RFC 4947






Network Working Group                                       G. Fairhurst
Request for Comments: 4947                        University of Aberdeen
Category: Informational                                  M.-J. Montpetit
                                       Motorola Connected Home Solutions
                                                               July 2007


  Address Resolution Mechanisms for IP Datagrams over MPEG-2 Networks

Status of This Memo



   This memo provides information for the Internet community.  It does
   not specify an Internet standard of any kind.  Distribution of this
   memo is unlimited.

Copyright Notice



   Copyright (C) The IETF Trust (2007).

Abstract



   This document describes the process of binding/associating IPv4/IPv6
   addresses with MPEG-2 Transport Streams (TS).  This procedure is
   known as Address Resolution (AR) or Neighbor Discovery (ND).  Such
   address resolution complements the higher-layer resource discovery
   tools that are used to advertise IP sessions.

   In MPEG-2 Networks, an IP address must be associated with a Packet ID
   (PID) value and a specific Transmission Multiplex.  This document
   reviews current methods appropriate to a range of technologies (such
   as DVB (Digital Video Broadcasting), ATSC (Advanced Television
   Systems Committee), DOCSIS (Data-Over-Cable Service Interface
   Specifications), and variants).  It also describes the interaction
   with well-known protocols for address management including DHCP, ARP,
   and the ND protocol.
















Fairhurst & Montpetit        Informational                      [Page 1]

RFC 4947       AR Mechanisms for IP over MPEG-2 Networks       July 2007


Table of Contents



   1. Introduction ....................................................3
      1.1. Bridging and Routing .......................................4
   2. Conventions Used in This Document ...............................7
   3. Address Resolution Requirements ................................10
      3.1. Unicast Support ...........................................12
      3.2. Multicast Support .........................................12
   4. MPEG-2 Address Resolution ......................................14
      4.1. Static Configuration ......................................15
           4.1.1. MPEG-2 Cable Networks ..............................15
      4.2. MPEG-2 Table-Based Address Resolution .....................16
           4.2.1. IP/MAC Notification Table (INT) and Its Usage ......17
           4.2.2. Multicast Mapping Table (MMT) and Its Usage ........18
           4.2.3. Application Information Table (AIT) and Its Usage ..18
           4.2.4. Address Resolution in ATSC .........................19
           4.2.5. Comparison of SI/PSI Table Approaches ..............19
      4.3. IP-Based Address Resolution for TS Logical Channels .......19
   5. Mapping IP Addresses to MAC/NPA Addresses ......................21
      5.1. Unidirectional Links Supporting Unidirectional
           Connectivity ..............................................22
      5.2. Unidirectional Links with Bidirectional Connectivity ......23
      5.3. Bidirectional Links .......................................25
      5.4. AR Server .................................................26
      5.5. DHCP Tuning ...............................................27
      5.6. IP Multicast AR ...........................................27
           5.6.1. Multicast/Broadcast Addressing for UDLR ............28
   6. Link Layer Support .............................................29
      6.1. ULE without a Destination MAC/NPA Address (D=1) ...........30
      6.2. ULE with a Destination MAC/NPA Address (D=0) ..............31
      6.3. MPE without LLC/SNAP Encapsulation ........................31
      6.4. MPE with LLC/SNAP Encapsulation ...........................31
      6.5. ULE with Bridging Header Extension (D=1) ..................32
      6.6. ULE with Bridging Header Extension and NPA Address (D=0) ..32
      6.7. MPE with LLC/SNAP & Bridging ..............................33
   7. Conclusions ....................................................33
   8. Security Considerations ........................................34
   9. Acknowledgments ................................................35
   10. References ....................................................35
      10.1. Normative References .....................................35
      10.2. Informative References ...................................36










Fairhurst & Montpetit        Informational                      [Page 2]

RFC 4947       AR Mechanisms for IP over MPEG-2 Networks       July 2007


1.  Introduction



   This document describes the process of binding/associating IPv4/IPv6
   addresses with MPEG-2 Transport Streams (TS).  This procedure is
   known as Address Resolution (AR), or Neighbor Discovery (ND).  Such
   address resolution complements the higher layer resource discovery
   tools that are used to advertise IP sessions.  The document reviews
   current methods appropriate to a range of technologies (DVB, ATSC,
   DOCSIS, and variants).  It also describes the interaction with well-
   known protocols for address management including DHCP, ARP, and the
   ND protocol.

   The MPEG-2 TS provides a time-division multiplexed (TDM) stream that
   may contain audio, video, and data information, including
   encapsulated IP Datagrams [RFC4259], defined in specification ISO/IEC
   138181 [ISO-MPEG2].  Each Layer 2 (L2) frame, known as a TS Packet,
   contains a 4 byte header and a 184 byte payload.  Each TS Packet is
   associated with a single TS Logical Channel, identified by a 13-bit
   Packet ID (PID) value that is carried in the MPEG-2 TS Packet header.

   The MPEG-2 standard also defines a control plane that may be used to
   transmit control information to Receivers in the form of System
   Information (SI) Tables [ETSI-SI], [ETSI-SI1], or Program Specific
   Information (PSI) Tables.

   To utilize the MPEG-2 TS as a L2 link supporting IP, a sender must
   associate an IP address with a particular Transmission Multiplex, and
   within the multiplex, identify the specific PID to be used.  This
   document calls this mapping an AR function.  In some AR schemes, the
   MPEG-2 TS address space is subdivided into logical contexts known as
   Platforms [ETSI-DAT].  Each Platform associates an IP service
   provider with a separate context that shares a common MPEG-2 TS
   (i.e., uses the same PID value).

   MPEG-2 Receivers may use a Network Point of Attachment (NPA)
   [RFC4259] to uniquely identify a L2 node within an MPEG-2
   transmission network.  An example of an NPA is the IEEE Medium Access
   Control (MAC) address.  Where such addresses are used, these must
   also be signalled by the AR procedure.  Finally, address resolution
   could signal the format of the data being transmitted, for example,
   the encapsulation, with any L2 encryption method and any compression
   scheme [RFC4259].

   The numbers of Receivers connected via a single MPEG-2 link may be
   much larger than found in other common LAN technologies (e.g.,
   Ethernet).  This has implications on design/configuration of the
   address resolution mechanisms.  Current routing protocols and some
   multicast application protocols also do not scale to arbitrarily



Fairhurst & Montpetit        Informational                      [Page 3]

RFC 4947       AR Mechanisms for IP over MPEG-2 Networks       July 2007


   large numbers of participants.  Such networks do not by themselves
   introduce an appreciable subnetwork round trip delay, however many
   practical MPEG-2 transmission networks are built using links that may
   introduce a significant path delay (satellite links, use of dial-up
   modem return, cellular return, etc.).  This higher delay may need to
   be accommodated by address resolution protocols that use this
   service.

1.1.  Bridging and Routing



   The following two figures illustrate the use of AR for a routed and a
   bridged subnetwork.  Various other combinations of L2 and L3
   forwarding may also be used over MPEG-2 links (including Receivers
   that are IP end hosts and end hosts directly connected to bridged LAN
   segments).

                           Broadcast Link AR
                           - - - - - - - - -
                           |               |
                           \/
                            1a            2b        2a
                   +--------+              +--------+
               ----+   R1   +----------+---+   R2   +----
                   +--------+ MPEG-2   |   +--------+
                              Link     |
                                       |   +--------+
                                       +---+   R3   +----
                                       |   +--------+
                                       |
                                       |   +--------+
                                       +---+   R4   +----
                                       |   +--------+
                                       |
                                       |

                      Figure 1: A routed MPEG-2 link

   Figure 1 shows a routed MPEG-2 link feeding three downstream routers
   (R2-R4).  AR takes place at the Encapsulator (R1) to identify each
   Receiver at Layer 2 within the IP subnetwork (R2, etc.).

   When considering unicast communication from R1 to R2, several L2
   addresses are involved:

   1a is the L2 (sending) interface address of R1 on the MPEG-2 link.
   2b is the L2 (receiving) interface address of R2 on the MPEG-2 link.
   2a is the L2 (sending) interface address of R2 on the next hop link.




Fairhurst & Montpetit        Informational                      [Page 4]

RFC 4947       AR Mechanisms for IP over MPEG-2 Networks       July 2007


   AR for the MPEG-2 link allows R1 to determine the L2 address (2b)
   corresponding to the next hop Receiver, router R2.

   Figure 2 shows a bridged MPEG-2 link feeding three downstream bridges
   (B2-B4).  AR takes place at the Encapsulator (B1) to identify each
   Receiver at L2 (B2-B4).  AR also takes place across the IP subnetwork
   allowing the Feed router (R1) to identify the downstream Routers at
   Layer 2 (R2, etc.).  The Encapsulator associates a destination
   MAC/NPA address with each bridged PDU sent on an MPEG-2 link.  Two
   methods are defined by ULE (Unidirectional Lightweight Encapsulation)
   [RFC4326]:

   The simplest method uses the L2 address of the transmitted frame.
   This is the MAC address corresponding to the destination within the
   L2 subnetwork (the next hop router, 2b of R2).  This requires each
   Receiver (B2-B4) to associate the receiving MPEG-2 interface with the
   set of MAC addresses that exist on the L2 subnetworks that it feeds.
   Similar considerations apply when IP-based tunnels support L2
   services (including the use of UDLR (Unidirectional Links)
   [RFC3077]).

   It is also possible for a bridging Encapsulator (B1) to encapsulate a
   PDU with a link-specific header that also contains the MAC/NPA
   address associated with a Receiver L2 interface on the MPEG-2 link
   (Figure 2).  In this case, the destination MAC/NPA address of the
   encapsulated frame is set to the Receiver MAC/NPA address (y), rather
   than the address of the final L2 destination.  At a different level,
   an AR binding is also required for R1 to associate the destination L2
   address 2b with R2.  In a subnetwork using bridging, the systems R1
   and R2 will normally use standard IETF-defined AR mechanisms (e.g.,
   IPv4 Address Resolution Protocol (ARP) [RFC826] and the IPv6 Neighbor
   Discovery Protocol (ND) [RFC2461]) edge-to-edge across the IP
   subnetwork.


















Fairhurst & Montpetit        Informational                      [Page 5]

RFC 4947       AR Mechanisms for IP over MPEG-2 Networks       July 2007


                                Subnetwork AR
                      - - - - - - - - - - - - - - - -
                      |                             |

                      |        MPEG-2 Link AR       |
                             - - - - - - - - -
                      |      |               |      |
                      \/     \/
                      1a      x              y      2b        2a
             +--------+  +----+              +----+  +--------+
         ----+   R1   +--| B1 +----------+---+ B2 +--+   R2   +----
             +--------+  +----+ MPEG-2   |   +----+  +--------+
                                Link     |
                                         |   +----+
                                         +---+ B3 +--
                                         |   +----+
                                         |
                                         |   +----+
                                         +---+ B4 +--
                                         |   +----+
                                         |

                       Figure 2: A bridged MPEG-2 link

   Methods also exist to assign IP addresses to Receivers within a
   network (e.g., stateless autoconfiguration [RFC2461], DHCP [RFC2131],
   DHCPv6 [RFC3315], and stateless DHCPv6 [RFC3736]).  Receivers may
   also participate in the remote configuration of the L3 IP addresses
   used in connected equipment (e.g., using DHCP-Relay [RFC3046]).

   The remainder of this document describes current mechanisms and their
   use to associate an IP address with the corresponding TS Multiplex,
   PID value, the MAC/NPA address and/or Platform ID.  A range of
   approaches is described, including Layer 2 mechanisms (using MPEG-2
   SI tables), and protocols at the IP level (including ARP [RFC826] and
   ND [RFC2461]).  Interactions and dependencies between these
   mechanisms and the encapsulation methods are described.  The document
   does not propose or define a new protocol, but does provide guidance
   on issues that would need to be considered to supply IP-based address
   resolution.











Fairhurst & Montpetit        Informational                      [Page 6]

RFC 4947       AR Mechanisms for IP over MPEG-2 Networks       July 2007


2.  Conventions Used in This Document



   AIT: Application Information Table specified by the Multimedia Home
   Platform (MHP) specifications [ETSI-MHP].  This table may carry
   IPv4/IPv6 to MPEG-2 TS address resolution information.

   ATSC: Advanced Television Systems Committee [ATSC].  A framework and
   a set of associated standards for the transmission of video, audio,
   and data using the ISO MPEG-2 standard [ISO-MPEG2].

   b: bit.  For example, one byte consists of 8-bits.

   B: Byte.  Groups of bytes are represented in Internet byte order.



   DSM-CC: Digital Storage Media Command and Control [ISO-DSMCC].  A
   format for the transmission of data and control information carried
   in an MPEG-2 Private Section, defined by the ISO MPEG-2 standard.

   DVB: Digital Video Broadcasting [DVB].  A framework and set of
   associated standards published by the European Telecommunications
   Standards Institute (ETSI) for the transmission of video, audio, and
   data, using the ISO MPEG-2 Standard.

   DVB-RCS: Digital Video Broadcast Return Channel via Satellite.  A
   bidirectional IPv4/IPv6 service employing low-cost Receivers
   [ETSI-RCS].

   DVB-S: Digital Video Broadcast for Satellite [ETSI-DVBS].

   Encapsulator: A network device that receives PDUs and formats these
   into Payload Units (known here as SNDUs) for output as a stream of TS
   Packets.

   Feed Router: The router delivering the IP service over a
   Unidirectional Link.

   INT: Internet/MAC Notification Table.  A unidirectional address
   resolution mechanism using SI and/or PSI Tables.

   L2: Layer 2, the link layer.

   L3: Layer 3, the IP network layer.

   MAC: Medium Access Control [IEEE-802.3].  A link layer protocol
   defined by the IEEE 802.3 standard (or by Ethernet v2).

   MAC Address: A 6-byte link layer address of the format described by
   the Ethernet IEEE 802 standard (see also NPA).



Fairhurst & Montpetit        Informational                      [Page 7]

RFC 4947       AR Mechanisms for IP over MPEG-2 Networks       July 2007


   MAC Header: The link layer header of the IEEE 802.3 standard
   [IEEE-802.3] or Ethernet v2.  It consists of a 6-byte destination
   address, 6-byte source address, and 2 byte type field (see also NPA,
   LLC (Logical Link Control)).

   MHP: Multimedia Home Platform.  An integrated MPEG-2 multimedia
   Receiver, that may (in some cases) support IPv4/IPv6 services
   [ETSI-MHP].

   MMT: Multicast Mapping Table (proprietary extension to DVB-RCS
   [ETSI-RCS] defining an AR table that maps IPv4 multicast addresses to
   PID values).

   MPE: Multiprotocol Encapsulation [ETSI-DAT], [ATSC-A90].  A  method
   that encapsulates PDUs, forming a DSM-CC Table Section.  Each Section
   is sent in a series of TS Packets using a single Stream (TS Logical
   Channel).

   MPEG-2: A set of standards specified by the Motion Picture Experts
   Group (MPEG), and standardized by the International Standards
   Organization (ISO/IEC 113818-1) [ISO-MPEG2], and ITU-T (in H.220).

   NPA: Network Point of Attachment.  A 6-byte destination address
   (resembling an IEEE MAC address) within the MPEG-2 transmission
   network that is used to identify individual Receivers or groups of
   Receivers [RFC4259].

   PAT: Program Association Table.  An MPEG-2 PSI control table.  It
   associates each program with the PID value that is used to send the
   associated PMT (Program Map Table).  The table is sent using the
   well-known PID value of 0x000, and is required for an MPEG-2
   compliant Transport Stream.

   PDU: Protocol Data Unit.  Examples of a PDU include Ethernet frames,
   IPv4 or IPv6 Datagrams, and other network packets.

   PID: Packet Identifier  [ISO-MPEG2].  A 13 bit field carried in the
   header of each TS Packet.  This identifies the TS Logical Channel to
   which a TS Packet belongs [ISO-MPEG2].  The TS Packets that form the
   parts of a Table Section, or other Payload Unit must all carry the
   same PID value.  A PID value of all ones indicates a Null TS Packet
   introduced to maintain a constant bit rate of a TS Multiplex.  There
   is no required relationship between the PID values used for TS
   Logical Channels transmitted using different TS Multiplexes.







Fairhurst & Montpetit        Informational                      [Page 8]

RFC 4947       AR Mechanisms for IP over MPEG-2 Networks       July 2007


   PMT: Program Map Table.  An MPEG-2 PSI control table that associates
   the PID values used by the set of TS Logical Channels/ Streams that
   comprise a program [ISO-MPEG2].  The PID value used to send the PMT
   for a specific program is defined by an entry in the PAT.

   Private Section: A syntactic structure constructed according to Table
   2-30 of [ISO-MPEG2].  The structure may be used to identify private
   information (i.e., not defined by [ISO-MPEG2]) relating to one or
   more elementary streams, or a specific MPEG-2 program, or the entire
   Transport Stream.  Other Standards bodies, e.g., ETSI and ATSC, have
   defined sets of table structures using the private_section structure.
   A Private Section is transmitted as a sequence of TS Packets using a
   TS Logical Channel.  A TS Logical Channel may carry sections from
   more than one set of tables.

   PSI: Program Specific Information [ISO-MPEG2].  PSI is used to convey
   information about services carried in a TS Multiplex.  It is carried
   in one of four specifically identified Table Section constructs
   [ISO-MPEG2], see also SI Table.

   Receiver: Equipment that processes the signal from a TS Multiplex and
   performs filtering and forwarding of encapsulated PDUs to the
   network-layer service (or bridging module when operating at the link
   layer).

   SI Table: Service Information Table [ISO-MPEG2].  In this document,
   this term describes a table that is been defined by another standards
   body to convey information about the services carried in a TS
   Multiplex.  A Table may consist of one or more Table Sections,
   however, all sections of a particular SI Table must be carried over a
   single TS Logical Channel [ISO-MPEG2].

   SNDU: Subnetwork Data Unit.  An encapsulated PDU sent as an MPEG-2
   Payload Unit.

   Table Section: A Payload Unit carrying all or a part of an SI or PSI
   Table [ISO-MPEG2].

   TS: Transport Stream [ISO-MPEG2], a method of transmission at the
   MPEG-2 level using TS Packets; it represents Layer 2 of the ISO/OSI
   reference model.  See also TS Logical Channel and TS Multiplex.

   TS Logical Channel: Transport Stream Logical Channel.  In this
   document, this term identifies a channel at the MPEG-2 level
   [ISO-MPEG2].  This exists at level 2 of the ISO/OSI reference model.
   All packets sent over a TS Logical Channel carry the same PID value
   (this value is unique within a specific TS Multiplex).  The term
   "Stream" is defined in MPEG-2 [ISO-MPEG2].  This describes the



Fairhurst & Montpetit        Informational                      [Page 9]

RFC 4947       AR Mechanisms for IP over MPEG-2 Networks       July 2007


   content carried by a specific TS Logical Channel (see ULE Stream).
   Some PID values are reserved (by MPEG-2) for specific signaling.
   Other standards (e.g., ATSC and DVB) also reserve specific PID
   values.

   TS Multiplex: In this document, this term defines a set of MPEG-2 TS
   Logical Channels sent over a single lower layer connection.  This may
   be a common physical link (i.e., a transmission at a specified symbol
   rate, FEC setting, and transmission frequency) or an encapsulation
   provided by another protocol layer (e.g., Ethernet, or RTP over IP).
   The same TS Logical Channel may be repeated over more than one TS
   Multiplex (possibly associated with a different PID value) [RFC4259],
   for example, to redistribute the same multicast content to two
   terrestrial TV transmission cells.

   TS Packet: A fixed-length 188B unit of data sent over a TS Multiplex
   [ISO-MPEG2].  Each TS Packet carries a 4B header.

   UDL: Unidirectional link: A one-way transmission link.  For example,
   and IP over DVB link using a broadcast satellite link.

   ULE: Unidirectional Lightweight Encapsulation.  A scheme that
   encapsulates PDUs, into SNDUs that are sent in a series of TS Packets
   using a single TS Logical Channel [RFC4326].

   ULE Stream: An MPEG-2 TS Logical Channel that carries only ULE
   encapsulated PDUs.  ULE Streams may be identified by definition of a
   stream_type in SI/PSI [RFC4326, ISO-MPEG2].

3.  Address Resolution Requirements



   The MPEG IP address resolution process is independent of the choice
   of encapsulation and needs to support a set of IP over MPEG-2
   encapsulation formats, including Multi-Protocol Encapsulation (MPE)
   ([ETSI-DAT], [ATSC-A90]) and the IETF-defined Unidirectional
   Lightweight Encapsulation (ULE) [RFC4326].

   The general IP over MPEG-2 AR requirements are summarized below:

      - A scalable architecture that may support large numbers of
        systems within the MPEG-2 Network [RFC4259].

      - A protocol version, to indicate the specific AR protocol in use
        and which may include the supported encapsulation method.

      - A method (e.g., well-known L2/L3 address/addresses) to identify
        the AR Server sourcing the AR information.




Fairhurst & Montpetit        Informational                     [Page 10]

RFC 4947       AR Mechanisms for IP over MPEG-2 Networks       July 2007


      - A method to represent IPv4/IPv6 AR information (including
        security mechanisms to authenticate the AR information to
        protect against address masquerading [RFC3756]).

      - A method to install AR information associated with clients at
        the AR Server (registration).

      - A method for transmission of AR information from an AR Server to
        clients that minimize the transmission cost (link-local
        multicast is preferable to subnet broadcast).

      - Incremental update of the AR information held by clients.

      - Procedures for purging clients of stale AR information.

   An MPEG-2 transmission network may support multiple IP networks.  If
   this is the case, it is important to recognize the scope within which
   an address is resolved to prevent packets from one addressed scope
   leaking into other scopes [RFC4259].  Examples of overlapping IP
   address assignments include:

      (i)   Private unicast addresses (e.g., in IPv4, 10/8 prefix;
            172.16/12 prefix; and 192.168/16 prefix).  Packets with
            these addresses should be confined to one addressed area.
            IPv6 also defines link-local addresses that must not be
            forwarded beyond the link on which they were first sent.

      (ii)  Local scope multicast addresses.  These are only valid
            within the local area (examples for IPv4 include:
            224.0.0/24; 224.0.1/24).  Similar cases exist for some IPv6
            multicast addresses [RFC2375].

      (iii) Scoped multicast addresses [RFC2365] and [RFC2375].
            Forwarding of these addresses is controlled by the scope
            associated with the address.  The addresses are only valid
            within an addressed area (e.g., the 239/8 [RFC2365]).

   Overlapping address assignments may also occur at L2, where the same
   MAC/NPA address is used to identify multiple Receivers [RFC4259]:

      (i)   An MAC/NPA unicast address must be unique within the
            addressed area.  The IEEE-assigned MAC addresses used in
            Ethernet LANs are globally unique.  If the addresses are not
            globally unique, an address must only be re-used by
            Receivers in different addressed (scoped) areas.






Fairhurst & Montpetit        Informational                     [Page 11]

RFC 4947       AR Mechanisms for IP over MPEG-2 Networks       July 2007


      (ii)  The MAC/NPA address broadcast address (a L2 address of all
            ones).  Traffic with this address should be confined to one
            addressed area.

      (iii) IP and other protocols may view sets of L3 multicast
            addresses as link-local.  This may produce unexpected
            results if frames with the corresponding multicast L2
            addresses are distributed to systems in a different L3
            network or multicast scope (Sections 3.2 and 5.6).

   Reception of unicast packets destined for another addressed area will
   lead to an increase in the rate of received packets by systems
   connected via the network.  Reception of the additional network
   traffic may contribute to processing load, but should not lead to
   unexpected protocol behaviour, providing that systems can be uniquely
   addressed at L2.  It does however introduce a potential Denial of
   Service (DoS) opportunity.  When the Receiver operates as an IP
   router, the receipt of such a packet can lead to unexpected protocol
   behaviour.

3.1.  Unicast Support



   Unicast address resolution is required at two levels.

   At the lower level, the IP (or MAC) address needs to be associated
   with a specific TS Logical Channel (PID value) and the corresponding
   TS Multiplex (Section 4).  Each Encapsulator within an MPEG-2 Network
   is associated with a set of unique TS Logical Channels (PID values)
   that it sources [ETSI-DAT, RFC4259].  Within a specific scope, the
   same unicast IP address may therefore be associated with more than
   one Stream, and each Stream contributes different content (e.g., when
   several different IP Encapsulators contribute IP flows destined to
   the same Receiver).  MPEG-2 Networks may also replicate IP packets to
   send the same content (Simulcast) to different Receivers or via
   different TS Multiplexes.  The configuration of the MPEG-2 Network
   must prevent a Receiver accepting duplicated copies of the same IP
   packet.

   At the upper level, the AR procedure needs to associate an IP address
   with a specific MAC/NPA address (Section 5).

3.2.  Multicast Support



   Multicast is an important application for MPEG-2 transmission
   networks, since it exploits the advantages of native support for link
   broadcast.  Multicast address resolution occurs at the network-level
   in associating a specific L2 address with an IP Group Destination
   Address (Section 5.6).  In IPv4 and IPv6 over Ethernet, this



Fairhurst & Montpetit        Informational                     [Page 12]

RFC 4947       AR Mechanisms for IP over MPEG-2 Networks       July 2007


   association is normally a direct mapping, and this is the default
   method also specified in both ULE [RFC4326] and MPE [ETSI-DAT].

   Address resolution must also occur at the MPEG-2 level (Section 4).
   The goal of this multicast address resolution is to allow a Receiver
   to associate an IPv4 or IPv6 multicast address with a specific TS
   Logical Channel and the corresponding TS Multiplex [RFC4259].  This
   association needs to permit a large number of active multicast
   groups, and should minimize the processing load at the Receiver when
   filtering and forwarding IP multicast packets (e.g., by distributing
   the multicast traffic over a number of TS Logical Channels).  Schemes
   that allow hardware filtering can be beneficial, since these may
   relieve the drivers and operating systems from discarding unwanted
   multicast traffic.

   There are two specific functions required for address resolution in
   IP multicast over MPEG-2 Networks:

   (i)  Mapping IP multicast groups to the underlying MPEG-2 TS Logical
        Channel (PID) and the MPEG-2 TS Multiplex at the Encapsulator.

   (ii) Provide signalling information to allow a Receiver to locate an
        IP multicast flow within an MPEG-2 TS Multiplex.

   Methods are required to identify the scope of an address when an
   MPEG-2 Network supports several logical IP networks and carries
   groups within different multicast scopes [RFC4259].

   Appropriate procedures need to specify the correct action when the
   same multicast group is available on separate TS Logical Channels.
   This could arise when different Encapsulators contribute IP packets
   with the same IP Group Destination Address in the ASM (Any-Source
   Multicast) address range.  Another case arises when a Receiver could
   receive more than one copy of the same packet (e.g., when packets are
   replicated across different TS Logical Channels or even different TS
   Multiplexes, a method known as Simulcasting [ETSI-DAT]).  At the IP
   level, the host/router may be unaware of this duplication and this
   needs to be detected by other means.

   When the MPEG-2 Network is peered to the multicast-enabled Internet,
   an arbitrarily large number of IP multicast group destination
   addresses may be in use, and the set forwarded on the transmission
   network may be expected to vary significantly with time.  Some uses
   of IP multicast employ a range of addresses to support a single
   application (e.g., ND [RFC2461], Layered Coding Transport (LCT)
   [RFC3451], and Wave and Equation Based Rate Control (WEBRC)
   [RFC3738]).  The current set of active addresses may be determined
   dynamically via a multicast group membership protocol (e.g., Internet



Fairhurst & Montpetit        Informational                     [Page 13]

RFC 4947       AR Mechanisms for IP over MPEG-2 Networks       July 2007


   Group Management Protocol (IGMP) [RFC3376] and Multicast Listener
   Discovery (MLD) [RFC3810]), via multicast routing (e.g., Protocol
   Independent Multicast (PIM) [RFC4601]) and/or other means (e.g.,
   [RFC3819] and [RFC4605]), however each active address requires a
   binding by the AR method.  Therefore, there are advantages in using a
   method that does not need to explicitly advertise an AR binding for
   each IP traffic flow, but is able to distribute traffic across a
   number of L2 TS Logical Channels (e.g., using a hash/mapping that
   resembles the mapping from IP addresses to MAC addresses [RFC1112,
   RFC2464]).  Such methods can reduce the volume of AR information that
   needs to be distributed, and reduce the AR processing.

   Section 5.6 describes the binding of IP multicast addresses to
   MAC/NPA addresses.

4.  MPEG-2 Address Resolution



   The first part of this section describes the role of MPEG-2
   signalling to identify streams (TS Logical Channels [RFC4259]) within
   the L2 infrastructure.

   At L2, the MPEG-2 Transport Stream [ISO-MPEG2] identifies the
   existence and format of a Stream, using a combination of two PSI
   tables: the Program Association Table (PAT) and entries in the
   program element loop of a Program Map Table (PMT).  PMT Tables are
   sent infrequently and are typically small in size.  The PAT is sent
   using the well-known PID value of 0X000.  This table provides the
   correspondence between a program_number and a PID value.  (The
   program_number is the numeric label associated with a program).  Each
   program in the Table is associated with a specific PID value, used to
   identify a TS Logical Channel (i.e., a TS).  The identified TS is
   used to send the PMT, which associates a set of PID values with the
   individual components of the program.  This approach de-references
   the PID values when the MPEG-2 Network includes multiplexors or re-
   multiplexors that renumber the PID values of the TS Logical Channels
   that they process.

   In addition to signalling the Receiver with the PID value assigned to
   a Stream, PMT entries indicate the presence of Streams using ULE and
   MPE to the variety of devices that may operate in the MPEG-2
   transmission network (multiplexors, remultiplexors, rate shapers,
   advertisement insertion equipment, etc.).

   A multiplexor or remultiplexor may change the PID values associated
   with a Stream during the multiplexing process, the new value being
   reflected in an updated PMT.  TS Packets that carry a PID value that
   is not associated with a PMT entry (an orphan PID), may, and usually
   will be dropped by ISO 13818-1 compliant L2 equipment, resulting in



Fairhurst & Montpetit        Informational                     [Page 14]

RFC 4947       AR Mechanisms for IP over MPEG-2 Networks       July 2007


   the Stream not being forwarded across the transmission network.  In
   networks that do not employ any intermediate devices (e.g., scenarios
   C,E,F of [RFC4259]), or where devices have other means to determine
   the set of PID values in use, the PMT table may still be sent (but is
   not required for this purpose).

   Although the basic PMT information may be used to identify the
   existence of IP traffic, it does not associate a Stream with an IP
   prefix/address.  The remainder of the section describes IP addresses
   resolution mechanisms relating to MPEG-2.

4.1.  Static Configuration



   The static mapping option, where IP addresses or flows are statically
   mapped to specific PIDs is the equivalent to signalling "out-of-
   band".  The application programmer, installing engineer, or user
   receives the mapping via some outside means, not in the MPEG-2 TS.
   This is useful for testing, experimental networks, small subnetworks
   and closed domains.

   A pre-defined set of IP addresses may be used within an MPEG-2
   transmission network.  Prior knowledge of the active set of addresses
   allows appropriate AR records to be constructed for each address, and
   to pre-assign the corresponding PID value (e.g., selected to optimize
   Receiver processing; to group related addresses to the same PID
   value; and/or to reflect a policy for usage of specific ranges of PID
   values).  This presumes that the PID mappings are not modified during
   transmission (Section 4).

   A single "well-known" PID is a specialization of this.  This scheme
   is used by current DOCSIS cable modems [DOCSIS], where all IP traffic
   is placed into the specified TS stream.  MAC filtering (and/or
   Section filtering in MPE) may be used to differentiate subnetworks.

4.1.1.  MPEG-2 Cable Networks



   Cable networks use a different transmission scheme for downstream
   (head-end to cable modem) and upstream (cable modem to head-end)
   transmissions.

   IP/Ethernet packets are sent (on the downstream) to the cable
   modem(s) encapsulated in MPEG-2 TS Packets sent on a single well-
   known TS Logical Channel (PID).  There is no use of in-band
   signalling tables.  On the upstream, the common approach is to use
   Ethernet framing, rather than IP/Ethernet over MPEG-2, although other
   proprietary schemes also continue to be used.





Fairhurst & Montpetit        Informational                     [Page 15]

RFC 4947       AR Mechanisms for IP over MPEG-2 Networks       July 2007


   Until the deployment of DOCSIS and EuroDOCSIS, most address
   resolution schemes for IP traffic in cable networks were proprietary,
   and did not usually employ a table-based address resolution method.
   Proprietary methods continue to be used in some cases where cable
   modems require interaction.  In this case, equipment at the head-end
   may act as gateways between the cable modem and the Internet.  These
   gateways receive L2 information and allocate an IP address.

   DOCSIS uses DHCP for IP client configuration.  The Cable Modem
   Terminal System (CMTS) provides a DHCP Server that allocates IP
   addresses to DOCSIS cable modems.  The MPEG-2 transmission network
   provides a L2 bridged network to the cable modem (Section 1).  This
   usually acts as a DHCP Relay for IP devices [RFC2131], [RFC3046], and
   [RFC3256].  Issues in deployment of IPv6 are described in [RFC4779].

4.2.  MPEG-2 Table-Based Address Resolution



   The information about the set of MPEG-2 Transport Streams carried
   over a TS Multiplex can be distributed via SI/PSI Tables.  These
   tables are usually sent periodically (Section 4).  This design
   requires access to and processing of the SI Table information by each
   Receiver [ETSI-SI], [ETSI-SI1].  This scheme reflects the complexity
   of delivering and coordinating the various Transport Streams
   associated with multimedia TV.  A TS Multiplex may provide AR
   information for IP services by integrating additional information
   into the existing control tables or by transmitting additional SI
   Tables that are specific to the IP service.

   Examples of MPEG-2 Table usage that allows an MPEG-2 Receiver to
   identify the appropriate PID and the multiplex associated with a
   specific IP address include:

   (i)   IP/MAC Notification Table (INT) in the DVB Data standard
         [ETSI-DAT].  This provides unidirectional address resolution of
         IPv4/IPv6 multicast addresses to an MPEG-2 TS.

   (ii)  Application Information Table (AIT) in the Multimedia Home
         Platform (MHP) specifications [ETSI-MHP].

   (iii) Multicast Mapping Table (MMT) is an MPEG-2 Table employed by
         some DVB-RCS systems to provide unidirectional address
         resolution of IPv4 multicast addresses to an MPEG-2 TS.

   The MMT and AIT are used for specific applications, whereas the INT
   [ETSI-DAT] is a more general DVB method that supports MAC, IPv4, and
   IPv6 AR when used in combination with the other MPEG-2 tables
   (Section 4).




Fairhurst & Montpetit        Informational                     [Page 16]

RFC 4947       AR Mechanisms for IP over MPEG-2 Networks       July 2007


4.2.1.  IP/MAC Notification Table (INT) and Its Usage



   The INT provides a set of descriptors to specify addressing in a DVB
   network.  The use of this method is specified for Multiprotocol
   Encapsulation (MPE) [ETSI-DAT].  It provides a method for carrying
   information about the location of IP/L2 flows within a DVB network.
   A Platform_ID identifies the addressing scope for a set of IP/L2
   streams and/or Receivers.  A Platform may span several Transport
   Streams carried by one or multiple TS Multiplexes and represents a
   single IP network with a harmonized address space (scope).  This
   allows for the coexistence of several independent IP/MAC address
   scopes within an MPEG-2 Network.

   The INT allows both fully-specified IP addresses and prefix matching
   to reduce the size of the table (and hence enhance signalling
   efficiency).  An IPv4/IPv6 "subnet mask" may be specified in full
   form or by using a slash notation (e.g., /127).  IP multicast
   addresses can be specified with or without a source (address or
   range), although if a source address is specified, then only the
   slash notation may be used for prefixes.

   In addition, for identification and security descriptors, the
   following descriptors are defined for address binding in INT tables:

   (i)   target_MAC_address_descriptor: A descriptor to describe a
         single or set of MAC addresses (and their mask).

   (ii)  target_MAC_address_range_descriptor: A descriptor that may be
         used to set filters.

   (iii) target_IP_address_descriptor: A descriptor describing a single
         or set of IPv4 unicast or multicast addresses (and their mask).

   (iv)  target_IP_slash_descriptor:  Allows definition and announcement
         of an IPv4 prefix.

   (v)   target_IP_source_slash_descriptor: Uses source and destination
         addresses to target a single or set of systems.

   (vi)  IP/MAC stream_location_descriptor: A descriptor that locates an
         IP/MAC stream in a DVB network.

   The following descriptors provide corresponding functions for IPv6
   addresses:

        target_IPv6_address_descriptor
        target_IPv6_slash_descriptor
        and target_IPv6_source_slash_descriptor



Fairhurst & Montpetit        Informational                     [Page 17]

RFC 4947       AR Mechanisms for IP over MPEG-2 Networks       July 2007


   The ISP_access_mode_descriptor allows specification of a second
   address descriptor to access an ISP via an alternative non-DVB
   (possibly non-IP) network.

   One key benefit is that the approach employs MPEG-2 signalling
   (Section 4) and is integrated with other signalling information.
   This allows the INT to operate in the presence of (re)multiplexors
   [RFC4259] and to refer to PID values that are carried in different TS
   Multiplexes.  This makes it well-suited to a Broadcast TV Scenario
   [RFC4259].

   The principal drawback is a need for an Encapsulator to introduce
   associated PSI/SI MPEG-2 control information.  This control
   information needs to be processed at a Receiver.  This requires
   access to information below the IP layer.  The position of this
   processing within the protocol stack makes it hard to associate the
   results with IP Policy, management, and security functions.  The use
   of centralized management prevents the implementation of a more
   dynamic scheme.

4.2.2.  Multicast Mapping Table (MMT) and Its Usage



   In DVB-RCS, unicast AR is seen as a part of a wider configuration and
   control function and does not employ a specific protocol.

   A Multicast Mapping Table (MMT) may be carried in an MPEG-2 control
   table that associates a set of multicast addresses with the
   corresponding PID values [MMT].  This table allows a DVB-RCS Forward
   Link Subsystem (FLSS) to specify the mapping of IPv4 and IPv6
   multicast addresses to PID values within a specific TS Multiplex.
   Receivers (DVB-RCS Return Channel Satellite Terminals (RCSTs)) may
   use this table to determine the PID values associated with an IP
   multicast flow that it requires to receive.  The MMT is specified by
   the SatLabs Forum [MMT] and is not currently a part of the DVB-RCS
   specification.

4.2.3.  Application Information Table (AIT) and Its Usage



   The DVB Multimedia Home Platform (MHP) specification [ETSI-MHP] does
   not define a specific AR function.  However, an Application
   Information Table (AIT) is defined that allows MHP Receivers to
   receive a variety of control information.  The AIT uses an MPEG-2
   signalling table, providing information about data broadcasts, the
   required activation state of applications carried by a broadcast
   stream, etc.  This information allows a broadcaster to request that a
   Receiver change the activation state of an application, and to direct





Fairhurst & Montpetit        Informational                     [Page 18]

RFC 4947       AR Mechanisms for IP over MPEG-2 Networks       July 2007


   applications to receive specific multicast packet flows (using IPv4
   or IPv6 descriptors).  In MHP, AR is not seen as a specific function,
   but as a part of a wider configuration and control function.

4.2.4.  Address Resolution in ATSC



   ATSC [ATSC-A54A] defines a system that allows transmission of IP
   packets within an MPEG-2 Network.  An MPEG-2 Program (defined by the
   PMT) may contain one or more applications [ATSC-A90] that include IP
   multicast streams [ATSC-A92].  IP multicast data are signalled in the
   PMT using a stream_type indicator of value 0x0D.  A MAC address list
   descriptor [SCTE-1] may also be included in the PMT.

   The approach focuses on applications that serve the transmission
   network.  A method is defined that uses MPEG-2 SI Tables to bind the
   IP multicast media streams and the corresponding Session Description
   Protocol (SDP) announcement streams to particular MPEG-2 Program
   Elements.  Each application constitutes an independent network.  The
   MPEG-2 Network boundaries establish the IP addressing scope.

4.2.5.  Comparison of SI/PSI Table Approaches



   The MPEG-2 methods based on SI/PSI meet the specified requirements of
   the groups that created them and each has their strength:  the INT in
   terms of flexibility and extensibility, the MMT in its simplicity,
   and the AIT in its extensibility.  However, they exhibit scalability
   constraints, represent technology specific solutions, and do not
   fully adopt IP-centric approaches that would enable easier use of the
   MPEG-2 bearer as a link technology within the wider Internet.

4.3.  IP-Based Address Resolution for TS Logical Channels



   As MPEG-2 Networks evolve to become multi-service networks, the use
   of IP protocols is becoming more prevalent.  Most MPEG-2 Networks now
   use some IP protocols for operations and control and data delivery.
   Address resolution information could also be sent using IP transport.
   At the time of writing there is no standards-based IP-level AR
   protocol that supports the MPEG-2 TS.

   There is an opportunity to define an IP-level method that could use
   an IP multicast protocol over a well-known IP multicast address to
   resolve an IP address to a TS Logical Channel (i.e., a Transport
   Stream).  The advantages of using an IP-based address resolution
   include:







Fairhurst & Montpetit        Informational                     [Page 19]

RFC 4947       AR Mechanisms for IP over MPEG-2 Networks       July 2007


   (i)   Simplicity:
         The AR mechanism does not require interpretation of L2 tables;
         this is an advantage especially in the growing market share for
         home network and audio/video networked entities.

   (ii)  Uniformity:
         An IP-based protocol can provide a common method across
         different network scenarios for both IP to MAC address mappings
         and mapping to TS Logical Channels (PID value associated with a
         Stream).

   (iii) Extensibility:
         IP-based AR mechanisms allow an independent evolution of the AR
         protocol.  This includes dynamic methods to request address
         resolution and the ability to include other L2 information
         (e.g., encryption keys).

   (iv)  Integration:
         The information exchanged by IP-based AR protocols can easily
         be integrated as a part of the IP network layer, simplifying
         support for AAA, policy, Operations and Management (OAM),
         mobility, configuration control, etc., that combine AR with
         security.

   The drawbacks of an IP-based method include:

   (i)   It can not operate over an MPEG-2 Network that uses MPEG-2
         remultiplexors [RFC4259] that modify the PID values associated
         with the TS Logical Channels during the multiplexing operation
         (Section 4).  This makes the method unsuitable for use in
         deployed broadcast TV networks [RFC4259].

   (ii)  IP-based methods can introduce concerns about the integrity of
         the information and authentication of the sender [RFC4259].
         (These concerns are also applicable to MPEG-2 Table methods,
         but in this case the information is confined to the L2 network,
         or parts of the network where gateway devices isolate the
         MPEG-2 devices from the larger Internet creating virtual MPEG-2
         private networks.) IP-based solutions should therefore
         implement security mechanisms that may be used to authenticate
         the sender and verify the integrity of the AR information as a
         part of a larger security framework.

   An IP-level method could use an IP multicast protocol running an AR
   Server (see also Section 5.4) over a well-known (or discovered) IP
   multicast address.  To satisfy the requirement for scalability to
   networks with a large number of systems (Section 1), a single packet
   needs to transport multiple AR records and define the intended scope



Fairhurst & Montpetit        Informational                     [Page 20]

RFC 4947       AR Mechanisms for IP over MPEG-2 Networks       July 2007


   for each address.  Methods that employ prefix matching are desirable
   (e.g., where a range of source/destination addresses are matched to a
   single entry).  It can also be beneficial to use methods that permit
   a range of IP addresses to be mapped to a set of TS Logical Channels
   (e.g., a hashing technique similar to the mapping of IP Group
   Destination Addresses to Ethernet MAC addresses [RFC1112] [RFC2464]).

5.  Mapping IP Addresses to MAC/NPA Addresses



   This section reviews IETF protocols that may be used to assign and
   manage the mapping of IP addresses to/from MAC/NPA addresses over
   MPEG-2 Networks.

   An IP Encapsulator requires AR information to select an appropriate
   MAC/NPA address in the SNDU header [RFC4259] (Section 6).  The
   information to complete this header may be taken directly from a
   neighbor/ARP cache, or may require the Encapsulator to retrieve the
   information using an AR protocol.  The way in which this information
   is collected will depend upon whether the Encapsulator functions as a
   Router (at L3) or a Bridge (at L2) (Section 1.1).

   Two IETF-defined protocols for mapping IP addresses to MAC/NPA
   addresses are the Address Resolution Protocol, ARP [RFC826], and the
   Neighbor Discovery protocol, ND [RFC2461], respectively for IPv4 and
   IPv6.  Both protocols are normally used in a bidirectional mode,
   although both also permit unsolicited transmission of mappings.  The
   IPv6 mapping defined in [RFC2464] can result in a large number of
   active MAC multicast addresses (e.g., one for each end host).

   ARP requires support for L2 broadcast packets.  A large number of
   Receivers can lead to a proportional increase in ARP traffic, a
   concern for bandwidth-limited networks.  Transmission delay can also
   impact protocol performance.

   ARP also has a number of security vulnerabilities.  ARP spoofing is
   where a system can be fooled by a rogue device that sends a
   fictitious ARP RESPONSE that includes the IP address of a legitimate
   network system and the MAC of a rogue system.  This causes legitimate
   systems on the network to update their ARP tables with the false
   mapping and then send future packets to the rogue system instead of
   the legitimate system.  Using this method, a rogue system can see
   (and modify) packets sent through the network.

   Secure ARP (SARP) uses a secure tunnel (e.g., between each client and
   a server at a wireless access point or router) [RFC4346].  The router
   ignores any ARP RESPONSEs not associated with clients using the
   secure tunnels.  Therefore, only legitimate ARP RESPONSEs are used




Fairhurst & Montpetit        Informational                     [Page 21]

RFC 4947       AR Mechanisms for IP over MPEG-2 Networks       July 2007


   for updating ARP tables.  SARP requires the installation of software
   at each client.  It suffers from the same scalability issues as the
   standard ARP.

   The ND protocol uses a set of IP multicast addresses.  In large
   networks, many multicast addresses are used, but each client
   typically only listens to a restricted set of group destination
   addresses and little traffic is usually sent in each group.
   Therefore, Layer 2 AR for MPEG-2 Networks must support this in a
   scalable manner.

   A large number of ND messages may cause a large demand for performing
   asymmetric operations.  The base ND protocol limits the rate at which
   multicast responses to solicitations can be sent.  Configurations may
   need to be tuned when operating with large numbers of Receivers.

   The default parameters specified in the ND protocol [RFC2461] can
   introduce interoperability problems (e.g., a failure to resolve when
   the link RTT (round-trip time) exceed 3 seconds) and performance
   degradation (duplicate ND messages with a link RTT > 1 second) when
   used in networks where the link RTT is significantly larger than
   experienced by Ethernet LANs.  Tuning of the protocol parameters
   (e.g., RTR_SOLICITATION_INTERVAL) is therefore recommended when using
   network links with appreciable delay (Section 6.3.2 of [RFC2461]).

   ND has similar security vulnerabilities to ARP.  The Secure Neighbor
   Discovery (SEND) [RFC3971] was developed to address known security
   vulnerabilities in ND [RFC3756].  It can also reduce the AR traffic
   compared to ND.  In addition, SEND does not require the configuration
   of per-host keys and can coexist with the use of both SEND and
   insecure ND on the same link.

   The ND Protocol is also used by IPv6 systems to perform other
   functions beyond address resolution, including Router Solicitation /
   Advertisement, Duplicate Address Detection (DAD), Neighbor
   Unreachability Detection (NUD), and Redirect.  These functions are
   useful for hosts, even when address resolution is not required.

5.1.  Unidirectional Links Supporting Unidirectional Connectivity



   MPEG-2 Networks may provide a Unidirectional Broadcast Link (UDL),
   with no return path.  Such links may be used for unicast applications
   that do not require a return path (e.g., based on UDP), but commonly
   are used for IP multicast content distribution.







Fairhurst & Montpetit        Informational                     [Page 22]

RFC 4947       AR Mechanisms for IP over MPEG-2 Networks       July 2007


                                           /-----\
                         MPEG-2 Uplink    /MPEG-2 \
                      ###################( Network )
                      #                   \       /
                 +----#------+             \--.--/
                 |  Network  |                |
                 |  Provider +                v MPEG-2 Downlink
                 +-----------+                |
                                        +-----v------+
                                        |   MPEG-2   |
                                        |  Receiver  |
                                        +------------+

                Figure 3: Unidirectional connectivity

   The ARP and ND protocols require bidirectional L2/L3 connectivity.
   They do not provide an appropriate method to resolve the remote
   (destination) address in a unidirectional environment.

   Unidirectional links therefore require a separate out-of-band
   configuration method to establish the appropriate AR information at
   the Encapsulator and Receivers.  ULE [RFC4326] defines a mode in
   which the MAC/NPA address is omitted from the SNDU.  In some
   scenarios, this may relieve an Encapsulator of the need for L2 AR.

5.2.  Unidirectional Links with Bidirectional Connectivity



   Bidirectional connectivity may be realized using a unidirectional
   link in combination with another network path.  Common combinations
   are a Feed link using MPEG-2 satellite transmission and a return link
   using terrestrial network infrastructure.  This topology is often
   known as a Hybrid network and has asymmetric network routing.

                                           /-----\
                         MPEG-2 uplink    /MPEG-2 \
                      ###################( Network )
                      #                   \       /
                 +----#------+             \--.--/
                 |  Network  |                |
                 |  Provider +-<-+            v MPEG-2 downlink
                 +-----------+   |            |
                                 |      +-----v------+
                                 +--<<--+   MPEG-2   |
                               Return   |  Receiver  |
                               Path     +------------+

                Figure 4: Bidirectional connectivity




Fairhurst & Montpetit        Informational                     [Page 23]

RFC 4947       AR Mechanisms for IP over MPEG-2 Networks       July 2007


   The Unidirectional Link Routing (UDLR) [RFC3077] protocol may be used
   to overcome issues associated with asymmetric routing.  The Dynamic
   Tunnel Configuration Protocol (DTCP) enables automatic configuration
   of the return path.  UDLR hides the unidirectional routing from the
   IP and upper layer protocols by providing a L2 tunnelling mechanism
   that emulates a bidirectional broadcast link at L2.  A network using
   UDLR has a topology where a Feed Router and all Receivers form a
   logical Local Area Network.  Encapsulating L2 frames allows them to
   be sent through an Internet Path (i.e., bridging).

   Since many unidirectional links employ wireless technology for the
   forward (Feed) link, there may be an appreciable cost associated with
   forwarding traffic on the Feed link.  Therefore, it is often
   desirable to prevent forwarding unnecessary traffic (e.g., for
   multicast this implies control of which groups are forwarded).  The
   implications of forwarding in the return direction must also be
   considered (e.g., asymmetric capacity and loss [RFC3449]).  This
   suggests a need to minimize the volume and frequency of control
   messages.

   Three different AR cases may be identified (each considers sending an
   IP packet to a next-hop IP address that is not currently cached by
   the sender):

   (i)   A Feed Router needs a Receiver MAC/NPA address.

         This occurs when a Feed Router sends an IP packet using the
         Feed UDL to a Receiver whose MAC/NPA address is unknown.  In
         IPv4, the Feed Router sends an ARP REQUEST with the IP address
         of the Receiver.  The Receiver that recognizes its IP address
         replies with an ARP RESPONSE to the MAC/NPA address of the Feed
         Router (e.g., using a UDLR tunnel).  The Feed Router may then
         address IP packets to the unicast MAC/NPA address associated
         with the Receiver.  The ULE encapsulation format also permits
         packets to be sent without specifying a MAC/NPA address, where
         this is desirable (Section 6.1 and 6.5).

   (ii)  A Receiver needs the Feed Router MAC/NPA address.

         This occurs when a Receiver sends an IP packet to a Feed Router
         whose MAC/NPA address is unknown.  In IPv4, the Receiver sends
         an ARP REQUEST with the IP address of the Feed Router (e.g.,
         using a UDLR tunnel).  The Feed Router replies with an ARP
         RESPONSE using the Feed UDL.  The Receiver may then address IP
         packets to the MAC/NPA address of the recipient.






Fairhurst & Montpetit        Informational                     [Page 24]

RFC 4947       AR Mechanisms for IP over MPEG-2 Networks       July 2007


   (iii) A Receiver needs another Receiver MAC/NPA address.

         This occurs when a Receiver sends an IP packet to another
         Receiver whose MAC/NPA address is unknown.  In IPv4, the
         Receiver sends an ARP REQUEST with the IP address of the remote
         Receiver (e.g., using a UDLR tunnel to the Feed Router).  The
         request is forwarded over the Feed UDL.  The target Receiver
         replies with an ARP RESPONSE (e.g., using a UDLR tunnel).  The
         Feed Router forwards the response on the UDL.  The Receiver may
         then address IP packets to the MAC/NPA address of the
         recipient.

   These 3 cases allow any system connected to the UDL to obtain the
   MAC/NPA address of any other system.  Similar exchanges may be
   performed using the ND protocol for IPv6.

   A long round trip delay (via the UDL and UDLR tunnel) impacts the
   performance of the reactive address resolution procedures provided by
   ARP, ND, and SEND.  In contrast to Ethernet, during the interval when
   resolution is taking place, many IP packets may be received that are
   addressed to the AR Target address.  The ARP specification allows an
   interface to discard these packets while awaiting the response to the
   resolution request.  An appropriately sized buffer would however
   prevent this loss.

   In case (iii), the time to complete address resolution may be reduced
   by the use of an AR Server at the Feed (Section 5.4).

   Using DHCP requires prior establishment of the L2 connectivity to a
   DHCP Server.  The delay in establishing return connectivity in UDLR
   networks that use DHCP, may make it beneficial to increase the
   frequency of the DTCP HELLO message.  Further information about
   tuning DHCP is provided in Section 5.5.

5.3.  Bidirectional Links



   Bidirectional IP networks can be and are constructed by a combination
   of two MPEG-2 transmission links.  One link is usually a broadcast
   link that feeds a set of remote Receivers.  Links are also provided
   from Receivers so that the combined link functions as a full duplex
   interface.  Examples of this use include two-way DVB-S satellite
   links and the DVB-RCS system.









Fairhurst & Montpetit        Informational                     [Page 25]

RFC 4947       AR Mechanisms for IP over MPEG-2 Networks       July 2007


5.4.  AR Server



   An AR Server can be used to distribute AR information to Receivers in
   an MPEG-2 Network.  In some topologies, this may significantly reduce
   the time taken for Receivers to discover AR information.

   The AR Server can operate as a proxy responding on behalf of
   Receivers to received AR requests.  When an IPv4 AR request is
   received (e.g., Receiver ARP REQUEST), an AR Server responds by
   (proxy) sending an AR response, providing the appropriate IP to
   MAC/NPA binding (mapping the IP address to the L2 address).

   Information may also be sent unsolicited by the AR Server using
   multicast/broadcast to update the ARP/neighbor cache at the Receivers
   without the need for explicit requests.  The unsolicited method can
   improve scaling in large networks.  Scaling could be further improved
   by distributing a single broadcast/multicast AR message that binds
   multiple IP and MAC/NPA addresses.  This reduces the network capacity
   consumed and simplifies client/server processing in networks with
   large numbers of clients.

   An AR Server can be implemented using IETF-defined Protocols by
   configuring the subnetwork so that AR Requests from Receivers are
   intercepted rather than forwarded to the Feed/broadcast link.  The
   intercepted messages are sent to an AR Server.  The AR Server
   maintains a set of MAC/NPA address bindings.  These may be configured
   or may learned by monitoring ARP messages sent by Receivers.
   Currently defined IETF protocols only allow one binding per message
   (i.e., there is no optimization to conserve L2 bandwidth).

   Equivalent methods could provide IPv6 AR.  Procedures for
   intercepting ND messages are defined in [RFC4389].  To perform an AR
   Server function, the AR information must also be cached.  A caching
   AR proxy stores the system state within a middle-box device.  This
   resembles a classic man-in-the-middle security attack; interactions
   with SEND are described in [SP-ND].

   Methods are needed to purge stale AR data from the cache.  The
   consistency of the cache must also be considered when the Receiver
   bindings can change (e.g., IP mobility, network topology changes, or
   intermittent Receiver connectivity).  In these cases, the use of old
   (stale) information can result in IP packets being directed to an
   inappropriate L2 address, with consequent packet loss.

   Current IETF-defined methods provide bindings of IP addresses to
   MAC/NPA, but do not allow the bindings to other L2 information
   pertinent to MPEG-2 Networks, requiring the use of other methods for




Fairhurst & Montpetit        Informational                     [Page 26]

RFC 4947       AR Mechanisms for IP over MPEG-2 Networks       July 2007


   this function (Section 4).  AR Servers can also be implemented using
   non-IETF AR protocols to provide the AR information required by
   Receivers.

5.5.  DHCP Tuning



   DHCP [RFC2131] and DHCPv6 [RFC3315] may be used over MPEG-2 Networks
   with bidirectional connectivity.  DHCP consists of two components: a
   protocol for delivering system-specific configuration parameters from
   a DHCP Server to a DHCP Client (e.g., default router, DNS server) and
   a mechanism for the allocation of network addresses to systems.

   The configuration of DHCP Servers and DHCP Clients should take into
   account the local link round trip delay (possibly including the
   additional delay from bridging, e.g., using UDLR).  A large number of
   clients can make it desirable to tune the DHCP lease duration and the
   size of the address pool.  Appropriate timer values should also be
   selected: the DHCP messages retransmission timeout, and the maximum
   delay that a DHCP Server waits before deciding that the absence of an
   ICMP echo response indicates that the relevant address is free.

   DHCP Clients may retransmit DHCP messages if they do not receive a
   response.  Some client implementations specify a timeout for the
   DHCPDISCOVER message that is small (e.g., suited to Ethernet delay,
   rather than appropriate to an MPEG-2 Network) providing insufficient
   time for a DHCP Server to respond to a DHCPDISCOVER retransmission
   before expiry of the check on the lease availability (by an ICMP Echo
   Request), resulting in potential address conflict.  This value may
   need to be tuned for MPEG-2 Networks.

5.6.  IP Multicast AR



   Section 3.2 describes the multicast address resolution requirements.
   This section describes L3 address bindings when the destination
   network-layer address is an IP multicast Group Destination Address.

   In MPE [ETSI-DAT], a mapping is specified for the MAC Address based
   on the IP multicast address for IPv4 [RFC1112] and IPv6 [RFC2464].
   (A variant of DVB (DVB-H) uses a modified MAC header [ETSI-DAT]).

   In ULE [RFC4326], the L2 NPA address is optional, and is not
   necessarily required when the Receiver is able to perform efficient
   L3 multicast address filtering.  When present, a mapping is defined
   based on the IP multicast address for IPv4 [RFC1112] and IPv6
   [RFC2464].






Fairhurst & Montpetit        Informational                     [Page 27]

RFC 4947       AR Mechanisms for IP over MPEG-2 Networks       July 2007


   The L2 group addressing method specified in [RFC1112] and [RFC2464]
   can result in more than one IP destination address being mapped to
   the same L2 address.  In Source-Specific Multicast, SSM [RFC3569],
   multicast groups are identified by the combination of the IP source
   and IP destination addresses.  Therefore, senders may independently
   select an IP group destination address that could map to the same L2
   address if forwarded onto the same L2 link.  The resulting addressing
   overlap at L2 can increase the volume of traffic forwarded to L3,
   where it then needs to be filtered.

   These considerations are the same as for Ethernet LANs, and may not
   be of concern to Receivers that can perform efficient L3 filtering.
   Section 3 noted that an MPEG-2 Network may need to support multiple
   addressing scopes at the network and link layers.  Separation of the
   different groups into different Transport Streams is one remedy (with
   signalling of IP to PID value mappings).  Another approach is to
   employ alternate MAC/NPA mappings to those defined in [RFC1112] and
   [RFC2464], but such mappings need to be consistently bound at the
   Encapsulator and Receiver, using AR procedures in a scalable manner.

5.6.1.  Multicast/Broadcast Addressing for UDLR



   UDLR is a Layer 2 solution, in which a Receiver may send
   multicast/broadcast frames that are subsequently forwarded natively
   by a Feed Router (using the topology in Figure 2), and are finally
   received at the Feed interface of the originating Receiver.  This
   multicast forwarding does not include the normal L3 Reverse Path
   Forwarding (RPF) check or L2 spanning tree checks, the processing of
   the IP Time To Live (TTL) field or the filtering of administratively
   scoped multicast addresses.  This raises a need to carefully consider
   multicast support.  To avoid forwarding loops, RFC 3077 notes that a
   Receiver needs to be configured with appropriate filter rules to
   ensure that it discards packets that originate from an attached
   network and are later received over the Feed link.

   When the encapsulation includes an MAC/NPA source address, re-
   broadcast packets may be filtered at the link layer using a filter
   that discards L2 addresses that are local to the Receiver.  In some
   circumstances, systems can send packets with an unknown (all-zero)
   MAC source address (e.g., IGMP Proxy Queriers [RFC4605]), where the
   source at L2 can not be determined at the Receiver.  These packets
   need to be silently discarded, which may prevent running the
   associated services on the Receiver.

   Some encapsulation formats also do not include an MAC/NPA source
   address (Table 1).  Multicast packets may therefore alternatively be
   discarded at the IP layer if their IP source address matches a local
   IP address (or address range).  Systems can send packets with an



Fairhurst & Montpetit        Informational                     [Page 28]

RFC 4947       AR Mechanisms for IP over MPEG-2 Networks       July 2007


   all-zero IP source address (e.g., BOOTP (bootstrap protocol)
   [RFC951], DHCP [RFC2131] and ND [RFC2461]), where the source at L3
   can not be determined at the Receiver these packets need to be
   silently discarded.  This may prevent running the associated services
   at a Receiver, e.g., participation in IPv6 Duplicate Address
   Detection or running a DHCP server.

6.  Link Layer Support



   This section considers link layer (L2) support for address resolution
   in MPEG-2 Networks.  It considers two issues: The code-point used at
   L2 and the efficiency of encapsulation for transmission required to
   support the AR method.  The table below summarizes the options for
   both MPE ([ETSI-DAT], [ATSC-A90]) and ULE [RFC4326] encapsulations.

   [RFC4840] describes issues and concerns that may arise when a link
   can support multiple encapsulations.  In particular, it identifies
   problems that arise when end hosts that belong to the same IP network
   employ different incompatible encapsulation methods.  An Encapsulator
   must therefore use only one method (e.g., ULE or MPE) to support a
   single IP network (i.e., set of IPv4 systems sharing the same subnet
   broadcast address or same IPv6 prefix).  All Receivers in an IP
   network must receive all IP packets that use a broadcast (directed to
   all systems in the IP network) or a local-scope multicast address
   (Section 3).  Packets with these addresses are used by many IP-based
   protocols including service discovery, IP AR, and routing protocols.
   Systems that fail to receive these packets can suffer connectivity
   failure or incorrect behaviour (e.g., they may be unable to
   participate in IP-based discovery, configuration, routing, and
   announcement protocols).  Consistent delivery can be ensured by
   transmitting link-local multicast or broadcast packets using the same
   Stream that is used for unicast packets directed to this network.  A
   Receiver could simultaneously use more than one L2 AR mechanism.
   This presents a potential conflict when the Receiver receives two
   different bindings for the same identifier.  When multiple systems
   advertise AR bindings for the same identifiers (e.g., Encapsulators),
   they must ensure that the advertised information is consistent.
   Conflicts may also arise when L2 protocols duplicate the functions of
   IP-based AR mechanisms.

   In ULE, the bridging format may be used in combination with the
   normal mode to address packets to a Receiver (all ULE Receivers are
   required to implement both methods).  Frames carrying IP packets
   using the ULE Bridging mode, that have a destination address
   corresponding to the MAC address of the Receiver and have an IP
   address corresponding to a Receiver interface, will be delivered to
   the IP stack of the Receiver.  All bridged IP multicast and broadcast
   frames will also be copied to the IP stack of the Receiver.



Fairhurst & Montpetit        Informational                     [Page 29]

RFC 4947       AR Mechanisms for IP over MPEG-2 Networks       July 2007


   Receivers must filter (discard) frames that are received with a
   source address that matches an address of the Receiver itself
   [802.1D].  It must also prevent forwarding frames already sent on a
   connected network.  For each network interface, it must therefore
   filter received frames where the frame source address matches a
   unicast destination address associated with a different network
   interface [802.1D].

   +-------------------------------+--------+----------------------+
   |                               | PDU    |L2 Frame Header Fields|
   | L2 Encapsulation              |overhead+----------------------+
   |                               |[bytes] |src mac|dst mac| type |
   +-------------------------------+--------+-------+-------+------+
   |6.1 ULE without dst MAC address| 8      |   -   |  -    | x    |
   |6.2 ULE with dst MAC address   | 14     |   -   |  x    | x    |
   |6.3 MPE without LLC/SNAP       | 16     |   -   |  x    | -    |
   |6.4 MPE with LLC/SNAP          | 24     |   -   |  x    | x    |
   |6.5 ULE with Bridging extension| 22     |   x   |  x    | x    |
   |6.6 ULE with Bridging & NPA    | 28     |   x   |  x    | x    |
   |6.7 MPE with LLC/SNAP&Bridging | 38     |   x   |  x    | x    |
   +-------------------------------+--------+-------+-------+------+

   Table 1: L2 Support and Overhead (x =supported, - =not supported)

   The remainder of the section describes IETF-specified AR methods for
   use with these encapsulation formats.  Most of these methods rely on
   bidirectional communications (see Sections 5.1, 5.2, and 5.3 for a
   discussion of this).

6.1.  ULE without a Destination MAC/NPA Address (D=1)



   The ULE encapsulation supports a mode (D=1) where the MAC/NPA address
   is not present in the encapsulated frame.  This mode may be used with
   both IPv4 and IPv6.  When used, the Receiver is expected to perform
   L3 filtering of packets based on their IP destination address
   [RFC4326].  This requires careful consideration of the network
   topology when a Receiver is an IP router, or delivers data to an IP
   router (a simple case where this is permitted arises in the
   connection of stub networks at a Receiver that have no connectivity
   to other networks).  Since there is no MAC/NPA address in the SNDU,
   ARP and the ND protocol are not required for AR.

   IPv6 systems can automatically configure their IPv6 network address
   based upon a local MAC address [RFC2462].  To use auto-configuration,
   the IP driver at the Receiver may need to access the MAC/NPA address
   of the receiving interface, even though this value is not being used
   to filter received SNDUs.




Fairhurst & Montpetit        Informational                     [Page 30]

RFC 4947       AR Mechanisms for IP over MPEG-2 Networks       July 2007


   Even when not used for AR, the ND protocol may still be required to
   support DAD, and other IPv6 network-layer functions.  This protocol
   uses a block of IPv6 multicast addresses, which need to be carried by
   the L2 network.  However, since this encapsulation format does not
   provide a MAC source address, there are topologies (e.g., Section
   5.6.1) where a system can not differentiate DAD packets that were
   originally sent by itself and were re-broadcast, from those that may
   have been sent by another system with the same L3 address.
   Therefore, DAD can not be used with this encapsulation format in
   topologies where this L2 forwarding may occur.

6.2.  ULE with a Destination MAC/NPA Address (D=0)



   The IPv4 Address Resolution Protocol (ARP) [RFC826] is identified by
   an IEEE EtherType and may be used over ULE [RFC4326].  Although no
   MAC source address is present in the ULE SNDU, the ARP protocol still
   communicates the source MAC (hardware) address in the ARP record
   payload of any query messages that it generates.

   The IPv6 ND protocol is supported.  The protocol uses a block of IPv6
   multicast addresses, which need to be carried by the L2 network.  The
   protocol uses a block of IPv6 multicast addresses, which need to be
   carried by the L2 network.  However, since this encapsulation format
   does not provide a MAC source address, there are topologies (e.g.,
   Section 5.6.1) where a system can not differentiate DAD packets that
   were originally sent by itself and were re-broadcast, from those that
   may have been sent by another system with the same L3 address.
   Therefore, DAD can not be used with this encapsulation format in
   topologies where this L2 forwarding may occur.

6.3.  MPE without LLC/SNAP Encapsulation



   This is the default (and sometimes only) mode specified by most MPE
   Encapsulators.  MPE does not provide an EtherType field and therefore
   can not support the Address Resolution Protocol (ARP) [RFC826].

   IPv6 is not supported in this encapsulation format, and therefore it
   is not appropriate to consider the ND protocol.

6.4.  MPE with LLC/SNAP Encapsulation



   The LLC/SNAP (Sub-Network Access Protocol) format of MPE provides an
   EtherType field and therefore may support ARP [RFC826].  There is no
   specification to define how this is performed.  No MAC source address
   is present in the SNDU, although the protocol communicates the source
   MAC address in the ARP record payload of any query messages that it
   generates.




Fairhurst & Montpetit        Informational                     [Page 31]

RFC 4947       AR Mechanisms for IP over MPEG-2 Networks       July 2007


   The IPv6 ND protocol is supported using The LLC/SNAP format of MPE.
   This requires specific multicast addresses to be carried by the L2
   network.  The IPv6 ND protocol is supported.  The protocol uses a
   block of IPv6 multicast addresses, which need to be carried by the L2
   network.  However, since this encapsulation format does not provide a
   MAC source address, there are topologies (e.g., Section 5.6.1) where
   a system can not differentiate DAD packets that were originally sent
   by itself and were re-broadcast, from those that may have been sent
   by another system with the same L3 address.  Therefore, DAD can not
   be used with this encapsulation format in topologies where this L2
   forwarding may occur.

6.5.  ULE with Bridging Header Extension (D=1)



   The ULE encapsulation supports a bridging extension header that
   supplies both a source and destination MAC address.  This can be used
   without an NPA address (D=1).  When no other Extension Headers
   precede this Extension, the MAC destination address has the same
   position in the ULE SNDU as that used for an NPA destination address.
   The Receiver may optionally be configured so that the MAC destination
   address value is identical to a Receiver NPA address.

   At the Encapsulator, the ULE MAC/NPA destination address is
   determined by a L2 forwarding decision.  Received frames may be
   forwarded or may be addressed to the Receiver itself.  As in other L2
   LANs, the Receiver may choose to filter received frames based on a
   configured MAC destination address filter.  ARP and ND messages may
   be carried within a PDU that is bridged by this encapsulation format.
   Where the topology may result in subsequent reception of re-
   broadcast copies of multicast frames that were originally sent by a
   Receiver (e.g., Section 5.6.1), the system must discard frames that
   are received with a source address that it used in frames sent from
   the same interface [802.1D].  This prevents duplication on the
   bridged network (e.g., this would otherwise invoke DAD).

6.6.  ULE with Bridging Header Extension and NPA Address (D=0)



   The combination of an NPA address (D=0) and a bridging extension
   header are allowed in ULE.  This SNDU format supplies both a source
   and destination MAC address and a NPA destination address (i.e.,
   Receiver MAC/NPA address).

   At the Encapsulator, the value of the ULE MAC/NPA destination address
   is determined by a L2 forwarding decision.  At the Receiver, frames
   may be forwarded or may be addressed to the Receiver itself.  As in
   other L2 LANs, the Receiver may choose to filter received frames
   based on a configured MAC destination address filter.  ARP and ND
   messages may be carried within a PDU that is bridged by this



Fairhurst & Montpetit        Informational                     [Page 32]

RFC 4947       AR Mechanisms for IP over MPEG-2 Networks       July 2007


   encapsulation format.  Where the topology may result in the
   subsequent reception of re-broadcast copies of multicast frames, that
   were originally sent by a Receiver (e.g., Section 5.6.1), the system
   must discard frames that are received with a source address that it
   used in frames sent from the same interface [802.1D].  This prevents
   duplication on the bridged network (e.g., this would otherwise invoke
   DAD).

6.7.  MPE with LLC/SNAP & Bridging



   The LLC/SNAP format MPE frames may optionally support an IEEE
   bridging header [LLC].  This header supplies both a source and
   destination MAC address, at the expense of larger encapsulation
   overhead.  The format defines two MAC destination addresses, one
   associated with the MPE SNDU (i.e., Receiver MAC address) and one
   with the bridged MAC frame (i.e., the MAC address of the intended
   recipient in the remote LAN).

   At the Encapsulator, the MPE MAC destination address is determined by
   a L2 forwarding decision.  There is currently no formal description
   of the Receiver processing for this encapsulation format.  A Receiver
   may forward frames or they may be addressed to the Receiver itself.
   As in other L2 LANs, the Receiver may choose to filter received
   frames based on a configured MAC destination address filter.  ARP and
   ND messages may be carried within a PDU that is bridged by this
   encapsulation format.  The MPE MAC destination address is determined
   by a L2 forwarding decision.  Where the topology may result in a
   subsequent reception of re-broadcast copies of multicast frames, that
   were originally sent by a Receiver (e.g., Section 5.6.1), the system
   must discard frames that are received with a source address that it
   used in frames sent from the same interface [802.1D].  This prevents
   duplication on the bridged network (e.g., this would otherwise invoke
   DAD).

7.  Conclusions



   This document describes addressing and address resolution issues for
   IP protocols over MPEG-2 transmission networks using both wired and
   wireless technologies.  A number of specific IETF protocols are
   discussed along with their expected behaviour over MPEG-2
   transmission networks.  Recommendations for their usage are provided.

   There is no single common approach used in all MPEG-2 Networks.  A
   static binding may be configured for IP addresses and PIDs (as in
   some cable networks).  In broadcast networks, this information is
   normally provided by the Encapsulator/Multiplexor and carried in
   signalling tables (e.g., AIT in MHP, the IP Notification Table, INT,




Fairhurst & Montpetit        Informational                     [Page 33]

RFC 4947       AR Mechanisms for IP over MPEG-2 Networks       July 2007


   of DVB and the DVB-RCS Multicast Mapping Table, and MMT).  This
   document has reviewed the status of these current address resolution
   mechanisms in MPEG-2 transmission networks and defined their usage.

   The document also considers a unified IP-based method for AR that
   could be independent of the physical layer, but does not define a new
   protocol.  It examines the design criteria for a method, with
   recommendations to ensure scalability and improve support for the IP
   protocol stack.

8.  Security Considerations



   The normal security issues relating to the use of wireless links for
   transmission of Internet traffic should be considered.

   L2 signalling in MPEG-2 transmission networks is currently provided
   by (periodic) broadcasting of information in the control plane using
   PSI/SI tables (Section 4).  A loss or modification of the SI
   information may result in an inability to identify the TS Logical
   Channel (PID) that is used for a service.  This will prevent
   reception of the intended IP packet stream.

   There are known security issues relating to the use of unsecured
   address resolution [RFC3756].  Readers are also referred to the known
   security issues when mapping IP addresses to MAC/NPA addresses using
   ARP [RFC826] and ND [RFC2461].  It is recommended that AR protocols
   support authentication of the source of AR messages and the integrity
   of the AR information, this avoids known security vulnerabilities
   resulting from insertion of unauthorized AR messages within a L2
   infrastructure.  For IPv6, the SEND protocol [RFC3971] may be used in
   place of ND.  This defines security mechanisms that can protect AR.

   AR protocols can also be protected by the use of L2 security methods
   (e.g., Encryption of the ULE SNDU [IPDVB-SEC]).  When these methods
   are used, the security of ARP and ND can be comparable to that of a
   private LAN: A Receiver will only accept ARP or ND transmissions from
   the set of peer senders that share a common group encryption and
   common group authentication key provided by the L2 key management.

   AR Servers (Section 5.4) are susceptible to the same kind of security
   issues as end hosts using unsecured AR.  These issues include
   hijacking traffic and denial-of-service within the subnet.  Malicious
   nodes within the subnet can take advantage of this property, and
   hijack traffic.  In addition, an AR Server is essentially a
   legitimate man-in-the-middle, which implies that there is a need to
   distinguish such proxies from unwanted man-in-the-middle attackers.
   This document does not introduce any new mechanisms for the




Fairhurst & Montpetit        Informational                     [Page 34]

RFC 4947       AR Mechanisms for IP over MPEG-2 Networks       July 2007


   protection of these AR functions (e.g., authenticating servers, or
   defining AR Servers that interoperate with the SEND protocol
   [SP-ND]).

9.  Acknowledgments



   The authors wish to thank the IPDVB WG members for their inputs and
   in particular, Rod Walsh, Jun Takei, and Michael Mercurio.  The
   authors also acknowledge the support of the European Space Agency.
   Martin Stiemerling contributed descriptions of scenarios,
   configuration, and provided extensive proof reading.  Hidetaka
   Izumiyama contributed on UDLR and IPv6 issues.  A number of issues
   discussed in the UDLR working group have also provided valuable
   inputs to this document (summarized in "Experiments with RFC 3077",
   July 2003).

10.  References



10.1.  Normative References



   [ETSI-DAT]    EN 301 192, "Specifications for Data Broadcasting",
                 v1.3.1, European Telecommunications Standards Institute
                 (ETSI), May 2003.

   [ETSI-MHP]    TS 101 812, "Digital Video Broadcasting (DVB);
                 Multimedia Home Platform (MHP) Specification", v1.2.1,
                 European Telecommunications Standards Institute (ETSI),
                 June 2002.

   [ETSI-SI]     EN 300 468, "Digital Video Broadcasting (DVB);
                 Specification for Service Information (SI) in DVB
                 systems", v1.7.1, European Telecommunications Standards
                 Institute (ETSI), December 2005.

   [ISO-MPEG2]   ISO/IEC IS 13818-1, "Information technology -- Generic
                 coding of moving pictures and associated audio
                 information -- Part 1: Systems", International
                 Standards Organization (ISO), 2000.

   [RFC826]      Plummer, D., "Ethernet Address Resolution Protocol: Or
                 Converting Network Protocol Addresses to 48.bit
                 Ethernet Address for Transmission on Ethernet
                 Hardware", STD 37, RFC 826, November 1982.

   [RFC1112]     Deering, S., "Host extensions for IP multicasting", STD
                 5, RFC 1112, August 1989.





Fairhurst & Montpetit        Informational                     [Page 35]

RFC 4947       AR Mechanisms for IP over MPEG-2 Networks       July 2007


   [RFC2461]     Narten, T., Nordmark, E., and W. Simpson, "Neighbor
                 Discovery for IP Version 6 (IPv6)", RFC 2461, December
                 1998.

   [RFC2464]     Crawford, M., "Transmission of IPv6 Packets over
                 Ethernet Networks", RFC 2464, December 1998.

   [RFC2131]     Droms, R., "Dynamic Host Configuration Protocol", RFC
                 2131, March 1997.

   [RFC3077]     Duros, E., Dabbous, W., Izumiyama, H., Fujii, N., and
                 Y. Zhang, "A Link-Layer Tunneling Mechanism for
                 Unidirectional Links", RFC 3077, March 2001.

   [RFC3315]     Droms, R., Bound, J., Volz, B., Lemon, T., Perkins, C.,
                 and M. Carney, "Dynamic Host Configuration Protocol for
                 IPv6 (DHCPv6)", RFC 3315, July 2003.

   [RFC3736]     Droms, R., "Stateless Dynamic Host Configuration
                 Protocol (DHCP) Service for IPv6", RFC 3736, April
                 2004.

   [RFC4326]     Fairhurst, G. and B. Collini-Nocker, "Unidirectional
                 Lightweight Encapsulation (ULE) for Transmission of IP
                 Datagrams over an MPEG-2 Transport Stream (TS)", RFC
                 4326, December 2005.

10.2.  Informative References



   [802.1D]      IEEE 802.1D, "IEEE Standard for Local and Metropolitan
                 Area Networks:  Media Access Control (MAC) Bridges",
                 IEEE, 2004.

   [802.3]       IEEE 802.3, "Local and metropolitan area networks-
                 Specific requirements Part 3: Carrier sense multiple
                 access with collision detection (CSMA/CD) access method
                 and physical layer specifications", IEEE Computer
                 Society, (also ISO/IEC 8802-3), 2002.

   [ATSC]        A/53C, "ATSC Digital Television Standard", Advanced
                 Television Systems Committee (ATSC), Doc. A/53C, 2004.

   [ATSC-A54A]   A/54A, "Guide to the use of the ATSC Digital Television
                 Standard", Advanced Television Systems Committee
                 (ATSC), Doc. A/54A, 2003.

   [ATSC-A90]    A/90, "ATSC Data Broadcast Standard", Advanced
                 Television Systems Committee (ATSC), Doc. A/90, 2000.



Fairhurst & Montpetit        Informational                     [Page 36]

RFC 4947       AR Mechanisms for IP over MPEG-2 Networks       July 2007


   [ATSC-A92]    A/92,  "Delivery of IP Multicast Sessions over ATSC
                 Data Broadcast", Advanced Television Systems Committee
                 (ATSC), Doc. A/92, 2002.

   [DOCSIS]      "Data-Over-Cable Service Interface Specifications,
                 DOCSIS 2.0, Radio Frequency Interface Specification",
                 CableLabs, document CM-SP-RFIv2.0-I10-051209, 2005.

   [DVB]         Digital Video Broadcasting (DVB) Project.
                 http://www.dvb.org.

   [ETSI-DVBS]   EN 301 421,"Digital Video Broadcasting (DVB);
                 Modulation and Coding for DBS satellite systems at
                 11/12 GHz", European Telecommunications Standards
                 Institute (ETSI).

   [ETSI-RCS]    EN 301 790, "Digital Video Broadcasting (DVB);
                 Interaction channel for satellite distribution
                 Systems", European Telecommunications Standards
                 Institute (ETSI).

   [ETSI-SI1]    TR 101 162, "Digital Video Broadcasting (DVB);
                 Allocation of Service Information (SI) codes for DVB
                 systems", European Telecommunications Standards
                 Institute (ETSI).

   [IPDVB-SEC]   H. Cruickshank, S. Iyengar, L. Duquerroy, P. Pillai,
                 "Security requirements for the Unidirectional
                 Lightweight Encapsulation (ULE) protocol", Work in
                 Progress, May 2007.

   [ISO-DSMCC]   ISO/IEC IS 13818-6, "Information technology -- Generic
                 coding of moving pictures and associated audio
                 information -- Part 6: Extensions for DSM-CC is a full
                 software implementation", International Standards
                 Organization (ISO), 2002.

   [LLC]         ISO/IEC 8802.2, "Information technology;
                 Telecommunications and information exchange between
                 systems; Local and metropolitan area networks; Specific
                 requirements; Part 2: Logical Link Control",
                 International Standards Organization (ISO), 1998.

   [MMT]         "SatLabs System Recommendations, Part 1, General
                 Specifications", Version 2.0, SatLabs Forum, 2006.
                 http://satlabs.org/pdf/
                 SatLabs_System_Recommendations_v2.0_general.pdf.




Fairhurst & Montpetit        Informational                     [Page 37]

RFC 4947       AR Mechanisms for IP over MPEG-2 Networks       July 2007


   [RFC951]      Croft, W. and J. Gilmore, "Bootstrap Protocol", RFC
                 951, September 1985.

   [RFC2365]     Meyer, D., "Administratively Scoped IP Multicast", BCP
                 23, RFC 2365, July 1998.

   [RFC2375]     Hinden, R. and S. Deering, "IPv6 Multicast Address
                 Assignments", RFC 2375, July 1998.

   [RFC2462]     Thomson, S. and T. Narten, "IPv6 Stateless Address
                 Autoconfiguration", RFC 2462, December 1998.

   [RFC3046]     Patrick, M., "DHCP Relay Agent Information Option", RFC
                 3046, January 2001.

   [RFC3256]     Jones, D. and R. Woundy, "The DOCSIS (Data-Over-Cable
                 Service Interface Specifications) Device Class DHCP
                 (Dynamic Host Configuration Protocol) Relay Agent
                 Information Sub-option", RFC 3256, April 2002.

   [RFC3376]     Cain, B., Deering, S., Kouvelas, I., Fenner, B., and A.
                 Thyagarajan, "Internet Group Management Protocol,
                 Version 3", RFC 3376, October 2002.

   [RFC3449]     Balakrishnan, H., Padmanabhan, V., Fairhurst, G., and
                 M. Sooriyabandara, "TCP Performance Implications of
                 Network Path Asymmetry", BCP 69, RFC 3449, December
                 2002.

   [RFC3451]     Luby, M., Gemmell, J., Vicisano, L., Rizzo, L.,
                 Handley, M., and J. Crowcroft, "Layered Coding
                 Transport (LCT) Building Block", RFC 3451, December
                 2002.

   [RFC3569]     Bhattacharyya, S., "An Overview of Source-Specific
                 Multicast (SSM)", RFC 3569, July 2003.

   [RFC3756]     Nikander, P., Kempf, J., and E. Nordmark, "IPv6
                 Neighbor Discovery (ND) Trust Models and Threats", RFC
                 3756, May 2004.

   [RFC3738]     Luby, M. and V. Goyal, "Wave and Equation Based Rate
                 Control (WEBRC) Building Block", RFC 3738, April 2004.

   [RFC3810]     Vida, R. and L. Costa, "Multicast Listener Discovery
                 Version 2 (MLDv2) for IPv6", RFC 3810, June 2004.





Fairhurst & Montpetit        Informational                     [Page 38]

RFC 4947       AR Mechanisms for IP over MPEG-2 Networks       July 2007


   [RFC3819]     Karn, P., Bormann, C., Fairhurst, G., Grossman, D.,
                 Ludwig, R., Mahdavi, J., Montenegro, G., Touch, J., and
                 L. Wood, "Advice for Internet Subnetwork Designers",
                 BCP 89, RFC 3819, July 2004.

   [RFC3971]     Arkko, J., Kempf, J., Zill, B., and P. Nikander,
                 "SEcure Neighbor Discovery (SEND)", RFC 3971, March
                 2005.

   [RFC4259]     Weis, B., "The Use of RSA/SHA-1 Signatures within
                 Encapsulating Security Payload (ESP) and Authentication
                 Header (AH)", RFC 4359, January 2006.

   [RFC4346]     Dierks, T. and E. Rescorla, "The Transport Layer
                 Security (TLS) Protocol Version 1.1", RFC 4346, April
                 2006.

   [RFC4389]     Thaler, D., Talwar, M., and C. Patel, "Neighbor
                 Discovery Proxies (ND Proxy)", RFC 4389, April 2006.

   [RFC4601]     Fenner, B., Handley, M., Holbrook, H., and I. Kouvelas,
                 "Protocol Independent Multicast - Sparse Mode (PIM-SM):
                 Protocol Specification (Revised)", RFC 4601, August
                 2006.

   [RFC4605]     Fenner, B., He, H., Haberman, B., and H. Sandick,
                 "Internet Group Management Protocol (IGMP) / Multicast
                 Listener Discovery (MLD)-Based Multicast Forwarding
                 ("IGMP/MLD Proxying")", RFC 4605, August 2006.

   [RFC4779]     Asadullah, S., Ahmed, A., Popoviciu, C., Savola, P.,
                 and J. Palet, "ISP IPv6 Deployment Scenarios in
                 Broadband Access Networks", RFC 4779, January 2007.

   [RFC4840]     Aboba, B., Davies, E., and D. Thaler, "Multiple
                 Encapsulation Methods Considered Harmful", RFC 4840,
                 April 2007.

   [SCTE-1]      "IP Multicast for Digital MPEG Networks", SCTE DVS
                 311r6, March 2002.

   [SP-ND]       Daley, G., "Securing Proxy Neighbour Discovery Problem
                 Statement", Work in Progress, February 2005.








Fairhurst & Montpetit        Informational                     [Page 39]

RFC 4947       AR Mechanisms for IP over MPEG-2 Networks       July 2007


Authors' Addresses



   Godred Fairhurst
   Department of Engineering
   University of Aberdeen
   Aberdeen, AB24 3UE
   UK

   EMail: gorry@erg.abdn.ac.uk
   URL: http://www.erg.abdn.ac.uk/users/gorry


   Marie-Jose Montpetit
   Motorola Connected Home Solutions
   Advanced Technology
   55 Hayden Avenue, 3rd Floor
   Lexington, Massachusetts  02421
   USA

   EMail: mmontpetit@motorola.com































Fairhurst & Montpetit        Informational                     [Page 40]

RFC 4947       AR Mechanisms for IP over MPEG-2 Networks       July 2007


Full Copyright Statement



   Copyright (C) The IETF Trust (2007).

   This document is subject to the rights, licenses and restrictions
   contained in BCP 78, and except as set forth therein, the authors
   retain all their rights.

   This document and the information contained herein are provided on an
   "AS IS" basis and THE CONTRIBUTOR, THE ORGANIZATION HE/SHE REPRESENTS
   OR IS SPONSORED BY (IF ANY), THE INTERNET SOCIETY, THE IETF TRUST AND
   THE INTERNET ENGINEERING TASK FORCE DISCLAIM ALL WARRANTIES, EXPRESS
   OR IMPLIED, INCLUDING BUT NOT LIMITED TO ANY WARRANTY THAT THE USE OF
   THE INFORMATION HEREIN WILL NOT INFRINGE ANY RIGHTS OR ANY IMPLIED
   WARRANTIES OF MERCHANTABILITY OR FITNESS FOR A PARTICULAR PURPOSE.

Intellectual Property



   The IETF takes no position regarding the validity or scope of any
   Intellectual Property Rights or other rights that might be claimed to
   pertain to the implementation or use of the technology described in
   this document or the extent to which any license under such rights
   might or might not be available; nor does it represent that it has
   made any independent effort to identify any such rights.  Information
   on the procedures with respect to rights in RFC documents can be
   found in BCP 78 and BCP 79.

   Copies of IPR disclosures made to the IETF Secretariat and any
   assurances of licenses to be made available, or the result of an
   attempt made to obtain a general license or permission for the use of
   such proprietary rights by implementers or users of this
   specification can be obtained from the IETF on-line IPR repository at
   http://www.ietf.org/ipr.

   The IETF invites any interested party to bring to its attention any
   copyrights, patents or patent applications, or other proprietary
   rights that may cover technology that may be required to implement
   this standard.  Please address the information to the IETF at
   ietf-ipr@ietf.org.

Acknowledgement



   Funding for the RFC Editor function is currently provided by the
   Internet Society.







Fairhurst & Montpetit        Informational                     [Page 41]