb82e36e25fd4eb6decfdd40b46c37425e6597c7444a1b6afb4e 19 0x6868305b4f40654460aad63af3cb9151ab67c775eaac5e5df90d3aea58dee141 20 0x16bc90219a36063a22889db810730a8b719c267d538cd28fa7c0d04f124c8580 21 0x3628f9cf1fbe3eb559854e3b1c06a4cd6a26906b4e2d2e70616a493bba2dc574 22 0x64abcc6759f1ce1ab57d41e17c2633f717064e35a7233a6682f8cf8e9538afec 23 0x01 Appendix H. Point Compression Point compression allows a shorter representation of affine points of an elliptic curve by exploiting algebraic relationships between the coordinate values based on the defining equation of the curve in question. Point decompression refers to the reverse process, where one tries and recover an affine point from its compressed representation and information on the domain parameters of the curve. Consequently, point compression followed by point decompression is the identity map. The description below makes use of an auxiliary function (the parity function), which we first define for prime fields GF(p), with p odd, and then extend to all fields GF(q), where q is an odd prime power. We assume each finite field to be unambiguously defined and known from context. Let y be a nonzero element of GF(q). If q:=p is an odd prime number, y and p-y can be uniquely represented as integers in the interval [1,p-1] and have odd sum p. Consequently, one can distinguish y from -y via the parity of this representation, i.e., via par(y):=y (mod 2). If q:=p^m, where p is an odd prime number and where m>0, both y and -y can be uniquely represented as vectors of length m, with coefficients in GF(p) (see Appendix B.2). In this case, the leftmost nonzero coordinate values of y and -y are in the same position and have representations in [1,p-1] with different parity. As a result, one can distinguish y from -y via the parity of the representation of this coordinate value. This extends the definition of the parity function to any odd-size field GF(q), where one defines par(0):=0. The value of the parity function is commonly called the parity bit. H.1. Point Compression for Weierstrass Curves If P:=(X, Y) is an affine point of the Weierstrass curve W_{a,b} defined over the field GF(q), then so is -P:=(X, -Y). Since the defining equation Y^2=X^2+a*X+b has at most two solutions with fixed X-value, one can represent P by its X-coordinate and one bit of information that allows one to distinguish P from -P, i.e., one can represent P as the ordered pair compr(P):=(X, par(Y)). If P is a point of order two, one can uniquely represent P by its X-coordinate alone, since Y=0 and has fixed parity. Conversely, given the ordered pair (X, t), where X is an element of GF(q) and where t=0 or t=1, and the domain parameters of the curve W_{a,b}, one can use the defining equation of the curve to try and determine candidate values for the Y-coordinate given X, by solving the quadratic equation Y^2:=alpha, where alpha:=X^3+a*X+b. If alpha is not a square in GF(q), this equation does not have a solution in GF(q) and the ordered pair (X, t) does not correspond to a point of this curve. Otherwise, there are two solutions, viz. Y=sqrt(alpha) and -Y. If alpha is a nonzero element of GF(q), one can uniquely recover the Y-coordinate for which par(Y):=t and, thereby, the point P:=(X, Y). This is also the case if alpha=0 and t=0, in which case Y=0 and the point P has order two. However, if alpha=0 and t=1, the ordered pair (X, t) does not correspond to the outcome of the pointcompression~~allows a shorter representation~~function. We extend the definitionof~~affine~~the point compression function to allpoints of~~an elliptic~~thecurveW_{a,b},by~~exploiting algebraic relationships between~~associatingthe~~coordinate~~(non-affine) point at infinity O with any ordered pair compr(O):=(X,0), where X is any element of GF(q) for which alpha:=X^3+a*X+b is not a square in GF(q), and recover this point accordingly. In this case, the point at infinity O can be represented by any ordered pair (X,0) of elements of GF(q) for which X^3+a*X+b is not a square in GF(q). Note that this ordered pair does not satisfy the defining equation of the curve in question. An application may fix a specific suitable value of X or choose multiple suchvalues~~based on~~and use this to encode additonal information. Further details are out of scope. H.2. Point Compression for Montgomery Curves If P:=(u, v) is an affine point of the Montgomery curve M_{A,B} defined over the field GF(q), then so is -P:=(u, -v). Sincethe defining equationB*v^2=u^3+A*u^2+u has at most two solutions with fixed u-value, one can represent P by its u-coordinate and one bitof~~the curve in question. Point decompression refers~~information that allows onetodistinguish P from -P, i.e., one can represent P asthe~~reverse process, where~~ordered pair compr(P):=(u, par(v)). If P is a point of order two,one~~tries~~can uniquely represent P by its u-coordinate alone, since v=0and~~recover~~has fixed parity. Conversely, given the ordered pair (u, t), where u isan~~affine point from its compressed representation~~element of GF(q) and where t=0 or t=1,and~~information on~~the domain parameters of the~~curve. Consequently, point compression followed by point decompression is the identity map. The description below makes~~curve M_{A,B}, one canusethe defining equationof~~an auxiliary function (the parity function), which we first define for prime fields GF(p), with p odd, and then extend~~the curveto~~all fields GF(q),~~try and determine candidate values for the v-coordinate given u, by solving the quadratic equation v^2:=alpha,where~~q~~alpha:=(u^3+A*u^2+u)/B. If alphais~~an odd prime power. We assume each finite field to be unambiguously defined~~not a square in GF(q), this equation does not have a solution in GF(q)and~~known from context. Let y be~~the ordered pair (u, t) does not correspond toa~~nonzero element~~pointof~~GF(q).~~this curve. Otherwise, there are two solutions, viz. v=sqrt(alpha) and -v.If~~q:=p~~alphais~~an odd prime number, y and p-y can be uniquely represented as integers in the interval [1,p-1] and have odd sum p. Consequently,~~a nonzero element of GF(q),one can~~distinguish y from -y via~~uniquely recoverthe~~parity of this representation, i.e., via par(y):=y (mod 2). If q:=p^m, where p~~v-coordinate for which par(v):=t and, thereby, the affine point P:=(u, v). Thisis~~an odd prime number~~also the case if alpha=0and~~where m>0, both y~~t=0, in which case v=0and~~-y can be uniquely represented as vectors~~the point P has order two. However, if alpha=0 and t=1, the ordered pair (u, t) does not correspond to the outcomeof~~length m, with coefficients in GF(p) (see Appendix B.2). In this case,~~the~~leftmost nonzero coordinate values~~point compression function. We extend the definitionof~~y~~the point compression function to all points of the curve M_{A,B}, by associating the (non-affine) point at infinity O with the ordered pair compr(O):=(0,1)and~~-y are in~~recover this point accordingly. (Note that this corresponds tothe~~same position~~case alpha=0and~~have representations in [1,p-1] with different parity. As a result, one~~t=1 above.) The point at infinity Ocan~~distinguish y from -y via~~be represented bythe~~parity~~ordered pair (0, 1)of~~the representation~~elementsofGF(q). Note thatthis~~coordinate value. This extends~~ordered pair does not satisfythe~~definition~~defining equationof the~~parity function to any odd-size field GF(q), where one defines par(0):=0. H.1.~~curve in question. H.3.Point Compression for~~Weierstrass~~Twisted EdwardsCurves If~~P:=(X, Y)~~P:=(x, y)is an affine point of the~~Weierstrass~~twisted Edwardscurve~~W_{a,b}~~E_{a,d}defined over the field GF(q), then so is~~-P:=(X, -Y).~~-P:=(-x, y).Since the defining equation~~Y^2=X^2+a*X+b~~a*x^2+y^2=1+d*x^2*y^2has at most two solutions with fixed~~X-value,~~y-value,one can represent P by its~~X-coordinate~~y-coordinateand one bit of information that allows one to distinguish P from -P, i.e., one can represent P as the ordered pair~~compr(P):=(X, par(Y)).~~compr(P):=(par(x), y).If P is a point of orderone ortwo, one can uniquely represent P by its~~X-coordinate~~y-coordinatealone, since~~Y=0~~x=0and has fixed parity. Conversely, given the ordered pair~~(X, t),~~(t, y),where~~X~~yis an element of GF(q) and where t=0 or t=1, and the domain parameters of the curve~~W_{a,b},~~E_{a,d},one can use the defining equation of the curve to try and determine candidate values for the~~Y-coordinate~~x-coordinategiven~~X,~~y,by solving the quadratic equation~~Y^2:=alpha,~~x^2:=alpha,where~~alpha:=X^3+a*X+b.~~alpha:=(1-y^2)/(a-d*y^2). (Here, observe that the denominator is nonzero for any point of E_{a,d}.)If alpha is not a square in GF(q), this equation does not have a solution in GF(q) and the ordered pair~~(X, t)~~(t, y)does not correspond to a point of this curve. Otherwise, there are two solutions, viz.~~Y=sqrt(alpha)~~x=sqrt(alpha)and~~-Y.~~-x.If alpha is a nonzero element of GF(q), one can uniquely recover the~~Y-coordinate~~x-coordinatefor which~~par(Y):=t~~par(x):=tand, thereby, theaffinepoint~~P:=(X, Y).~~P:=(x, y).This is also the case if alpha=0 and t=0, in which case~~Y=0~~x=0and the point P has orderone ortwo. However, if alpha=0 and t=1, the ordered pair~~(X, t)~~(t, y)does not correspond to the outcome of the~~point compression function. We extend~~point compression function. Note that the point compression function is defined for all points of the twisted Edwards curve E_{a,d}. Here, the identity element O:=(0,1) is associated with the compressed point compr(O):=(0,1). (Note that this corresponds to the case alpha=0 and t=0 above.) We extend the definition of the compression function further, to also include a special marker element 'btm', by associating this marker element with the ordered pair compr(btm):=(1,1) and recover this marker element accordingly. (Note that this corresponds to the case alpha=0 and t=1 above.) The marker element 'btm' can be represented by the ordered pair (1,1) of elements of GF(q). Note that this ordered pair does not satisfy the defining equation of the curve in question. Appendix I. Data Conversions This section introduces various data conversion routines that are useful when representing integers, finite field elements, and curve points as binary or octet strings. I.1. Strings and String Operations The string over some alphabet S consisting of the symbols x_{l-1}, x_{l-2}, ..., x_1, x_0 (each in S), in this order, is denoted by str(x_{l-1}, x_{l-2}, ..., x_1, x_0). The length of this string (over S) is the number of symbols it contains (here: l). The empty string is the (unique) string of length l=0. The right-concatenation of two strings X and Y (defined over the same alphabet) isthe~~definition~~string Z consistingof the~~point compression function to all points~~symbolsofX (inthe~~curve W_{a,b},~~same order) followedby~~associating~~the~~(non-affine) point at infinity O with any ordered pair compr(O):=(X,0), where X is any element~~symbolsof~~GF(q) for which alpha:=X^3+a*X+b is not a square in GF(q), and recover this point accordingly. In this case,~~Y (inthe~~point at infinity O can be represented by any ordered pair (X,0) of elements~~same order). The lengthof~~GF(q) for which X^3+a*X+b~~the resulting string Zis~~not a square in GF(q). Note that this ordered pair does not satisfy~~the~~defining equation~~sumof the~~curve in question. An application may fix a specific suitable value~~lengthsof X~~or choose multiple such values~~and~~use this to encode additonal information. Further details are out of scope. H.2. Point Compression for Montgomery Curves If P:=(u, v) is an affine point of the Montgomery curve M_{A,B} defined over the field GF(q), then so~~Y. This string operationis~~-P:=(u, -v). Since the defining equation B*v^2=u^3+A*u^2+u has at most two solutions with fixed u-value, one can represent P~~denotedby~~its u-coordinate and one bit of information that allows one to distinguish P from -P, i.e., one can represent P as the ordered pair compr(P):=(u, par(v)). If P~~Z:=X||Y. The string Xiscalleda~~point~~prefixof~~order two, one can uniquely represent P by its u-coordinate alone, since v=0 and has fixed parity. Conversely, given~~Z;the~~ordered pair (u, t), where u~~string Y a postfix. The t-prefix of a string Z of length lis~~an element~~its unique prefix Xof~~GF(q) and where t=0 or t=1, and~~length t;the~~domain parameters~~t-postfix its unique postfix Yoflength t (where in both cases t is an integer inthe~~curve M_{A,B}, one~~interval [0,l]). Onecan~~use~~define these notions as well if t is outsidethe~~defining equation of~~interval [0,l] by stipulating that a t-prefix or t-postfix isthe~~curve to try~~empty string if t is negativeand~~determine candidate values for the v-coordinate given u, by solving~~that it isthe~~quadratic equation v^2:=alpha, where alpha:=(u^3+A*u^2+u)/B. If alpha~~entire string Z if tis~~not~~larger than l. Sometimes,a~~square in GF(q), this equation does not have~~t-prefix ofa~~solution in GF(q) and the ordered pair (u, t) does not correspond to~~string Z is denoted by trunc-left(Z,t);a~~point~~t-postfix by trunc- right(Z,t). A string X is called a substringof~~this curve. Otherwise, there are two solutions, viz. v=sqrt(alpha) and -v. If alpha~~Z if itis a~~nonzero element~~prefixof~~GF(q), one can uniquely recover the v-coordinate for which par(v):=t and, thereby,~~some postfix of Z. The string resulting from prependingthe~~affine point P:=(u, v). This~~string Y with Xis~~also~~the~~case if alpha=0 and t=0,~~string X||Y. An octet (or byte) is an integerin~~which case v=0 and~~the~~point P has order two. However, if alpha=0 and t=1,~~interval [0,256). An octet string is a string, wherethe~~ordered pair (u, t) does not correspond to~~alphabet isthe~~outcome~~setofall octets. A binary string (or bit string, for short) is a string, wherethe~~point compression function. We extend~~alphabet isthe~~definition~~set {0,1}ofbinary digits. Note thatthe~~point compression function to all points~~length of a string is defined in termsof the~~curve M_{A,B}, by associating~~underlying alphabet, as arethe~~(non-affine) point at infinity O with~~operations inthe~~ordered pair compr(O):=(0,1)~~previous paragraph. I.2. Conversion between Bit Stringsand~~recover this point accordingly. (Note that this~~Integers (BS2I, I2BS) There is a 1-1 correspondence between bit strings of length l and integers in the interval [0, 2^l), where the bit string X:=str(x_{l-1}, x_{l-2}, ..., x_1, x_0)corresponds to the~~case alpha=0 and t=1 above.) The point at infinity O can be represented by~~integer x:=x_{l-1}*2^{l-1} + x_{l-2}*2^{l-2} + ... + x_1*2 + x_0*1. (If l=0,the~~ordered pair (0, 1) of elements of GF(q).~~empty bit string corresponds to the integer zero.)Note that~~this ordered pair does not satisfy the defining equation of~~whilethe~~curve in question. H.3. Point Compression for Twisted Edwards Curves If P:=(x, y)~~mapping from bit strings to integersis~~an affine point of the twisted Edwards curve E_{a,d} defined over~~uniquely defined,the~~field GF(q), then so~~inverse mapping from integers to bit stringsis~~-P:=(-x, y). Since the defining equation a*x^2+y^2=1+d*x^2*y^2 has at most two solutions with fixed y-value, one~~not, since any non-negative integer smaller than 2^tcan~~represent P by its y-coordinate and one~~be represented as abitstringof~~information that allows one~~length at least t (dueto~~distinguish P~~leading zero coefficients in base 2 representation). The latter representation is called tight if the bit string representation has minimal length. This defines the mapping BS2Ifrom~~-P, i.e., one can represent P as~~bit strings to integers andthe~~ordered pair compr(P):=(par(x), y). If P is a point~~mapping I2BS(x,l) from non-negative integers smaller than 2^l to bit stringsof~~order one or two, one can uniquely represent P by its y-coordinate alone, since x=0~~length l. I.3. Conversion between Octet Stringsand~~has fixed parity. Conversely, given the ordered pair (t, y), where y~~Integers (OS2I, I2OS) Thereis~~an element~~a 1-1 correspondence between octet stringsof~~GF(q) and where t=0 or t=1,~~length landintegers inthe~~domain parameters of~~interval [0, 256^l), wherethe~~curve E_{a,d}, one can use~~octet string X:=str(X_{l-1}, X_{l-2}, ..., X_1, X_0) corresponds tothe~~defining equation of~~integer x:=X_{l-1}*256^{l-1} + X^{l-2}*256^{l-2} + ... + X_1*256 + X_0*1. (If l=0,the~~curve~~empty string correspondsto~~try and determine candidate values for~~the~~x-coordinate given y, by solving the quadratic equation x^2:=alpha, where alpha:=(1-y^2)/(a-d*y^2). (Here, observe~~integer zero.) Notethatwhilethe~~denominator~~mapping from octet strings to integersis~~nonzero for~~uniquely defined, the inverse mapping from integers to octet strings is not, sinceany~~point~~non-negative integer smaller than 256^t can be represented as an octet stringof~~E_{a,d}.) If alpha is not a square in GF(q), this equation does not have a solution~~length at least t (due to leading zero coefficientsin~~GF(q)~~base 256 representation). The latter representation is called tight if the octet string representation has minimal length. This defines the mapping OS2I from octet strings to integersand the~~ordered pair (t, y) does not correspond~~mapping I2OS(x,l) from non-negative integers smaller than 256^lto~~a point~~octet stringsof~~this curve. Otherwise, there are two solutions, viz. x=sqrt(alpha)~~length l. I.4. Conversion between Octet Stringsand~~-x. If alpha~~Bit Strings (OS2BS, BS2OS) Thereis a~~nonzero element~~1-1 correspondence between octet stringsof~~GF(q), one can uniquely recover the x-coordinate for which par(x):=t and, thereby,~~length l and bit strings of length 8*l, wherethe~~affine point P:=(x, y). This is also~~octet string X:=str(X_{l-1}, X_{l-2}, ..., X_1, X_0) corresponds tothe~~case if alpha=0 and t=0, in which case x=0 and~~right-concatenation ofthe~~point P has order one or two. However, if alpha=0 and t=1,~~8-bit strings x_{l-1}, x_{l-2}, ..., x_1, x_0, where each octet X_i corresponds tothe~~ordered pair (t, y) does not correspond~~8-bit string x_i accordingto the~~outcome~~mappingof~~the point compression function.~~Appendix I.2 above.Note that the~~point compression function~~mapping from octet strings to bit stringsisuniquelydefined~~for all points~~and so is the inverse mapping from bit strings to octet strings, if one prepends each bit string with the smallest number of 0 bits so as to result in a bit stringoflength divisible by eight (i.e., one uses pre-padding). This definesthe~~twisted Edwards curve E_{a,d}. Here,~~mapping OS2BS from octet strings to bit strings andthe~~identity element O:=(0,1) is associated with~~corresponding mapping BS2OS from bit strings to octet strings. When we refer to a specific bit position in an octet string, this indicatesthe~~compressed point compr(O):=(0,1). (Note that~~corresponding position, whenthis~~corresponds to~~octet string is viewed as a bit string usingthe~~case alpha=0~~OS2BS mapping above. I.5. Conversion between Field Elementsand~~t=0 above.) We extend the definition~~Octet Strings (FE2OS, OS2FE) There is a 1-1 correspondence between elementsof the~~compression function further, to also include~~fixed finite field GF(q), where q:=p^m, where p isa~~special marker element 'btm', by associating this marker element with the ordered pair compr(btm):=(1,1)~~prime numberand~~recover this marker~~where m>0, and vectors of length m, with coefficients in GF(p), where eachelement~~accordingly. (Note that this corresponds~~x of GF(q) is a vector (x_{m-1}, x_{m-2}, ..., x_1, x_0) accordingto the~~case alpha=0 and t=1 above.) The marker~~conventions of Appendix B.2. In this case, this fieldelement~~'btm'~~can beuniquelyrepresented by the~~ordered pair (1,1)~~right-concatenationof~~elements~~the octet strings X_{m-1}, X_{m-2}, ..., X_1, X_0, where each octet string X_i corresponds to the integer x_i in the interval [0,p-1] according to the mappingof~~GF(q).~~Appendix I.3 above.Note that~~this ordered pair does not satisfy~~boththe~~defining equation of~~mapping from field elements to octet strings andthe~~curve in question. Appendix I. Data Conversions The~~inverse mapping from octet strings to field elements are only uniquely defined if each octetstring~~over some alphabet S consisting of~~X_i hasthe~~symbols x_{l-1}, x_{l-2}, ..., x_1, x_0 (each in S), in this order, is denoted by str(x_{l-1}, x_{l-2}, ..., x_1, x_0). The length of this string (over S) is~~same fixed size (e.g.,the~~number of symbols it contains (here: l). The empty string~~smallest integer l so that 256^l >= p) and if all integers are reduced modulo p. If so, the latter representation is called tight if lisminimal so that 256^l >= p. This definesthe~~(unique) string of length l=0. The right-concatenation of two~~mapping FE2OS(x,l) from field elements to octetstrings~~X~~and~~Y (defined over~~the~~same alphabet)~~mapping OS2FE(X,l) from octet strings to field elements, where the underlying fieldisimplicit and assumed to be known from context. In this case,theoctetstring~~Z consisting of the symbols of X (in~~has length l*m. (Observe that with tight representations,the~~same order) followed~~parameter l is uniquely definedby the~~symbols~~characteristic pof~~Y (in~~the~~same order).~~field GF(q) in question.)The~~length of the resulting string Z~~OS2FE(X,l) mappingiscalled strict if it operates asthe~~sum of~~OS2FE(X,l) function, except that it fails whenever it would require at least one modular reduction. Notice thatthe~~lengths of X and Y. This string operation is denoted~~tight FE2OS mapping followedby~~Z:=X||Y. The string X~~the strict OS2FE mappingis~~called a prefix of Z;~~the~~string Y a postfix. The t-prefix~~identity map (and, hence, OS2FE never fails in this case). I.6. Conversion between Elementsof~~a string~~Z~~of length l~~mod n and Octet Strings (ZnE2OS, OS2ZnE) Thereis~~its unique prefix X~~a 1-1 correspondence between elementsof~~length t;~~the~~t-postfix its unique postfix Y~~set Z_nof~~length t (where in both cases t is an integer~~integers modulo n and integersin the interval~~[0,l]). One can define these notions as well if t~~[0,n), where each element x of Z_nis~~outside~~uniquely represented bythe~~interval [0,l]~~integer x mod n. In this case, x mod n can be uniquely representedby~~stipulating~~the octet string X according to the mapping of Appendix I.3 above. Notethat~~a t-prefix or t-postfix is~~boththe~~empty string if t is negative~~mapping from elements of Z_n to octet stringsand~~that it is~~the~~entire string Z if t is larger than l. Sometimes, a t-prefix~~inverse mapping from octet strings to elementsof~~a~~Z_n are only uniquely defined if the octetstring~~Z is denoted by trunc-left(Z,t);~~hasa~~t-postfix by trunc- right(Z,t). A string X~~fixed size (e.g., the smallest integer l so that 256^l >= n) and if all integers are first reduced modulo n. If so, the latter representationis called~~a substring of Z~~tightif~~it~~lis~~a prefix~~minimal so that 256^l >= n. This defines the mapping ZnE2OS(x,l) from elementsof~~some postfix~~Z_n to octet strings and the mapping OS2ZnE(X,l) from octet strings to elementsof~~Z. The string resulting~~Z_n, where the underlying modulus n is implicit and assumed to be knownfrom~~prepending~~context. In this case,theoctetstring~~Y~~has length l. (Observe thatwith~~X is~~tight representations,the~~string X||Y. An octet~~parameter lis~~an integer in~~uniquely defined bythe~~interval [0,256). An octet string~~parameter n in question.) The OS2ZnE(X,l) mappingis~~a string, where~~called strict if it operates asthe~~alphabet is~~OS2ZnE(X,l) function, except that it fails whenever it would require at least one modular reduction. Notice thatthe~~set of all octets. A binary string (or bit string, for short) is a string, where~~tight ZnE2OS mapping followed bythe~~alphabet~~strict OS2ZnE mappingis the~~set {0,1}.~~identity map (and, hence, ZnE2OS never fails in this case).Note that~~the length of a string~~if nis~~defined in terms of the underlying alphabet,~~a prime number p, the conversions ZnE2OS and FE2OS are consistent,as are~~the operations in the previous paragraph. I.1. Conversion between Bit Strings~~OS2ZnEand~~Integers (BS2I, I2BS) There~~OS2FE. This is, however, no longer the case if nis a~~1-1 correspondence between bit strings~~strict prime power. The conversion rules for composite n values may be useful, e.g., when encoding RSA parameters (or elementsof~~length l~~any other non-prime size set Z_n, for that matter). I.7. Ordering Conventions One can consider various representation functions, depending on bit- orderingand~~integers in the interval [0, 2^l),~~octet-ordering conventions. The description below makes use of an auxiliary function (the reversion function),where the~~bit~~reverse of thestring X:=str(x_{l-1}, x_{l-2}, ..., x_1, x_0)~~corresponds to the integer x:=x_{l-1}*2^{l-1} + x_{l-2}*2^{l-2} + ... + x_1*2 + x_0*1. (If l=0, the empty bit string corresponds to the integer zero.) Note that while the mapping from bit strings to integers~~is~~uniquely defined, the inverse mapping from integers~~definedto~~bit strings is not, since any non-negative integer smaller than 2^t can~~be~~represented as a bit string of length at least t (due to leading zero coefficients in base 2 representation). The latter representation is called tight if~~the~~bit~~string~~representation has minimal length. This defines the mapping BS2I from bit strings to integers~~X':=rev(X):=str(x_0, x_1, ..., x_{l-2}, x_{l-1}). Below, we use this reversion function with binary and octet strings. We now describe representations in most-significant-bit first (msb) or least-significant-bit first (lsb) orderandthose in most- significant-byte first (MSB) or least-significant-byte first (LSB) order. One distinguishesthe~~mapping I2BS(x,l) from non-negative integers smaller than 2^l to bit strings~~following octet-string representationsof~~length l. I.2. Conversion between Octet Strings~~integersand~~Integers (OS2I, I2OS) There is a 1-1 correspondence between octet strings of length l~~field elements: 1. MSB, msb: represent field elementsand integers~~in the interval [0, 256^l), where~~as above, yieldingthe octet string~~X:=str(X_{l-1},~~str(X_{l-1},X_{l-2}, ..., X_1,~~X_0) corresponds to the integer x:=X_{l-1}*256^{l-1} + X^{l-2}*256^{l-2} + ... + X_1*256 + X_0*1. (If l=0, the empty string corresponds to the integer zero.) Note that while the mapping from octet strings to integers is uniquely defined,~~X_0). 2. MSB, lsb: reversethe~~inverse mapping from integers to octet strings is not, since any non-negative integer smaller than 256^t can be represented as an octet string~~bit-orderof~~length at least t (due to leading zero coefficients in base 256 representation). The latter representation is called tight if~~each octet, viewed as 8-bit string, yieldingthe octet string~~representation has minimal length. This defines the mapping OS2I from octet strings to integers and~~str((rev(X_{l-1}), rev(X_{l-2}), ..., rev(X_1), rev(X_0)). 3. LSB, lsb: reversethe~~mapping I2OS(x,l) from non-negative integers smaller than 256^l to octet strings of length l. I.3. Conversion between Octet Strings and Bit Strings (OS2BS, BS2OS) There is a 1-1 correspondence between~~octet~~strings of length l~~stringand~~bit strings~~bit-orderof~~length 8*l, where~~each octet, yieldingthe octet string~~X:=str(X_{l-1}, X_{l-2},~~str(rev(X_{0}), rev(X_{1}),...,~~X_1, X_0) corresponds to~~rev(X_{l-2}), rev(X_{l-1})). 4. LSB, msb: reversethe~~right-concatenation of~~octet string, yieldingthe~~8-bit strings x_{l-1}, x_{l-2}, ..., x_1, x_0, where each~~octet~~X_i corresponds to~~string str(X_{0}, X_{1}, ..., X_{l-2}, X_{l-1}). Thus,the~~8-bit~~2-octetstring~~x_i according to~~"07e3" representsthe~~mapping of Appendix I.1 above. Note that~~integer 2019 (=0x07e3) in MSB/msb order,the~~mapping from octet strings to bit strings is uniquely defined~~integer 57,543 (0xe0c7) in MSB/lsb order, the integer 51,168 (0xc7e0) in LSB/lsb order,and~~so is~~the~~inverse mapping from bit strings to octet strings, if one prepends each bit string~~integer 58,119 (=0xe307) in LSB/msb order. Note that,with the~~smallest number of 0 bits so~~above data conversions, there is still some ambiguityas to~~result in~~how to represent an integer or a field element asa bit string~~of length divisible by eight (i.e., one uses pre-padding). This defines the mapping OS2BS from~~oroctet~~strings to bit strings and the corresponding mapping BS2OS from bit strings~~string (dueto~~octet strings. I.4.~~leading zeros). However, tight representations (as defined above) are non-ambiguous. (Note, in particular, that tightness implies that elements of GF(q) are always uniquely represented.) I.8.Conversion~~between Field Elements~~Between Curve Pointsand Octet Strings~~(FE2OS, OS2FE) There~~For each of the curve models we consider, each affine pointis~~a 1-1 correspondence between~~an ordered pair (X, Y) whose coordinates areelements of~~the fixed~~afinite field~~GF(q), where q:=p^m, where p is a prime number and where m>0,~~GF(q)and~~vectors of length m, with coefficients~~that satisfy the defining equation for the curvein~~GF(p),~~question. Each compressed point is an ordered pair (X,t) (for Weierstrass curves and Montgomery curves) or (t, X) (for twisted Edwards curves)where~~each~~X is anelement~~x~~of GF(q)and where tis~~a vector (x_{m-1}, x_{m-2}, ..., x_1, x_0) according to the conventions~~an elementof{0,1} (seeAppendix~~B.2. In this case, this field element can be uniquely~~H). The affine point (X, Y) isrepresented by the~~right-concatenation of~~ordered pair whose coordinates arethe octet~~strings X_{m-1}, X_{m-2}, ..., X_1, X_0, where each octet~~string~~X_i corresponds to the integer x_i in~~representations ofthe~~interval [0,p-1] according to~~elements X and Y of GF(q), respectively, usingthetight FE2OSmapping of Appendix~~I.2 above.~~I.5.Note~~that both the mapping from field elements to~~that, since we use a tight representation, this results in a pair ofoctet strings~~and~~(each of length l*m), wherethe~~inverse mapping from octet strings to field elements~~parameters l and mare~~only~~uniquely defined~~if each octet string X_i has~~bythe~~same fixed size (e.g.,~~field GF(q) in question. The inverse mapping results by convertingthe~~smallest integer l so that 256^l >= p)~~firstand~~if all integers are reduced modulo p. If so, the latter representation is called tight if l is minimal so that 256^l >= p. This defines~~second coordinate of this pair (each an octet string of length l*m) to, respectively,the~~mapping FE2OS(x,l) from field~~elements~~to octet strings~~XandY of GF(q) viathestrict OS2FEmapping~~OS2FE(X,l) from octet strings to field elements, where the underlying field~~of Appendix I.5. Note that if itis~~implicit and assumed to be~~not a prioriknown~~from context. In this case,~~whetherthe~~octet string has length l*m. (Observe~~input to this inverse mapping actually represents an affine curve point, one should checkthat~~with tight representations, the parameter l~~thisis~~uniquely defined by the characteristic p~~indeed an ordered pairof~~the field GF(q) in question.) I.5. Conversion between Elements~~octet strings (eachof~~Z mod n~~length l*m)and~~Octet Strings (ZnE2OS, OS2ZnE) There is a 1-1 correspondence between elements~~that the output (X,Y) -- if defined -- is indeed an affine pointof the~~set Z_n of integers modulo n and integers~~curvein~~the interval [0,n),~~question,where~~each element x of Z_n~~this mapping fails if either condition is not satisfied. The compressed point (X,t) or (t, X)is~~uniquely~~represented by the~~integer x mod n. In this case, x mod n can be uniquely represented by~~ordered pair whose coordinates arethe octet stringrepresentations of the parity bit t in {0,1} and the elementX~~according to~~of GF(q), respectively, usingthetight FE2OSmapping of Appendix~~I.2 above.~~I.5.Note~~that both the mapping from elements~~that, since we use tight representations, this results in an ordered pairof~~Z_n to~~octet strings(of length 1andl*m, respectively), wherethe~~inverse mapping from octet strings to elements of Z_n~~parameters l and mare~~only~~uniquely defined~~if~~bythe~~octet string has a fixed size (e.g.,~~field GF(q) in question. The inverse mapping results by convertingthe~~smallest integer l so that 256^l >= n) and if all integers are~~first~~reduced modulo n. If so, the latter representation is called tight if l is minimal so that 256^l >= n. This defines the mapping ZnE2OS(x,l) from elements~~and second coordinateof~~Z_n to~~this pair (each anoctet~~strings~~string, of length 1 and l*m, respectively) to, respectively, the element t of {0,1}and the~~mapping OS2ZnE(X,l) from octet strings to elements~~element Xof~~Z_n, where~~GF(q) viathe~~underlying modulus n is implicit~~strict OS2FE mapping of Appendix I.5,and~~assumed to be known from context. In~~representingthis~~case, the octet string has length l. (Observe that with tight representations,~~asthe~~parameter l is uniquely defined by~~compressed point (X, t) or (t, X) according tothe~~parameter n~~curve modelin~~question.)~~question.Note that if~~n~~itisnota~~prime number p,~~priori known whetherthe~~conversions ZnE2OS~~input to this inverse mapping actually represents a compressed curve point, one should check that this is indeed an ordered pair of octet strings (of length 1and~~FE2OS are consistent, as are OS2ZnE~~l*m, respectively)and~~OS2FE. This is, however, no longer~~thatthe~~case~~output (X, t) or (t, X) --if~~n~~defined --isindeeda~~strict prime power. The conversion rules for composite n values may be useful, e.g., when encoding RSA parameters (or elements~~compressed pointof~~any other non-prime size set Z_n,~~the curve in question, using the point decompression processfor~~that matter). I.6. Ordering Conventions One can consider various representation functions, depending on bit- ordering and octet-ordering conventions.~~this curve (see Appendix H), where this mapping fails if either condition is not satisfied. NOTE 1:The~~description below makes use~~representationsof~~an auxiliary function (the reversion function), which we define both for bit strings~~affineandcompressed points above are as ordered pairs ofoctet strings.~~For a bit string [octet string] X:=str(x_{l-1}, x_{l-2}, ..., x_1, x_0),~~In practice, one often represents these as octet strings instead, via right-concatenation ofits~~reverse~~coordinates (in left-to-right order). Since each coordinate has known length, this operationisreversible. When appropriate, we refer tothelatter as the octet (rather than the pair) representation of a point. NOTE 2: The octet representation of compressed points above identifies the paritybit~~string [octet string] X':=rev(X):=str(x_0, x_1, ..., x_{l-2}, x_{l-1}). We now describe representations in most-significant-bit first (msb) or least-significant-bit first (lsb) order and those in most- significant-byte first (MSB) or least-significant-byte first (LSB) order. One distinguishes~~t ofthe~~following octet-string~~curve point in question via the 1-octetrepresentations oftheintegers0and~~field elements:~~1.~~MSB, msb: represent field elements and integers as above, yielding the octet string str(X_{l-1}, X_{l-2}, ..., X_1, X_0). 2. MSB, lsb: reverse~~Obviously, other 1-1 mappings are also possible. As an example, with [SEC1],the~~bit-order of each octet, viewed~~parity bit t is represented by 0x02 or 0x03 depending on whether t=0 or t=1, respectively The same [SEC1] specification represents affine pointsas~~8-bit string, yielding the~~above (asoctet~~string str((rev(X_{l-1}), rev(X_{l-2}), ..., rev(X_1), rev(X_0)). 3. LSB, lsb: reverse~~string), but prepends this withthe~~octet string~~1-octet prefix 0x04,and~~bit-order~~represents the identity elementof~~each octet, yielding~~the~~octet~~curve as the 1-octetstring~~str(rev(X_{0}), rev(X_{1}), ..., rev(X_{l-2}), rev(X_{l-1})). 4. LSB, msb: reverse~~0x00. This variable-size point representation hasthe~~octet string, yielding~~property that its 1-octet prefix identifies whether it encodes an affine curve point, a compressed point (including parity bit), orthe~~octet string str(X_{0}, X_{1}, ..., X_{l-2}, X_{l-1}). Thus,~~identity element, whilethe~~2-octet string "07e3" represents~~remainder of this representation uniquely determinesthe~~integer 2019 (=0x07e3) in MSB/msb order,~~curve point's value. Whilethe~~integer 57,543 (0xe0c7)~~descriptionin~~MSB/lsb order,~~[SEC1] only applies to Weierstrass curves,the~~integer 51,168 (0xc7e0) in LSB/lsb order,~~description above applies to each of the curve models we consider (i.e., these apply to Montgomery curvesandtwisted Edwards curves as well). Collectively, we simply refer to this as the "SEC1" point representation. Note that elements of a prime field GF(p), where p is a 255-bit prime number, have a tight representation as a 32-byte string, where a fixed bit position is always set to zero. (This is the leftmost bit position of this octet string if one followsthe~~integer 58,119 (=0xe307) in LSB/msb order. Note that, with~~MSB/msb representation conventions.) This allowsthe~~above data conversions, there is still some ambiguity as to how~~parity bit of a compressed point (see Appendix H)to~~represent an integer or~~be encoded in this bit position and, thereby, allowsa~~field~~compressed point and anelement~~as a bit string or~~of GF(p) to be represented by anoctet string~~(due to leading zeros). However, tight representations (as defined above) are non-ambiguous. (Note,~~of the same length. This is called the "squeezed" point representation. (We will use this squeezed representationin~~particular, that tightness implies that~~Appendix J.) Obviously, other representations (e.g., those ofelements of~~GF(q)~~Z_n) may also have fixed bit values in certain positions, which can be used to squeeze-in additional information. Further detailsare~~always uniquely represented.) Note~~out of scope. Noticethat elements of a prime field GF(p), where p is a~~255-bit~~prime~~number,~~number with bit-size m divisible by eight, have a tight representation as an (m/8)-byte string, but do nothave abit position that is always set to zero. Thus, in this case, one cannot represent a compressed point as an octet string of the same length as an element of GF(p). However, one can still encode this as an octet string of length (m/8)+1 (see Note 1 above). If one uses right-concatenation as in Note 1, but (for historial reasons) represents the parity bit t of the compressed point in question by 0x00 or 0x80 depending on whether t=0 or t=1, respectively, this is again called the "squeezed' representation (despite this being somewhat a misnomer, since each point is now represented as an octet string that is one octet longer than thetight representationof elements of GF(p)). Notice that this representation corresponds to the compressed point representation of Appendix I.8as~~a 32-byte~~octetstring,~~where a fixed~~but with the bit-ordering in the 1-octet representation of t reversed. (Note that this puts the paritybit~~position is always set to zero. (This is~~t inthe leftmost bit position of~~this~~theoctet string if one follows the MSB/msb representation conventions.)~~This allows the parity bit of a compressed point (see Appendix H) to be encoded in~~We will usethis~~bit position and, thereby, allows a compressed point and a field element of GF(p) to be represented by an octet string of the same length. This is called the~~squeezed~~point representation. Obviously, other representations (e.g., those of elements of Z_n) may also have fixed bit values~~represenationin~~certain positions, which can be used to squeeze-in additional information. Further details are out of scope.~~Appendix O.Appendix J. Representation Examples Curve25519 Family Members We present some examples of computations using the curves introduced inAppendix E and Appendix G ofthis document. In each case, we indicate the values of P, k*P, and (k+1)*P, where P is a fixed multiple (here: 2019) of~~the base point~~the base point of the curve in question and where the private key k is the integer k 45467544759954639344191351164156560595299236761702065033670739677 691372543056 (=0x6485b7e6 cd83e5c2 0d5dbfe4 f915494d 9cf5c65d 778c32c3 c08d5abd 15e29c50). In the examples below, each curve point is represented using the "squeezed" point representation (see Appendix I.8), whereby each point is represented as a 32-octet string, where the ordering convention (see Appendix I.7) depends on the underlying curve model in question. Here, points of a Weierstrass curve are represented in tight MSB/msb-order, points of a Montgomery curve in tight LSB/msb- order, and points of a twisted Edwards curve in tight LSB/lsb-order. For points that are a public key, the corresponding private keys are represented as 32-octet strings, using the same (tight) ordering conventions as with the public keys. For affine points, we also give the tight representation of eachofits coordinates, usingthe~~curve in question and where~~same ordering conventions as used withthe~~private key k is~~squeezed point representation. For further details, seethe~~integer k 45467544759954639344191351164156560595299236761702065033670739677 691372543056 (=0x6485b7e6 cd83e5c2 0d5dbfe4 f915494d 9cf5c65d 778c32c3 c08d5abd 15e29c50).~~examples themselves.J.1. Example with Curve25519 Pm=(u, v), k*Pm=(u1, v1), and (k+1)*Pm=(u2, v2) with Curve25519: u 53025657538808013645618620393754461319535915376830819974982289332 088255623750 (=0x753b7566 df35d574 4734142c 9abf931c ea290160 aa75853c 7f972467 b7f13246). v 53327798092436462013048370302019946300826511459161905709144645521 233690313086 (=0x75e676ce deee3b3c 12942357 22f1d884 ac06de07 330fb07b ae35ca26 df75417e). u1 42039618818474335439333192910143029294450651736166602435248528442 691717668056 (=0x5cf194be f0bdd6d6 be58e18a 8f16740a ec25f4b0 67f7980a 23bb6468 88bb9cd8). v1 76981661982917351630937517222412729130882368858134322156485762195 67913357634 (=0x110501f6 1dff511e d6c4e9b9 bfd5acbe 8bf043b8 c3e381dd f5771306 479ad142). u2 34175116482377882355440137752573651838273760818624557524643126101 82464621878 (=0x078e3e38 41c3e0d0 373e5454 ecffae33 2798b10a 55c72117 62629f97 f1394d36). v2 43046985853631671610553834968785204191967171967937842531656254539 962663994648 (=0x5f2bbb06 f7ec5953 2c2a1a62 21124585 1d2682e0 cc37307e fbc17f7f 7fda8518). As suggested in Appendix C.2, the v-coordinate of k*Pm can be indirectly computed from the u-coordinates of Pm, k*Pm, and (k+1)*Pm, and the v-coordinate of Pm, which allows computation of the entire point k*Pm (and not just its u-coordinate) if k*Pm is computed using the Montgomery ladder (as, e.g., [RFC7748] recommends), since that algorithm computes both u1 and u2 and the v-coordinate of the point Pm may be available from context. The representation of k and the compressed representations of Pm and k*Pm in tight LSB/msb-order are given by repr(k) 0x509ce215 bd5a8dc0 c3328c77 5dc6f59c 4d4915f9 e4bf5d0d c2e583cd e6b78564 repr(Pm) 0x4632f1b7 6724977f 3c8575aa 600129ea 1c93bf9a 2c143447 74d535df 66753b75; repr(k*Pm) 0xd89cbb88 6864bb23 0a98f767 b0f425ec 0a74168f 8ae158be d6d6bdf0 be94f15c, where the leftmost bit of the rightmost octet indicates the parity of the v-coordinate of the point of Curve25519 in question (which, in this case, are both zero, since v and v1 are even). See Appendix H.2 and Appendix I for further detail on (squeezed) point compression. The scalar representation and (squeezed) point representation illustrated above are consistent with the representations specified in [RFC7748], except that in [RFC7748] only an affine point's u-coordinate is represented (i.e., the v-coordinate of any point is always implicitly assumed to have an even value) and that the representation of the point at infinity is not specified. Another difference is that [RFC7748] allows non-unique representations of some elements of GF(p), whereas our representation conventions do not (since tight). A randomized representation (t1, t2) of the point k*Pm in tight LSB/ msb order is given by t1 409531317901122685707535715924445398426503483189854716584 37762538294289253464 (=0x5844b232 8c4586dc 62f593c5 599c2a8c e61ba893 bb052de6 77510a42 b3a68a5a) t2 451856098332889407421278004628150814449259902023388533929 08848927625430980881 (=0x11598452 e65138dc ce948d7e d8f46a18 b640722c 8e170957 751b7729 1b26e663), where this representation is defined in Appendix K.5 and uses the mapping of Appendix K.3.2 with the default square root function. J.2. Example with Edwards25519 Pe=(x, y), k*Pe=(x1, y1), and (k+1)*Pe=(x2, y2) with Edwards25519: x 25301662348702136092602268236183361085863932475593120475382959053 365387223252 (=0x37f03bc0 1070ed12 d3218f8b ba1abb74 fd6b94eb 62033d09 83851e21 d6a460d4). y 54434749145175762798550436656748568411099702168121592090608501578 942019473360 (=0x7858f9e7 6774ed8e 23d614d2 36715fc7 56813b02 9aa13c18 960705c5 b3a30fd0). x1 42966967796585460733861724865699548279978730460766025087444502812 416557284873 (=0x5efe7124 465b5bdb b364bb3e e4f106e2 18d59b36 48f4fe83 c11afc91 785d7e09). y1 46006463385134057167371782068441558951541960707376246310705917936 352255317084 (=0x65b6bc49 985badaf bc5fdd96 fb189502 35d5effd 540b439d 60508827 80bc945c). x2 42629294840915692510487991904657367226900127896202625319538173473 104931719808 (=0x5e3f536a 3be2364a 1fa775a3 5f8f65ae 93f4a89d 81a04a2e 87783748 00120a80). y2 29739282897206659585364020239089516293417836047563355347155817358 737209129078 (=0x41bfd66e 64bdd801 c581a720 f48172a8 187445fa 350924a2 c92c791e 38d57876). The representation of k and the compressed representations of Pe and k*Pe in tight LSB/lsb-order are given by repr(k) =0x0a3947a8 bd5ab103 c34c31ee ba63af39 b292a89f 27fdbab0 43a7c1b3 67eda126; repr(Pe) =0x0bf0c5cd a3a0e069 183c8559 40dc816a e3fa8e6c 4b286bc4 71b72ee6 e79f1a1e; repr(k*Pe) =0x3a293d01 e4110a06 b9c2d02a bff7abac 40a918df 69bbfa3d f5b5da19 923d6da7, where the rightmost bit of the rightmost octet indicates the parity of the x-coordinate of the point of Edwards25519 in question (which, in this case, are zero and one, respectively, since x is even and x1 is odd). See Appendix H.3 and Appendix I for further detail on (squeezed) point compression. The scalar representation and (squeezed) point representation illustrated above are fully consistent with the representations specified in [RFC8032]. Note that, contrary to [RFC7748], [RFC8032] requires unique representations of all elements of GF(p). A randomized representation (t1, t2) of the point k*Pe in tight LSB/ lsb order is given by t1 577913017083163641949634219017190182170288776648725395935 97750427519399254040 (=0x181a32c5 10e06dbc ea321882 f3519055 535e289e 8faac654 82e26f61 aded23fe) t2 454881407940919718426608573125377401686255068210624245884 05479716220480287974 (=0x672e36c5 ae353073 cdfac343 e8297b05 1b010d0f 5b1016db dd4baf54 28068926), where this representation is defined in Appendix K.5 and uses the mapping of Appendix K.3.3 with the default square root function and underlying isomorphic mapping between Edwards25519 and Curve25519 of Appendix E.2. J.3. Example with Wei25519 Pw=(X, Y), k*Pw=(X1, Y1), and (k+1)*Pw=(X2, Y2) with Wei25519: X 14428294459702615171094958724191825368445920488283965295163094662 783879239338 (=0x1fe62011 89e0801e f1debed7 456a3dc7 94d3ac0b 55202fe7 2a41cf12 629e56aa). Y 53327798092436462013048370302019946300826511459161905709144645521 233690313086 (=0x75e676ce deee3b3c 12942357 22f1d884 ac06de07 330fb07b ae35ca26 df75417e). X1 34422557393689369648095312405803933433606568476197477554293337733 87341283644 (=0x079c3f69 9b688181 69038c35 39c11eb5 96d09f5b 12a242b4 ce660f13 3368c13c). Y1 76981661982917351630937517222412729130882368858134322156485762195 67913357634 (=0x110501f6 1dff511e d6c4e9b9 bfd5acbe 8bf043b8 c3e381dd f5771306 479ad142). X2 22716193187790487472805844610038683159372373526135883092373909944 834653057415 (=0x3238e8e2 ec6e8b7a e1e8feff 97aa58dd d2435bb5 0071cbc2 0d0d4a42 9be67187). Y2 43046985853631671610553834968785204191967171967937842531656254539 962663994648 (=0x5f2bbb06 f7ec5953 2c2a1a62 21124585 1d2682e0 cc37307e fbc17f7f 7fda8518). The representation of k and the compressed representations of Pw and k*Pw in tight MSB/msb-order are given by repr(k) =0x6485b7e6 cd83e5c2 0d5dbfe4 f915494d 9cf5c65d 778c32c3 c08d5abd 15e29c50; repr(Pw) =0x1fe62011 89e0801e f1debed7 456a3dc7 94d3ac0b 55202fe7 2a41cf12 629e56aa; repr(k*Pw) =0x079c3f69 9b688181 69038c35 39c11eb5 96d09f5b 12a242b4 ce660f13 3368c13c, where the leftmost bit of the leftmost octet indicates the parity of the Y-coordinate of the point of Wei25519 in question (which, in this case, are both zero, since Y and Y1 are even). See Appendix H.1 and Appendix I for further detail on (squeezed) point compression. The scalar representation is consistent with the representations specified in [SEC1]; the (squeezed) point representation illustrated above is "new". For completeness, we include a SEC1-consistent representation of the point Pw in affine format and in compressed format below. The SEC1-compliant affine representation of the point Pw in tight MSB/msb-order is given by aff(Pw) =0x04 1fe62011 89e0801e f1debed7 456a3dc7 94d3ac0b 55202fe7 2a41cf12 629e56aa 75e676ce deee3b3c 12942357 22f1d884 ac06de07 330fb07b ae35ca26 df75417e, whereas the SEC1-compliant compressed representation of the point Pw in tight MSB/msb-order is given by compr(Pw) =0x02 1fe62011 89e0801e f1debed7 456a3dc7 94d3ac0b 55202fe7 2a41cf12 629e56aa; The SEC1-compliant uncompressed format aff(Pw) of an affine point Pw corresponds to the right-concatenation of its X- and Y-coordinates, each in tight MSB/msb-order, prepended by the string 0x04, where the reverse procedure is uniquely defined, since elements of GF(p) have a unique fixed-size representation. The (squeezed) compressed format repr(Pw) corresponds to the SEC1-compliant compressed format by extracting the parity bit t from the leftmost bit of the leftmost octet of repr(Pw), replacing the bit position by the value zero, and prepending the octet string with 0x02 or 0x03, depending on whether t=0 or t=1, respectively, where the reverse procedure is uniquely defined, since GF(p) is a 255-bit prime field. For further details, see [SEC1]. Note that, due to the bit-size of the prime p, the squeezed compressed format repr(Pw) is one octet shorter than the SEC1-compliant compressed format compr(Pw). A randomized representation (t1, t2) of the point k*Pw in tight MSB/ msb order is given by t1 446363445988889734093446280484122107283059206243307955388 84223152228795899590 (=0x62af4697 4dd469ac 96c64809 c16c8517 b6a0cee5 40ba0e2e 6dd2b36a fcc75ec6) t2 213890166610228613105792710708385961712211281744756216061 11930888059603107561 (=0x2f49c121 8fed7912 031157ee ae066507 a972320b 6180e267 4025b006 2e67bee9), where this representation is defined in Appendix K.5 and uses the mapping of Appendix K.3.1 with the default square root function. J.4. Example with Wei25519.2 Pw2=(X, Y), k*Pw2=(X1, Y1), and (k+1)*Pw2=(X2, Y2) with Wei25519.2: X 17830493209951148331008014701079988862634531394137235438571836389 227198459763 (=0x276bb396 d766b695 bfe60ab1 3c0260dd c09f5bcf 7b3ca47c f21c8672 d1ecaf73). Y 21064492012933896105338241940477778461866060481408222122979836206 137075789640 (=0x2e921479 5ad47af7 784831de 572ed8e9 7e20e137 cc67378c 184ca19f f9136f48). X1 65470988951686461979789632362377759464688342154017353834939203791 39281908968 (=0x0e7986d2 e94354ab 8abd8806 3154536a 4dcf8e6e 65557183 e242192d 3b87f4e8). Y1 51489590494292183562535790579480033229043271539297275888817125227 35262330110 (=0x0b623521 c1ff84bc 1522ff26 3376796d be77fcad 1fcabc28 98f1be85 d7576cfe). X2 83741788501517200942826153677682120998854086551751663061374935388 3494226693 (=0x01d9f633 b2ac2606 9e6e93f7 6917446c 2b27c16f 729121d7 709c0a58 00ef9b05). Y2 42567334190622848157611574766896093933050043101247319937794684825 168161540336 (=0x5e1c41e1 fb74e41b 3a19ce50 e1b2caf7 7cabcbb3 0c1c1474 a4fd13e6 6c4c08f0). The representation of k and the compressed representations of Pw2 and k*Pw2 in tight MSB/msb-order are given by repr(k) =0x6485b7e6 cd83e5c2 0d5dbfe4 f915494d 9cf5c65d 778c32c3 c08d5abd 15e29c50; repr(Pw2) =0x276bb396 d766b695 bfe60ab1 3c0260dd c09f5bcf 7b3ca47c f21c8672 d1ecaf73; repr(k*Pw2) =0x0e7986d2 e94354ab 8abd8806 3154536a 4dcf8e6e 65557183 e242192d 3b87f4e8, where the leftmost bit of the leftmost octet indicates the parity of the Y-coordinate of the point of Wei25519.2 in question (which, in this case, are both zero, since Y and Y1 are even). See Appendix Appendix H.1 and Appendix I for further detail on (squeezed) point compression. A randomized representation (t1, t2) of the point k*Pw2 in tight MSB/ msb order is given by t1 416669672354928148679758598803660112405431159793278161879 36189858804289581274 (=0x5c1eaaef 80f9d4af 33c119fc c99acd58 f81e7d69 999c7048 e4043a77 87a930da) t2 361115271162391608083096560179337391059615651279123199921 18531180247832114098 (=0x4fd66668 e7174775 de44c852 92df8cfe b9832ef8 2570b3b8 fe5ec21a b2d4b3b2), where this representation is defined in Appendix K.5 and uses the mapping of Appendix K.3.1 with the default square root function. J.5. Example with Wei25519.-3 Pw3=(X, Y), k*Pw3=(X1, Y1), and (k+1)*Pw3=(X2, Y2) with Wei25519.-3: X 14780197759513083469009623947734627174363231692126610860256057394 455099634096 (=0x20ad4ba4 612f0586 221787b0 d01ba46c d1d8cd5a 0348ef00 eb4c9272 03ca71b0). Y 45596733430378470319805536538617129933663237960146030424392249401 952949482817 (=0x64ced628 e982648e 4bfcf30c 71c4d267 ba48b0ce fee20062 b43ef4c9 73f7b541). X1 47362979975244556396292400751828272600887612546997532158738958926 60745725532 (=0x0a78a650 a39995ef dcf4de88 940d4ce9 5b2ca35c c5d70e06 63b8455e 2e04e65c). Y1 30318112837157047703426636957515037640997356617656007157255559136 153389790354 (=0x4307719a 20d08741 58d5889e 8c8ec27e 246b0342 55f8fd62 dbc9ca09 e79c7492). X2 23778942085873786433506063022059853212880296499622328201295446580 293591664363 (=0x3492677e 6ae9d1c3 e08f908b 61033f3d 4e8322c9 fba6da81 2c95b067 9b1486eb). Y2 44846366394651736248316749170687053272682847823018287439056537991 969511150494 (=0x632624d4 ab94c83a 796511c0 5f5412a3 876e56d2 ed18eca3 21b95bef 7bf9939e). The representation of k and the compressed representations of Pw3 and k*Pw3 in tight MSB/msb-order are given by repr(k) =0x6485b7e6 cd83e5c2 0d5dbfe4 f915494d 9cf5c65d 778c32c3 c08d5abd 15e29c50; repr(Pw3) =0xa0ad4ba4 612f0586 221787b0 d01ba46c d1d8cd5a 0348ef00 eb4c9272 03ca71b0; repr(k*Pw3) =0x0a78a650 a39995ef dcf4de88 940d4ce9 5b2ca35c c5d70e06 63b8455e 2e04e65c, where the leftmost bit of the leftmost octet indicates the parity of the Y-coordinate of the point of Wei25519.-3 in question (which, in this case, are one and zero, respectively, since Y is odd and Y1 is even). See Appendix H.1 and Appendix I for further detail on (squeezed) point compression. A randomized representation (t1, t2) of the point k*Pw3 in tight MSB/ msb order is given by t1 573714937613596601525680684642155667097217474964816246889 88981227297409008259 (=0x7ed71d5f 566d2259 99bdb404 bfb9d6cf d2e86ccb 1894d4a6 c75e3c69 e5eb0283) t2 269945781324580189815142015663892935722419453863927287235 57891665397640090729 (=0x3bae63c8 70f60de0 c2e35f94 d24220f1 bb6efd00 37625869 f84923de ff4c5469), where this representation is defined in Appendix K.5 and uses the mapping of Appendix K.3.1 with the default square root function. Appendix K. Auxiliary Functions This section illustrates how one could implement common routines, such as taking square roots and inverses in finite fields, and how to map field elements to curve points and to curve points that avoid some outliers. K.1. Square Roots in GF(q) Square roots are easy to compute in GF(q) if q = 3 (mod 4) (see Appendix K.1.1) or if q = 5 (mod 8) (see Appendix K.1.2). Details on how to compute square roots for other values of q are out of scope. If square roots are easy to compute in GF(q), then so are these in GF(q^2). K.1.1. Square Roots in GF(q), where q = 3 (mod 4) If y is a nonzero element of GF(q) and z:=y^{(q-3)/4}, then y is a square in GF(q) only if y*z^2=1. If y*z^2=1, z is a square root of 1/y and y*z is a square root of y in GF(q). K.1.2. Square Roots in GF(q), where q = 5 (mod 8) If y is a nonzero element of GF(q) and z:=y^{(q-5)/8}, then y is a square in GF(q) only if y^2*z^4=1. a. If y*z^2=+1, z is a square root of 1/y and y*z is a square root of y in GF(q); b. If y*z^2=-1, i*z is a square root of 1/y and i*y*z is a square root of y in GF(q). Here, i is an element of GF(q) for which i^2=-1 (e.g., i:=2^{(q-1)/4}). This field element can be precomputed. K.2. Inversion If y is an integer and gcd(y,n)=1, one can efficiently compute 1/y (mod n) via the extended Euclidean Algorithm (see Section 2.2.5 of [GECC]). One can use this algorithm as well to compute the inverse of a nonzero element y of a prime field GF(p), since gcd(y,p)=1. The inverse of a nonzero element y of GF(q) can be computed as 1/y:=y^{q-2} (since y^{q-1}=1). If inverses are easy to compute in GF(q), then so are these in GF(q^2). Further details are out of scope. The inverses of two nonzero elements y1 and y2 of GF(q) can be computed by first computing the inverse z of y1*y2 and by subsequently computing y2*z=:1/y1 and y1*z=:1/y2. K.3. Mappings to Curve Points One can map elements of GF(q) that are not a square in GF(q) to points of a Weierstrass curve (see Appendix K.3.1), to points of a Montgomery curve (see Appendix K.3.2), or to points of a twisted Edwards curve (see Appendix K.3.3), under some mild conditions on the domain parameters. Full details on mappings that apply if these conditions are not satisfied are out of scope. K.3.1. Mapping to Points of Weierstrass Curve The description below assumes that the domain parameters a and b of the Weierstrass curve W_{a,b} are nonzero. For ease of exposition, we define f(z):=z^3+a*z+b. (Note that for an affine point (X,Y) of W_{a,b} one has Y^2=f(X).) If t is an element of GF(q) that is not a square in GF(q) and that is unequal to -1, then the element X:=(-b/a)*(1+1/(t+t^2)) is the unique solution of the equation f(t*X)=t^3*f(X) and is nonzero. Consequently, either X or X':=t*X is the x-coordinate of an affine point of W{a,b}, depending on whether f(X) is a square in GF(q). a. If f(X) is a square in GF(q) and Y:=sqrt(f(X)), then t is mapped to the point P(t):=(X, Y); b. If f(X) is not a square in GF(q) and Y':=sqrt(f(X')), then t is mapped to the point P(t):=(X', -Y'). Formally, this mapping is not properly defined, since a nonzero square y:=x^2 in GF(q) has two solutions, viz. x and -x; it is properly defined, however, if one designates for each element in GF(q) that is a square in GF(q) precisely one square root as "the" square root of this element. Note that always picking the square root with zero parity (see Appendix H) satisfies this condition (henceforth called the default square root function). If -1 is not a square in GF(q), this element is mapped to the point at infinity O of W_{a,b}. The set of points of W_{a,b} that arises this way has size roughly 3/8 of the order of the curve and each such point arises as image of one or two t values. Further details are out of scope. NOTE 1: If -1 is not a square in GF(q), the mapping above yields the point at infinity for t=-1. One can modify this mapping to always yield an affine point, by mapping the element -1 to, e.g., the base point G of W_{a,b} and leaving the remainder of the mapping the same. Suitability of such a modification is application-specific. Details are out of scope. NOTE 2: The description above assumes that the domain parameters a and b of the Weierstrass curve are nonzero. If this is not the case, one can often find an isogenous curve W_{a',b'} for which the domain parameters a' and b' are nonzero. If so, one can map elements of GF(q) that are not a square in GF(q) to points of W_{a,b} via function composition, where one uses the mapping above to arrive at a point of W_{a',b'} and where one subsequently uses the dual isogeny from W_{a',b'} to W_{a,b} to arrive at a point of W_{a,b}. As an example, one can show that if a is zero and if -4*b is a cube in GF(q) (such as is the case with, e.g., the "BitCoin" curve secp256k1 [SEC2]), this curve is 3-isogenous to a curve with this property and the strategy above applies (for an example with secp256k1, see Appendix L). Further details are out of scope. K.3.2. Mapping to Points of Montgomery Curve The description below assumes that the domain parameter A of the Montgomery curve M_{A,B} is nonzero. For ease of exposition, we define f(z):=z^3+A*z^2+z. (Note that for an affine point (u,v) of M_{A,B} one has B*v^2=f(u).) If t is an element of GF(q) that is not a square in GF(q) and that is unequal to -1, then the element u:=-(1+1/t)/A is the unique nonzero solution of the equation f(t*u)=t^3*f(u). Consequently, either u or u':=t*u is the u-coordinate of an affine point of M{A,B}, depending on whether f(u)/B is a square in GF(q). a. If f(u)/B is a square in GF(q) and v:=sqrt(f(u)/B), then t is mapped to the point P(t):=(u, v); b. If f(u)/B is a not a square in GF(q) and v':=sqrt(f(u')/B), then t is mapped to the point P(t):=(u', -v'). As before, formally, this mapping is not properly defined, since a nonzero square y:=x^2 in GF(q) has two solutions, viz. x and -x; it is properly defined, however, if one designates for each element in GF(q) that is a square in GF(q) precisely one square root as "the" square root of this element. Note that always picking the square root with zero parity (see Appendix H) satisfies this condition (henceforth called the default square root function). If -1 is not a square in GF(q), this element is mapped to the point at infinity O of M_{A,B}. The set of points of M_{A,B} that arises this way has size roughly 1/2 of the order of the curve and each such point arises as image of precisely one t value. Further details are out of scope. NOTE 1: If -1 is not a square in GF(q), the mapping above yields the point at infinity for t=-1. One can modify this mapping to always yield an affine point, by mapping the element -1 to, e.g., the base point G of M_{A,B} and leaving the remainder of the mapping the same. Suitability of such a modification is application-specific. Details are out of scope. NOTE 2: The description above assumes that the domain parameter A of the Montgomery curve is nonzero. If this is not the case, the curve is a Weierstrass curve for which the domain parameter b is zero and Note 2 of Appendix K.3.1 applies. If q = 3 (mod 4), an even simpler approach is possible, where one modifies the construction above and simply takes u:=t and u':=-t (which works, since -1 is not a square in GF(q) and f(-t)=-f(t)). In this case, this construction can be extended to all elements t of GF(q) and, if so, yields a 1-1 mapping between GF(q) and all affine curve points. K.3.3. Mapping to Points of Twisted Edwards Curve One can map elements of GF(q) that are not a square in GF(q) to points of the twisted Edwards curve E_{a,d} via function composition, where one uses the mapping of Appendix K.3.1 to arrive at a point of the Weierstrass curve W_{a,b} and where one subsequently uses the isomorphic mapping between twisted Edwards curves and Weierstrass curves of Appendix D.3 to arrive at a point of E_{a,d}. Another mapping is obtained by function composition, where one instead uses the mapping of Appendix K.3.2 to arrive at a point of the Montgomery curve M_{A,B} and where one subsequently uses the isomorphic mapping between twisted Edwards curves and Montgomery curves of Appendix D.1 to arrive at a point of E_{a,d}. Obviously, one can use function composition (now using the respective pre-images - if these exist) to realize the pre-images of either mapping. K.4. Mappings to High-Order Curve Points Appendix K.3 described how one can map elements of GF(q) that are not a square in GF(q) to points of a Weierstrass curve, to points of a Montgomery curve, or to points of a twisted Edwards curve, under some mild conditions on the domain parameters. Below, we use the mappings of that appendix and the parity function par(.) specified in Appendix H to construct mappings to high-order curve points only (i.e., mappings that avoid points in the small subgroup, see Appendix B.1). We consider mappings to high-order points of a Weierstrass curve (see Appendix K.4.1), to high-order points of a Montgomery curve (see Appendix K.4.2), and to high-order points of a twisted Edwards curve (see Appendix K.4.3). As before, full details on mappings that apply if the mild conditions on the domain parameters are not satisfied are out of scope. K.4.1. Mapping to High-Order Points of Weierstrass Curve The description below assumes that the domain parameters a and b of the Weierstrass curve W_{a,b} are nonzero. For ease of exposition, we define f(z):=z^3+a*z+b. (Note that for an affine point (X,Y) of W_{a,b} one has Y^2=f(X).) If t is an element of GF(q) that is not a square in GF(q) and that is unequal to -1, the mapping of Appendix K.3.1 yields an affine point P(t):=(X, Y) of W_{a,b}. Let P0:=(X0, Y0) be a fixed affine point of W_{a,b} for which neither P0, P0 + P(t), nor P0 - P(t) is in the small subgroup of W_{a,b}. (Note that this implies that P0 and P(t) are distinct affine points of the curve and that these are not each other's inverse.) For binary digit s, the point Q(t,s) is now defined as follows: a. If par(Y0*Y)=s, then the pair (t,s) is mapped to the point Q(t,s):=P0 + P(t); b. If par(Y0*Y)<>s, then the pair (t,s) is mapped to the point Q(t,s):=P0 - P(t). Note that this mapping is properly defined as long as the fixed point P0 (the so-called "curve offset") alluded to above indeed exists. In cases of practical interest that we are aware of, this is indeed the case (see, e.g., Table 1). If -1 is not a square in GF(q), the pair (-1,s) is mapped to the affine point P0 of W_{a,b} (irrespective of the value of s). The set of points of W_{a,b} that arises this way has size roughly 3/8 of the order of the curve and each such point arises as image of up to four values of the pair (t,s). Further details are out of scope. From the group law for Weierstrass curves (see Appendix C.1) it follows that one can express the coordinates of Q(t,s), with t<>-1, in terms of the X-coordinates of P0 and P(t) and the product of their Y-coordinates. (Here, observe that Y0*Y is a square root of f(X0)*f(X).) Thus, Q(t,s) can be computed without the need to fully compute P(t). K.4.2. Mapping to High-Order Points of Montgomery Curve The description below assumes that the domain parameters A and B of the Montgomery curve M_{A,B} are nonzero. For ease of exposition, we define f(z):=z^3+A*z^2+z. (Note that for an affine point (u,v) of M_{A,B} one has B*v^2=f(u).) If t is an element of GF(q) that is not a square in GF(q) and that is unequal to -1, the mapping of Appendix K.3.2 yields an affine point P(t):=(u, v) of M_{A,B}. Let P0:=(u0, v0) be a fixed affine point of M_{A,B} for which neither P0, P0 + P(t), nor P0 - P(t) is in the small subgroup of M_{A,B}. (Note that this implies that P0 and P(t) are distinct affine points of the curve and that these are not each other's inverse.) For binary digit s, the point Q(t,s) is now defined as follows: a. If par(B*v0*v)=s, then the pair (t,s) is mapped to the point Q(t,s):=P0 + P(t); b. If par(B*v0*v)<>s, then the pair (t,s) is mapped to the point Q(t,s):=P0 - P(t). Note that this mapping is properly defined as long as the fixed point P0 (the so-called "curve offset") alluded to above indeed exists. In cases of practical interest that we are aware of, this is indeed the case (see, e.g., Table 1). If -1 is not a square in GF(q), the pair (-1,s) is mapped to the affine point P0 of M_{A,B} (irrespective of the value of s). The set of points of M_{A,B} that arises this way has size roughly 1/2 of the order of the curve and each such point arises as image of up to two values of the pair (t,s). Further details are out of scope. From the group law for Montgomery curves (see Appendix C.2) it follows that one can express the coordinates of Q(t,s), with t<>-1, in terms of the u-coordinates of P0 and P(t) and the product of their v-coordinates. (Here, observe that B*v0*v is a square root of f(u0)*f(u).) Thus, Q(t,s) can be computed without the need to fully compute P(t). +----------------------------+------------------------+ | Curve | Fixed curve offset P0 | +----------------------------+------------------------+ | NIST P-224 [FIPS-186-4] | Base point (Gx,Gy) | | NIST P-256 [FIPS-186-4] | P0:=(0,y) with y even | | NIST P-384 [FIPS-186-4] | P0:=(0,y) with y even | | NIST P-521 [FIPS-186-4] | P0:=(0,y) with y even | | Brainpool224r1 [RFC5639] | Base point (Gx, Gy) | | Brainpool256r1 [RFC5639] | Base point (Gx, Gy) | | Brainpool320r1 [RFC5639] | Base point (Gx, Gy) | | Brainpool384r1 [RFC5639] | Base point (Gx, Gy) | | Brainpool512r1 [RFC5639] | P0:=(3,y), y even | | Curve25519 [RFC7748] | P0:=(90,v), v even | | Curve448 [RFC7748] | P0:=(50,v), v even | | Wei25519 [Appendix E.3] | P0:=(3,y), y even | | Wei25519.2 [Appendix G.3] | P0:=(244,y), y even | | Wei25519.-3 [Appendix G.3] | P0:=(41,y), y even | | secp256k1.m [Appendix L.3] | P0:=(0,y), y even | +----------------------------+------------------------+ Table 1: Curve offsets P0 for mappings that avoid low-order points K.4.3. Mapping to High-Order Points of Twisted Edwards Curve One can map elements of GF(q) that are not a square in GF(q) to points of the twisted Edwards curve E_{a,d} via function composition, where one uses the mapping of Appendix K.4.1 to arrive at a point of the Weierstrass curve W_{a,b} that does not have low order and where one subsequently uses the isomorphic mapping between twisted Edwards curves and Weierstrass curves of Appendix D.3 to arrive at a point of E_{a,d} with this property. Another mapping is obtained by function composition, where one instead uses the mapping of Appendix K.4.2 to arrive at a point of the Montgomery curve M_{A,B} that does not have low order and where one subsequently uses the isomorphic mapping between twisted Edwards curves and Montgomery curves of Appendix D.1 to arrive at a point of E_{a,d} with this property. Obviously, one can use function composition (now using the respective pre-images - if these exist) to realize the pre-images of either mapping. K.5. Randomized Representation of Curve Points The mappings of Appendix K.3 allow one to represent a curve point Q as a specific element t of GF(q), provided this point arises as a point in the range of the mapping at hand. For Montgomery curves and twisted Edwards curves, this covers roughly half of the curve points; for Weierstrass curves, roughly 3/8 of the curve points. One can extend the mappings above, by mapping a pair (t1, t2) of inputs to the point Q:=P2(t1, t2):=P(t1) + P(t2). In this case, each curve point has roughly q/4 representations as an ordered pair (t1, t2) on average. In fact, one can show that if the input pairs are generated uniformly at random, then the corresponding curve points follow a distribution that is also (statistically indistinguishable from) a uniform distribution, and vice-versa. Here, each pair (t1, t2) deterministically yields a curve point, whereas for each curve point Q, a randomized algorithm yields an ordered pair (t1, t2) of pre- images of Q, where the expected number of randomized pre-images one has to try is small (four if one uses the mapping of Appendix K.3.1; two if one uses the mapping of Appendix K.3.2). For further details, see Algorithm 1 of [Tibouchi]. Similar properties hold if one uses the mappings of Appendix K.4 (rather than those of Appendix K.3): in this case, the mapping allows one to represent a curve point Q as a specific element (t,s) of GF(q)x{0,1}, provided this point arises as a point in the range of the mapping at hand. For Montgomery curves and twisted Edwards curves, this covers roughly half of the curve points; for Weierstrass curves, roughly 3/8 of the curve points. One can extend the mappings above, by mapping a pair ((t1,s1), (t2,s2)) of inputs to the point Q:=Q2((t1,s1), (t2,s2)):=Q(t1,s1) - Q(t2,s2). In this case, each curve point has roughly q representations as an ordered pair ((t1,s1), (t2,s2)) on average. In fact, one can show that if the input pairs are generated uniformly at random, then the corresponding curve points follow a distribution that is also (statistically indistinguishable from) a uniform distribution, and vice-versa. Here, each pair ((t1,s1), (t2,s2)) deterministically yields a curve point, whereas for each curve point Q, a randomized algorithm yields an ordered pair ((t1,s1), (t2,s2)) of pre-images of Q, where the expected number of randomized pre-images one has to try is small (four if one uses the mapping of Appendix K.4.1; two if one uses the mapping of Appendix K.4.2). Further details are out of scope. NOTE 1: The main difference between the two constructions above is that that the first construction uses the mappings to curve points described in Appendix K.3, while the second construction uses the mappings to high-order curve points described in Appendix K.4. Note that Q2((t1,s1), (t2,s2)) assumes all values (+/-) P(t1) (+/-) P(t2) if one considers all possible values for the binary digits s1 and s2. (This, thereby, includes the value P2(t1, t2).) NOTE 2: The second mapping above operates on input pairs (t,s), where t is an element of GF(q) that is not a square in GF(q) and where s is a binary digit from the set {0,1}. One can use these mappings to produce a mapping from the nonzero elements of GF(q) to points of a curve via function composition, where one first maps any nonzero element u of GF(q) to the pair (t,s):=(delta*u^2, par(u)), where delta is a fixed element of GF(q) that is not a square in GF(q), and where one subsequently applies any of the mappings above to yield a point of the curve in question. The resulting mapping from the nonzero elements of GF(q) to high-order curve points can be extended further to one that operates on all elements of GF(q) by mapping 0 to any fixed high-order point of the curve in question (e.g., to the fixed "curve offset" P0 of this curve [henceforth called the default completed mapping]). Depending on application-specific criteria, yet other function compositions may be preferred. For the first mapping above, one can use a similar function composition, where one simply drops the binary digit s and maps 0 to the point at infinity or any other suitable curve point. Further details are out of scope. Appendix L. Curve secp256k1 and Friend This section illustrates how isogenies can be used to yield curves with specific properties (here, for illustrated for the "BitCoin" curve secp256k1). L.1. Curve Definition and Alternative Representation The elliptic curve secp256k1 is the Weierstrass curve W_{a,b} defined over the prime field GF(p), with p:=2^256-2^32-2^9-2^8-2^7-2^6-2^4-1, where a:=0 and b:=7. This curve has order h*n, where h=1 and where n is a prime number. For this curve, domain parameter a is zero, whereas b is not. The quadratic twist of this curve has order h1*n1, where h1 is a 37-bit integer and where n1 is a prime number. For this curve, the base point is the point (GX, GY). The curve secp256k1 is 3-isogenous to the Weierstrass curve secp256k1.m defined over GF(p), which has nonzero domain parameters a and b and has as base point the pair (GmX,GmY), where parameters are as specified in Appendix L.3 and where the related mappings are as specified in Appendix L.2. L.2. Switching Between Representations Each affine point (X,Y) of secp256k1 corresponds to the point (X',Y'):=(u(X)/w(X)^2,Y*v(X)/w(X)^3) of secp256k1.m, where u, v, and w are the polynomials with coefficients in GF(p) as defined in Appendix L.4.1, while the point at infinity of secp256k1 corresponds to the point at infinity of secp256k1.m. Under this isogenous mapping, the base point (GX,GY) of secp256k1 corresponds to the base point (GmX,GmY) of secp256k1.m. The dual isogeny maps the affine point (X',Y') of secp256k1.m to the affine point (X,Y):=(u'(X')/w'(X')^2,Y'*v'(X')/w'(X')^3) of secp256k1, where u', v', and w' are the polynomials with coefficients in GF(p) as defined in Appendix L.4.2, while mapping the point at infinity O of secp256k1.m to the point at infinity O of secp256k1. Under this dual isogenous mapping, the base point (GmX, GmY) of secp256k1.m corresponds to a multiple of the base point (GX, GY) of secp256k1, where this multiple is l=3 (the degree of the isogeny; see the description in Appendix F.4). Note that this isogenous map (and its dual) primarily involves the evaluation of three fixed polynomials involving the x-coordinate, which takes roughly 10 modular multiplications (or less than 1% relative incremental cost compared to the cost of an elliptic curve scalar multiplication). L.3. Domain Parameters The parameters of the curve sec256k1 and the corresponding 3-isogenous curve sec256k1.m are as indicated below. Here, the domain parameters of the curve secp256k1 are as specified in [SEC2]; the domain parameters of secp256k1.m are "new". General parameters (for all curves): p 2^256-2^32-2^9-2^8-2^7-2^6-2^4-1 (=0xffffffff ffffffff ffffffff ffffffff ffffffff ffffffff fffffffe fffffc2f) h 1 n 11579208923731619542357098500868790785283756427907490438260516314 1518161494337 (=0xffffffff ffffffff ffffffff fffffffe baaedce6 af48a03b bfd25e8c d0364141) h1 23479460174521 (=0x1a 9bfcab89) n1 10131766773001318469008702396060356387381009972480920692566974370 31 (=0x099ee564 ea5d84f5 08913936 a761b0d5 d792a426 a7779817 ae2f5b67) Weierstrass curve-specific parameters (for secp256k1): a 0 (=0x00) b 7 (=0x07) GX 55066263022277343669578718895168534326250603453777594175500187360 389116729240 (=0x79be667e f9dcbbac 55a06295 ce870b07 029bfcdb 2dce28d9 59f2815b 16f81798) GY 32670510020758816978083085130507043184471273380659243275938904335 757337482424 (=0x483ada77 26a3c465 5da4fbfc 0e1108a8 fd17b448 a6855419 9c47d08f fb10d4b8) Weierstrass curve-specific parameters (for secp256k1.m): a 93991599167772749909245591943117186381494883464374162770646538702 960816911535 (=0xcfcd5c21 75e2ef7d ccdce737 770b7381 5a2f13c5 09035ca2 54a14ac9 f08974af) b 1771 (=0x06eb) GmX 26591621185618668069038227574782692264471832498547635565821216767 730887659845 (=0x3aca5300 959fa1d0 baf78dcf f77a616f 395e586d 67aced0a 88798129 0c279145) GmY 67622516283223102233819216063319565850973524550533340939716651159 860372686848 (=0x9580fce5 3a170f4f b744579f f3d62086 12cd6a23 3e2de237 f976c6a7 8611c800) L.4. Isogeny Details The isogeny and dual isogeny are both isogenies with degree l=3. Both are specified by a triple of polynomials u, v, and w (resp. u', v', and w') of degree 3, 3, and 1, respectively, with coefficients in GF(p). The coeffients of each of these polynomials are specified in Appendix L.4.1 (for the isogeny) and in Appendix L.4.2 (for the dual isogeny). For each polynomial in variable x, the coefficients are tabulated as sequence of coefficients of x^0, x^1, x^2, ..., in hexadecimal format. L.4.1. Isogeny Parameters L.4.1.1. Coefficients of u(x) 0 0x54 1 0xa4d89db3ed06c81e6143ec2eca9f761d8d17260dc229e1da1f73f714506872a9 2 0xcc58ffccbd9febb4a66222c7d1311d988d88c0624bcd68ec4c758a8e67dfd99b 3 0x01 L.4.1.2. Coefficients of v(x) 0 0x1c 1 0x94c7bc69befd17f2fae2e3ebf24df1f355d181fa1a8056103ba9baad4b40f029 2 0xb2857fb31c6fe18ef993342bb9c9ac64d44d209371b41d6272b04fd61bcfc851 3 0x01 L.4.1.3. Coefficients of w(x) 0 0xe62c7fe65ecff5da53311163e8988ecc46c4603125e6b476263ac546b3efeae5 1 0x01 L.4.2. Dual Isogeny Parameters L.4.2.1. Coefficients of u'(x) 0 0x8e38e38e38e38e38e38e38e38e38e38e38e38e38e38e38e38e38e38daaaaa8c7 1 0x44cd5cd7ce55a801725891578fbe7356bd936355fd0e2f538797cecff7a37244 2 0x668d0011162006c3c889f4680f9a4b77d0d26a89e6bb87b13bd8d1cfdd600a41 3 0x8e38e38e38e38e38e38e38e38e38e38e38e38e38e38e38e38e38e38daaaaa88c L.4.2.2. Coefficients of v'(x) 0 0x4bda12f684bda12f684bda12f684bda12f684bda12f684bda12f684b8e38e23c 1 0x519ba9c1f48f68054def6a410f0fa6e8b71c6c3b4a8958324681f6508c01fada 2 0xb34680088b100361e444fa3407cd25bbe8693544f35dc3d89dec68e76eb00338 3 0x2f684bda12f684bda12f684bda12f684bda12f684bda12f684bda12f38e38d84 L.4.2.3. Coefficients of w'(x) 0 0x4d7a804ce3901e71066ccbd44636539b2bb2df6c8e4be29d8d4fb028e43033de 1 0x01 Appendix M. Curve448 and Cousins This section introduces curves related to Curve448 and explains their relationships. M.1. Curve Definition and Alternative Representations The elliptic curve Curve448 is the Montgomery curve M_{A,B} defined over the prime field GF(p), with p:=2^{448}-2^{224}-1, where A:=156326 and B:=1. This curve has order h*n, where h=4 and where n is a prime number. For this curve, A^2-4 is not a square in GF(p), whereas A-2 is. The quadratic twist of this curve has order h1*n1, where h1=4 and where n1 is a prime number. For this curve, the base point is the point (Gu, Gv), where Gu=5 and where Gv is an even integer in the interval [0, p-1]. This curve has the same group structure as (is "isomorphic" to) the twisted Edwards curve E_{a,d} defined over GF(p), with as base point the point (Gx, Gy), where parameters are as specified in Appendix M.3. This curve is denoted as Ed448. For this curve, the parameter a is a square in GF(p), whereas d is not, so the group laws of Appendix C.3 apply. The curve is also isomorphic to the elliptic curve W_{a,b} in short- Weierstrass form defined over GF(p), with as base point the point (GX, GY), where parameters are as specified in Appendix M.3. This curve is denoted as Wei448. M.2. Switching between Alternative Representations Each affine point (u, v) of Curve448 corresponds to the point (X, Y):=(u + A/3, v) of Wei448, while the point at infinity of Curve448 corresponds to the point at infinity of Wei448. (Here, we used the mappings of Appendix D.2 and that B=1.) Under this mapping, the base point (Gu, Gv) of Curve448 corresponds to the base point (GX, GY) of Wei448. The inverse mapping maps the affine point (X, Y) of Wei448 to (u, v):=(X - A/3, Y) of Curve448, while mapping the point at infinity of Wei448 to the point at infinity of Curve448. Note that this mapping involves a simple shift of the first coordinate and can be implemented via integer-only arithmetic as a shift of -(p-A)/3 for the isomorphic mapping and a shift of (p-A)/3 for its inverse, where delta=(p-A)/3 is the element of GF(p) defined by delta 24227957476520229684977460262933484478454712022910602009383006 63935374427222435908954654612328921819766962948206145457870178326 72736371 (=0x55555555 55555555 55555555 55555555 55555555 55555555 55555554 ffffffff ffffffff ffffffff ffffffff ffffffff ffffffff ffff3473). (Note that, depending on the implementation details of the field arithmetic, one may have to shift the result by +p or -p if this integer is not in the interval [0,p-1].) The curve Ed448 is isomorphic to the curve Curve448, where the base point (Gu, Gv) of Curve448 corresponds to the base point (Gx,Gy) of Ed448 and where the point at infinity and the point (0,0) of order two of Curve448 correspond to, respectively, the point (0, 1) and the point (0, -1) of order two of Ed448 and where each other point (u, v) of Curve448 corresponds to the point (c*u/v, (u+1)/(u-1)) of Ed448, where c is the element of GF(p) defined by c sqrt((A-2)/B) 19788846729546443953835400975385803825683515259105980214819977919 60874042320025157136042631277930307478554244641856917664538448351 92428 (=0x45b2c5f7 d649eed0 77ed1ae4 5f44d541 43e34f71 4b71aa96 c945af01 2d182975 0734cde9 faddbda4 c066f7ed 54419ca5 2c85de1e 8aae4e6c). (Here, we used the mapping of Appendix D.1 and normalized this using the mapping of Appendix F.1 (where the element s of that appendix is set to c above).) The inverse mapping from Ed448 to Curve448 is defined by mapping the point (0, 1) and the point (0, -1) of order two of Ed448 to, respectively, the point at infinity and the point (0,0) of order two of Curve448 and having each other point (x, y) of Ed448 correspond to the point ((y + 1)/(y - 1), c*(y + 1)/((y-1)*x)) of Curve448. The curve Ed448 is isomorphic to the Weierstrass curve Wei448, where the base point (Gx, Gy) of Ed448 corresponds to the base point (GX,GY) of Wei448 and where the identity element (0,1) and the point (0,-1) of order two of Ed448 correspond to, respectively, the point at infinity O and the point (A/3, 0) of order two of Wei448 and where each other point (x, y) of Ed448 corresponds to the point (X, Y):=((y+1)/(y-1)+A/3, c*(y+1)/((y-1)*x)) of Wei448, where c was defined before. (Here, we used the mapping of Appendix D.3.) The inverse mapping from Wei448 to Ed448 is defined by mapping the point at infinity O and the point (A/3, 0) of order two of Wei448 to, respectively, the identity element (0,1) and the point (0,-1) of order two of Ed448 and having each other point (X, Y) of Wei448 correspond to the point (c*(X-A/3)/Y, (X-A/3+1)/(X-A/3-1)) of Ed448. Note that these mappings can be easily realized if points are represented in projective coordinates, using a few field multiplications only, thus allowing switching between alternative curve representations with negligible relative incremental cost. M.3. Domain Parameters The parameters of the Montgomery curve and the corresponding isomorphic curves in twisted Edwards curve and short-Weierstrass form are as indicated below. Here, the domain parameters of the Montgomery curve Curve448 and of the twisted Edwards curve Ed448 are as specified in [RFC7748]; the domain parameters of Wei448 are "new". IMPORTANT NOTE: the supposed base point of Ed448 specified in [RFC7748] is incorrect, since it has order 2*n, and - in the notation below - that point is the point (Gx,-Gy)=-(Gx, Gy)+(0,-1). The birational map in that document is also incorrect. General parameters (for all curve models): p 2^{448}-2^{224}-1 (=0xffffffff ffffffff ffffffff ffffffff ffffffff ffffffff fffffffe ffffffff ffffffff ffffffff ffffffff ffffffff ffffffff ffffffff) h 4 n 18170968107390172263733095197200113358841034017182951507037254979 51460039615395857161957552916923759633102937090916623047737558596 49779 (=2^{446} - 0x8335dc16 3bb124b6 5129c96f de933d8d 723a70aa dc873d6d 54a7bb0d) h1 4 n1 18170968107390172263733095197200113358841034017182951507037254979 51601601218258006270024365576458970017341485218301563757529931495 32941 (=2^{446} + 0x0335dc16 3bb124b6 5129c96f de933d8d 723a70aa dc873d6d 54a7bb0d) Montgomery curve-specific parameters (for Curve448): A 156326 (=0x0262a6) B 1 (=0x01) Gu 5 (=0x05) Gv 35529392678556817526412750206378333480897639938771427183188089843 51690887869674100029326737658645509101427741472681058389855952906 06362 (=0x7d235d12 95f5b1f6 6c98ab6e 58326fce cbae5d34 f55545d0 60f75dc2 8df3f6ed b8027e23 46430d21 1312c4b1 50677af7 6fd7223d 457b5b1a) Edwards curve-specific parameters (for Ed448): a 1 (0x01) d 39082/39081 = (A+2)/(A-2) (=611975850744529176160423220965553317543219696871016626328968936 41508786004263647489178559928366602041476867897998937814706546281 5545017) (=0xd78b4bdc 7f0daf19 f24f38c2 9373a2cc ad461572 42a50f37 809b1da3 412a12e7 9ccc9c81 264cfe9a d0809970 58fb61c4 243cc32d baa156b9) Gx 34539749303972951637400860415053741026665526007518329021640697028 16456950736723444304817877593406332217083915834240417889241245677 00732 (=0x79a70b2b 70400553 ae7c9df4 16c792c6 1128751a c9296924 0c25a07d 728bdc93 e21f7787 ed697224 9de732f3 8496cd11 69871309 3e9c04fc) Gy 3/2 36341936214780344527466190394400226717682068034365903014074509959 03061640833653863431981918493382729650444422309218186805267490091 82721 (=0x7fffffff ffffffff ffffffff ffffffff ffffffff ffffffff ffffffff 80000000 00000000 00000000 00000000 00000000 00000000 00000001) Weierstrass curve-specific parameters (for Wei448): a 48455914953040459369954920525866968956909424045821204018766013278 70748854444871817909309224657843639533925896412290915740356571996 37535 (=0xaaaaaaaa aaaaaaaa aaaaaaaa aaaaaaaa aaaaaaaa aaaaaaaa aaaaaaa9 ffffffff ffffffff ffffffff ffffffff ffffffff fffffffe 1a76d41f) b 26919952751689144094419400292148316087171902247678446677092229599 28193808024928787727394013698802021963292164673494953191916856645 13904 (=0x5ed097b4 25ed097b 425ed097 b425ed09 7b425ed0 97b425ed 097b425e 71c71c71 c71c71c7 1c71c71c 71c71c71 c71c71c7 1c72c87b 7cc69f70) GX 48455914953040459369954920525866968956909424045821204018766013278 70748854444871817909309224657843639533925896412290915740356653456 29073 (=0xaaaaaaaa aaaaaaaa aaaaaaaa aaaaaaaa aaaaaaaa aaaaaaaa aaaaaaaa 00000000 00000000 00000000 00000000 00000000 00000000 0000cb91) GY 35529392678556817526412750206378333480897639938771427183188089843 51690887869674100029326737658645509101427741472681058389855952906 06362 (=0x7d235d12 95f5b1f6 6c98ab6e 58326fce cbae5d34 f55545d0 60f75dc2 8df3f6ed b8027e23 46430d21 1312c4b1 50677af7 6fd7223d 457b5b1a) Appendix N. Further Cousins of Curve448 This section introduces some further curves related to Curve448 and explains their relationships. N.1. Further Alternative Representations The Weierstrass curve Wei448 is isomorphic to the Weierstrass curve Wei448.1 defined over GF(p), with as base point the pair (G1X,G1Y), and isogenous to the Weierstrass curve Wei448.-3 defined over GF(p), with as base point the pair (G3X, G3Y), where parameters are as specified in Appendix N.3 and where the related mappings are as specified in Appendix N.2. The Edwards curve Ed448 is isogenous to the Edwards curve Edwards448 defined over GF(p), with as base point the pair (G1x,G1y), where parameters are as specified in Appendix N.3 and where the related mappings are as specified in Appendix N.2. N.2. Further Switching Each affine point (X, Y) of Wei448 corresponds to the point (X', Y'):=(X*s^2,Y*s^3) of Wei448.1, where s is the element of GF(p) defined by s 52322274343677442779379520589771028818568404587729117919590511061 93509510238347880134473888687471465216641846232724641298954890800 00881 (=0xb848cd01 981d2f83 f2829b42 eb86914e 88f44c9d 05dcbdff dbdd1e56 c4674bc8 d6d90d91 862a38f5 ca797ca7 f21c05cf a7ac32bf d2ca0171), while the point at infinity of Wei448 corresponds to the point at infinity of Wei448.1. (Here, we used the mapping of Appendix F.3.) Under this mapping, the base point (GX, GY) of Wei448 corresponds to the base point (G1X,G1Y) of Wei448.1. The inverse mapping maps the affine point (X', Y') of Wei448.1 to (X,Y):=(X'/s^2,Y'/s^3) of Wei448, while mapping the point at infinity O of Wei448.1 to the point at infinity O of Wei448. Note that this mapping (and its inverse) involves a modular multiplication of both coordinates with fixed constants s^2 and s^3 (respectively, 1/s^2 and 1/s^3), which can be precomputed. Each affine point (X,Y) of Wei448 for which the Y-coordinate is nonzero (i.e., each point with order larger than two) corresponds to the point (X',Y'):=(X1*t^2,Y1*t^3) of Wei448.-3, where (X1,Y1)=(u(X)/w(X),Y*v(X)/w(X)^2), where u, v, and w are the polynomials with coefficients in GF(p) as defined in Appendix N.4.1 and where t is the element of GF(p) defined by t 23579450751475691430882365546539966269774125426758968522698856022 13378944265540874438945283200254318223329383397068961863760712339 07365 (=0x530c9a1d7cf071d09646b83db246626b4e57ba5d6a791bef761972543209d c5c20d81498d5ab8d7a2fb22507ca68c040a6c82eb3b6c7aaa5), while the point at infinity and the point (A/3,0) of order two of Wei448 corresponds to the point at infinity of Wei448.-3. (Here, we used the isogenous mapping of Appendix F.4.) Under this isogenous mapping, the base point (GX,GY) of Wei448 corresponds to the base point (G3X,G3Y) of Wei448.-3. The dual isogeny maps the affine point (X',Y') of Wei448.-3 to the affine point (X,Y):=(u'(X1)/w'(X1),Y1*v'(X1)/w'(X1)^2) of Wei448, where (X1,Y1)=(X'/t^2,Y'/t^3) and where u', v', and w' are the polynomials with coefficients in GF(p) as defined in Appendix N.4.2, while mapping the point at infinity O of Wei448.-3 to the point at infinity O of Wei448. Under this dual isogenous mapping, the base point (G3X, G3Y) of Wei448.-3 corresponds to a multiple of the base point (GX, GY) of Wei448, where this multiple is l=2 (the degree of the isogeny; see the description in Appendix F.4). Note that this isogenous map (and its dual) primarily involves the evaluation of three fixed polynomials involving the x-coordinate, which takes only a few modular multiplications (less than 0.5% relative incremental cost compared to the cost of an elliptic curve scalar multiplication). Each point (x1,y1) of Edwards448 corresponds to the point (x,y) of Ed448, where x = c*x1*y1/(1-d1*x1^2*y1^2) = c*x1*y1/(2-x1^2-y1^2) and y =(1 + d1*x1^2*y1^2)/(y1^2-x1^2) = -(x1^2+y1^2)/(x1^2-y1^2). (Here, we used the 4-isogenous mapping of Appendix F.4. Under this isogenous mapping, the base point (G1x, G1y) of Edwards448 corresponds to the base point (Gx,Gy) of Ed448. The dual isogeny maps each point (x,y) of Ed448 to the point (x1,y1) of Edwards448, where x1 = (4*x*y/c)/(y^2-x^2) and y1 = (1 - d*x^2*y^2)/(1 + d*x^2*y^2) = (2-x^2-y^2)/(x^2+y^2). Under this dual isogenous mapping, the base point (Gx, Gy) of Ed448 corresponds to a multiple of the base point (G1x, G1y) of Edwards448, where this multiple is l=4 (the degree of the isogeny; see the description in Appendix F.4). Note that this isogenous map (and its dual) primarily involves the evaluation of three fixed polynomials, which takes only a few multiplications (less than 0.5% relative incremental cost compared to the cost of an elliptic curve scalar multiplication). The point (0,1) and the point (0,-1) of order two of Edwards448 correspond to, respectively, the point at infinity and the point (0,0) of order two of Curve448, while each other point (x1,y1) of Edwards448 corresponds to the point (u,v) of Curve448, where u = y1^2/x1^2 and v = y1*(2-x1^2-y1^2)/x1^3. Under this isogenous mapping, the base point (G1x, G1y) of Edwards448 corresponds to the base point (Gu,Gv) of Curve448. The dual isogeny maps both the point at infinity and the point (0,0) of order two of Curve448 to the point (0,1) of Edwards448, while each other point (u,v) of Curve448 corresponds to the point (x1,y1) of Edwards448, where x1 = 4*(u^2-1)*v/((u^2-1)^2+4*v^2) and y1 = u*((u^2-1)^2-4*v^2)/(2*(u^2+1)*v^2-u*(u^2-1)^2). Under this dual isogenous mapping, the base point (Gu, Gv) of Curve448 corresponds to a multiple of the base point (G1x, G1y) of Edwards448, where this multiple is l=4 (the degree of the isogeny; see above). N.3. Further Domain Parameters The parameters of the Weierstrass curve with a=1 that is isomorphic with Wei448 and the parameters of the Weierstrass curve with a=-3 that is isogenous with Wei448 are as indicated below. Both domain parameter sets can be exploited directly to derive more efficient point addition formulae, should an implementation facilitate this. The domain parameters of the twisted Edwards curve Edwards448 are as specified in [RFC7748]. General parameters: same as for Wei448 (see Appendix M.3) Weierstrass curve-specific parameters (for Wei448.1, i.e., with a=1): a 1 (=0x01) b 65961281701807170531944804985907990287225248056560036392380945951 38183088507635437786021044927715119224497407914895790669345268896 52743 (=0xe8528596 bfbcbac9 7ebdbe4e 9683e25c 73a5ff37 6c4cd400 5a75c425 8e3eb05a 9f6f8c24 24cb5aa9 0dcf9fa4 cab6691d 5530347c 28437207) G1X 19236211982508211644805033459306273038523230481309141518540414163 72091186292458482231912460243257247478684005448999746809691007995 9723 (=0x06c672d5 b5bae33b 010fa210 9de7937a 95db8ffc 043c507f 5e0d07a1 25382eaf 13f5fc3b 75db2614 6e6d002f d8364ed6 c9bc8fbf bbda22ab) G1Y 30319443056877169804488072384563064288675576234196773667920807567 79177927858755621958756222206632465988308466319556948821775845861 64158 (=0x6ac9c53c 767cd3ae cbf904a1 2923502f 115355d1 6ae8911c 5c92f612 aa854455 d1e6d29f 4db4ddea 519a174f c0dd2505 ec3328ba 250a07be) Weierstrass curve-specific parameters (for Wei448.-3, i.e., with a=- 3): a -3 (=0xffffffff ffffffff ffffffff ffffffff ffffffff ffffffff fffffffe ffffffff ffffffff ffffffff ffffffff ffffffff ffffffff fffffffc) b 69993768681000150084833669961900533067383335592494498709534693464 91314250731583068774689950893229681024927315747794587331422088592 54465 (=0xf686723d 80e29d06 2d00a9f1 3305b698 85790019 cca78035 9dac226b efb1ae21 125397dd 16f255b0 cc5d18e5 43582a1c af90dfe2 c0aeaec1) G3X 40677474994869876470916133424311516856662407970799424837841348421 87696274665113140719001227030116551378877280368526334985627104680 88795 (=0x8f452c6b dc3265dd 580b2638 59a02b20 198cc020 1dd7fba1 8b431694 4a936052 fb4e4a41 93d01fa5 5fb5c732 7393208b 8170f3f2 be78d3db) G3Y 54594210970205994927260789585006437115117066846498189378285031510 90310290468347714929366106635470978666795512446629051235704504868 06147 (=0xc0494f90 461db11c 35fb7646 8349399a ae230351 11330cce b7473244 ab63c955 cf6ec02f 2656b439 44b19f4b 52eef12e 73026bbc 84444683) Edwards curve-specific parameters (for Edwards448): a 1 (0x01) d1 -39081 = -(A-2)/4 (=726838724295606890549323807888004534353641360687318060281490199 18061232816673077268639638369867654593008888446184363736105349801 8326358) (=0xffffffff ffffffff ffffffff ffffffff ffffffff ffffffff fffffffe ffffffff ffffffff ffffffff ffffffff ffffffff ffffffff ffff6756) G1x 22458004029592430018760433409989603624678964163256413424612546168 69504154674060329090291928693579532825780320751464461736746026352 47710 (=0x4f1970c6 6bed0ded 221d15a6 22bf36da 9e146570 470f1767 ea6de324 a3d3a464 12ae1af7 2ab66511 433b80e1 8b00938e 2626a82b c70cc05e) G1y 29881921007848149267601793044393067343754404015408024209592824137 23315061898358760035368786554187847339823032335034625005315450628 32660 (=0x693f4671 6eb6bc24 88762037 56c9c762 4bea7373 6ca39840 87789c1e 05a0c2d7 3ad3ff1c e67c39c4 fdbd132c 4ed7c8ad 9808795b f230fa14) N.4. Isogeny Details The isogeny and dual isogeny are both isogenies with degree l=2. Both are specified by a triple of polynomials u, v, and w (resp. u', v', and w') of degree 2, 2, and 1, respectively, with coefficients in GF(p). The coeffients of each of these polynomials are specified in Appendix N.4.1 (for the isogeny) and in Appendix N.4.2 (for the dual isogeny). For each polynomial in variable x, the coefficients are tabulated as sequence of coefficients of x^0, x^1, x^2, ..., in hexadecimal format. N.4.1. Isogeny Parameters N.4.1.1. Coefficients of u(x) 0 0x01 1 0x55555555555555555555555555555555555555555555555555555554ffffffff ffffffffffffffffffffffffffffffffffffffffffff3473 2 0x01 N.4.1.2. Coefficients of v(x) 0 0x1c71c71c71c71c71c71c71c71c71c71c71c71c71c71c71c71c71c71c55555555 5555555555555555555555555555555555555555f72db94a 1 0xaaaaaaaaaaaaaaaaaaaaaaaaaaaaaaaaaaaaaaaaaaaaaaaaaaaaaaa9ffffffff fffffffffffffffffffffffffffffffffffffffffffe68e6 2 0x01 N.4.1.3. Coefficients of w(x) 0 0x55555555555555555555555555555555555555555555555555555554ffffffff ffffffffffffffffffffffffffffffffffffffffffff3473 1 0x01 N.4.2. Dual Isogeny Parameters N.4.2.1. Coefficients of u'(x) 0 0x016c26e0e8 1 0x5555555555555555555555555555555555555555555555555555555500000000 0000000000000000000000000000000000000000000065c6 2 0x3fffffffffffffffffffffffffffffffffffffffffffffffffffffffc0000000 000000000000000000000000000000000000000000000000 N.4.2.2. Coefficients of v'(x) 0 0x8e38e38e38e38e38e38e38e38e38e38e38e38e38e38e38e38e38e38daaaaaaaa aaaaaaaaaaaaaaaaaaaaaaaaaaaaaaaaaaaaaaaa45836c31 1 0x5555555555555555555555555555555555555555555555555555555500000000 0000000000000000000000000000000000000000000065c6 2 0x1fffffffffffffffffffffffffffffffffffffffffffffffffffffffe0000000 000000000000000000000000000000000000000000000000 N.4.2.3. Coefficients of w'(x) 0 0x5555555555555555555555555555555555555555555555555555555500000000 000000000000000000000000000000000000000000019719 1 0x01 Appendix O. Representation Examples Curve448 Family Members We present some examples of computations using the curves introduced inAppendix M and Appendix N ofthis document. In each case, we indicate the values of P, k*P, and (k+1)*P, where P is a fixed multiple (here: 2019) of the base point of the curve in question and where the private key k is the integer k 62662039304523906689788124833289384446202946474440057655160773695 63756342505410402166230018620066482794080866641616932013327623579 01952 (=0xdcb3bbb9 e42d7aca fe62052d 902123c7 0872b984 4c1e199f 7c5d37bd 1171102b c20a6352 d9c91886 29b685de 51441e84 3afe2665 5251aa80).In the examples below, each curve point is represented using the compressed point representation (see Appendix I.8), but where (for historical reasons) the parity bit t of the compressed curve point in question is represented by 0x00 or 0x80 depending on whether t=0 or t=1, respectively. Notice that this representation corresponds to the compressed point representation of Appendix I.8, but with the bit-ordering in the 1-octet representation of t reversed. (Note that this puts the parity bit t in the leftmost bit position of the octet string if one follows the MSB/msb representation conventions.) For simplicity, this representation is again called the "squeezed" representation, although each point is now represented as a 57-octet string and, thereby, one octet longer than the tight representation of elements of GF(p). As before, the ordering convention (see Appendix I.7) depends on the underlying curve model in question. Here, points of a Weierstrass curve are represented in tight MSB/msb- order, points of a Montgomery curve in tight LSB/msb-order, and points of a twisted Edwards curve in tight LSB/lsb-order. For points that are a public key, the corresponding private keys are represented as 56-octet strings, using the same ordering conventions as with the public keys. For affine points, we also give the tight representation of each of its coordinates (as 56-octet strings), using the same ordering conventions as used with the squeezed point representation. For further details, see the examples themselves.O.1. Example with Curve448 Pm=(u, v), k*Pm=(u1, v1), and (k+1)*Pm=(u2, v2) with Curve448: u 53298594738299085772373536080133483236673782578895339676785179923 90764298300090102709453866054695061082746243636045110750296444932 27715 (=0xbbb91ba3 b0ef74c3 214394b4 d8f0d32d c4a92193 5f573009 39fd86a3 8d54be2a 4d63380b 692381bb ed7339fd dca7b0cd a80166fe 18c086c3). v 30578727850066757341435137807347775064915058999485530015946871157 86794631407274936870618580714107931661730999350222644894729285604 97149 (=0x6bb38e82 8d52337f 6f0395ef dc16c776 52162f5e 309112ae fc7401bf 0cfb0499 eb1ed555 bf507ebc c33b4753 2d6dc6c5 d68dea1c c1e4c1fd). u1 64579461799301726935877447646800238443923683299745374127971411973 12515295161791889743228049222279188968365877164188075095074418806 82513 (=0xe37497bf 9f704689 54ec6537 cbbe91d0 3ffcdcdb 8b707253 a2212cdb e020ba9a 0bf65a1d 5d9a128a f85c63a2 79a00139 7aca56db 15335011). v1 55735504615964066386264989698774850924544182484936624265048483231 35693859362627880184586282439234602798023594054611737412667543758 11547 (=0xc44e5e0f 2c254d23 1dc082db 77175e8c fd37793c 22ebe200 77905a5f 750b3c9f 4a95d4d5 4e1a1e54 d2d31689 4249252d 0c8b1c45 1c1481db). u2 32685564331119553171673802371596819258307818641496728161547328225 07595618587323619256769558535630624960575212644680149034661008254 8876 (=0x0b831eca 9c6215b0 5d830361 4013732f 7a9dd07f ebb9441e 49129264 eb724f44 dc53671c ffabb9ee 0c02aa74 b083cd82 a821a4cf 6f6d8c8c). v2 57682103223585233918507344950062950306770296215271320612937204938 77499282103483092990510136901415757273082719665657484294344333591 20741 (=0xcb2988a4 6e37f9a9 a7a1255b 2fd2eea9 82308e7c eb8e18b8 2252175f fd416a10 5984c6b8 36470e48 31879293 8f6139c6 f96164cb 14010965). As suggested in Appendix C.2, the v-coordinate of k*Pm can be indirectly computed from the u-coordinates of Pm, k*Pm, and (k+1)*Pm, and the v-coordinate of Pm, which allows computation of the entire point k*Pm (and not just its u-coordinate) if k*Pm is computed using the Montgomery ladder (as, e.g., [RFC7748] recommends), since that algorithm computes both u1 and u2 and the v-coordinate of the point Pm may be available from context. The representation of k and the compressed representations of Pm and k*Pm in tight LSB/msb-order are given by repr(k) 0x80aa5152 6526fe3a 841e4451 de85b629 8618c9d9 52630ac2 2b107111 bd375d7c 9f191e4c 84b97208 c7232190 2d0562fe ca7a2de4 b9bbb3dc; repr(Pm) 0xc386c018 fe6601a8 cdb0a7dc fd3973ed bb812369 0b38634d 2abe548d a386fd39 0930575f 9321a9c4 2dd3f0d8 b4944321 c374efb0 a31bb9bb 80; repr(k*Pm) 0x11503315 db56ca7a 3901a079 a2635cf8 8a129a5d 1d5af60b 9aba20e0 db2c21a2 5372708b dbdcfc3f d091becb 3765ec54 8946709f bf9774e3 80, where the leftmost bit of the rightmost octet indicates the parity of the v-coordinate of the point of Curve448 in question (which, in this case, are both one, since v and v1 are odd). See Appendix H.2 and Appendix I for further detail on (squeezed) point compression. The scalar representation and (squeezed) point representation illustrated above are consistent with the representations specified in [RFC7748], except that in [RFC7748] only an affine point's u-coordinate is represented (i.e., the v-coordinate of any point is always implicitly assumed to have an even value) and that the representation of the point at infinity is not specified. (Note that due to the bit-size of the prime p, the lossless representation requires an additional octet compared to the lossy representation without v-coordinate.) Another difference is that [RFC7748] allows non-unique representations of some elements of GF(p), whereas our representation conventions do not (since tight). A randomized representation (t1, t2) of the point k*Pm in tight LSB/ msb order is given by t1 642695971489808425948939115432957219707501931105169269237 122551860533279049805112466411050091592893048844749561382 909707113070546618079 (=0xdf86cb83 ae1ca6e6 da6afbaf afbb2fc0 606a136f 80eea078 c868a5d7 7e638d09 99518385 65250cf1 9c034f96 1fa28f54 f3016600 68335de2) t2 569275737967591640709387827593956375775147481657775744720 460881642951497067363381071471046477130052706607411985560 522861593611384288817 (=0x3176361c 580a7bcd d7880d84 aba10bc6 57010328 afb728cc 2016461b 246bef46 0eb4bb04 8c1a3616 c3f74a56 3cc1790f 6472256b ca3481c8), where this representation is defined in Appendix K.5 and uses the mapping of Appendix K.3.2 with the default square root function. O.2. Example with Ed448 Pe=(x, y), k*Pe=(x1, y1), and (k+1)*Pe=(x2, y2) with Ed448: x 12711234107145442394649604543297947887906244696692372551963816418 93066253979844478364753304240794498368174540810674220788120782656 62747 (=0x2cc52fd1 6370554f 00c0f73f 64bda240 f5950177 d9033f6d 74acd12d 68c79a51 315f556f 240973f9 e5f71ed7 9314ee9d c87f0b1b bcc0fd1b). y 69251010954633529003803699627438795111055087299023774963200632446 22677618700964599963790149315020469517869703738619380660774687159 85238 (=0xf3e8bb95 c9675fd0 0c388fc5 e96cfbc7 3c19d945 76849979 34c4ab60 73c4a763 c2a89bac d3879838 f4de11a3 3a4710c2 396dea1d cc012956). x1 69268794439088733926883958090256942256857349796922332363888137509 71700910417786272464007666020220956482611896297610096130434552586 39205 (=0xf3f8c472 ca2e730b 05cc9092 f9d40956 029113e3 e92c2d55 76406db2 c2903721 62f43371 1c0ec80c f8d7222d 1d701467 9da18531 0fb5bb65). y1 50516707418203531159001223293623288296803299598968490915066154362 78541820739332329525138363312119838075438487384161435963107103409 09734 (=0xb1eccbfc 5f5f92e8 d9129d14 b721c524 96fc1b1f a4c17c5f e4979b0c 763f34ba 91299376 d2499220 19b05f56 c3bb6b5d ac988271 287d7aa6). x2 67287262124444231243108222498849910362455590990935326363062127166 04126947894055981270997819628982374416022607672923451356182938105 87868 (=0xecfe1a4f a4cd7e2f 19afcf16 1ce2198f 0a850beb 41afa209 94741609 5b1a858a 8e9548f5 011d188e d50484d3 119103f6 8bcd5ba2 a6e3e8dc). y2 13744276256057290540518554008940700979716578667786691114397525367 92684542875757407063179870154307882588988293167000249160114881659 30341 (=0x3068a338 4016ebfd a229ac73 b5c30bba ff67e183 71d1185f 19dfbbee 28478baf 9034ebad 51407f01 35162743 c2c234bc 2d484c13 552ea565). The representation of k and the compressed representations of Pe and k*Pe in tight LSB/lsb-order are given by repr(k) =0x01558a4a a6647f5c 2178228a 7ba16d94 6118939b 4ac65043 d4088e88 bdecba3e f9987832 219d4e10 e3c48409 b4a0467f 535eb427 9dddcd3b; repr(Pe) =0x6a948033 b857b69c 4308e25c c5887b2f 1c19e1cb 35d91543 c6e523ce 06d5232c 9e99216e a29b983c e3df3697 a3f11c30 0bfae693 a9dd17cf 01; repr(k*Pe) =0x655ebe14 8e411935 bad6ddc3 6afa0d98 0449924b 6ec99489 5d2cfc6e 30d9e927 fa3e8325 f8d83f69 24a384ed 28b9489b 1749fafa 3fd3378d 01, where the rightmost bit of the rightmost octet indicates the parity of the x-coordinate of the point of Ed448 in question (which, in this case, are both one, since x and x1 are odd). See Appendix H.3 and Appendix I for further detail on (squeezed) point compression. The scalar representation and (squeezed) point representation illustrated above are fully consistent with the representations specified in [RFC8032]. Note that, contrary to [RFC7748], [RFC8032] requires unique representations of all elements of GF(p). A randomized representation (t1, t2) of the point k*Pe in tight LSB/ lsb order is given by t1 397357047759003459380102071532091085834125520561197668989 747600577137881485970346806080038194336473483709104865191 806326006691504231547 (=0xde295d0e 5efceb9b f43967ca be45a54b a1f75bdd a4b1b1b3 b24a8d1d f2056329 e506867e c968aa8b 866017e4 f0cbc343 2cf8e7fa 0b202fd1) t2 711800301530600330791068062467600183663589340593884950808 136091389056251997893995894309660827763434071897306280320 151044063120296064809 (=0x94ecb72a 069a5322 e62d9357 c49d5664 1c351611 d1f361a8 cbb8a12c f410e821 4fbe8e02 8d85d404 399b4c7c 5a6a72ce deef7b08 96302d5f), where this representation is defined in Appendix K.5 and uses the mapping of Appendix K.3.3 with the default square root function and underlying isomorphic mapping between Ed448 and Curve448 of Appendix M.2. O.3. Example with Wei448 Pw=(X, Y), k*Pw=(X1, Y1), and (k+1)*Pw=(X2, Y2) with Wei448: X 29070637261778856087396075817199998758219070555984737667402173284 55389871077654193754799253725773241315783295429899652880118118204 91344 (=0x6663c64e 5b9a1f6d cbee3f5f 839b7dd8 6f53cc3e 0a01dab3 e4a8314e 8d54be2a 4d63380b 692381bb ed7339fd dca7b0cd a80166fe 18c15250). Y 30578727850066757341435137807347775064915058999485530015946871157 86794631407274936870618580714107931661730999350222644894729285604 97149 (=0x6bb38e82 8d52337f 6f0395ef dc16c776 52162f5e 309112ae fc7401bf 0cfb0499 eb1ed555 bf507ebc c33b4753 2d6dc6c5 d68dea1c c1e4c1fd). X1 40351504322781497250899987383866753965468971276834772118588405333 77140867939355980788573436893357369201402928958042617224896092079 46142 (=0x8e1f426a 4a1af133 ff970fe2 76693c7a eaa78786 361b1cfe 4ccbd786 e020ba9a 0bf65a1d 5d9a128a f85c63a2 79a00139 7aca56db 15341b9e). Y1 55735504615964066386264989698774850924544182484936624265048483231 35693859362627880184586282439234602798023594054611737412667543758 11547 (=0xc44e5e0f 2c254d23 1dc082db 77175e8c fd37793c 22ebe200 77905a5f 750b3c9f 4a95d4d5 4e1a1e54 d2d31689 4249252d 0c8b1c45 1c1481db). X2 51724471386152414687122300763026650882740205909970876834920746101 21508416303604179834986180511406702029983417676758930643822754281 77944 (=0xb62dc975 470cc05b 082dae0b eabe1dda 25487b2a 9663eec8 f3bd3d0e eb724f44 dc53671c ffabb9ee 0c02aa74 b083cd82 a821a4cf 6f6e5818). Y2 57682103223585233918507344950062950306770296215271320612937204938 77499282103483092990510136901415757273082719665657484294344333591 20741 (=0xcb2988a4 6e37f9a9 a7a1255b 2fd2eea9 82308e7c eb8e18b8 2252175f fd416a10 5984c6b8 36470e48 31879293 8f6139c6 f96164cb 14010965). The representation of k and the compressed representations of Pw and k*Pw in tight MSB/msb-order are given by repr(k) =0xdcb3bbb9 e42d7aca fe62052d 902123c7 0872b984 4c1e199f 7c5d37bd 1171102b c20a6352 d9c91886 29b685de 51441e84 3afe2665 5251aa80; repr(Pw) =0x80 6663c64e 5b9a1f6d cbee3f5f 839b7dd8 6f53cc3e 0a01dab3 e4a8314e 8d54be2a 4d63380b 692381bb ed7339fd dca7b0cd a80166fe 18c15250; repr(k*Pw) =0x80 8e1f426a 4a1af133 ff970fe2 76693c7a eaa78786 361b1cfe 4ccbd786 e020ba9a 0bf65a1d 5d9a128a f85c63a2 79a00139 7aca56db 15341b9e, where the leftmost bit of the leftmost octet indicates the parity of the Y-coordinate of the point of Wei448 in question (which, in this case, are both one, since Y and Y1 are odd). See Appendix H.1 and Appendix I for further detail on (squeezed) point compression. The scalar representation is consistent with the representations specified in [SEC1]; the (squeezed) point representation illustrated above is "new". For completeness, we include a SEC1-consistent representation of the point Pw in affine format and in compressed format below. The SEC1-compliant affine representation of the point Pw in tight MSB/msb-order is given by aff(Pw) =0x6663c64e 5b9a1f6d cbee3f5f 839b7dd8 6f53cc3e 0a01dab3 e4a8314e 8d54be2a 4d63380b 692381bb ed7339fd dca7b0cd a80166fe 18c15250 6bb38e82 8d52337f 6f0395ef dc16c776 52162f5e 309112ae fc7401bf 0cfb0499 eb1ed555 bf507ebc c33b4753 2d6dc6c5 d68dea1c c1e4c1fd, whereas the SEC1-compliant compressed representation of the point Pw in tight MSB/msb-order is given by compr(Pw) =0x03 6663c64e 5b9a1f6d cbee3f5f 839b7dd8 6f53cc3e 0a01dab3 e4a8314e 8d54be2a 4d63380b 692381bb ed7339fd dca7b0cd a80166fe 18c15250. The SEC1-compliant uncompressed format aff(Pw) of an affine point Pw corresponds to the right-concatenation of its X- and Y-coordinates, each in tight MSB/msb-order, prepended by the string 0x04, where the reverse procedure is uniquely defined, since elements of GF(p) have a unique fixed-size representation. The (squeezed) compressed format repr(Pw) corresponds to the SEC1-compliant compressed format by extracting the parity bit t from the leftmost bit of the leftmost octet of repr(Pw), and replacing this leftmost octet with 0x02 or 0x03, depending on whether t=0 or t=1, respectively, where the reverse procedure is uniquely defined. For further details, see [SEC1]. Note that, due to the bit-size of the prime p, the squeezed compressed format repr(Pw) and the SEC1-compliant compressed format compr(Pw) have the same size. A randomized representation (t1, t2) of the point k*Pw in tight MSB/ msb order is given by t1 655783099225353926682910498535559663266263823350679216116 172951494291735730803127024621397533084891460609898061397 896825551162064841608 (=0xe6f93655 2765628b accfe61c 7dc6a594 e06fb243 70195ded 74d88a53 fdedc2e8 077e0eff 62fa6a80 fa26b499 1f8796f5 21f2f03b f7e92b88) t2 357918241879339174086992006475988394618511927120788596330 507910466738735762660894972854331591097934354210992993787 402433561014235472657 (=0x7e0ffcaf 7add27bc bb723629 95fdedd0 8769f676 78d953bc 0d38f4f6 d63a59dc 00f2d55a a4db7dab 16364503 591edcb1 e095a577 43dea311), where this representation is defined in Appendix K.5 and uses the mapping of Appendix K.3.1 with the default square root function. O.4. Example with Wei448.-3 Pw3=(X, Y), k*Pw3=(X1, Y1), and (k+1)*Pw3=(X2, Y2) with Wei448.-3: X 54121793865726175505902038600562190720650456678500106168173285986 99999531708218763586616425010404811083912084906688745035466757984 48968 (=0xbe9f5a23 51709e13 d5ad50c2 a27be8ee 1b051970 2580d5c3 c2de7f75 3010635e d89ef547 8b67dc54 16d63c5b 1cc1116f dd453515 71b39b48). Y 14962282101304548030627835311887275833718070818965306362006934455 59168773381983445709256615887526455657034051121085622763637035580 12661 (=0x34b2dcc4 92d6a940 e6249c14 122d0ba4 5dc040e9 3f060d8f a65fa300 eb3cc969 25188b59 2d31039c f7a8e14a 48320a32 efe9b42b 986afef5). X1 18808295916646645825216065847266150404062470629833854840155953858 63091795696773741607659794828181692381790403935750135247605982648 6547 (=0x069fdd7c 2ec1ecbf d3cd0e27 1e8110c6 d2e478f2 aa393928 64a5511e da0b8dc7 3834fd57 b5ef8527 361a8176 c6da44ee 63701c0c f49d7d13). Y1 12212945244064471634326466576257313927639904273911210953487761656 77684161144865373513143868308041748047828401098060667767703779846 85920 (=0x2b03e68e b61581c4 9f977443 3e1ddc63 976f8f1d cdb185ee 9c53328d b425973d 359bbc09 468645c4 0996a2c7 fda561be acb4d0b5 745ab760). X2 58672976485086436102048679093716482249296622848351051568512020319 97872083950108489407370832733527154843728068195507632886574086695 12670 (=0xcea6f66e e741e7b3 ee50acd4 bd6eacbf 821fab72 bf5fe85b 8f614af9 04aff677 15e820b9 e4bcc159 f67a97f3 2c176d2c d9b7cdeb f753f3de). Y2 63661899992109030051219177516378471383513217472497460517936503629 79522840238080543318627428149249774773108009447466292682661818280 41265 (=0xe0394408 ed2b4efb b6b6ac7e bc815516 fdf31a6e d32db3f9 54cd8ac1 c7ddf0cc e7507688 a70f219a 57eef863 49003560 66747ca3 00105a31). The representation of k and the compressed representations of Pw3 and k*Pw3 in tight MSB/msb-order are given by repr(k) =0xdcb3bbb9 e42d7aca fe62052d 902123c7 0872b984 4c1e199f 7c5d37bd 1171102b c20a6352 d9c91886 29b685de 51441e84 3afe2665 5251aa80; repr(Pw3) =0x80 be9f5a23 51709e13 d5ad50c2 a27be8ee 1b051970 2580d5c3 c2de7f75 3010635e d89ef547 8b67dc54 16d63c5b 1cc1116f dd453515 71b39b48; repr(k*Pw3) =0x00 069fdd7c 2ec1ecbf d3cd0e27 1e8110c6 d2e478f2 aa393928 64a5511e da0b8dc7 3834fd57 b5ef8527 361a8176 c6da44ee 63701c0c f49d7d13, where the leftmost bit of the leftmost octet indicates the parity of the Y-coordinate of the point of Wei448.-3 in question (which, in this case, are one and zero, respectively, since Y is odd and Y1 is even). See Appendix H.1 and Appendix I for further detail on (squeezed) point compression. A randomized representation (t1, t2) of the point k*Pw3 in tight MSB/ msb order is given by t1 450833060883286904091316612794941178576639837300736625958 696097131313213727115363096930063001237631586932727905179 306828042642854311987 (=0x9ec9ba07 3fb2bb5e 9dbee995 067ce094 63601ecd 325f0930 aea79cb8 745fa71d 4caa37ee f04fab67 ab2de747 4ac0a025 830f4828 429cf833) t2 339205723274519707955026734148022275762579914421865223818 363622725164496136165251928391223173879522521195772276587 373445978123589677750 (=0x7778c1f9 9d900633 d161d7ea a963ddad e9101d3f f4f04710 623d2a51 6ca10133 3db9ccc3 86df9271 fbb72740 77f79dd1 9aed0bfb e3bc72b6), where this representation is defined in Appendix K.5 and uses the mapping of Appendix K.3.1 with the default square root function. O.5. Example with Edwards448 Pe1=(x, y), k*Pe1=(x1, y1), and (k+1)*Pe1=(x2, y2) with Edwards448: x 70320395893028961673046639985409870226249442701760956079298956688 26896600999421897751877804946848997852325361659665744287620719558 67733 (=0xf7acf3ca b79b29c2 aa44863d 9edaeca4 8c90ad84 e460df42 7dd9ab59 1bd8a844 07cb3419 59309b33 1e22bfa1 a2d37e10 e2e42a1f 170f0855). y 70628706854857281648863291487942166052137991441320055237644304464 58787938273165391464653528929699350754224243613996187734424074211 98773 (=0xf8c2f181 3bceee8e 085ecd70 d1b6aa4c ea9b95bd 8f36ab44 c79e9124 1ea625b7 f9f5ec57 89cc5af2 a2eb255a b252b874 509dc0d9 685841b5). x1 38125875041649701211705790554244713713134918749445854542272999596 74058986304488795258334978838809456257721496105769894880185657328 40277 (=0x864880b9 e1900c68 ba4a545a 6fe2b161 62dcc3b9 fa218e4b feba9828 5cee5193 f2c989f6 c3b94eb6 2914dce7 b4818e4d 8fc8d51f 05a13355). y1 11060653846610182753991162627427631707898421166839907726978369444 53337541552746428662176632660036639406375548888849623833963458813 1154 (=0x03e54af3 7f4cf5e6 5f1e2acd 5c4a4554 76adc652 b198ab2a 719e5aa9 ee749871 0193da82 ab6d000b f55836b1 0615653f 69514297 f4459f52). x2 15620503788413497044804517304021524439062374489822547728508337937 50606335270276724725939683726318058744384611584731365019896485812 8760 (=0x05806f71 95e85352 ef3960ac 1ff9cf6c 3c99e0ee 2e75edfc a133cafc 4a4b5fbf e4339859 c5fa123b 70ad2faf 7584ab9d 264540e7 7d560978). y2 40019917514121727463122190125689377890703570698337158159153510836 68442386516751945577468473801561261386285902585868517988506010293 44096 (=0x8cf44811 3cec6e07 d1bbe9f5 4062075c 6fec0ac5 31272dce 1f446aeb d895373d e312c18d 6a345755 2861e014 0cc23158 a46ace4c 9ca21b60). The representation of k and the compressed representations of Pe1 and k*Pe1 in tight LSB/lsb-order are given by repr(k) =0x01558a4a a6647f5c 2178228a 7ba16d94 6118939b 4ac65043 d4088e88 bdecba3e f9987832 219d4e10 e3c48409 b4a0467f 535eb427 9dddcd3b; repr(Pe1) =0xad821a16 9b03b90a 2e1d4a4d 5aa4d745 4f5a3391 ea37af9f eda46578 248979e3 22d56cf1 bda9d957 32556d8b 0eb37a10 717773dc 818f431f 01; repr(k*Pe1) =0x4af9a22f e9428a96 fca6a860 8d6c1aaf d000b6d5 415bc980 8e192e77 955a798e 54d5198d 4a63b56e 2aa2523a b35478fa 67af32fe cf52a7c0 01, where the rightmost bit of the rightmost octet indicates the parity of the x-coordinate of the point of Edwards448 in question (which, in this case, are both one, since x and x1 are odd). See Appendix H.3 and Appendix I for further detail on (squeezed) point compression. The scalar representation and (squeezed) point representation illustrated above are fully consistent with the representations specified in [RFC8032]. Note that, contrary to [RFC7748], [RFC8032] requires unique representations of all elements of GF(p). A randomized representation (t1, t2) of the point k*Pe1 in tight LSB/ lsb order is given by t1 125390048858887400104074787879402833851854739339836093733 734638776755983021034212058415891288350265701101219981698 849086128138510420407 (=0xed921f3d 6ea4e452 dd06e783 782cbeb3 c5847a79 d9e6b993 bd387cf5 feeddafe af8c038d f2732362 92724d37 273eedfc f2ab2499 98a79434) t2 365268494484253132875102676783560666625109899133767696106 602723958322248430160651314075127005631031993354968950936 71875730862008188281 (=0x9ebc28c0 86176a1a c7f0cf71 ca5f2a8f 908bb27b e85c0bbd 1641c052 e542f7d3 88e18886 5afdca32 8df45408 8b6da28c 0bc09d83 309ebb30), where this representation is defined in Appendix K.5 and uses the mapping of Appendix K.3.3 with the default square root function and underlying 4-isogenous mapping between Edwards448 and Curve448 of Appendix N.2. Author's Address Rene Struik Struik Security Consultancy Email: rstruik.ext@gmail.com