CoRE Working Group                                             M. Tiloca
Internet-Draft                                                   RISE SICS AB
Intended status: Standards Track                             G. Selander
Expires: December 30, 2018 April 25, 2019                                     F. Palombini
                                                             Ericsson AB
                                                                 J. Park
                                             Universitaet Duisburg-Essen
                                                           June 28,
                                                        October 22, 2018

           Group OSCORE - Secure group communication Group Communication for CoAP
                  draft-ietf-core-oscore-groupcomm-02
                draft-ietf-core-oscore-groupcomm-03

Abstract

   This document describes a mode for protecting group communication
   over the Constrained Application Protocol (CoAP).  The proposed mode
   relies on Object Security for Constrained RESTful Environments
   (OSCORE) and the CBOR Object Signing and Encryption (COSE) format.
   In particular, it is defined defines how OSCORE should be is used in a group communication
   setting, while fulfilling the same security requirements for request messages group
   requests and related response messages. responses.  Source authentication of all messages
   exchanged within the group is
   ensured, provided by means of digital signatures
   produced through private keys
   of by the sender endpoints and embedded in the protected CoAP messages.

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 December 30, 2018. April 25, 2019.

Copyright Notice

   Copyright (c) 2018 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.  Introduction  . . . . . . . . . . . . . . . . . . . . . . . .   3
     1.1.  Terminology . . . . . . . . . . . . . . . . . . . . . . .   4
   2.  OSCORE Security Context . . . . . . . . . . . . . . . . . . .   5
     2.1.  Management of Group Keying Material . . . . . . . . . . .   7
     2.2.  Wrap-Around of Partial IVs  . . . . . . . . . . . . . . .   8
   3.  The COSE Object . . . . . . . . . . . . . . . . . . . . . . .   7
     3.1.  Example: Request   8
   4.  OSCORE Header Compression . . . . . . . . . . . . . . . . . .   9
     4.1.  Encoding of the OSCORE Option Value . . . . . . . . . . .   9
     3.2.  Example: Response
     4.2.  Encoding of the OSCORE Payload  . . . . . . . . . . . . .  10
     4.3.  Examples of Compressed COSE Objects . . . . . . . . . . .  10
   4.
   5.  Message Processing Binding, Sequence Numbers, Freshness and Replay
       Protection  . . . . . . . . . . . . . . . . . . . . .  10
     4.1.  Protecting the Request . . . .  11
     5.1.  Synchronization of Sender Sequence Numbers  . . . . . . .  12
   6.  Message Processing  . . . . . . .  10
     4.2.  Verifying . . . . . . . . . . . . . .  12
     6.1.  Protecting the Request  . . . . . . . . . . . . . . . . . .  11
     4.3.  Protecting  13
     6.2.  Verifying the Response Request . . . . . . . . . . . . . . . . .  11
     4.4.  Verifying .  13
     6.3.  Protecting the Response . . . . . . . . . . . . . . . . .  11
   5.  Synchronization of Sequence Numbers  13
     6.4.  Verifying the Response  . . . . . . . . . . . . .  12
   6. . . . .  14
   7.  Responsibilities of the Group Manager . . . . . . . . . . . .  12
   7.  14
   8.  Security Considerations . . . . . . . . . . . . . . . . . . .  13
     7.1.  15
     8.1.  Group-level Security  . . . . . . . . . . . . . . . . . .  14
     7.2.  15
     8.2.  Uniqueness of (key, nonce)  . . . . . . . . . . . . . . .  14
     7.3.  Collision  16
     8.3.  Management of Group Identifiers Keying Material . . . . . . . . . . .  16
     8.4.  Update of Security Context and Key Rotation . .  14
   8.  IANA Considerations . . . . .  17
     8.5.  Collision of Group Identifiers  . . . . . . . . . . . . .  17
   9.  IANA Considerations . . . . .  15
   9.  Acknowledgments . . . . . . . . . . . . . . . .  18
     9.1.  OSCORE Flag Bits Registry . . . . . . .  15 . . . . . . . . .  18
   10. References  . . . . . . . . . . . . . . . . . . . . . . . . .  15  18
     10.1.  Normative References . . . . . . . . . . . . . . . . . .  15  18
     10.2.  Informative References . . . . . . . . . . . . . . . . .  16  19
   Appendix A.  Assumptions and Security Objectives  . . . . . . . .  18  20
     A.1.  Assumptions . . . . . . . . . . . . . . . . . . . . . . .  18  20
     A.2.  Security Objectives . . . . . . . . . . . . . . . . . . .  20  21
   Appendix B.  List of Use Cases  . . . . . . . . . . . . . . . . .  21  22
   Appendix C.  Example of Group Identifier Format . . . . . . . . .  23  24
   Appendix D.  Set-up of New Endpoints  . . . . . . . . . . . . . .  24
     D.1.  Join Process  25
   Appendix E.  Examples of Synchronization Approaches . . . . . . .  26
     E.1.  Best-Effort Synchronization . . . . . . . . . . . . . . .  24
     D.2.  Provisioning and Retrieval of Public Keys  26
     E.2.  Baseline Synchronization  . . . . . . . .  27
     D.3.  Group Joining Based on the ACE Framework . . . . . . . .  29
   Appendix E.  Examples of Synchronization Approaches  26
     E.3.  Challenge-Response Synchronization  . . . . . . .  29
     E.1.  Best-Effort Synchronization . . . .  27
   Appendix F.  No Verification of Signatures  . . . . . . . . . . .  29
     E.2.  Baseline Synchronization  28
   Appendix G.  Document Updates . . . . . . . . . . . . . . . .  30
     E.3.  Challenge-Response Synchronization . .  29
     G.1.  Version -02 to -03  . . . . . . . . .  30

   Appendix F.  No Verification of Signatures . . . . . . . . . .  29
     G.2.  Version -01 to -02  .  32
   Appendix G.  Document Updates . . . . . . . . . . . . . . . . . .  32
     G.1.  30
     G.3.  Version -01 -00 to -02 -01  . . . . . . . . . . . . . . . . . . .  32
     G.2.  Version -00 to -01  31
   Acknowledgments . . . . . . . . . . . . . . . . . . . . . .  33 . . .  31
   Authors' Addresses  . . . . . . . . . . . . . . . . . . . . . . .  34  32

1.  Introduction

   The Constrained Application Protocol (CoAP) [RFC7252] is a web
   transfer protocol specifically designed for constrained devices and
   networks [RFC7228].

   Group communication for CoAP [RFC7390] addresses use cases where
   deployed devices benefit from a group communication model, for
   example to reduce latencies and latencies, improve performance. performance and reduce bandwidth
   utilisation.  Use cases include lighting control, integrated building
   control, software and firmware updates, parameter and configuration
   updates, commissioning of constrained networks, and emergency
   multicast (see Appendix B).  Furthermore, [RFC7390] recognizes the
   importance to introduce a secure mode for CoAP group communication.
   This specification defines such a mode.

   Object Security for Constrained RESTful Environments
   (OSCORE)[I-D.ietf-core-object-security] describes a security protocol
   based on the exchange of protected CoAP messages.  OSCORE builds on
   CBOR Object Signing and Encryption (COSE) [RFC8152] and provides end-
   to-end encryption, integrity, and replay protection and binding of
   response to request between a sending
   endpoint sender and a receiving endpoint possibly involving intermediary
   endpoints. receipient, also in the
   presence of intermediaries.  To this end, a CoAP message is protected
   by including its payload (if any), certain options, and header fields
   in a COSE object, which finally replaces the authenticated and encrypted
   fields in the protected message.

   This document describes group defines Group OSCORE, providing end-to-end security of
   CoAP messages exchanged between members of a group. group, and preserving
   independence of transport layer.  In particular, the described
   approach defines how OSCORE should be used in a group communication
   setting, so that end-to-end security is assured by using in the same security method. way as
   OSCORE for unicast communication.  That is, end-to-end security is assured
   provided for (multicast) CoAP multicast requests sent by a client
   endpoints to the group group,
   and for related CoAP responses sent as reply by multiple server endpoints. servers.  Group
   OSCORE provides source authentication of all CoAP messages exchanged
   within the group, by means of digital signatures produced through
   private keys of sender devices and embedded in the protected CoAP
   messages.

   As in OSCORE, it is still possible to simultaneously rely on DTLS
   [RFC6347] to protect hop-
   by-hop hop-by-hop communication between a sender endpoint and a
   proxy (and vice versa), and between a proxy and a recipient endpoint (and vice
   versa).  Note that DTLS cannot be used to secure messages sent over
   multicast.

1.1.  Terminology

   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.

   Readers are expected to be familiar with the terms and concepts
   described in CoAP [RFC7252] including "endpoint", "client", "server",
   "sender" and "recipient"; group communication for CoAP [RFC7390];
   COSE and counter signatures [RFC8152].

   Readers are also expected to be familiar with the terms and concepts
   for protection and processing of CoAP messages through OSCORE, such
   as "Security Context" and "Master Secret", defined in
   [I-D.ietf-core-object-security].

   Terminology for constrained environments, such as "constrained
   device", "constrained-node network", is defined in [RFC7228].

   This document refers also to the following terminology.

   o  Keying material: data that is necessary to establish and maintain
      secure communication among endpoints.  This includes, for
      instance, keys and IVs [RFC4949].

   o  Group: a set of endpoints that share group keying material and
      security parameters (Common Context of the group's Security Context, see Section 2).  That is, the  The term
      group used in this specification refers thus to a "security
      group", not to be confused with network/
      multicast groups network/multicast group or
      application groups. group.

   o  Group Manager (GM): entity responsible for a set of OSCORE groups. group.  Each endpoint
      in a group securely communicates securely with the respective GM, which is not
      neither required to be an actual group member and nor to take part in
      the group communication.  The full list of responsibilities of the
      Group Manager is provided in Section 6. 7.

   o  Silent server: member of a group that never replies back after
      receiving request messages. responds to requests.
      Note that a silent server can act as a client, the two roles are
      independent.

   o  Group ID: group Identifier (Gid): identifier assigned to the group.  Group IDs are
      Identifiers should be unique within the set of groups of a same given
      Group Manager.

   o  Endpoint ID: Sender ID of the endpoint, as defined Manager, in
      [I-D.ietf-core-object-security].  An Endpoint ID is provided order to an
      endpoint upon joining a group, is valid only within that group,
      and is unique within the same group.  Endpoints which avoid collisions.  In case they are
      configured only as silent servers do not have an Endpoint ID.
      not, the considerations in Section 8.5 apply.

   o  Group request: CoAP request message sent by a client endpoint in the group
      to all server endpoints servers in that group.

   o  Source authentication: evidence that a received message in the
      group originated from a specifically specific identified group member.  This
      also provides assurances assurance that the message was not tampered with by
      anyone, be it a different legitimate group member or by an endpoint
      which is not a non-group group member.

2.  OSCORE Security Context

   To support group communication secured with OSCORE, each endpoint
   registered as member of a group maintains a Security Context as
   defined in Section 3 of [I-D.ietf-core-object-security]. [I-D.ietf-core-object-security], extended as
   defined below.  Each endpoint in a group stores: makes use of:

   1.  one Common Context, shared by all the endpoints in the a given group.
       In particular:

       *  All the endpoints in the group agree on the same COSE AEAD
          algorithm.

       *  The ID Context parameter stores contains the Group ID Gid of the group, which
          is used to retrieve the Security Context for processing
          messages intended to the group's endpoints of the group (see
          Section 4). 6).  The choice of the Group ID for a given group's Security
          Context Gid is application specific.
          An example of specific formatting of the Group ID that would follow this
          specification Gid is given in
          Appendix C.  It is the role of the  The application needs to specify how to handle
          possible collisions. collisions between Gids, see Section 8.5.

       *  A new parameter Counter Signature Algorithm is included, and
          its included.  Its
          value identifies the digital signature algorithm used for source
          authenticating messages sent within the group, by means of to
          compute a counter signature on the COSE object (see
          Section 4.5 of [RFC8152]). [RFC8152]) which provides source authentication
          within the group.  Its value is immutable once the Common
          Context is established.  All the
          endpoints in the group agree on the same counter signature
          algorithm.  The list of supported signature algorithms is part
          of the group communication policy and MUST include the EdDSA signature algorithm ed25519 [RFC8032].
          [RFC8032] is mandatory to implement.

   2.  one Sender Context, unless the endpoint is configured exclusively
       as silent server.  The Sender Context is used to secure outgoing
       group
       messages and is initialized according to Section 3 of
       [I-D.ietf-core-object-security], once the endpoint has joined the
       group.  In practice, the symmetric keying material in the  The Sender Context of the sender a given endpoint is shared with matches the
       corresponding Recipient Context in all the recipient endpoints that have received group messages receiving a
       protected message from that same sender endpoint.  Besides, in addition to
       what is defined in [I-D.ietf-core-object-security], the Sender
       Context stores also the endpoint's public-private key pair. private key.

   3.  one Recipient Context for each distinct endpoint from which group
       messages are received, used to process such incoming messages.  The
       recipient endpoint creates a new may generate the Recipient Context upon receiving an
       incoming message from another endpoint in the group for the first
       time (see Section 4.2 6.2 and Section 4.4).  In
       practice, the symmetric keying material in a given 6.4).  Each Recipient Context of
       matches the recipient endpoint is shared with Sender Context of the associated
       sender endpoint from which group protected
       messages are received.  Besides, in addition to what is defined
       in [I-D.ietf-core-object-security], each Recipient Context stores
       also the public key of the associated other endpoint from which
       group
       messages are received.

   The table in Figure 1 overviews the new information included in the
   OSCORE Security Context, with respect to what defined in Section 3 of
   [I-D.ietf-core-object-security].

         +---------------------------+-----------------------------+
         |      Context portion      |       New information       |
         +---------------------------+-----------------------------+
         |                           |                             |
         |      Common Context       | Counter signature algorithm |
         |                           |                             |
         |      Sender Context       | Endpoint's own private key  |
         |                           |                             |
         |      Sender Context       | Endpoint's own public key   |
         |                           |                             |
         |  Each Recipient Context   | Public key of the           |
         |                           | associated other endpoint   |
         |                           |                             |
         +---------------------------+-----------------------------+

            Figure 1: Additions to the OSCORE Security Context

   Upon receiving a secure CoAP message, a recipient endpoint relies on uses the sender endpoint's sender's
   public key, in order to verify the counter signature conveyed in of the COSE Object.
   Object (see Section 3).

   If not already stored in the Recipient Context associated to the
   sender endpoint,
   sender, the recipient endpoint retrieves the public key from
   a the Group
   Manager, which collects public keys upon endpoints' joining, acts as
   trusted key repository.  In such a case, repository and ensures the correct binding association between
   the sender endpoint and the retrieved public key must be
   assured, and the identifier of the sender, for instance by
   means of public key certificates.

   It is RECOMMENDED that the Group Manager collects public keys and
   provides them to group members upon request as described in
   [I-D.tiloca-ace-oscoap-joining], where the join process is based on
   the ACE framework for Authentication and Authorization in constrained
   environments [I-D.ietf-ace-oauth-authz].  Further
   discussion details about how
   public keys can be handled and retrieved in the group is provided in Appendix D.2.

   The out of the
   scope of this document.

   An endpoint receives its own Sender Key/IV stored in ID from the Group Manager upon
   joining the group.  That Sender Context ID is valid only within that group,
   and is unique within the Recipient
   Keys/IVs stored group.  An endpoint uses its own Sender ID
   (together with other data) to generate unique AEAD nonces for
   outgoing messages, as in [I-D.ietf-core-object-security].  Endpoints
   which are configured only as silent servers do not have a Sender ID.

   The Sender/Recipient Keys and the Recipient Contexts Common IV are derived according to
   the same scheme defined in Section Sections 3.2 and 5.2 of
   [I-D.ietf-core-object-security].  The mandatory-to-implement HKDF and
   AEAD algorithms for Group OSCORE are the same as in
   [I-D.ietf-core-object-security].

2.1.  Management of Group Keying Material

   The approach described in this specification should take into account
   the risk of compromise of group members.

   In particular, the adoption
   of key management schemes order to establish a new Security Context in a group, a new Group
   Identifier (Gid) for secure revocation that group and renewal a new value for the Master Secret
   parameter MUST be distributed.  An example of
   Security Contexts and Gid format supporting
   this operation is provided in Appendix C.  Then, each group member
   re-derives the keying material should be considered. stored in its own Sender Context and
   Recipient Contexts as described in Section 2, using the updated Gid.

   Consistently with the security assumptions in Appendix A.1, it is
   RECOMMENDED to adopt a group key management scheme, and securely
   distribute a new value for the Gid and for the Master Secret
   parameter of the group's Security Context, before a new joining
   endpoint is added to the group or after a currently present endpoint
   leaves the group.  This is necessary in order to preserve backward security
   and forward security in the group.

   In particular, a new Group Identifier (Gid) for that group and a new
   value for group, if the application requires it.

   The specific approach used to distribute the new Gid and Master
   Secret parameter must also be distributed.  An
   example to the group is out of the scope of Group Identifier format supporting this operation document.
   However, it is
   provided in Appendix C.  Then, each RECOMMENDED that the Group Manager supports the
   distribution of the new Gid and Master Secret parameter to the group member re-derives
   according to the
   keying material stored Group Rekeying Process described in
   [I-D.tiloca-ace-oscoap-joining].

2.2.  Wrap-Around of Partial IVs

   A client can eventually experience a wrap-around of its own Sender Context and Recipient
   Contexts
   Sequence Number, which is used as described Partial IV in Section 2, using outgoing requests and
   incremented after each request.  When this happens, the updated Group
   Identifier.

   Especially OSCORE
   Security Context MUST be renewed in dynamic, large-scale, groups where endpoints can join
   and leave at any time, it is important that the considered group key
   management scheme is efficient and highly scalable group, in order to avoid
   reusing nonces with the same keys.

   Therefore, when experiencing a wrap-around of its own Sender Sequence
   Number, a group
   size, member MUST NOT transmit further group requests until
   a new OSCORE Security Context has been derived.  In particular, the
   endpoint SHOULD inform the Group Manager of the occurred wrap-around,
   in order to limit the impact on performance due to trigger a prompt renewal of the OSCORE Security
   Context and keying material update. Context.

3.  The COSE Object

   When creating a protected CoAP message, an endpoint

   Building on Section 5 of [I-D.ietf-core-object-security], this
   section defines how to use COSE [RFC8152] to wrap and protect data in
   the group
   computes the COSE object using original message.  OSCORE uses the untagged COSE_Encrypt0
   structure
   [RFC8152] as defined in Section 5 of [I-D.ietf-core-object-security], with an Authenticated Encryption with Additional Data
   (AEAD) algorithm.  For Group OSCORE, the following modifications. modifications
   apply:

   o  The value of the 'kid' parameter external_aad in the 'unprotected' field of
      response messagess SHALL be set to the Endpoint ID of the endpoint
      transmitting the message, i.e. the Sender ID.

   o  The 'unprotected' field SHALL additionally include the following
      parameter:

      *  CounterSignature0 : its value is set to the counter signature
         of the COSE object, computed by the endpoint by means of its
         own private key as described in Section 4.5 of [RFC8152].  The
         presence of this parameter is explicitly signaled, by using the
         reserved sixth least significant bit of the first byte of flag
         bits in the value of the OSCORE Option (see Section 6.1 of
         [I-D.ietf-core-object-security]).

   o  The Additional Authenticated Data (AAD) considered to compute the
      COSE object is
      extended with the counter signature algorithm used to protect group sign
      messages.  In particular, compared with reference to Section 5.4 of
      [I-D.ietf-core-object-security], the 'algorithms' array in the external_aad SHALL
      aad_array MUST also include 'alg_countersign', which contains the
      Counter Signature Algorithm from the Common Context (see
      Section 2).  This external_aad structure is used both for the
      encryption process producing the ciphertext (see Section 5.3 of
      [RFC8152]) and for the signing process producing the counter
      signature, as defined below.

 external_aad = bstr .cbor aad_array

 aad_array = [
    ...
    oscore_version : uint,
    algorithms : [alg_aead : int / tstr , alg_countersign : int / tstr],
    ...
    request_kid : bstr,
    request_piv : bstr,
    options : bstr
 ]

   o  The OSCORE compression defined in Section 6 of
      [I-D.ietf-core-object-security] is used, with the following
      additions for the encoding value of the OSCORE Option.

      *  The fourth least significant bit of 'kid' parameter in the first byte 'unprotected' field of flag bits
         SHALL
      response messages MUST be set to 1, to indicate the presence Sender ID of the endpoint
      transmitting the message.  That is, unlike in

      [I-D.ietf-core-object-security], the 'kid' parameter for is always
      present in all messages, i.e. both group requests and responses.

      *

   o  The fifth least significant bit of the first byte of flag bits 'unprotected' field MUST be additionally include the following
      parameter:

      *  CounterSignature0 : its value is set to 1 for group requests, to indicate the presence counter signature
         of the 'kid context' parameter COSE object, computed by the sender using its own
         private key as described in Appendix A.2 of [RFC8152].  In
         particular, the OSCORE Sig_structure contains the external_aad as
         defined above and the ciphertext of the COSE_Encrypt0 object as
         payload.

4.  OSCORE Header Compression

   The OSCORE compression defined in Section 6 of
   [I-D.ietf-core-object-security] is used, with the following additions
   for the encoding of the OSCORE Option and the OSCORE Payload.

4.1.  Encoding of the OSCORE Option Value

   Analogously to [I-D.ietf-core-object-security], the value of the
   OSCORE option SHALL contain the OSCORE flag bits, the Partial IV
   parameter, the kid context parameter (length and value), and the kid
   parameter, with the following modifications:

   o  The first byte, containing the OSCORE flag MAY bits, has the following
      encoding modifications:

      *  The fourth least significant bit MUST be set to 1 in every
         message, to indicate the presence of the 'kid' parameter for
         all group requests and responses.  That is, unlike in
         [I-D.ietf-core-object-security], the 'kid' parameter is always
         present in all messages.

      *  The sixth fifth least significant bit of MUST be set to 1 for group
         requests, to indicate the first byte presence of flag bits
         is originally marked as reserved the 'kid context'
         parameter in
         [I-D.ietf-core-object-security] and its usage is defined the compressed COSE object.  The 'kid context' MAY
         be present in
         this specification.  This responses if the application requires it.  In
         such a case, the kid context flag MUST be set to 1.

      *  The sixth least significant bit is set to 1 if the
         'CounterSignature0' parameter is present, or to 0 otherwise.
         In order to ensure source authentication of group messages as
         described in this specification, this bit SHALL MUST be set to 1.

      *

   The flag bits are registered in the OSCORE Flag Bits registry
   specified in Section 13.7 of [I-D.ietf-core-object-security] and in
   Section 9.1 of this Specification.

   o  The 'kid context' value encodes the Group Identifier value (Gid)
      of the group's Security Context.

      *  The following q bytes (q given by the Counter Signature
         Algorithm specified in the Security Context) encode the value
         of the 'CounterSignature0' parameter including the counter
         signature of the COSE object.

      *

   o  The remaining bytes in the OSCORE Option value encode the value of
      the 'kid' parameter, which is always present both in group
      requests and in responses.

            0 1 2 3 4 5 6 7 <----------- n bytes -----------> <-- 1 byte -->
    +-+-+-+-+-+-+-+-+---------------------------------+--------------+ ------------>
           +-+-+-+-+-+-+-+-+----------------------------------+
           |0 0|1|h|1|  n  |       Partial IV (if any)        |  s (if any)  |
    +-+-+-+-+-+-+-+-+---------------------------------+--------------+
           +-+-+-+-+-+-+-+-+----------------------------------+

            <-- 1 byte --> <------ s bytes ------> <--------- q bytes --------->
    -----------------------+-----------------------------+-----------+
       kid context = Gid
           +--------------+-----------------------+-----------+
           |      CounterSignature0  s (if any)  |   kid context = Gid   |    kid    |
    -----------------------+-----------------------------+-----------+
           +--------------+-----------------------+-----------+

                       Figure 2: OSCORE Option Value

3.1.  Example: Request

   Request

4.2.  Encoding of the OSCORE Payload

   The payload of the OSCORE message SHALL encode the ciphertext of the
   COSE object concatenated with kid = 0x25, Partial IV = 5 the value of the CounterSignature0 (if
   present) of the COSE object, computed as in Appendix A.2 of
   [RFC8152].

4.3.  Examples of Compressed COSE Objects

   This section covers a list of OSCORE Header Compression examples for
   group requests and kid context = 0x44616c,
   assuming responses.  The examples assume that the
   COSE_Encrypt0 object is set (which means the CoAP message and
   cryptographic material is known).  Note that the examples do not
   include the full CoAP unprotected message or the full security
   context, but only the input necessary to the compression mechanism,
   i.e. the COSE_Encrypt0 object.  The output is the compressed COSE
   object as defined in Section 4 and divided into two parts, since the
   object is transported in two CoAP fields: OSCORE option and payload.

   The examples assume that the label for the new kid context defined in
   [I-D.ietf-core-object-security] has value 10.  COUNTERSIGN is the
   CounterSignature0 byte string as described in Section 3 and is 64
   bytes long in this example.  The ciphertext in this example is 14
   bytes long.

   1.  Request with ciphertext = 0xaea0155667924dff8a24e4cb35b9, kid =
       0x25, Partial IV = 5 and kid context = 0x44616c

   Before compression (96 bytes):

   [
   h'',
   { 4:h'25', 6:h'05', 10:h'44616c', 9:COUNTERSIGN },
   h'aea0155667924dff8a24e4cb35b9'
   ]

   After compression (85 bytes):

   Flag byte: 0b00111001 = 0x39

   Option Value: 39 05 03 44 61 6c COUNTERSIGN 25 (7 bytes + size of
    COUNTERSIGN) bytes)

   Payload: ae a0 15 56 67 92 4d ff 8a 24 e4 cb 35 b9 COUNTERSIGN
   (14 bytes)

3.2.  Example: Response bytes + size of COUNTERSIGN)

   1.  Response with ciphertext = 60b035059d9ef5667c5a0710823b, kid = 0x52.  COUNTERSIGN is the CounterSignature0 byte
   string as described in Section 3
       0x52 and is 64 bytes long in this
   example.  The ciphertext in this example is 14 bytes long. no Partial IV.

   Before compression (88 bytes):

   [
   h'',
   { 4:h'52', 9:COUNTERSIGN },
   h'60b035059d9ef5667c5a0710823b'
   ]

   After compression (80 bytes):

   Flag byte: 0b00101000 = 0x28

   Option Value: 28 COUNTERSIGN 52 (2 bytes + size of COUNTERSIGN) bytes)

   Payload: 60 b0 35 05 9d 9e f5 66 7c 5a 07 10 82 3b COUNTERSIGN
   (14 bytes)

4. bytes + size of COUNTERSIGN)

5.  Message Processing

   Each request message Binding, Sequence Numbers, Freshness and response message is protected Replay Protection

   The requirements and processed
   as specified in [I-D.ietf-core-object-security], with the
   modifications properties described in the following sections.  The following
   security objectives are fulfilled, as further discussed in
   Appendix A.2: data replay protection, group-level data
   confidentiality, source authentication, message integrity, and
   message ordering.

   Furthermore, endpoints in the group locally perform error handling
   and processing Section 7 of invalid messages according
   [I-D.ietf-core-object-security] also apply to the same principles
   adopted OSCORE used in [I-D.ietf-core-object-security].  However, a receiver
   endpoint MUST stop processing and silently reject any group
   communication.  In particular, group OSCORE provides message binding
   of responses to requests, which
   is malformed provides relative freshness of
   responses, and does not follow the format replay protection of requests.  More details about
   error processing for replay detection in group OSCORE are specified
   in Section 3,
   without sending back any error message.  This prevents servers from
   replying with multiple error messages to a client sending a group
   request, so avoiding the risk 6 of flooding this specification.  The mechanisms describing replay
   protection and possibly congesting the
   group.

4.1.  Protecting the Request

   A client transmits a secure freshness of Observe notifications do not apply to
   group request OSCORE, as described in Section 8.1
   of [I-D.ietf-core-object-security], with the following modifications.

   o  In step 2, the 'algorithms' array in the Additional Authenticated
      Data Observe is modified as described in Section 3.

   o  In step 4, the encoding not defined for group settings.

5.1.  Synchronization of Sender Sequence Numbers

   Upon joining the compressed COSE object is modified
      as described in Section 3.

4.2.  Verifying group, new servers are not aware of the Request

   Upon receiving Sender
   Sequence Number values currently used by different clients to
   transmit group requests.  This means that, when such servers receive
   a secure group request, request from a server proceeds as described
   in Section 8.2 of [I-D.ietf-core-object-security], with given client for the following
   modifications.

   o  In step 2, the decoding of the compressed COSE object first time, they
   are not able to verify if that request is modified
      as described in Section 3.  If the received Recipient ID ('kid')
      does fresh and has not match been
   replayed or (purposely) delayed.  The same holds when a server loses
   synchronization with any Recipient Context Sender Sequence Numbers of clients, for the retrieved Group
      ID ('kid context'), then the server creates instance
   after a new Recipient
      Context, initializes it according device reboot.

   The exact way to Section 3 of
      [I-D.ietf-core-object-security], address this issue is application specific, and includes the client's public
      key.

   o  In step 4, the 'algorithms' array in
   depends on the Additional Authenticated
      Data particular use case and its synchronization
   requirements.  The list of methods to handle synchronization of
   Sender Sequence Numbers is modified as described in Section 3.

   o  In step 6, the server also verifies the counter signature using
      the public key part of the client from group communication policy,
   and different servers can use different methods.

   Requests sent over Multicast must be Non-Confirmable (Section 8.1 of
   [RFC7252]), as overall most of the associated Recipient
      Context.

4.3.  Protecting messages sent within a group.
   Thus, senders should store their outgoing messages for an amount of
   time defined by the Response

   A server application and sufficient to correctly handle
   possible retransmissions.

   Appendix E describes three possible approaches that has received a secure group can be considered
   for synchronization of sequence numbers.

6.  Message Processing

   Each request may reply with a
   secure response, which message and response message is protected and processed
   as described specified in Section 8.3 of [I-D.ietf-core-object-security], with the following modifications.

   o  In step 2,
   modifications described in the 'algorithms' following sections.  The following
   security objectives are fulfilled, as further discussed in
   Appendix A.2: data replay protection, group-level data
   confidentiality, source authentication, message integrity, and
   message ordering.

   Furthermore, endpoints in the group locally perform error handling
   and processing of invalid messages according to the same principles
   adopted in [I-D.ietf-core-object-security].  However, a recipient
   MUST stop processing and silently reject any message which is
   malformed and does not follow the format specified in Section 3, or
   which is not cryptographically validated in a successful way.  Either
   case, the recipient MUST NOT send back any error message.  This
   prevents servers from replying with multiple error messages to a
   client sending a group request, so avoiding the risk of flooding and
   possibly congesting the group.

   As per [RFC7252][RFC7390], group requests sent over multicast must be
   Non-confirmable.  However, this does not prevent the acknowledgment
   of Confirmable group requests in non-multicast environments.

6.1.  Protecting the Request

   A client transmits a secure group request as described in Section 8.1
   of [I-D.ietf-core-object-security], with the following modifications.

   o  In step 2, the 'algorithms' array in the Additional Authenticated
      Data is modified as described in Section 3.

   o  In step 4, the encryption of the COSE object is modified as
      described in Section 3.  The encoding of the compressed COSE
      object is modified as described in Section 3.

4.4. 4.

6.2.  Verifying the Response Request

   Upon receiving a secure response message, the client group request, a server proceeds as described
   in Section 8.4 8.2 of [I-D.ietf-core-object-security], with the following
   modifications.

   o  In step 2, the decoding of the compressed COSE object is modified
      as described in follows
      Section 3. 4.  If the received Recipient ID ('kid') does not match
      with any Recipient Context for the retrieved Group
      ID Gid ('kid context'),
      then the client server creates a new Recipient Context, initializes it
      according to Section 3 of [I-D.ietf-core-object-security], and includes also
      retrieving the server's client's public key.

   o  In step 3, 4, the 'algorithms' array in the Additional Authenticated
      Data is modified as described in Section 3.

   o  In step 5, 6, the client server also verifies the counter signature using
      the public key of the server client from the associated Recipient
      Context.

5.  Synchronization of Sequence Numbers

   Upon joining the group, new servers are not aware of

6.3.  Protecting the sequence
   number values currently used by different clients to transmit Response

   A server that has received a secure group
   requests.  This means that, when such servers receive a secure group
   request from a given client for the first time, they are not able to
   verify if that request is fresh and has not been replayed.  The same
   holds when a server loses synchronization may reply with sequence numbers of
   clients, for instance after a device reboot.

   The exact way to address this issue depends on the specific use case
   and its synchronization requirements.  The list of methods to handle
   synchronization of sequence numbers
   secure response, which is part of the group
   communication policy, and different servers can use different
   methods.  Appendix E describes three possible approaches that can be
   considered.

6.  Responsibilities protected as described in Section 8.3 of the Group Manager

   The Group Manager is responsible for performing
   [I-D.ietf-core-object-security], with the following tasks: modifications.

   o  Creating and managing OSCORE groups.  This includes  In step 2, the assignment
      of a Group ID to every newly created group, as well 'algorithms' array in the Additional Authenticated
      Data is modified as ensuring
      uniqueness of Group IDs within described in Section 3.

   o  In step 4, the set encryption of its OSCORE groups.

   o  Defining policies for authorizing the joining COSE object is modified as
      described in Section 3.  The encoding of its OSCORE
      groups.  Such policies can be enforced by a third party, which the compressed COSE
      object is modified as described in Section 4.

6.4.  Verifying the Response

   Upon receiving a trust relation with secure response message, the Group Manager and enforces join
      policies on behalf client proceeds as
   described in Section 8.4 of [I-D.ietf-core-object-security], with the Group Manager.
   following modifications.

   o  Driving  In step 2, the join process to add new endpoints as group members.

   o decoding of the compressed COSE object is modified
      as described in Section 3.  If the received Recipient ID ('kid')
      does not match with any Recipient Context for the retrieved Gid
      ('kid context'), then the client creates a new Recipient Context,
      initializes it according to Section 3 of
      [I-D.ietf-core-object-security], also retrieving the server's
      public key.

   o  In step 3, the 'algorithms' array in the Additional Authenticated
      Data is modified as described in Section 3.

   o  In step 5, the client also verifies the counter signature using
      the public key of the server from the associated Recipient
      Context.

7.  Responsibilities of the Group Manager

   The Group Manager is responsible for performing the following tasks:

   1.   Creating and managing OSCORE groups.  This includes the
        assignment of a Gid to every newly created group, as well as
        ensuring uniqueness of Gids within the set of its OSCORE groups.

   2.   Defining policies for authorizing the joining of its OSCORE
        groups.  Such policies can be enforced locally by the Group
        Manager, or by a third party in a trust relation with the Group
        Manager and entrusted to enforce join policies on behalf of the
        Group Manager.

   3.   Driving the join process to add new endpoints as group members.

   4.   Establishing Security Common Contexts and providing them to
        authorized group members during the join process, together with
        a corresponding Security Sender Context.

   o

   5.   Generating and managing Endpoint Sender IDs within its OSCORE groups, as
        well as assigning and providing them to new endpoints during the
        join process.  This includes ensuring uniqueness of Endpoints Sender IDs
        within each of its OSCORE groups.

   o

   6.   Defining a set of supported signature algorithms as part of the communication policy of for each of its OSCORE groups,
        and signalling it to new endpoints during the join process.

   o  Defining the methods to handle loss of synchronization with
      sequence numbers as part of the communication policy of each of
      its OSCORE groups, and signaling the one(s) to use to new
      endpoints during the join process.

   o

   7.   Renewing the Security Context of an OSCORE group upon membership
        change, by revoking and renewing common security parameters and
        keying material (rekeying).

   o

   8.   Providing the management keying material that a new endpoint
        requires to participate in the rekeying process, consistently consistent with
        the key management scheme used in the group joined by the new
        endpoint.

   o

   9.   Updating the Group ID Gid of its OSCORE groups, upon renewing the
        respective Security Context.

   The Group Manager may additionally be responsible for the following
   tasks:

   o

   10.  Acting as trusted key repository, in order to store handle the public keys of
        the members of its OSCORE groups, and provide providing such public keys
        to other members of the same group upon request.  This
      specification recommends that the Group Manager is entrusted to
      perform this task.

   o  Acting as network router device where endpoints register to
      correctly receive group messages sent to the multicast IP address  The actual
        storage of that group.

   o  Autonomously and locally enforcing access policies to authorize
      new endpoints public keys may be entrusted to join its OSCORE groups.

7. a separate secure
        storage device.

8.  Security Considerations

   The same security considerations from OSCORE (Section 11 of
   [I-D.ietf-core-object-security]) apply to this specification.
   Additional security aspects to be taken into account are discussed
   below.

7.1.

8.1.  Group-level Security

   The approach described in this document relies on commonly shared
   group keying material to protect communication within a group.  This
   means that messages
   has the following implications.

   o  Messages are encrypted at a group level (group-level data
      confidentiality), i.e. they can be decrypted by any member of the
      group, but not by an external adversary or other external
      entities.

   In addition,

   o  The AEAD algorithm provides only group authentication, i.e. it is required
      ensures that all a message sent to a group members are trusted, i.e.
   they do has been sent by a member
      of that group, but not by the alleged sender.  This is why source
      authentication of messages sent to a group is ensured through a
      counter signature, which is computed by the sender using its own
      private key and then appended to the message payload.

   Note that, even if an endpoint is authorized to be a group member and
   to take part in group communications, there is a risk that it behaves
   inappropriately.  For instance, it can forward the content of group
   messages in the group to unauthorized entities.  However, in many use
   cases, the devices in the group belong to a common authority and are
   configured by a commissioner (see Appendix B).

7.2. B), which results in a
   practically limited risk and enables a prompt detection/reaction in
   case of misbehaving.

8.2.  Uniqueness of (key, nonce)

   The proof for uniqueness of (key, nonce) pairs in Appendix D.3 of
   [I-D.ietf-core-object-security] is also valid in group communication
   scenarios.  That is, given an OSCORE group:

   o  Uniqueness of Sender IDs within the group is enforced by the Group
      Manager.

   o  Case  The case A is limited to in Appendix D.3 of [I-D.ietf-core-object-security] for
      messages including a Partial IV concerns only group requests, and
      same considerations hold. from [I-D.ietf-core-object-security] apply
      here as well.

   o  Case  The case B applies to in Appendix D.3 of [I-D.ietf-core-object-security] for
      messages not including a Partial IV concerns all group responses,
      and same considerations hold.

   It follows that from [I-D.ietf-core-object-security] apply
      here as well.

   As a consequence, each message encrypted/decrypted with the same
   Sender Key is processed by using a different (ID_PIV, PIV) pair.
   This means that nonces used by any fixed encrypting endpoint are
   unique.  Thus, each message is processed with a different (key,
   nonce) pair.

7.3.  Collision

8.3.  Management of Group Identifiers

   In case endpoints are deployed Keying Material

   The approach described in multiple groups managed by
   different non-synchronized Group Managers, it is possible for Group
   Identifiers of different groups to coincide.  However, this does not
   impair specification should take into account
   the security risk of the AEAD algorithm. compromise of group members.  In fact, as long as the Master Secret is different particular, this
   document specifies that a key management scheme for different secure revocation
   and renewal of Security Contexts and group keying material should be
   adopted.

   Especially in dynamic, large-scale, groups where endpoints can join
   and this condition holds over leave at any time, it is important that the considered group key
   management scheme is efficient and as long as highly scalable with the Sender
   IDs within a group are unique, it follows that AEAD keys
   size, in order to limit the impact on performance due to the Security
   Context and nonces
   are different among different groups.

8.  IANA Considerations

   This document keying material update.

8.4.  Update of Security Context and Key Rotation

   A group member can receive a message shortly after the group has no actions for IANA.

9.  Acknowledgments

   The authors sincerely thank Stefan Beck, Rolf Blom, Carsten Bormann,
   Esko Dijk, Klaus Hartke, Richard Kelsey, John Mattsson, Jim Schaad,
   Ludwig Seitz been
   rekeyed, and Peter van der Stok for their feedback new security parameters and comments.

   The work on this document has keying material have been partly supported
   distributed by the EIT-
   Digital High Impact Initiative ACTIVE.

10.  References

10.1.  Normative References

   [I-D.ietf-core-object-security]
              Selander, G., Mattsson, J., Palombini, F., and L. Seitz,
              "Object Security for Constrained RESTful Environments
              (OSCORE)", draft-ietf-core-object-security-13 (work Group Manager.  In the following two cases, this
   may result in
              progress), June 2018.

   [RFC2119]  Bradner, S., "Key words for use in RFCs to Indicate
              Requirement Levels", BCP 14, RFC 2119,
              DOI 10.17487/RFC2119, March 1997,
              <https://www.rfc-editor.org/info/rfc2119>.

   [RFC7252]  Shelby, Z., Hartke, K., and C. Bormann, "The Constrained
              Application Protocol (CoAP)", RFC 7252,
              DOI 10.17487/RFC7252, June 2014,
              <https://www.rfc-editor.org/info/rfc7252>.

   [RFC8032]  Josefsson, S. 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 and Encryption (COSE)",
              RFC 8152, DOI 10.17487/RFC8152, July 2017,
              <https://www.rfc-editor.org/info/rfc8152>.

   [RFC8174]  Leiba, B., "Ambiguity of 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>.

10.2.  Informative References

   [I-D.ietf-ace-dtls-authorize]
              Gerdes, S., Bergmann, O., Bormann, C., Selander, G., and
              L. Seitz, "Datagram Transport Layer Security (DTLS)
              Profile for Authentication and Authorization for
              Constrained Environments (ACE)", draft-ietf-ace-dtls-
              authorize-03 (work in progress), March 2018.

   [I-D.ietf-ace-oauth-authz]
              Seitz, L., Selander, G., Wahlstroem, E., Erdtman, S., and
              H. Tschofenig, "Authentication and Authorization for
              Constrained Environments (ACE) using the OAuth 2.0
              Framework (ACE-OAuth)", draft-ietf-ace-oauth-authz-12
              (work in progress), May 2018.

   [I-D.ietf-ace-oscore-profile]
              Seitz, L., Palombini, F., Gunnarsson, M., and G. Selander,
              "OSCORE profile of the Authentication and Authorization
              for Constrained Environments Framework", draft-ietf-ace-
              oscore-profile-01 (work in progress), March 2018.

   [I-D.ietf-core-echo-request-tag]
              Amsuess, C., Mattsson, J., and G. Selander, "Echo and
              Request-Tag", draft-ietf-core-echo-request-tag-01 (work in
              progress), March 2018.

   [I-D.palombini-ace-key-groupcomm]
              Palombini, F. and M. Tiloca, "Key Provisioning for Group
              Communication using ACE", draft-palombini-ace-key-
              groupcomm-01 (work in progress), June 2018.

   [I-D.somaraju-ace-multicast]
              Somaraju, A., Kumar, S., Tschofenig, H., and W. Werner,
              "Security for Low-Latency Group Communication", draft-
              somaraju-ace-multicast-02 (work in progress), October
              2016.

   [I-D.tiloca-ace-oscoap-joining]
              Tiloca, M. and J. Park, "Joining OSCORE groups in ACE",
              draft-tiloca-ace-oscoap-joining-03 (work in progress),
              March 2018.

   [RFC2093]  Harney, H. and C. Muckenhirn, "Group Key Management
              Protocol (GKMP) Specification", RFC 2093,
              DOI 10.17487/RFC2093, July 1997,
              <https://www.rfc-editor.org/info/rfc2093>.

   [RFC2094]  Harney, H. and C. Muckenhirn, "Group Key Management
              Protocol (GKMP) Architecture", RFC 2094,
              DOI 10.17487/RFC2094, July 1997,
              <https://www.rfc-editor.org/info/rfc2094>.

   [RFC2627]  Wallner, D., Harder, E., and R. Agee, "Key Management for
              Multicast: Issues and Architectures", RFC 2627,
              DOI 10.17487/RFC2627, June 1999,
              <https://www.rfc-editor.org/info/rfc2627>.

   [RFC3376]  Cain, B., Deering, S., Kouvelas, I., Fenner, B., and A.
              Thyagarajan, "Internet Group Management Protocol, Version
              3", RFC 3376, DOI 10.17487/RFC3376, October 2002,
              <https://www.rfc-editor.org/info/rfc3376>.

   [RFC3740]  Hardjono, T. and B. Weis, "The Multicast Group Security
              Architecture", RFC 3740, DOI 10.17487/RFC3740, March 2004,
              <https://www.rfc-editor.org/info/rfc3740>.

   [RFC3810]  Vida, R., Ed. and L. Costa, Ed., "Multicast Listener
              Discovery Version 2 (MLDv2) for IPv6", RFC 3810,
              DOI 10.17487/RFC3810, June 2004,
              <https://www.rfc-editor.org/info/rfc3810>.

   [RFC4046]  Baugher, M., Canetti, R., Dondeti, L., and F. Lindholm,
              "Multicast Security (MSEC) Group Key Management
              Architecture", RFC 4046, DOI 10.17487/RFC4046, April 2005,
              <https://www.rfc-editor.org/info/rfc4046>.

   [RFC4301]  Kent, S. and K. Seo, "Security Architecture for the
              Internet Protocol", RFC 4301, DOI 10.17487/RFC4301,
              December 2005, <https://www.rfc-editor.org/info/rfc4301>.

   [RFC4535]  Harney, H., Meth, U., Colegrove, A., and G. Gross,
              "GSAKMP: Group Secure Association Key Management
              Protocol", RFC 4535, DOI 10.17487/RFC4535, June 2006,
              <https://www.rfc-editor.org/info/rfc4535>.

   [RFC4944]  Montenegro, G., Kushalnagar, N., Hui, J., and D. Culler,
              "Transmission of IPv6 Packets over IEEE 802.15.4
              Networks", RFC 4944, DOI 10.17487/RFC4944, September 2007,
              <https://www.rfc-editor.org/info/rfc4944>.

   [RFC4949]  Shirey, R., "Internet Security Glossary, Version 2",
              FYI 36, RFC 4949, DOI 10.17487/RFC4949, August 2007,
              <https://www.rfc-editor.org/info/rfc4949>.

   [RFC6282]  Hui, J., Ed. and P. Thubert, "Compression Format for IPv6
              Datagrams over IEEE 802.15.4-Based Networks", RFC 6282,
              DOI 10.17487/RFC6282, September 2011,
              <https://www.rfc-editor.org/info/rfc6282>.

   [RFC6347]  Rescorla, E. and N. Modadugu, "Datagram Transport Layer
              Security Version 1.2", RFC 6347, DOI 10.17487/RFC6347,
              January 2012, <https://www.rfc-editor.org/info/rfc6347>.

   [RFC6749]  Hardt, D., Ed., "The OAuth 2.0 Authorization Framework",
              RFC 6749, DOI 10.17487/RFC6749, October 2012,
              <https://www.rfc-editor.org/info/rfc6749>.

   [RFC7228]  Bormann, C., Ersue, M., and A. Keranen, "Terminology for
              Constrained-Node Networks", RFC 7228,
              DOI 10.17487/RFC7228, May 2014,
              <https://www.rfc-editor.org/info/rfc7228>.

   [RFC7390]  Rahman, A., Ed. and E. Dijk, Ed., "Group Communication for
              the Constrained Application Protocol (CoAP)", RFC 7390,
              DOI 10.17487/RFC7390, October 2014,
              <https://www.rfc-editor.org/info/rfc7390>.

Appendix A.  Assumptions and Security Objectives

   This section presents a set of assumptions and security objectives
   for the approach described in this document.

A.1.  Assumptions

   The following assumptions are assumed to be already addressed and are
   out of the scope of this document.

   o  Multicast communication topology: this document considers both
      1-to-N (one sender and multiple recipients) and M-to-N (multiple
      senders and multiple recipients) communication topologies.  The
      1-to-N communication topology is the simplest group communication
      scenario that would serve the needs of a typical low-power and
      lossy network (LLN).  Examples of use cases that benefit from
      secure group communication are provided in Appendix B.

      In a 1-to-N communication model, only a single client transmits
      data to the group, in the form of request messages; in an M-to-N
      communication model (where M and N do not necessarily have the
      same value), M group members are clients.  According to [RFC7390],
      any possible proxy entity is supposed to know about the clients in
      the group and to not perform aggregation of response messages from
      multiple servers.  Also, every client expects and is able to
      handle multiple response messages associated to a same request
      sent to the group.

   o  Group size: security solutions for group communication should be
      able to adequately support different and possibly large groups.
      The group size is the current number of members in a group.  In
      the use cases mentioned in this document, the number of clients
      (normally the controlling devices) is expected to be much smaller
      than the number of servers (i.e. the controlled devices).  A
      security solution for group communication that supports 1 to 50
      clients would be able to properly cover the group sizes required
      for most use cases that are relevant for this document.  The
      maximum group size is expected to be in the range of 2 to 100
      devices.  Groups larger than that should be divided into smaller
      independent groups, e.g. by grouping lights in a building on a per
      floor basis.

   o  Communication with the Group Manager: an endpoint must use a
      secure dedicated channel when communicating with the Group
      Manager, even when not registered as group member.  In particular,
      communications with the Group Manager occuring during the join
      process to become a group member must also be secured.

   o  Establishment and management of Security Contexts: an OSCORE
      Security Context must be established among the group members.  In
      particular, a Common Context must be provided to a new joining
      endpoint together with a corresponding Sender Context.  On the
      other hand, Recipient Contexts are locally and individually
      derived by each group member.  A secure mechanism must be used to
      generate, revoke and (re-)distribute keying material, multicast
      security policies and security parameters in the group.  The
      actual establishment and management of the Security Context is out
      of the scope of this document, and it is anticipated that an
      activity in IETF dedicated to the design of a generic key
      management scheme will include this feature, preferably based on
      [RFC3740][RFC4046][RFC4535].

   o  Multicast data security ciphersuite: all group members must agree
      on a ciphersuite to provide authenticity, integrity and
      confidentiality of messages in the group.  The ciphersuite is
      specified as part of the Security Context.

   o  Backward security: a new device joining the group should not have
      access to any old Security Contexts used before its joining.  This
      ensures that a new group member is not able to decrypt
      confidential data sent before it has joined the group.  The
      adopted key management scheme should ensure that the Security
      Context is updated to ensure backward confidentiality.  The actual
      mechanism to update the Security Context and renew the group
      keying material upon a group member's joining has to be defined as
      part of the group key management scheme.

   o  Forward security: entities that leave the group should not have
      access to any future Security Contexts or message exchanged within
      the group after their leaving.  This ensures that a former group
      member is not able to decrypt confidential data sent within the
      group anymore.  Also, it ensures that a former member is not able
      to send encrypted and/or integrity protected messages to the group
      anymore.  The actual mechanism to update the Security Context and
      renew the group keying material upon a group member's leaving has
      to be defined as part of the group key management scheme.

A.2.  Security Objectives

   The approach described in this document aims at fulfilling the
   following security objectives:

   o  Data replay protection: replayed group request messages or
      response messages must be detected.

   o  Group-level data confidentiality: messages sent within the group
      shall be encrypted if privacy sensitive data is exchanged within
      the group.  This document considers group-level data
      confidentiality since messages are encrypted at a group level,
      i.e. in such a way that they can be decrypted by any member of the
      group, but not by an external adversary or other external
      entities.

   o  Source authentication: messages sent within the group shall be
      authenticated.  That is, it is essential to ensure that a message
      is originated by a member of the group in the first place, and in
      particular by a specific member of the group.

   o  Message integrity: messages sent within the group shall be
      integrity protected.  That is, it is essential to ensure that a
      message has not been tampered with by an external adversary or
      other external entities which are not group members.

   o  Message ordering: it must be possible to determine the ordering of
      messages coming from a single sender endpoint.  In accordance with
      OSCORE [I-D.ietf-core-object-security], this results in providing
      relative freshness of group requests and absolute freshness of
      responses.  It is not required to determine ordering of messages
      from different sender endpoints.

Appendix B.  List of Use Cases

   Group Communication for CoAP [RFC7390] provides the necessary
   background for multicast-based CoAP communication, with particular
   reference to low-power and lossy networks (LLNs) and resource
   constrained environments.  The interested reader is encouraged to
   first read [RFC7390] to understand the non-security related details.
   This section discusses a number of use cases that benefit from secure
   group communication.  Specific security requirements for these use
   cases are discussed in Appendix A.

   o  Lighting control: consider a building equipped with IP-connected
      lighting devices, switches, and border routers.  The devices are
      organized into groups according to their physical location in the
      building.  For instance, lighting devices and switches in a room
      or corridor can be configured as members of a single group.
      Switches are then used to control the lighting devices by sending
      on/off/dimming commands to all lighting devices in a group, while
      border routers connected to an IP network backbone (which is also
      multicast-enabled) can be used to interconnect routers in the
      building.  Consequently, this would also enable logical groups to
      be formed even if devices in the lighting group may be physically
      in different subnets (e.g. on wired and wireless networks).
      Connectivity between lighting devices may be realized, for
      instance, by means of IPv6 and (border) routers supporting 6LoWPAN
      [RFC4944][RFC6282].  Group communication enables synchronous
      operation of a group of connected lights, ensuring that the light
      preset (e.g. dimming level or color) of a large group of
      luminaires are changed at the same perceived time.  This is
      especially useful for providing a visual synchronicity of light
      effects to the user.  As a practical guideline, events within a
      200 ms interval are perceived as simultaneous by humans, which is
      necessary to ensure in many setups.  Devices may reply back to the
      switches that issue on/off/dimming commands, in order to report
      about the execution of the requested operation (e.g.  OK, failure,
      error) and their current operational status.  In a typical
      lighting control scenario, a single switch is the only entity
      responsible for sending commands to a group of lighting devices.
      In more advanced lighting control use cases, a M-to-N
      communication topology would be required, for instance in case
      multiple sensors (presence or day-light) are responsible to
      trigger events to a group of lighting devices.  Especially in
      professional lighting scenarios, misaligned Security Contexts between the roles of client sender and server
      are configured by the lighting commissioner, and devices strictly
      follow those roles.

   o  Integrated building control: enabling Building Automation and
      Control Systems (BACSs) to control multiple heating, ventilation
      and air-conditioning units to pre-defined presets.  Controlled
      units can be organized into groups in order to reflect their
      physical position in
   recipient.

   In the building, e.g. devices in first case, the same room
      can be configured as members of a single group.  As sender protects a practical
      guideline, events within intervals of seconds are typically
      acceptable.  Controlled units are expected to possibly reply back
      to message using the BACS issuing control commands, in order to report about old
   Security Context, i.e. before having installed the
      execution of new Security
   Context.  However, the requested operation (e.g.  OK, failure, error)
      and their current operational status.

   o  Software and firmware updates: software and firmware updates often
      comprise quite a large amount of data.  This can overload a LLN
      that is otherwise typically used recipient receives the message after having
   installed the new Security Context, hence not being able to deal with only small amounts
      of data, on an infrequent base.  Rather than sending software and
      firmware updates as unicast messages correctly
   process it.  A possible way to each individual device,
      multicasting such updated data ameliorate this issue is to preserve
   the old, recent, Security Context for a maximum amount of time
   defined by the application.  By doing so, the recipient can still try
   to process the received message using the old retained Security
   Context as second attempt.  Note that a larger former (compromised) group
   member can take advantage of devices at
      once displays this by sending messages protected with
   the old retained Security Context.  Therefore, a number conservative
   application policy should not admit the storage of benefits.  For instance, it can
      significantly reduce old Security
   Contexts.

   In the network load and decrease second case, the overall
      time latency for propagating this data sender protects a message using the new
   Security Context, but the recipient receives that request before
   having installed the new Security Context.  Therefore, the recipient
   would not be able to all devices.  Even if correctly process the complete whole update request and hence discards
   it.  If the recipient receives the new Security Context shortly after
   that and the sender endpoint uses CoAP retransmissions, the former
   will still be able to receive and correctly process itself is secured, securing the
      individual messages is important, in case updates consist of
      relatively large amounts of data. message.  In fact, checking individual
      received data piecemeal
   any case, the recipient should actively ask the Group Manager for tampering avoids that devices store
      large amounts of partially corrupted data and that they detect
      tampering hereof only an
   updated Security Context according to an application-defined policy,
   for instance after all data has been received.  Devices
      receiving software and firmware updates a given number of unsuccessfully decrypted
   incoming messages.

8.5.  Collision of Group Identifiers

   In case endpoints are expected to possibly
      reply back, deployed in order multiple groups managed by
   different non-synchronized Group Managers, it is possible for Group
   Identifiers of different groups to provide coincide.  That can also happen if
   the application can not guarantee unique Group Identifiers within a feedback about
   given Group Manager.  However, this does not impair the execution security of
   the update operation (e.g.  OK, failure, error) AEAD algorithm.

   In fact, as long as the Master Secret is different for different
   groups and their current
      operational status.

   o  Parameter this condition holds over time, and configuration update: by means of multicast
      communication, it is possible to update as long as the settings of Sender
   IDs within a group of
      similar devices, both simultaneously and efficiently.  Possible
      parameters are related, for instance, to network load management
      or network access controls.  Devices receiving parameter and
      configuration updates unique, AEAD keys are expected to possibly reply back, different among
   different groups.

9.  IANA Considerations

   Note to
      provide a feedback about the execution RFC Editor: Please replace all occurrences of "[[this
   document]]" with the update operation
      (e.g.  OK, failure, error) and their current operational status.

   o  Commissioning RFC number of LLNs systems: a commissioning device this specification.

9.1.  OSCORE Flag Bits Registry

   The entry with Bit Position TBD is
      responsible for querying all devices added to the "OSCORE Flag Bits"
   registry.

+--------------+-------------+---------------------+-------------------+
| Bit Position |     Name    |     Description     |   Specification   |
+--------------+-------------+---------------------+-------------------+
|     TBD      | Counter     | Set to 1 if counter | [[this document]] |
|              | Signature   | signature present   |                   |
|              |             | in the local network or a
      selected subset of them, compressed   |                   |
|              |             | COSE object         |                   |
+--------------+-------------+---------------------+-------------------+

10.  References

10.1.  Normative References

   [I-D.ietf-core-object-security]
              Selander, G., Mattsson, J., Palombini, F., and L. Seitz,
              "Object Security for Constrained RESTful Environments
              (OSCORE)", draft-ietf-core-object-security-15 (work in order
              progress), August 2018.

   [RFC2119]  Bradner, S., "Key words for use in RFCs to discover their presence, Indicate
              Requirement Levels", BCP 14, RFC 2119,
              DOI 10.17487/RFC2119, March 1997,
              <https://www.rfc-editor.org/info/rfc2119>.

   [RFC7252]  Shelby, Z., Hartke, K., and
      be aware of their capabilities, default configuration, C. Bormann, "The Constrained
              Application Protocol (CoAP)", RFC 7252,
              DOI 10.17487/RFC7252, June 2014,
              <https://www.rfc-editor.org/info/rfc7252>.

   [RFC8032]  Josefsson, S. and
      operating conditions.  Queried devices displaying similarities in
      their capabilities 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 and features, or sharing a common physical
      location can be configured as members Encryption (COSE)",
              RFC 8152, DOI 10.17487/RFC8152, July 2017,
              <https://www.rfc-editor.org/info/rfc8152>.

   [RFC8174]  Leiba, B., "Ambiguity of a single group.  Queried
      devices are expected to reply back to the commissioning device, Uppercase vs Lowercase in
      order to notify their presence, and provide the requested
      information and their current operational status.

   o  Emergency multicast: a particular emergency related information
      (e.g. natural disaster) is generated RFC
              2119 Key Words", BCP 14, RFC 8174, DOI 10.17487/RFC8174,
              May 2017, <https://www.rfc-editor.org/info/rfc8174>.

10.2.  Informative References

   [I-D.ietf-ace-oauth-authz]
              Seitz, L., Selander, G., Wahlstroem, E., Erdtman, S., and multicast by an emergency
      notifier,
              H. Tschofenig, "Authentication and relayed to multiple devices.  The latters may reply
      back to Authorization for
              Constrained Environments (ACE) using the emergency notifier, OAuth 2.0
              Framework (ACE-OAuth)", draft-ietf-ace-oauth-authz-16
              (work in order to provide their feedback
      and local information related to the ongoing emergency.  This kind
      of setups should additionally rely on a fault tolerance multicast
      algorithm, such as MPL.

Appendix C.  Example of Group Identifier Format

   This section provides an example of how the Group Identifier (Gid)
   can be specifically formatted.  That is, the Gid can be composed of
   two parts, namely a Group Prefix progress), October 2018.

   [I-D.ietf-core-echo-request-tag]
              Amsuess, C., Mattsson, J., and a Group Epoch.

   The Group Prefix is constant over time G. Selander, "Echo and is uniquely defined
              Request-Tag", draft-ietf-core-echo-request-tag-02 (work in the
   set of all the groups associated to the same Group Manager.  The
   choice of the Group Prefix
              progress), June 2018.

   [I-D.somaraju-ace-multicast]
              Somaraju, A., Kumar, S., Tschofenig, H., and W. Werner,
              "Security for a given group's Security Context is
   application specific.  The size of the Group Prefix directly impact
   on the maximum number of distinct groups under the same Group
   Manager.

   The Low-Latency Group Epoch is set to 0 upon the group's initialization, and is
   incremented by 1 upon completing each renewal of the Security Context Communication", draft-
              somaraju-ace-multicast-02 (work in progress), October
              2016.

   [I-D.tiloca-ace-oscoap-joining]
              Tiloca, M., Park, J., and keying material F. Palombini, "Key Management
              for OSCORE Groups in the group (see Section 2.1).  In particular,
   once a new Master Secret has been distributed to the group, all the
   group members increment by 1 the Group Epoch ACE", draft-tiloca-ace-oscoap-
              joining-05 (work in the Group Identifier progress), October 2018.

   [RFC4944]  Montenegro, G., Kushalnagar, N., Hui, J., and D. Culler,
              "Transmission of that group.

   As an example, a 3-byte Group Identifier can be composed of: i) a
   1-byte Group Prefix '0xb1' interpreted as a raw byte string; IPv6 Packets over IEEE 802.15.4
              Networks", RFC 4944, DOI 10.17487/RFC4944, September 2007,
              <https://www.rfc-editor.org/info/rfc4944>.

   [RFC4949]  Shirey, R., "Internet Security Glossary, Version 2",
              FYI 36, RFC 4949, DOI 10.17487/RFC4949, August 2007,
              <https://www.rfc-editor.org/info/rfc4949>.

   [RFC6282]  Hui, J., Ed. and P. Thubert, "Compression Format for IPv6
              Datagrams over IEEE 802.15.4-Based Networks", RFC 6282,
              DOI 10.17487/RFC6282, September 2011,
              <https://www.rfc-editor.org/info/rfc6282>.

   [RFC6347]  Rescorla, E. and ii)
   a 2-byte Group Epoch interpreted as an unsigned integer ranging from
   0 to 65535.  Then, after having established the N. Modadugu, "Datagram Transport Layer
              Security Common
   Context 61532 times in Version 1.2", RFC 6347, DOI 10.17487/RFC6347,
              January 2012, <https://www.rfc-editor.org/info/rfc6347>.

   [RFC7228]  Bormann, C., Ersue, M., and A. Keranen, "Terminology for
              Constrained-Node Networks", RFC 7228,
              DOI 10.17487/RFC7228, May 2014,
              <https://www.rfc-editor.org/info/rfc7228>.

   [RFC7390]  Rahman, A., Ed. and E. Dijk, Ed., "Group Communication for
              the group, its Group Identifier will assume
   value '0xb1f05c'.

   As discussed in Section 7.3, if endpoints are deployed in multiple
   groups managed by different non-synchronized Group Managers, it is
   possible that Group Identifiers of different groups coincide at some
   point in time.  In this case, a recipient endpoint has to handle
   coinciding Group Identifiers, Constrained Application Protocol (CoAP)", RFC 7390,
              DOI 10.17487/RFC7390, October 2014,
              <https://www.rfc-editor.org/info/rfc7390>.

Appendix A.  Assumptions and has to try using different OSCORE Security Contexts to process an incoming message, until the right one
   is found Objectives

   This section presents a set of assumptions and security objectives
   for the message is correctly verified.  Therefore, it is
   favourable that Group Idenfiers from different Group Managers have a
   size that result approach described in a small probability of collision.  How small this
   probability should be is up document.

A.1.  Assumptions

   The following assumptions are assumed to system designers.

Appendix D.  Set-up of New Endpoints

   An endpoint joins a group by explicitly interacting with the
   responsible Group Manager.  Communications between a joining endpoint
   and the Group Manager rely on the CoAP protocol and must be secured.
   Specific details on how to secure communications between joining
   endpoints already addressed and a Group Manager are
   out of scope.

   In order to receive multicast messages sent to the group, a joining
   endpoint has to register with scope of this document.

   o  Multicast communication topology: this document considers both
      1-to-N (one sender and multiple recipients) and M-to-N (multiple
      senders and multiple recipients) communication topologies.  The
      1-to-N communication topology is the simplest group communication
      scenario that would serve the needs of a typical low-power and
      lossy network router device
   [RFC3376][RFC3810], signaling its intent to receive packets sent to
   the multicast IP address (LLN).  Examples of use cases that group.  As benefit from
      secure group communication are provided in Appendix B.

      In a particular case, the
   Group Manager can also act as such 1-to-N communication model, only a network router device.  Upon
   joining single client transmits
      data to the group, endpoints are not required to know how many and
   what endpoints are active in the same group.

   Furthermore, in order to participate form of request messages; in the secure group
   communication, an endpoint needs to be properly initialized upon
   joining the group.  In particular, the Group Manager provides keying
   material M-to-N
      communication model (where M and parameters to a joining endpoint, which can then
   initialize its own Security Context (see Section 2).

   The following Appendix D.1 provides an example describing how such
   information can be provided to an endpoint upon joining a group
   through N do not necessarily have the responsible Group Manager.  Then, Appendix D.2 discusses
   how public keys of
      same value), M group members can be handled and made available are clients.  According to
   group members.  Finally, Appendix D.3 overviews how [RFC7390],
      any possible proxy entity is supposed to know about the ACE framework
   for Authentication clients in
      the group and Authorization in constrained environments
   [I-D.ietf-ace-oauth-authz] can be possibly used to support such a
   join process.

D.1.  Join Process

   An endpoint requests not perform aggregation of response messages from
      multiple servers.  Also, every client expects and is able to
      handle multiple response messages associated to join a group by sending a confirmable CoAP
   POST same request
      sent to the group.

   o  Group Manager responsible size: security solutions for that group.  This
   join request can reflect group communication should be
      able to adequately support different and possibly large groups.
      The group size is the format current number of the Key Distribution Request
   message defined members in Section 4.1 of [I-D.palombini-ace-key-groupcomm].
   Besides, it can be addressed to a CoAP resource associated to that
   group and carries group.  In
      the following information.

   o  Group identifier: use cases mentioned in this document, the Group Identifier (Gid) number of clients
      (normally the group, as
      known controlling devices) is expected to be much smaller
      than the joining endpoint at this point in time.  This may not
      fully coincide with number of servers (i.e. the Gid currently associated controlled devices).  A
      security solution for group communication that supports 1 to the group,
      e.g. if it includes a dynamic component.  This information can 50
      clients would be
      mapped able to properly cover the first element of the 'scope' parameter of the Key
      Distribution Request message defined group sizes required
      for most use cases that are relevant for this document.  The
      maximum group size is expected to be in Section 4.1 the range of
      [I-D.palombini-ace-key-groupcomm]. 2 to 100
      devices.  Groups larger than that should be divided into smaller
      independent groups.

   o  Role:  Communication with the exact role Group Manager: an endpoint must use a
      secure dedicated channel when communicating with the Group
      Manager, also when not registered as group member.

   o  Provisioning and management of Security Contexts: an OSCORE
      Security Context must be established among the joining endpoint group members.  A
      secure mechanism must be used to generate, revoke and
      (re-)distribute keying material, multicast security policies and
      security parameters in the group.
      Possible values are: "client", "server", "silent server", "client
      and server", or "client  The actual provisioning and silent server".  This information can
      be mapped to the second element
      management of the 'scope' parameter Security Context is out of the
      Key Distribution Request message defined in Section 4.1 scope of
      [I-D.palombini-ace-key-groupcomm]. this
      document.

   o  Retrieval flag: indication of interest  Multicast data security ciphersuite: all group members must agree
      on a ciphersuite to receive the public keys provide authenticity, integrity and
      confidentiality of the endpoints currently in the group, as included messages in the
      following join response.  This flag must not be present if the
      Group Manager group.  The ciphersuite is not configured to store the public keys
      specified as part of group
      members, or if the Security Context.

   o  Backward security: a new device joining endpoint is configured exclusively as
      silent server for the group should not have
      access to join. any old Security Contexts used before its joining.  This information can be
      mapped to the 'get_pub_keys' parameter of the Key Distribution
      Request message defined in Section 4.1 of
      [I-D.palombini-ace-key-groupcomm].

   o  Identity credentials: information elements to enforce source
      authentication of
      ensures that a new group messages from member is not able to decrypt
      confidential data sent before it has joined the joining endpoint, such
      as its public key. group.  The exact content depends on whether
      adopted key management scheme should ensure that the Group
      Manager Security
      Context is configured updated to store ensure backward confidentiality.  The actual
      mechanism to update the public keys of group members.
      If this is Security Context and renew the case, this information is omitted if it group
      keying material upon a group member's joining has been
      provided to be defined as
      part of the same Group Manager upon previously joining group key management scheme.

   o  Forward security: entities that leave the
      same group should not have
      access to any future Security Contexts or a different message exchanged within
      the group under its control. after their leaving.  This information ensures that a former group
      member is
      also omitted if not able to decrypt confidential data sent within the joining endpoint
      group anymore.  Also, it ensures that a former member is configured exclusively as
      silent server for the joined group.  Appendix D.2 discusses
      additional details on provisioning of public keys and other
      information not able
      to enforce source authentication of joining
      endpoints's messages.  This information can be mapped send encrypted and/or integrity protected messages to the
      'client_cred' parameter of group
      anymore.  The actual mechanism to update the Key Distribution Request message Security Context and
      renew the group keying material upon a group member's leaving has
      to be defined in Section 4.1 as part of [I-D.palombini-ace-key-groupcomm]. the group key management scheme.

A.2.  Security Objectives

   The Group Manager approach described in this document aims at fulfilling the
   following security objectives:

   o  Data replay protection: replayed group request messages or
      response messages must be able to verify that detected.

   o  Group-level data confidentiality: messages sent within the joining endpoint group
      shall be encrypted if privacy sensitive data is
   authorized to become a member of exchanged within
      the group.  To this end, the Group
   Manager can directly authorize the joining endpoint, or expect it to
   provide authorization evidence previously obtained from  This document considers group-level data
      confidentiality since messages are encrypted at a trusted
   entity.  Appendix D.3 describes how this group level,
      i.e. in such a way that they can be achieved decrypted by
   leveraging the ACE framework for Authentication and Authorization in
   constrained environments [I-D.ietf-ace-oauth-authz].

   In case any member of successful authorization check, the Group Manager
   generates
      group, but not by an Endpoint ID assigned to the joining endpoint, before
   proceeding with the rest of the join process.  Instead, in case the
   authorization check fails, the Group Manager aborts the join process.
   Further details about external adversary or other external
      entities.

   o  Source authentication: messages sent within the authorization of joining endpoint are out
   of scope.

   As discussed in Section 2.1, group shall be
      authenticated.  That is, it is recommended essential to ensure that the Security
   Context is renewed before the joining endpoint receives the group
   keying material and becomes a new active member of the group.  This message
      is achieved originated by securely distributing a new Master Secret and a new
   Group Identifier to member of the endpoints currently present group in the same
   group.

   Once renewed the Security Context first place, and in the group, the Group Manager
   replies to the joining endpoint with
      particular by a CoAP response carrying the
   following information.  This join response can reflect the format specific member of the Key Distribution Response message defined in Section 4.2 of
   [I-D.palombini-ace-key-groupcomm]. group.

   o  Security Common Context:  Message integrity: messages sent within the OSCORE Security Common Context
      associated group shall be
      integrity protected.  That is, it is essential to the joined ensure that a
      message has not been tampered with by an external adversary or
      other external entities which are not group (see Section 2).  This information
      can members.

   o  Message ordering: it must be mapped possible to determine the 'key' parameter ordering of the Key Distribution
      Response message defined
      messages coming from a single sender.  In accordance with OSCORE
      [I-D.ietf-core-object-security], this results in Section 4.2 providing
      relative freshness of group requests and absolute freshness of
      [I-D.palombini-ace-key-groupcomm].

   o  Endpoint ID: the Endpoint ID associated to the joining endpoint.
      This information
      responses.  It is not included in case 'Role' in required to determine ordering of messages
      from different senders.

Appendix B.  List of Use Cases

   Group Communication for CoAP [RFC7390] provides the join
      request necessary
   background for multicast-based CoAP communication, with particular
   reference to low-power and lossy networks (LLNs) and resource
   constrained environments.  The interested reader is equal encouraged to "silent server".  This information can be
      mapped
   first read [RFC7390] to understand the 'clientID' parameter within the 'key' parameter non-security related details.
   This section discusses a number of
      the Key Distribution Response message defined use cases that benefit from secure
   group communication.  Specific security requirements for these use
   cases are discussed in Section 4.2 of
      [I-D.palombini-ace-key-groupcomm]. Appendix A.

   o  Member public keys: the public keys of the endpoints currently
      present  Lighting control: consider a building equipped with IP-connected
      lighting devices, switches, and border routers.  The devices are
      organized into groups according to their physical location in the group.  This includes: the public keys of the non-
      silent servers currently
      building.  For instance, lighting devices and switches in the group, if the joining endpoint is a room
      or corridor can be configured (also) as client; and the public keys members of a single group.
      Switches are then used to control the clients
      currently lighting devices by sending
      on/off/dimming commands to all lighting devices in the a group, if the joining endpoint is configured
      (also) as server or silent server.  This information is omitted in
      case the Group Manager while
      border routers connected to an IP network backbone (which is not configured also
      multicast-enabled) can be used to store interconnect routers in the public keys
      of group members or
      building.  Consequently, this would also enable logical groups to
      be formed even if the 'Retrieval flag' was not present devices in the
      join request.  Appendix D.2 discusses additional details on
      provisioning public keys upon joining the lighting group and may be physically
      in different subnets (e.g. on retrieving
      public keys wired and wireless networks).

      Connectivity between lighting devices may be realized, for
      instance, by means of IPv6 and (border) routers supporting 6LoWPAN
      [RFC4944][RFC6282].  Group communication enables synchronous
      operation of a group members.  This information can be mapped to
      the 'pub_keys' parameter of connected lights, ensuring that the Key Distribution Response message
      defined in Section 4.2 light
      preset (e.g. dimming level or color) of [I-D.palombini-ace-key-groupcomm].

   o  Group policies: a list large group of key words indicating the particular
      policies enforced in
      luminaires are changed at the group. same perceived time.  This includes, is
      especially useful for instance, the
      method to achieve synchronization providing a visual synchronicity of sequence numbers among group
      members (see Appendix E), as well as light
      effects to the rekeying protocol used user.  As a practical guideline, events within a
      200 ms interval are perceived as simultaneous by humans, which is
      necessary to ensure in many setups.  Devices may reply back to
      renew the keying material
      switches that issue on/off/dimming commands, in order to report
      about the execution of the requested operation (e.g.  OK, failure,
      error) and their current operational status.  In a typical
      lighting control scenario, a single switch is the only entity
      responsible for sending commands to a group (see Section 2.1).  This
      information can of lighting devices.
      In more advanced lighting control use cases, a M-to-N
      communication topology would be mapped required, for instance in case
      multiple sensors (presence or day-light) are responsible to
      trigger events to a group of lighting devices.  Especially in
      professional lighting scenarios, the 'group_policies' parameter roles of client and server
      are configured by the
      Key Distribution Response message defined in Section 4.2 of
      [I-D.palombini-ace-key-groupcomm]. lighting commissioner, and devices strictly
      follow those roles.

   o  Management keying material: the set of administrative keying
      material used  Integrated building control: enabling Building Automation and
      Control Systems (BACSs) to participate control multiple heating, ventilation
      and air-conditioning units to pre-defined presets.  Controlled
      units can be organized into groups in order to reflect their
      physical position in the group rekeying process run by
      the Group Manager (see Section 2.1).  The specific elements of
      this management keying material depend on the group rekeying
      protocol used building, e.g. devices in the group.  For instance, this same room
      can simply consist
      in be configured as members of a group key encryption key and single group.  As a pairwise symmetric key shared
      between the joining endpoint and the Group Manager, in case GKMP
      [RFC2093][RFC2094] is used.  Instead, if key-tree based rekeying
      protocols like LKH [RFC2627] are used, it can consist in the set practical
      guideline, events within intervals of symmetric keys associated seconds are typically
      acceptable.  Controlled units are expected to the key-tree leaf representing the
      group member up possibly reply back
      to the key-tree root representing the group key
      encryption key.  This information can be mapped BACS issuing control commands, in order to report about the
      'mgt_key_material' parameter
      execution of the Key Distribution Response
      message defined in Section 4.2 of
      [I-D.palombini-ace-key-groupcomm].

D.2.  Provisioning requested operation (e.g.  OK, failure, error)
      and their current operational status.

   o  Software and firmware updates: software and Retrieval firmware updates often
      comprise quite a large amount of Public Keys

   As mentioned in Section 6, it is recommended data.  This can overload a LLN
      that the Group Manager
   acts as trusted key repository, so storing public keys is otherwise typically used to deal with only small amounts
      of group
   members data, on an infrequent base.  Rather than sending software and providing them
      firmware updates as unicast messages to other members of the same group upon
   request.  In each individual device,
      multicasting such updated data to a case, larger group of devices at
      once displays a joining endpoint provides its own public
   key to the Group Manager, as 'Identity credentials' number of benefits.  For instance, it can
      significantly reduce the join
   request, when joining the group (see Appendix D.1).

   After that, network load and decrease the Group Manager should verify that overall
      time latency for propagating this data to all devices.  Even if
      the joining endpoint
   actually owns complete whole update process itself is secured, securing the associated private key, for instance by performing
   a proof-of-possession challenge-response, whose details are out
      individual messages is important, in case updates consist of
   scope.
      relatively large amounts of data.  In case fact, checking individual
      received data piecemeal for tampering avoids that devices store
      large amounts of failure, the Group Manager performs up partially corrupted data and that they detect
      tampering hereof only after all data has been received.  Devices
      receiving software and firmware updates are expected to possibly
      reply back, in order to provide a pre-
   defined maximum number of retries, after which it aborts feedback about the join
   process.

   In case execution of successful challenge-response,
      the Group Manager stores update operation (e.g.  OK, failure, error) and their current
      operational status.

   o  Parameter and configuration update: by means of multicast
      communication, it is possible to update the received public key as associated settings of a group of
      similar devices, both simultaneously and efficiently.  Possible
      parameters are related, for instance, to network load management
      or network access controls.  Devices receiving parameter and
      configuration updates are expected to possibly reply back, to
      provide a feedback about the joining endpoint execution of the update operation
      (e.g.  OK, failure, error) and its
   Endpoint ID.  From then on, that public key will be available their current operational status.

   o  Commissioning of LLNs systems: a commissioning device is
      responsible for
   secure and trusted delivery to other endpoints querying all devices in the group.  A
   possible approach for local network or a group member
      selected subset of them, in order to retrieve the public key discover their presence, and
      be aware of
   other group members is described their capabilities, default configuration, and
      operating conditions.  Queried devices displaying similarities in Section 7
      their capabilities and features, or sharing a common physical
      location can be configured as members of
   [I-D.palombini-ace-key-groupcomm].

   Finally, a single group.  Queried
      devices are expected to reply back to the Group Manager sends commissioning device, in
      order to notify their presence, and provide the join response requested
      information and their current operational status.

   o  Emergency multicast: a particular emergency related information
      (e.g. natural disaster) is generated and multicast by an emergency
      notifier, and relayed to multiple devices.  The latters may reply
      back to the joining
   endpoint, as described emergency notifier, in Appendix D.1.

   The joining endpoint does not have order to provide its own public key if
   that already occurred upon previously joining their feedback
      and local information related to the same or ongoing emergency.  This kind
      of setups should additionally rely on a fault tolerance multicast
      algorithm, such as MPL.

Appendix C.  Example of Group Identifier Format

   This section provides an example of how the Group Identifier (Gid)
   can be specifically formatted.  That is, the Gid can be composed of
   two parts, namely a different
   group under Group Prefix and a Group Epoch.

   The Group Prefix is constant over time and is uniquely defined in the
   set of all the groups associated to the same Group Manager.  However, separately for each
   group under its control,  The
   choice of the Group Manager maintains an updated list Prefix for a given group's Security Context is
   application specific.  The size of active Endpoint IDs associated to the respective endpoint's public
   key.

   Instead, in case the Group Manager does not act as trusted key
   repository, Prefix directly impact
   on the following exchange with maximum number of distinct groups under the same Group Manager can occur
   during the join process.

   1.  The joining endpoint signs its own certificate by using its own
       private key.
   Manager.

   The certificate includes also the identifier of Group Epoch is set to 0 upon the
       issuer Certification Authority (CA).  There group's initialization, and is no restriction on
   incremented by 1 upon completing each renewal of the Certificate Subject included Security Context
   and keying material in the joining endpoint's
       certificate.

   2.  The joining endpoint specifies group (see Section 2.1).  In particular,
   once a new Master Secret has been distributed to the signed certificate as
       'Identity credentials' group, all the
   group members increment by 1 the Group Epoch in the join request (Appendix D.1).  The
       joining endpoint can optionally specify also a list Group Identifier
   of public key
       repositories storing its own certificate.  In such that group.

   As an example, a case, this
       information 3-byte Group Identifier can be mapped composed of: i) a
   1-byte Group Prefix '0xb1' interpreted as a raw byte string; and ii)
   a 2-byte Group Epoch interpreted as an unsigned integer ranging from
   0 to 65535.  Then, after having established the 'pub_keys_repos' parameter of
       the Key Distribution Request message defined in Section 4.1 of
       [I-D.palombini-ace-key-groupcomm].

   3.  When processing Security Common
   Context 61532 times in the join request, group, its Group Identifier will assume
   value '0xb1f05c'.

   Using an immutable Group Prefix for a group assumes that enough time
   elapses between two consecutive usages of the same Group Manager first
       validates Epoch value
   in that group.  This ensures that the certificate by verifying Gid value is temporally unique
   during the signature lifetime of a given message.  Thus, the
       issuer CA, expected highest
   rate for addition/removal of group members and then verifies the signature consequent group
   rekeying should be taken into account for a proper dimensioning of
   the joining
       endpoint.

   4.  The Group Manager stores the association between the Certificate
       Subject Epoch size.

   As discussed in Section 8.5, if endpoints are deployed in multiple
   groups managed by different non-synchronized Group Managers, it is
   possible that Group Identifiers of the joining endpoint's certificate different groups coincide at some
   point in time.  In this case, a recipient has to handle coinciding
   Group Identifiers, and has to try using different OSCORE Security
   Contexts to process an incoming message, until the pair {Group
       ID, Endpoint ID of right one is found
   and the joining endpoint}. If received message is correctly verified.  Therefore, it is favourable
   that Group Identifiers from the
       joining endpoint, the different Group Manager also stores the list of
       public key repositories storing the certificate of the joining
       endpoint.

   When Managers have a group member X wants size that
   result in a small probability of collision.  How small this
   probability should be is up to retrieve the public key system designers.

Appendix D.  Set-up of another
   group member Y in the same group, the endpoint X proceeds as follows.

   1.  The New Endpoints

   An endpoint X contacts joins a group by explicitly interacting with the
   responsible Group Manager, specifying the pair
       {Group ID, Endpoint ID Manager.  When becoming members of a group,
   endpoints are not required to know how many and what endpoints are in
   the same group.

   Communications between a joining endpoint Y}.

   2.  The Group Manager provides the endpoint X with the Certificate
       Subject CS from the certificate of endpoint Y.  If available, and the Group Manager provides the endpoint X also with rely
   on the list CoAP protocol and must be secured.  Specific details on how to
   secure communications between joining endpoints and a Group Manager
   are out of
       public key repositories storing the certificate scope of the endpoint
       Y.

   3. this document.

   The endpoint X retrieves the certificate of Group Manager must verify that the joining endpoint X from a
       key repository storing it, by using is authorized
   to join the group.  To this end, the Certificate Subject CS.

D.3. Group Joining Based on Manager can directly
   authorize the ACE Framework

   The join process joining endpoint, or expect it to register an endpoint as a new member of provide authorization
   evidence previously obtained from a group
   can be based on trusted entity.  Further details
   about the ACE framework for Authentication and
   Authorization in constrained environments [I-D.ietf-ace-oauth-authz],
   built on re-use authorization of OAuth 2.0 [RFC6749]. joining endpoints are out of scope.

   In particular, the approach described in
   [I-D.tiloca-ace-oscoap-joining] uses case of successful authorization check, the ACE framework Group Manager
   generates a Sender ID assigned to delegate the authentication and authorization of joining endpoints to an
   Authorization Server in a trust relation endpoint, before
   proceeding with the rest of the join process.  That is, the Group Manager.  At
   Manager provides the same time, it allows a joining endpoint to establish a secure
   channel with the Group Manager, by leveraging protocol-specific
   profiles of ACE, such as [I-D.ietf-ace-oscore-profile] and
   [I-D.ietf-ace-dtls-authorize], to achieve communication security,
   proof-of-possession and server authentication.

   More specifically keying material and with reference
   parameters to initialize the terminology defined in
   OAuth 2.0:

   o  The joining endpoint acts as ACE Client;

   o OSCORE Security Context (see Section 2).
   The Group Manager acts as ACE Resource Server, with different CoAP
      resources for different groups it is responsible for;

   o  An Authorization Server enables and enforces authorized access actual provisioning of keying material and parameters to the
   joining endpoint to is out of the Group Manager and its CoAP resources
      paired with groups to join.

   Messages exchanged among scope of this document.

   It is RECOMMENDED that the participants follow join process adopts the formats defined approach described
   in [I-D.palombini-ace-key-groupcomm].  Both the joining endpoint [I-D.tiloca-ace-oscoap-joining] and based on the Group Manager have to adopt secure communication also ACE framework for any
   message exchange with the
   Authentication and Authorization Server.  To this end,
   different alternatives are possible, such as OSCORE, DTLS [RFC6347]
   or IPsec [RFC4301]. in constrained environments
   [I-D.ietf-ace-oauth-authz].

Appendix E.  Examples of Synchronization Approaches

   This section describes three possible approaches that can be
   considered by server endpoints to synchronize with sender sequence
   numbers of client endpoints sending group requests.

E.1.  Best-Effort Synchronization

   Upon receiving a group request from a client, a server does not take
   any action to synchonize with the sender sequence number of that
   client.  This provides no assurance at all as to message freshness,
   which can be acceptable in non-critical use cases.

E.2.  Baseline Synchronization

   Upon receiving a group request from a given client for the first
   time, a server initializes its last-seen sender sequence number in
   its Recipient Context associated to that client.  However, the server
   drops the group request without delivering it to the application
   layer.  This provides a reference point to identify if future group
   requests from the same client are fresher than the last one received.

   A replay time interval exists, between when a possibly replayed or
   delayed message is originally transmitted by a given client and the
   first authentic fresh message from that same client is received.
   This can be acceptable for use cases where servers admit such a
   trade-off between performance and assurance of message freshness.

E.3.  Challenge-Response Synchronization

   A server performs a challenge-response exchange with a client, by
   using the Echo Option for CoAP described in Section 2 of
   [I-D.ietf-core-echo-request-tag] and consistently with what specified
   in according to Section 7.5.2 of
   [I-D.ietf-core-object-security].

   That is, upon receiving a group request from a particular client for
   the first time, the server processes the message as described in
   Section 4.2 6.2 of this specification, but, even if valid, does not
   deliver it to the application.  Instead, the server replies to the
   client with a 4.03 Forbidden response message including an Echo
   Option, and stores the option value included therein.

   Upon receiving a 4.03 Forbidden response that includes an Echo Option
   and originates from a verified group member, a client sends a request
   as a unicast message addressed to the same server, echoing the Echo
   Option value.  In particular, the client does not necessarily resend
   the same group request, but can instead send a more recent one, if
   the application permits it.  This makes it possible for the client to
   not retain previously sent group requests for full retransmission,
   unless the application explicitly requires otherwise.  In either
   case, the client uses the sender sequence number value currently
   stored in its own Sender Context.  If the client stores group
   requests for possible retransmission with the Echo Option, it should
   not store a given request for longer than a pre-configured time
   interval.  Note that the unicast request echoing the Echo Option is
   correctly treated and processed as a group message, since the 'kid context'
   field including the Group Identifier of the OSCORE group is still
   present in the OSCORE Option as part of the COSE object (see
   Section 3).

   Upon receiving the unicast request including the Echo Option, the
   server verifies that the option value equals the stored and
   previously sent value; otherwise, the request is silently discarded.
   Then, the server verifies that the unicast request has been received
   within a pre-configured time interval, as described in
   [I-D.ietf-core-echo-request-tag].  In such a case, the request is
   further processed and verified; otherwise, it is silently discarded.
   Finally, the server updates the Recipient Context associated to that
   client, by setting the Replay Window according to the Sequence Number
   from the unicast request conveying the Echo Option.  The server
   either delivers the request to the application if it is an actual
   retransmission of the original one, or discards it otherwise.
   Mechanisms to signal whether the resent request is a full
   retransmission of the original one are out of the scope of this
   specification.

   In case it does not receive a valid unicast request including the
   Echo Option within the configured time interval, the server endpoint
   should perform the same challenge-response upon receiving the next
   group request from that same client.

   A server should not deliver group requests from a given client to the
   application until one valid request from that same client has been
   verified as fresh, as conveying an echoed Echo Option
   [I-D.ietf-core-echo-request-tag].  Also, a server may perform the
   challenge-response described above at any time, if synchronization
   with sender sequence numbers of clients is (believed to be) lost, for
   instance after a device reboot.  It is the role of the application to
   define under what circumstances sender sequence numbers lose
   synchronization.  This can include a minimum gap between the sender
   sequence number of the latest accepted group request from a client
   and the sender sequence number of a group request just received from
   the same client.  A client has to be always ready to perform the challenge-
   response
   challenge-response based on the Echo Option in case a server starts
   it.

   Note that endpoints configured as silent servers are not able to
   perform the challenge-response described above, as they do not store
   a Sender Context to secure the 4.03 Forbidden response to the client.
   Therefore, silent servers should adopt alternative approaches to
   achieve and maintain synchronization with sender sequence numbers of
   clients.

   This approach provides an assurance of absolute message freshness.
   However, it can result in an impact on performance which is
   undesirable or unbearable, especially in large groups where many
   endpoints at the same time might join as new members or lose
   synchronization.

Appendix F.  No Verification of Signatures

   There are some application scenarios using group communication that
   have particularly strict requirements.  One example of this is the
   requirement of low message latency in non-emergency lighting
   applications [I-D.somaraju-ace-multicast].  For those applications
   which have tight performance constraints and relaxed security
   requirements, it can be inconvenient for some endpoints to verify
   digital signatures in order to assert source authenticity of received
   group
   messages.  In other cases, the signature verification can be deferred
   or only checked for specific actions.  For instance, a command to
   turn a bulb on where the bulb is already on does not need the
   signature to be checked.  In such situations, the counter signature
   needs to be included anyway as part of the group message, so that an
   endpoint that needs to validate the signature for any reason has the
   ability to do so.

   In this specification, it is NOT RECOMMENDED that endpoints do not
   verify the counter signature of received group messages.  However, it is
   recognized that there may be situations where it is not always
   required.  The consequence of not doing the signature validation is
   that security in the group is based only on the group-authenticity of
   the shared keying material used for encryption.  That is, endpoints
   in the group have evidence that a received message has been
   originated by a group member, although not specifically identifiable
   in a secure way.  This can violate a number of security requirements,
   as the compromise of any element in the group means that the attacker
   has the ability to control the entire group.  Even worse, the group
   may not be limited in scope, and hence the same keying material might
   be used not only for light bulbs but for locks as well.  Therefore,
   extreme care must be taken in situations where the security
   requirements are relaxed, so that deployment of the system will
   always be done safely.

Appendix G.  Document Updates

   RFC EDITOR: PLEASE REMOVE THIS SECTION.

G.1.  Version -02 to -03

   o  Revised structure and phrasing for improved readability and better
      alignment with draft-ietf-core-object-security.

   o  Added discussion on wrap-Around of Partial IVs (see Section 2.2).

   o  Separate sections for the COSE Object (Section 3) and the OSCORE
      Header Compression (Section 4).

   o  The countersignature is now appended to the encrypted payload of
      the OSCORE message, rather than included in the OSCORE Option (see
      Section 4).

   o  Extended scope of Section 5, now titled " Message Binding,
      Sequence Numbers, Freshness and Replay Protection".

   o  Clarifications about Non-Confirmable messages in Section 5.1
      "Synchronization of Sender Sequence Numbers".

   o  Clarifications about error handling in Section 6 "Message
      Processing".

   o  Compacted list of responsibilities of the Group Manager in
      Section 7.

   o  Revised and extended security considerations in Section 8.

   o  Added IANA considerations for the OSCORE Flag Bits Registry in
      Section 9.

   o  Revised Appendix D, now giving a short high-level description of a
      new endpoint set-up.

G.2.  Version -01 to -02

   o  Terminology has been made more aligned with RFC7252 and draft-
      ietf-core-object-security: i) "client" and "server" replace the
      old "multicaster" and "listener", respectively; ii) "silent
      server" replaces the old "pure listener".

   o  Section 2 has been updated to have the Group Identifier stored in
      the 'ID Context' parameter defined in draft-ietf-core-object-
      security.

   o  Section 3 has been updated with the new format of the Additional
      Authenticated Data.

   o  Major rewriting of Section 4 to better highlight the differences
      with the message processing in draft-ietf-core-object-security.

   o  Added Sections 7.2 and 7.3 discussing security considerations
      about uniqueness of (key, nonce) and collision of group
      identifiers, respectively.

   o  Minor updates to Appendix A.1 about assumptions on multicast
      communication topology and group size.

   o  Updated Appendix C on format of group identifiers, with practical
      implications of possible collisions of group identifiers.

   o  Updated Appendix D.2, adding a pointer to draft-palombini-ace-key-
      groupcomm about retrieval of nodes' public keys through the Group
      Manager.

   o  Minor updates to Appendix E.3 about Challenge-Response
      synchronization of sequence numbers based on the Echo option from
      draft-ietf-core-echo-request-tag.

G.2.

G.3.  Version -00 to -01

   o  Section 1.1 has been updated with the definition of group as
      "security group".

   o  Section 2 has been updated with:

      *  Clarifications on etablishment/derivation of security contexts.

      *  A table summarizing the the additional context elements
         compared to OSCORE.

   o  Section 3 has been updated with:

      *  Examples of request and response messages.

      *  Use of CounterSignature0 rather than CounterSignature.

      *  Additional Authenticated Data including also the signature
         algorithm, while not including the Group Identifier any longer.

   o  Added Section 6, listing the responsibilities of the Group
      Manager.

   o  Added Appendix A (former section), including assumptions and
      security objectives.

   o  Appendix B has been updated with more details on the use cases.

   o  Added Appendix C, providing an example of Group Identifier format.

   o  Appendix D has been updated to be aligned with draft-palombini-
      ace-key-groupcomm.

Acknowledgments

   The authors sincerely thank Stefan Beck, Rolf Blom, Carsten Bormann,
   Esko Dijk, Klaus Hartke, Rikard Hoeglund, Richard Kelsey, John
   Mattsson, Jim Schaad, Ludwig Seitz and Peter van der Stok for their
   feedback and comments.

   The work on this document has been partly supported by the EIT-
   Digital High Impact Initiative ACTIVE.

Authors' Addresses

   Marco Tiloca
   RISE SICS AB
   Isafjordsgatan 22
   Kista  SE-16440 Stockholm
   Sweden

   Email: marco.tiloca@ri.se

   Goeran Selander
   Ericsson AB
   Torshamnsgatan 23
   Kista  SE-16440 Stockholm
   Sweden

   Email: goran.selander@ericsson.com

   Francesca Palombini
   Ericsson AB
   Torshamnsgatan 23
   Kista  SE-16440 Stockholm
   Sweden

   Email: francesca.palombini@ericsson.com

   Jiye Park
   Universitaet Duisburg-Essen
   Schuetzenbahn 70
   Essen  45127
   Germany

   Email: ji-ye.park@uni-due.de