Mobile Ad hoc Networking (MANET)                              T. Clausen
Internet-Draft                          LIX, Ecole Polytechnique, France
Expires: September 7, December 28, 2006                                   C. Dearlove
                                         BAE Systems Advanced Technology
                                                                  Centre
                                                              P. Jacquet
                                                 Project Hipercom, INRIA
                                                  The OLSRv2 Design Team
                                                     MANET Working Group
                                                           March 6,
                                                           June 26, 2006

          The Optimized Link-State Routing Protocol version 2
                       draft-ietf-manet-olsrv2-01
                       draft-ietf-manet-olsrv2-02

Status of this Memo

   By submitting this Internet-Draft, each author represents that any
   applicable patent or other IPR claims of which he or she is aware
   have been or will be disclosed, and any of which he or she becomes
   aware will be disclosed, in accordance with Section 6 of BCP 79.

   Internet-Drafts are working documents of the Internet Engineering
   Task Force (IETF), its areas, and its working groups.  Note that
   other groups may also distribute working documents as Internet-
   Drafts.

   Internet-Drafts are draft documents valid for a maximum of six months
   and may be updated, replaced, or obsoleted by other documents at any
   time.  It is inappropriate to use Internet-Drafts as reference
   material or to cite them other than as "work in progress."

   The list of current Internet-Drafts can be accessed at
   http://www.ietf.org/ietf/1id-abstracts.txt.

   The list of Internet-Draft Shadow Directories can be accessed at
   http://www.ietf.org/shadow.html.

   This Internet-Draft will expire on September 7, December 28, 2006.

Copyright Notice

   Copyright (C) The Internet Society (2006).

Abstract

   This document describes version 2 of the Optimized Link State Routing
   (OLSRv2) protocol for mobile ad hoc networks.  The protocol is embodies
   an optimization of the classical link state algorithm tailored to the
   requirements of a mobile wireless LAN.

   The key optimization of OLSRv2 is that of multipoint relays,
   providing an efficient mechanism for network-wide broadcast of link-
   state information. information (i.e. reducing the cost of performing a network-
   wide link-state broadcast).  A secondary optimization is, is that OLSRv2
   employs partial link-state information: each node maintains
   information of about all destinations, but only a subset of links.  This
   allows that only select nodes diffuse link-state advertisements (i.e.
   reduces the number of network-wide link-state broadcasts) and that
   these advertisements contain only a subset of links (i.e. reduces the
   size of each network-wide link-state broadcast).  The partial link-state link-
   state information thus obtained still allows each OLSRv2 node to at
   all times maintain optimal (in terms of number of hops) routes to all
   destinations in the network.

   OLSRv2 imposes minimum requirements to the network by not requiring
   sequenced or reliable transmission of control traffic.  Furthermore,
   the only interaction between OLSRv2 and the IP stack is routing table
   management.

   OLSRv2 is particularly suitable for large and dense networks as the
   technique of MPRs works well in this context.

Table of Contents

   1.  Introduction . . . . . . . . . . . . . . . . . . . . . . . . .  5
     1.1
     1.1.  Terminology  . . . . . . . . . . . . . . . . . . . . . . .  6
     1.2
     1.2.  Applicability Statement  . . . . . . . . . . . . . . . . .  7  6
   2.  Protocol Overview and Functioning  . . . . . . . . . . . . . .  9
     2.1  8
     2.1.  Protocol Extensibility . . . . . . . . . . . . . . . . . . 11 10
   3.  Processing and Forwarding Repositories . . . . . . . . . . . . 13
     3.1 11
     3.1.  Received Message Set . . . . . . . . . . . . . . . . . . . 13
     3.2 . . . . 11
     3.2.  Fragment Set . . . . . . . . . . . . . . . . . . . . . . . 13
     3.3 11
     3.3.  Processed Set  . . . . . . . . . . . . . . . . . . . . . . 14
     3.4 12
     3.4.  Forwarded Set  . . . . . . . . . . . . . . . . . . . . . . 14
     3.5 12
     3.5.  Relay Set  . . . . . . . . . . . . . . . . . . . . . . . . 14 12
   4.  Packet Processing and Message Forwarding . . . . . . . . . . . 16
     4.1 14
     4.1.  Actions when Receiving an OLSRv2 Packet  . . . . . . . . . 16
     4.2 14
     4.2.  Actions when Receiving an OLSRv2 Message . . . . . . . . . 16
     4.3 14
     4.3.  Message Considered for Processing  . . . . . . . . . . . . 16
     4.4 15
     4.4.  Message Considered for Forwarding  . . . . . . . . . . . . 18 17
   5.  Information Repositories . . . . . . . . . . . . . . . . . . . 21
     5.1 20
     5.1.  Neighborhood Information Base  . . . . . . . . . . . . . . 21
       5.1.1 20
       5.1.1.  Link Set . . . . . . . . . . . . . . . . . . . . . . . 21
       5.1.2   2-Hop Neighbor 20
       5.1.2.  MPR Set  . . . . . . . . . . . . . . . . . . 22
       5.1.3   Neighborhood Address Association Set . . . . . . . . . 23
       5.1.4 21
       5.1.3.  MPR Selector Set . . . . . . . . . . . . . . . . . . . . . . . 23
       5.1.5   MPR Selector Set . . . 21
     5.2.  Topology Information Base  . . . . . . . . . . . . . . . . 23
       5.1.6 21
       5.2.1.  Advertised Neighbor Set  . . . . . . . . . . . . . . . 23
     5.2   Topology Information Base 21
       5.2.2.  ANSN History Set . . . . . . . . . . . . . . . . 24
       5.2.1 . . . 22
       5.2.3.  Topology Set . . . . . . . . . . . . . . . . . . . . . 24
       5.2.2 22
       5.2.4.  Attached Network Set . . . . . . . . . . . . . . . . . 24
       5.2.3 23
       5.2.5.  Routing Set  . . . . . . . . . . . . . . . . . . . . . 25 23
   6.  OLSRv2 Control Message Structures  . . . . . . . . . . . . . . 26
     6.1 24
     6.1.  General OLSRv2 Message TLVs  . . . . . . . . . . . . . . . 26
       6.1.1 24
       6.1.1.  VALIDITY_TIME TLV  . . . . . . . . . . . . . . . . . . 26
       6.1.2   INTERVAL_TIME TLV  . . . . . . . . . . . . . . 24
     6.2.  HELLO Messages . . . . 27
     6.2   Local Interface Blocks . . . . . . . . . . . . . . . . . . 28
     6.3 25
       6.2.1.  HELLO Messages . . . . . . . . . . . . Message OLSRv2 Message TLVs  . . . . . . . . . . 28
       6.3.1 26
       6.2.2.  HELLO Message: Message OLSRv2 Address Block TLVs  . . . . . . . . . . . . . 29
       6.3.2   HELLO Message: Address Blocks TLVs 26
     6.3.  TC Messages  . . . . . . . . . . 29
     6.4   TC Messages . . . . . . . . . . . . . 27
     6.4.  TC Message: OLSRv2 Address Block TLVs  . . . . . . . . . . 30 27
   7.  HELLO Message Generation . . . . . . . . . . . . . . . . . . . 31
     7.1 29
     7.1.  HELLO Message: Transmission  . . . . . . . . . . . . . . . 33 29
   8.  HELLO Message Processing . . . . . . . . . . . . . . . . . . . 34
     8.1 30
     8.1.  Populating the Link MPR Selector Set  . . . . . . . . . . . . . 30
     8.2.  Symmetric Neighborhood and 2-Hop Neighborhood Changes  . . 31
   9.  TC Message Generation  . . 34
     8.2   Populating the 2-Hop Neighbor Set . . . . . . . . . . . . 36
     8.3   Populating the MPR Selector Set . . . . . . 32
     9.1.  TC Message: Transmission . . . . . . . 37
     8.4   Neighborhood and 2-Hop Neighborhood Changes . . . . . . . 38
   9. . . . 33
   10. TC Message Generation Processing  . . . . . . . . . . . . . . . . . . . . 40
     9.1 34
     10.1. Single TC Message: Transmission Message Processing . . . . . . . . . . . . . . . 34
       10.1.1. Populating the ANSN History Set  . . 41
   10.   TC Message Processing . . . . . . . . . 35
       10.1.2. Populating the Topology Set  . . . . . . . . . . 42
     10.1  Checking Freshness & Validity of a TC message  . . . . . . 42
     10.2  Updating 35
       10.1.3. Populating the Topology Attached Network Set  . . . . . . . . . 36
     10.2. Completed TC Message Processing  . . . . . . . 43
     10.3  Purging Old Entries from the Topology Set  . . . . . . . . 44
     10.4  Updating 37
       10.2.1. Purging the Attached Networks Topology Set . . . . . . . . . . . . 44
     10.5 . . . 37
       10.2.2. Purging Old Entries from the Attached Network Set . . . . 45
     10.6  Processing Unfragmented TC Messages  . . . . . . . . . . . 45
     10.7  Processing Partially or Wholly Self-Contained
           Fragmented TC Messagess  . . . . . . . . . . . . . . . . . 45 37
   11. Populating the MPR Set . . . . . . . . . . . . . . . . . . . 47 . 38
   12. Populating Derived Sets  . . . . . . . . . . . . . . . . . . 48
     12.1 . 39
     12.1. Populating the Relay Set . . . . . . . . . . . . . . . . . 48
     12.2 39
     12.2. Populating the Advertised Neighbor Set . . . . . . . . . . 48 39
   13.   Populating the Neighborhood Address Association Set  . . . . 49
   14. Routing Table Calculation  . . . . . . . . . . . . . . . . . 50
   15. . 40
   14. Proposed Values for Constants  . . . . . . . . . . . . . . . 53
     15.1  Message Intervals  . . . . . . . . . . . . 44
     14.1. Neighborhood Discovery Constants . . . . . . . . 53
     15.2  Holding Times . . . . . 44
     14.2. Message Intervals  . . . . . . . . . . . . . . . . . 53
     15.3  Willingness . . . 44
     14.3. Holding Times  . . . . . . . . . . . . . . . . . . . . 53
     15.4  Time . . 44
     14.4. Willingness  . . . . . . . . . . . . . . . . . . . . . . . 44
   15. Sequence Numbers . . 54
   16.   Representing Time . . . . . . . . . . . . . . . . . . . . . 55
   17. 45
   16. IANA Considerations  . . . . . . . . . . . . . . . . . . . . 56
     17.1 . 46
     16.1. Multicast Addresses  . . . . . . . . . . . . . . . . . . . 56
     17.2 46
     16.2. Message Types  . . . . . . . . . . . . . . . . . . . . . . 56
     17.3 46
     16.3. TLV Types  . . . . . . . . . . . . . . . . . . . . . . . . 56
   18. 46
   17. References . . . . . . . . . . . . . . . . . . . . . . . . . 57
       Authors' Addresses . . . 48
     17.1. Normative References . . . . . . . . . . . . . . . . . . . 58
   A.  Example Heuristic for Calculating MPRs 48
     17.2. Informative References . . . . . . . . . . . . 59
   B.  Example Algorithms for Generating Control Traffic . . . . . . 62
     B.1 48
   Appendix A.   Example Algorithm Heuristic for Generating HELLO messages Calculating MPRs . . . . . 62
     B.2   Example Algorithm for Generating TC messages . . 49
   Appendix B.   Heuristics for Generating Control Traffic  . . . . . 63 52
   Appendix C.   Protocol and Port Number . . . . . . . . . . . . . . . . . . . 65 53
   Appendix D.   Packet and Message Layout  . . . . . . . . . . . . . 54
   Appendix D.1. OLSRv2 Packet Format . . . . . 66
     D.1   OLSRv2 Packet Format . . . . . . . . . . . . 54
   Appendix E.   Node Configuration . . . . . . . 66
   E.  Node Configuration . . . . . . . . . . 59
   Appendix F.   Jitter . . . . . . . . . . . . 73
   F.  Security Considerations . . . . . . . . . . . 60
   Appendix G.   Security Considerations  . . . . . . . . 74
     F.1   Confidentiality . . . . . . 63
   Appendix G.1. Confidentiality  . . . . . . . . . . . . . . . 74
     F.2   Integrity . . . 63
   Appendix G.2. Integrity  . . . . . . . . . . . . . . . . . . . . . 74
     F.3 63
   Appendix G.3. Interaction with External Routing Domains  . . . . . . . . 75
     F.4 64
   Appendix G.4. Node Identity  . . . . . . . . . . . . . . . . . . . . . . 76
   G. 65
   Appendix H.   Flow and Congestion Control  . . . . . . . . . . . . 66
   Appendix I.   Contributors . . . . . 77
   H.  Sequence Numbers . . . . . . . . . . . . . . . . 67
   Appendix J.   Acknowledgements . . . . . . . 78
   I.  Contributors . . . . . . . . . . . 68
   Authors' Addresses . . . . . . . . . . . . . . 79
   J.  Acknowledgements . . . . . . . . . . . . . . . . . . . . . . . 80 69
   Intellectual Property and Copyright Statements . . . . . . . . 81 . . 70

1.  Introduction

   The Optimized Link State Routing Protocol protocol version 2 (OLSRv2) is an
   update to OLSRv1 as published in RFC3626 [1].  Compared to RFC3626,
   OLSRv2 retains the same basic mechanisms and algorithms, while
   providing an even more flexible signaling framework and some
   simplification of the messages being exchanged.  Also, OLSRv2 takes
   care to accomodate accommodate both IPv4 and IPv6 addresses in a compact
   fashion.

   OLSRv2 is developed for mobile ad hoc networks.  It operates as a
   table driven, proactive protocol, i.e. it exchanges topology
   information with other nodes of the network regularly.  Each node
   selects a set of its neighbor nodes as "MultiPoint Relays" (MPRs).
   In OLSRv2, only
   Only nodes that are selected as such MPRs are then responsible for
   forwarding control traffic intended for diffusion into the entire
   network.  MPRs provide an efficient mechanism for flooding control
   traffic by reducing the number of transmissions required.

   Nodes selected as MPRs also have a special responsibility when
   declaring link state information in the network.  Indeed, the only
   requirement for OLSRv2 to provide shortest path routes to all
   destinations is that MPR nodes declare link-state information for
   their MPR selectors.  Additional available link-state information may
   be utilized, e.g., e.g. for redundancy.

   Nodes which have been selected as multipoint relays by some neighbor
   node(s) announce this information periodically in their control
   messages.  Thereby a node announces to the network that it has
   reachability to the nodes which have selected it as an MPR.  Thus, as
   well as being used to facilitate efficient flooding, MPRs are also
   used for route calculation from any given node to any destination in
   the network.

   A node selects MPRs from among its one hop neighbors with
   "symmetric", i.e., i.e. bi-directional, linkages.  Therefore, selecting
   the route
   routes through MPRs automatically avoids the problems associated with
   data packet transfer over uni-directional links (such as the problem
   of not getting link-layer acknowledgments for data packets at each
   hop, for link-layers employing this technique for unicast traffic).

   OLSRv2 is developed to work independently from other protocols.
   Likewise, OLSRv2 makes no assumptions about the underlying link-
   layer.  However, OLSRv2 may use link-layer information and
   notifications when available and applicable.

   OLSRv2, as OLSRv1, inherits the concept of forwarding and relaying
   from HIPERLAN (a MAC layer protocol) which is standardized by ETSI
   [5].

1.1
   [6].

1.1.  Terminology

   The keywords "MUST", "MUST NOT", "REQUIRED", "SHALL", "SHALL NOT",
   "SHOULD", "SHOULD NOT", "RECOMMENDED", "MAY", and "OPTIONAL" in this
   document are to be interpreted as described in RFC2119 [2].

   MANET specific terminology is to be interpreted as described in [3]
   and [4].

   Additionally, this document uses the following terminology:

   node - a A MANET router which implements the Optimized Link State
      Routing protocol version 2 as specified in this document.

   OLSRv2 interface - A network device participating in a MANET interface, running OLSRv2.  A node may have several OLSRv2 interfaces, each interface
      assigned one or more IP addresses.

   symmetric strict 2-hop neighbor - A node X symmetric 2-hop neighbor which is
      not a symmetric 1-hop neighbor of node Y if node Y can hear node X
      (i.e., and is not a link exists from an OLSRv2 interface on node X to an
      OLSRv2 interface on node Y).  A 2-hop neighbor may also be called only
      through a symmetric 1-hop neighbor.

   2-hop neighbor - A with willingness WILL_NEVER.
      (If node X Z is a symmetric 2-hop neighbor of node X then there is a
      node Y if such that node X Z is a symmetric 1-hop neighbor of node Y
      and node Y is a symmetric 1-hop neighbor of node Y, but is not X. If node Y itself. Z is a
      symmetric strict 2-hop neighbor - of node X then there is such a 2-hop neighbor
      node Y with willingness which is not a neighbor WILL_NEVER.)

   symmetric strict 2-hop neighborhood - The set of the node, and is not a symmetric strict
      2-hop neighbor only through a neighbor with
      willingness WILL_NEVER. neighbors of node X.

   multipoint relay (MPR) - A node which is selected by its symmetric
      1-hop neighbor, node X, to "re-transmit" all the broadcast
      messages that it receives from node X, provided that the message
      is not a duplicate, and that the time to live hop limit field of the message is
      greater than one.

   multipoint relay

   MPR selector (MPR selector, MS) - A node which has selected its symmetric 1-hop
      neighbor, node X, as one of its multipoint
      relays, will be called MPRs is an MPR selector of node X.

   link - A link

1.2.  Applicability Statement

   OLSRv2 is a pair of OLSRv2 interfaces from two different
      nodes, where at least one interface proactive routing protocol for mobile ad hoc networks
   (MANETs) [7], [8].  It is able well suited to hear (i.e. receive
      traffic from) the other.

   symmetric link - A link where both interfaces are able to hear (i.e.
      receive messages from) the other.

   asymmetric link - A link which is not symmetric.

   symmetric 1-hop neighborhood - The symmetric 1-hop neighborhood of
      any node X is the set of nodes which have at least one symmetric
      link to node X.

   symmetric 2-hop neighborhood - The symmetric 2-hop neighborhood of
      node X is the set of nodes, excluding node X itself, which have a
      symmetric link to the symmetric 1-hop neighborhood of X.

   symmetric strict 2-hop neighborhood - The symmetric strict 2-hop
      neighborhood of node X is the set of nodes in its symmetric 2-hop
      neighborhood that are neither in its symmetric 1-hop neighborhood
      nor reachable only through a symmetric 1-hop neighbor of node X
      with willingness WILL_NEVER.

1.2  Applicability Statement

   OLSRv2 is a proactive routing protocol for mobile ad hoc networks
   (MANETs) [6], [7].  It is well suited to large and dense networks of
   mobile nodes, as large and dense networks of
   mobile nodes, as the optimization achieved using the MPRs works well
   in this context.  The larger and more dense a network, the more
   optimization can be achieved as compared to the classic link state
   algorithm.  OLSRv2 uses hop-by-hop routing, i.e., i.e. each node uses its
   local information to route packets.

   As OLSRv2 continuously maintains routes to all destinations in the
   network, the protocol is beneficial for traffic patterns where the
   traffic is random and sporadic between a large subset of nodes, and
   where the [source, destination] pairs are changing over time: no
   additional control traffic need be generated in this situation since
   routes are maintained for all known destinations at all times.  Also,
   since routes are maintained continously, continuously, traffic is subject to no
   delays due to buffering/route-discovery.  This continued route
   maintenance may be done using periodic message exchange, as detailed
   in this specification, or triggered by external events if available.

   OLSRv2 supports nodes which have multiple interfaces which
   participate in the MANET.  OLSRv2, additionally, supports nodes which
   have non-MANET interfaces which can serve as (if configured to do so)
   gateways towards other networks.

   The message exchange format, contained in previous versions of this
   specification, has been factored out to an independant independent specification
   [4],
   [3], which is used for carrying OLSRv2 control signals.  OLSRv2 is
   thereby able to accommodate allow for extensions via "external" and "internal"
   extensibility.  External extensibility implies that a protocol
   extension may specify and exchange new message types types, formatted
   according to [3], which can be forwarded and delivered correctly according to [4]. correctly.
   Internal extensibility implies, implies that a protocol extension may define
   additional attributes to be carried embedded in the standard OLSRv2
   control
   messages, messages detailed in this specification, using the TLV
   mechanism specified in [3], while these OLSRv2 control messages with
   additional attributes can still be correctly understood by all OLSRv2
   nodes.

   The OLSRv2 neighborhood discovery protocol using HELLO messages has
   likewise been factored out to an independent specification [4].  This
   neighborhood discovery protocol serves to ensure that each OLSRv2
   node has available continuously updated information repositories
   describing the node's 1-hop and 2-hop neighbors. [4] uses the message
   format specified in [3], and hence is extensible as described above.

2.  Protocol Overview and Functioning

   OLSRv2 is a proactive routing protocol for mobile ad hoc networks.
   The protocol inherits the stability of a link state algorithm and has
   the advantage of having routes immediately available when needed due
   to its proactive nature.  OLSRv2 is an optimization over the
   classical link state protocol, tailored for mobile ad hoc networks.
   The main tailoring and optimizations of OLSRv2 are:

   o  periodic, unacknowledged transmission of all control messages;

   o  optimized flooding for global link-state information diffusion;

   o  partial topology maintenance -- - each node will know of all
      destinations and knows only a subset of
      the links in the network.

   More specifically, OLSRv2 consists of the following main components:

   o  A general and flexible signaling framework, allowing network, sufficient for
      information exchange between OLSRv2 nodes.  This framework allows
      for both local information a minimum hop route to
      all destinations.

   Using the message exchange (between neighboring nodes) format [3] and global information exchange using an optimized flooding
      mechanism denoted "MPR flooding".

   o  A specification of local signaling, denoted HELLO messages.  HELLO
      messages in the neighborhood discovery
   protocol [4], OLSRv2 serve to:

      *  discover links also contains the following main components:

   o  a TLV, to adjacent OLSR nodes;

      *  perform bidirectionality check on be included within the discovered links;

      *  advertise neighbors and hence discover 2-hop neighbors;

      *  signal MPR selection. HELLO messages are emitted periodically, thereby of [4], allowing nodes a
      node to
      continuously track changes in their local neighborhoods. signal MPR selection;

   o  A  an optimized flooding mechanism for global information exchange,
      denoted "MPR flooding";

   o  a specification of global signaling, denoted TC (Topology Control)
      messages.  TC messages in OLSRv2 serve to:

      *  inject link-state information into the entire network. network;

      *  inject addresses of hosts and networks for which they may serve
         as a gateway into the entire network.

      *  allow nodes with multiple interface addresses to ensure that
         nodes within two hops can associate these addresses with a
         single node for efficient MPR Set determination.

      TC messages are emitted periodically, thereby allowing nodes to
      continuously track global changes in the network.

   Thus, through periodic exchange

   The use of HELLO messages, [4] allows a node is able to
   acquire and maintain information about continuously track changes to its immediate neighborhood.
   This includes information about immediate neighbors, as well as nodes
   which are
   local topology up to two hops away.  By HELLO messages being exchanged
   periodically,  This allows a node learns about changes in the neighborhood (new
   nodes emerging, old nodes disappearing) without requiring explicit
   mechanisms for doing so.

   Based on the local topology information, acquired through the
   periodic exchange of HELLO messages, an OLSRv2 node is able to make
   provisions for ensuring optimized flooding, denoted "MPR flooding",
   as well as injection of link-state information into the network.
   This is done through the notion of Multipoint Relays. MultiPoint Relays (MPRs).

   The idea of multipoint relays MPRs is to minimize the overhead of flooding messages in
   the network by reducing redundant retransmissions of messages in the
   same region.  Each node in the network selects an MPR Set, a set of
   nodes in its symmetric 1-hop neighborhood which may retransmit its
   messages.  This
   set of selected neighbor nodes is called the "Multipoint Relay" (MPR)
   Set of that node.  The 1-hop neighbors of a node N which are *NOT* not in its MPR
   set, set
   receive and process broadcast messages messages, but do not retransmit
   broadcast messages received from node N. that node.  The MPR Set of a node is
   selected such that it covers (in terms of radio range) all symmetric
   strict 2-hop nodes.  The MPR Set of N, denoted as MPR(N), is then an
   arbitrary X
   may be any subset of the its symmetric 1-hop neighborhood of N which
   satisfies the following condition: such that every
   node in the its symmetric strict 2-hop neighborhood of N MUST have has a symmetric link towards MPR(N).
   to a node in the MPR Set of node X. The MPR Set of a node may thus be
   said to "cover" the node's symmetric strict 2-hop neighborhood.  The
   smaller a MPR Set, the fewer times messages are forwarded and the
   less resulting control traffic overhead results from
   the routing protocol. [7] overhead. [8] gives an analysis and
   example of MPR selection algorithms.  Notice,  Note that as long as the
   condition above is satisfied, any algorithm selecting MPR Sets is
   acceptable in terms of implementation interoperability.

   Each node maintains information about the set of symmetric 1-hop
   neighbors that have selected it as MPR.  This set is called the "Multipoint Relay
   Selector Set" (MPR MPR
   Selector Set) Set of a the node.  A node obtains this information from periodic HELLO messages received from an
   MPR TLV which is inserted into the neighbors. HELLO message exchange of [4].

   Each node also maintains a Relay Set, which is the set of nodes for
   which a node is to relay broadcast traffic.  The Relay Set is derived
   from the MPR Selector Set in that the Relay Set MUST contain all the
   nodes in the MPR Selector set and MAY contain additional nodes.

   A broadcast message, intended to be diffused in the whole network,
   coming from any of the nodes in the Relay Set of node N is assumed to
   be retransmitted by node N, if N has not received it yet.  This set
   can change over time (e.g., when a node selects another MPR Set) and
   is indicated by the selector nodes in their HELLO messages.

   Using the MPR flooding mechanism, link-state information can be
   injected into the network.  For this purpose, a node maintains an
   Advertised Neighbor Set which MUST contain all the nodes in the MPR
   selector set and MAY contain additional nodes.  If the Advertised
   Neighbor Set of a node is non-empty, TC messages, containing the
   links between the node and the nodes it is reported in TC messages
   generated by that node.  If the Advertised Neighbor Set, Set is empty, TC
   messages are not generated, generated by that node, unless needed for gateway reporting
   reporting, or multiple
   interface address association (if for a short period to accelerate the latter case only, with minimal
   scope). removal of
   unwanted links.

   OLSRv2 is designed to work in a completely distributed manner and
   does not depend on any central entity.  The protocol does not require
   reliable transmission of control messages: each node sends control
   messages periodically, and can therefore sustain a reasonable loss of
   some such messages.  Such losses may occur frequently in radio
   networks due to collisions or other transmission problems.

   Also,

   OLSRv2 does not require sequenced delivery of messages.  Each control
   message contains a sequence number which is incremented for each
   message.  Thus the recipient of a control message can, if required,
   easily identify which information is more recent - even if messages
   have been re-ordered while in transmission.  Furthermore,
   OLSRv2 provides support for protocol extensions such as sleep mode
   operation, multicast-routing etc.  Such extensions may be introduced
   as additions to the protocol without breaking backwards compatibility
   with earlier versions.

   OLSRv2 does not require any changes to the format of IP packets.
   Thus packets, any
   existing IP stack can be used as is: OLSRv2 only interacts with
   routing table management.  OLSR sends its own control messages using UDP.

2.1

2.1.  Protocol Extensibility

   This specification defines and uses two

   OLSRv2 uses the neighborhood discovery mechanism of [4], and
   specifies additionally one message types, HELLO type, TC, and TC.  As for OLSRv1 [1] extensions to a number of TLVs.
   All messages exchanged by [4] and by OLSRv2 may define new use and comply with the
   extensible message types to carry additional information.  This may be
   considered as exchange format of [3], thus OLSR provides both
   "external" extensibility.  New extensibility (addition of new message types are
   divided into two ranges, those which may be added by standards
   actions (with types up to 127) as in OLSRv1
   [1]) and those made available for private/
   local use (with types 128 "internal" extensibility (addition of information to 255).

   All new
   existing messages must be syntactically OLSRv2 messages, through TLVs) as defined described in
   [4].  (Some additional constraints to that specification are added
   for OLSRv2 packets and messages, requiring full packet and [3].

   Those nodes which do not recognize a new message
   headers.)  Note that type ("external
   extensibility") will ignore this message type for processing, but
   will correctly forward the message, if it is required to include one or more blocks
   of unstructured data specified in such a the message (possibly as its only content)
   this may be achieved by including each block as
   header.  Those nodes which do not recognize a single message TLV
   block, with an appropriately newly defined message TLV.  (Like message
   types, TLV types are divided into those up to 127 which may be type
   ignore the added
   by standards action, and those from 128 to 255 available for private/
   local use.)

   A network may contain nodes both aware of, and unaware of, any new
   message types.  The originator of a message can control whether a
   message flooded through TLV, but process (if the network message type is forwarded by nodes which are
   unaware of recognized)
   the message type, thus reaching all nodes in correctly, as well as forwards the network,
   or is only flooded by nodes which recognise message, if specified
   in the message type.

   OLSRv2 also supports an alternative, and more powerful, extension
   mechanism which was not supported by OLSRv1, that of adding new
   information to an already defined message type, whilst still leaving
   the predefined information unchanged and usable, including by a node
   which does not recognise the new information.  This may be considered
   to be "internal" extensibility of a message.

   The mechanism for this extensibility is the use of TLV (type-length-
   value) structures in the message format defined in [4] to carry
   information associated with either the message as a whole, or with
   one or more addresses carried in the message.  The messages defined
   in this specification carry two types of addresses, those of the
   originating node's own interfaces participating in OLSRv2, and those
   of neighbouring nodes or networks to which it has a route.  (New
   message types may define other relationships to addresses which they
   carry.)  All information associated with these addresses, or the
   message as a whole, in messages defined in this specification is in
   TLV format; additional TLVs may be defined and added to these
   messages.

   Those nodes which do not recognise newly defined TLV types ignore the
   added TLVs.  (This is facilitated by that the TLVs defined in this
   specification, or in [4], have the lowest type numbers and that TLVs
   must be included in type order, as specified in [4].)  It is
   important that newly defined TLV types permit this behaviour.

3.  Processing header.

3.  Processing and Forwarding Repositories

   The following data-structures data structures are employed in order to ensure that a
   message is processed at most once and is forwarded at most once per
   interface of a node, and that fragmented content is treated
   correctly.

3.1

3.1.  Received Message Set

   Each node maintains, for each OLSRv2 interface it possesses, interface, a set of signatures
   of messages, which have been received over that interface, in the
   form of "Received Tuples":

      (RX_type, RX_addr, RX_orig_addr, RX_seq_number, RX_time)

   where:

   RX_type is the received message type, or zero if the received message
      sequence number is not type-specific.

   RX_addr type-specific;

   RX_orig_addr is the originator address of the received message;

   RX_seq_number is the message sequence number of the received message;

   RX_time specifies the time at which this record Received Tuple expires and
      *MUST* be removed.

   In a node, this is denoted the "Received Message Set" for that interface.

3.2

3.2.   Fragment Set

   Each node stores messages containing fragmented content until all
   fragments are received and the message processing can be completed,
   in the form of "Fragment Tuples":

      (FG_message, FG_time)

   where:

   FG_message is the message containing fragmented content;

   FG_time specifies the time at which this record Fragment Tuple expires and
      MUST be removed.

   In a node, this is denoted the "Fragment Set".

3.3

3.3.  Processed Set

   Each node maintains a set of signatures of messages which have been
   processed by the node, in the form of "Processed Tuples":

      (P_type, P_addr, P_orig_addr, P_seq_number, P_time)

   where:

   P_type is the processed message type, or zero if the processed
      message sequence number is not type-specific.

   P_addr type-specific;

   P_orig_addr is the originator address of the processed message;

   P_seq_number is the message sequence number of the processed message;

   P_time specifies the time at which this record Processed Tuple expires and
      *MUST* be removed.

   In a node, this is denoted the "Processed Set".

3.4

3.4.  Forwarded Set

   Each node maintains a set of signatures of messages which have been
   retransmitted/forwarded by the node, in the form of "Forwarded
   Tuples":

      (FW_type, FW_addr, FW_orig_addr, FW_seq_number, FW_time)

   where:

   FW_type is the forwarded message type, or zero if the forwarded
      message sequence number is not type-specific.

   FW_addr type-specific;

   FW_orig_addr is the originator address of the forwarded message;

   FW_seq_number is the message sequence number of the forwarded
      message;

   FW_time specifies the time at which this record Forward Tuple expires and
      *MUST* be removed.

   In a node, this is denoted the "Forwarded Set".

3.5

3.5.  Relay Set

   Each node maintains a set of neighbor interface addresses for which
   it is to relay flooded messages, in the form of "Relay Tuples":

      (RY_if_addr)

      (RY_iface_addr)

   where:

   RY_if_addr

   RY_iface_addr is the address of a neighbor interface for which the
      node SHOULD relay flooded messages.  This MUST include a prefix
      length.

   In a node, this is denoted the "Relay Set".

4.  Packet Processing and Message Forwarding

   Upon

   On receiving a basic packet, as defined in [3], a node examines each
   of the message headers.  If the message type is known to the node,
   the message is processed locally according to the specifications for
   that message type.  The message is also independently evaluated for
   forwarding.

4.1

4.1.  Actions when Receiving an OLSRv2 Packet

   Upon

   On receiving a packet, a node MUST perform the following task: tasks:

   1.  If  The packet may be fully parsed on reception, or the packet contains no and
       its messages (i.e. may be parsed only as required.  (It is possible to
       parse the packet length is
       less than header, or equal determine its absence, without
       parsing any messages.  It is possible to the size of divide the packet header) or into
       messages without even fully parsing their headers.  It is
       possible to determine whether a message is to be forwarded, and
       to forward it, without parsing its body.  It is possible to
       determine whether a message is to be processed without parsing
       its body.  It is possible to determine if that processing may be
       delayed because the
       packet cannot message is a fragment by inspecting the first
       few octets of the message body without fully parsing it.)

   2.  If parsing fails at any point the relevant entity (packet or
       message) MUST be parsed into messages, silently discarded, other parts of the packet MUST
       (up to the whole packet) MAY be silently discarded.

   2.  Otherwise, discarded;

   3.  Otherwise if the packet header is present and it contains a
       packet TLV block, then each TLV in it is processed according to
       its type;

   4.  Otherwise each message in the packet packet, if any, is treated
       according to Section 4.2.

4.2

4.2.  Actions when Receiving an OLSRv2 Message

   A node MUST perform the following tasks for each received OLSRv2
   message:

   1.  If the received OLSRv2 message header cannot be correctly parsed
       according to the specification in [4], [3], or if the node recognizes
       from the originator address of the message that the message is an interface address of
       one which the receiving node itself originated, then the message
       MUST be silently discarded;

   2.  Otherwise:

       1.  If the received message is of a known type then the message
           is considered for processing according to Section 4.3; 4.3, AND;

       2.  If for the received message TTL (<hop-limit> + <hop-count>) > 0, and if either the
           message is of a known type, or bit 3 of the message semantics
           octet in the message header is clear, as indicated in [4], 1,
           then the message is considered for forwarding according to
           Section 4.4.

4.3

4.3.  Message Considered for Processing

   If a message (the "current message") is considered for processing,
   the following tasks MUST be performed:

   1.  If an entry exists in the Processed Set where:

       *  P_type == the message type, type of the current message, or 0 if the
          typedep bit 2 of in the message semantics octet (in the message
          header) of the current message is clear, cleared ('0'), AND;

       *  P_addr  P_orig_addr == the originator address of the current message,
          AND;

       *  P_seq_number == the message sequence number of the current
          message.

       then the current message MUST NOT be processed.

   2.  Otherwise:

       1.  Create an entry in the Processed Set with:

           +  P_type = the message type, type of the current message, or 0 if
              the typedep bit 2 of in the message semantics octet (in the
              message header) of the current message is clear; cleared ('0');

           +  P_addr  P_orig_addr = originator address of the current message;

           +  P_seq_number = sequence number of the current message;

           +  P_time = current time + P_HOLD_TIME.

       2.  If the current message does not contain a valid message TLV of type
           Fragment
           with Type == FRAGMENTATION (or if it does and the indicated
           number of fragments is one) then process the message fully
           according to its type.

       3.  Otherwise:

           1.  If the current message does not contain a valid message is "wholly or partially self-contained" as
               indicated by its Fragment
               TLV with Type == CONT_SEQ_NUM then process the current message as far as possible
               MUST be silently discarded;

           2.  Otherwise if the current message is "partially or wholly
               self-contained", as indicated by the notselfcont bit in
               the Value field of the TLV with Type == FRAGMENTATION
               being cleared ('0'), then process the current message as
               far as possible according to its type;

           2.

           3.  If the Fragment Set includes any messages Fragment Tuples with:

               -  either the typedepseq bit in the Value field of the
                  TLV with Type == FRAGMENTATION in the same current message
                  is cleared ('0') OR message type of FG_message ==
                  message type of current message, AND;

               -  originator address and of FG_message == originator address
                  of current message, AND;

               -  content sequence number as (the Value field of the
                  message TLV with Type == CONT_SEQ_NUM) of FG_message
                  == content sequence number of current message, and message; AND

               -  either fragment number (from the same Value field of the
                  TLV with Type == FRAGMENTATION) in FG_message ==
                  fragment number or a
               different of current message OR number of
                  fragments (from the Value field of the TLV with Type
                  == FRAGMENTATION) of FG_message != number of fragments, fragments
                  of current message;

               then remove these messages
               are Fragment Tuples from the Fragment Set;

           3.

           4.  If the Fragment Set includes messages Fragment Tuples containing
               all the remaining fragments of the same overall message
               as the current message (i.e. message, i.e. if the number of messages Fragment
               Tuples such that:

               -  either the typedepseq bit in the
               Fragment Set Value field of the
                  TLV with Type == FRAGMENTATION in the same current message
                  is cleared ('0') OR message type of FG_message ==
                  message type of current message, AND;

               -  originator address and of FG_message == originator address
                  of current message, AND;

               -  content sequence number as (the Value field of the
                  message TLV with Type == CONT_SEQ_NUM) of FG_message
                  == content sequence number of current message

               is equal to the
               current message's number (number of fragments of fragments, current message, less
               one) then all of these messages Fragment Tuples are removed from
               the Fragment Set and their messages processed according
               to their type (taking account of any previous processing if
               of any which are partially or all were wholly or
               partially self-contained);

           4.

           5.  Otherwise, the current message a Fragment Tuple is added to the Fragment Set
               with a

               -  FG_message = current message;

               -  FG_time of FG_HOLD_TIME (possibly = current time + FG_HOLD_TIME;

               possibly replacing an
               identical and previous a previously received instance of the
               same
               fragment of the same content).

4.4 fragment.

4.4.  Message Considered for Forwarding

   If a message is considered for forwarding then if forwarding, and it is either of a
   message type defined in this document, document or of an unknown message type type,
   then it MUST use the following algorithm.  A message type not defined
   in this document may MAY specify the use of this, or another algorithm.
   (Such an other algorithm MAY use the Received Set for the receiving
   interface, it SHOULD use the Forwarded Set similarly to the following
   algorithm.)

   If a message is considered for forwarding according to this
   algorithm, the following tasks MUST be performed:

   1.  If there is no symmetric link in the Link Set between the
       receiving interface and the sending interface (as indicated by the source interface of
       the IP datagram containing the message) does not match (taking
       into account any address prefix of) any N_neighbor_iface_addr in
       any Symmetric Neighbor Tuple, then the message MUST be silently
       discarded.

   2.  Otherwise:

       1.  If an entry exists in the Received Set for the receiving
           interface, where:

           +  RX_type == the message type, or 0 if the typedep bit 2 of in
              the message semantics octet (in the message header) is clear,
              cleared ('0'), AND;

           +  RX_addr  RX_orig_addr == the originator address of the received
              message, AND;
           +  RX_seq_number == the sequence number of the received
              message.

           then the message MUST be silently discarded.

       2.  Otherwise:

           1.  Create an entry in the Received Set for the receiving
               interface with:

               -  RX_type = the message type, or 0 if the typedep bit 2 of in
                  the message semantics octet (in the message header) is
                  clear;
                  cleared ('0');

               -  RX_addr  RX_orig_addr = originator address of the message;

               -  RX_seq_number = sequence number of the message;

               -  RX_time = current time + RX_HOLD_TIME.

           2.  If an entry exists in the Forwarded Set where:

               -  FW_type == the message type, or 0 if the typedep bit 2 of
                  in the message semantics octet (in the message header)
                  is
                  clear; cleared ('0');

               -  FW_addr  FW_orig_addr == the originator address of the received
                  message, AND;

               -  FW_seq_number == the sequence number of the received
                  message.

               then the message MUST be silently discarded.

           3.  Otherwise if an entry exists in the a Relay Set, where
               RY_if_addr == source Tuple exists whose RY_iface_addr
               matches (taking into account any address of prefix) the message
               sending interface (as indicated by the source interface
               of the IP datagram containing the message):

               1.  Create an entry in the Forwarded Set with:

                   o  FW_type = the message type, or 0 if the typedep
                      bit 2 of in the message semantics octet (in the message
                      header) is
                      clear; cleared ('0');

                   o  FW_addr  FW_orig_addr = originator address of the message;
                   o  FW_seq_number = sequence number of the message;

                   o  FW_time = current time + FW_HOLD_TIME.

               2.  The message header is modified as follows:

                   o  Decrement <hop-limit> in the message TTL header by 1;

                   o  Increment <hop-count> in the message hop count header by 1;

               3.  Transmit the message on all OLSRv2 interfaces of the
                   node.

   Messages are retransmitted in the format specified by [4] [3] with the
   All-OLSRv2-Multicast
   ALL-MANET-NEIGHBORS address (see Section 17.1) 16.1) as destination IP
   address.

5.  Information Repositories

   The purpose of OLSRv2 is to determine the Routing Set, which may be
   used to update IP's Routing Table, providing "next hop" routing
   information for IP datagrams.  In order to accomplish this, OLSRv2
   maintains
   uses a number of protocol sets, sets: the information repository of
   the protocol.  These sets are updated, directly or indirectly, Neighborhood Information Base,
   provided by the
   exchange of messages between nodes [4], is in the network.  In turn the
   contents of these messages are largely determined OLSRv2 augmented by the contents of
   a part of the information repositories, the Neighbourhood allowing MPR
   selection and signaling.  Additionally, OLSRv2 specifies a Topology
   Information Base, which contains information about describes the 1- information used for and 2- hop
   neighbourhoods of
   acquired through TC message exchange - in other words: the topology
   base represents the network topology graph as seen from each node.  The remaining part of

   Addresses (other than originator addresses) recorded in the information
   repository,
   Neighborhood Information Base and the Topology Information Base (including the Routing Set)
   contains information about the network which is not constrained MUST
   all be recorded with prefix lengths, in order to allow comparison
   with addresses received in HELLO and TC messages.  For the node's neighbourhood.  The Topology
   Information Base is updated
   by the OLSRv2 messages defined in this document, it is not used applies to
   define their contents.  The process of information exchange which
   leads to the population of the Neighbourhood Information Base A_neighbor_iface_addr,
   T_dest_iface_addr, T_last_iface_addr, AN_net_addr, AN_gw_iface_addr,
   R_dest_addr, R_dest_addr, R_next_iface_addr and R_local_iface_addr,
   but not AH_orig_addr.  For the
   Topology Neighborhood Information Base is started using only the node's own OLSRv2
   interface addresses and host and network associated addresses.  These
   are not affected by the exchange of the OLSRv2 messages defined in
   this document.

5.1 see [4].

5.1.  Neighborhood Information Base

   The neighborhood information base stores information about links
   between local interfaces and interfaces on adjacent nodes.

5.1.1  In
   addition to the sets described in [4], OLSRv2 adds an element to each
   Link Set

   A node records Tuple to allow a set of "Link Tuples":

     (L_local_iface_addr, L_neighbor_iface_addr,
      L_SYM_time, L_ASYM_time, L_willingness, L_time).

   where:

   L_local_iface_addr is node to record the interface address willingness of a 1-hop
   neighbor node to be selected as an MPR.  Also, OLSRv2 adds an MPR Set
   and an MPR Selector Set to the local node;

   L_neighbor_iface_addr Neighborhood Information Base.  The
   MPR Set is the interface address used by a node to record which of its symmetric 1-hop
   neighbors are selected as MPRs, and the neighbor node;

   L_SYM_time MPR Selector Set is the time until used by a
   node to record which of its symmetric 1-hop neighbors have selected
   it as MPR.  Thus the link MPR Set is considered symmetric;

   L_ASYM_time used, in addition to what is
   specified in [4], when generating HELLO messages, and the time until which the neighbor interface MPR
   Selector Set is
      considered heard; populated, in addition to what is specified in [4]
   when processing HELLO messages.

5.1.1.  Link Set

   The Link Tuples, specified in [4] are augmented by an element,
   L_willingness:

   L_willingness is the nodes node's willingness to be selected as an MPR;

   L_time specifies when this record expires and *MUST* be removed.

              +-------------+-------------+--------------+
              | L_SYM_time  | L_ASYM_time | L_STATUS     |
              +-------------+-------------+--------------+
              | Expired     | Expired     | LOST         |
              |             |             |              |
              | Not Expired | Expired     | SYMMETRIC    |
              |             |             |              |
              | Not Expired | Not Expired | SYMMETRIC    |
              |             |             |              |
              | Expired     | Not Expired | ASYMMETRIC   |
              +-------------+-------------+--------------+

                                  Table 1

   The status remaining elements of the link, denoted L_STATUS, can be derived based on the
   fields L_SYM_time and L_ASYM_time as defined in Table 1.

   In a node, the set of Link Tuples is denoted the "Link Set".

5.1.2  2-Hop Neighbor are as specified in [4].

5.1.2.  MPR Set

   A node records maintains a set of "2-Hop Neighbor Tuples"

    (N2_local_iface_addr, N2_neighbor_iface_addr, N2_2hop_iface_addr, N2_time)

   describing symmetric links between its neighbors and the symmetric
   2-hop neighborhood.

   N2_local_iface_addr is the address all of the local OLSRv2 interface over addresses with
   which the information was received;

   N2_neighbor_iface_addr is the interface address of a neighbor;

   N2_2hop_iface_addr is the interface address of a 2-hop neighbor with node has a symmetric link to the node with interface address
      N_neighbor_iface_addr;

   N2_time specifies the time at which the tuple expires and *MUST* be
      removed.

   In a node, the set of 2-Hop Neighbor Tuples is denoted the "2-Hop
   Neighbor Set".

5.1.3  Neighborhood Address Association Set

   A node maintains, for each 1-hop and 2-hop neighbor with multiple
   addresses participating in the OLSRv2 network, a "Neighborhood
   Address Association Tuple", representing that "these addresses belong
   to the same node".

       (NA_neighbor_addr_list, NA_time)

   NA_neighbor_iface_addr_list is the list of interface addresses which are of the 1-hop or 2-hop neighbor node;

   NA_time specifies the time at symmetric
   neighbors which the tuple expires and *MUST* be
      removed.

   In a node, the set of Neighborhood Address Association Tuples node has selected as MPRs.  This is denoted the "Neighborhood Address Association
   "MPR Set".

5.1.4

5.1.3.  MPR Selector Set

   A node maintains a set of neighbors which are selected as MPRs.
   Their "MPR Selector Tuples" containing all of the
   OLSRv2 interface addresses are listed in with which the MPR Set.

5.1.5  MPR Selector Set

   A node maintains, for each interface has a symmetric link
   and which are of an 1-hop neighbor symmetric neighbors which has have selected it as MPR, an "MPR Selector Tuple", representing the an
   interface of the neighbor
   node which have selected it as an MPR.

       (MS_neighbor_if_addr,

       (MS_neighbor_iface_addr, MS_time)

   MS_neighbor_if_addr

   MS_neighbor_iface_addr specifies the an OLSRv2 interface address with
      which the node has a symmetric link and which is of a 1-hop
      neighbor,
      symmetric neighbor which has selected the node as an MPR;

   MS_time specifies the time at which the tuple this MPR Selector Tuple expires
      and *MUST* be removed.

   Notice that if

   In a MPR selector node has multiple interface addresses, node, the set of MPR Selector Set will contain one tuple Tuples is denoted the "MPR
   Selector Set".

5.2.  Topology Information Base

   The Topology Information Base stores information, required for each the
   generation and processing of TC messages.  The Advertised Neighbor
   Set contains OLSRv2 interface
   address addresses of symmetric 1-hop neighbors
   which are to be reported in TC messages.  The Topology Set and
   Attached Network Set both record information received through TC
   messages.  Thus the MPR selector.

5.1.6 Advertised Neighbor Set is used for generating TC
   messages, while the Topology Set and Attached Network Set are
   populated when processing TC messages.

   Additionally, a Routing Set is maintained, derived from the
   information recorded in the Neighborhood Information Base, Topology
   Set and Attached Network Set.

5.2.1.  Advertised Neighbor Set

   A node maintains a set of neighbor OLSRv2 interface addresses, addresses of symmetric
   1-hop neighbors, which are to be advertised through TC messages:

       (A_neighbor_iface_addr)
   For this set, an Advertised Neighbor Set Sequence Number (ASSN) (ANSN) is
   maintained.  Each time the Advertised Neighbor Set is updated, the
   ASSN
   ANSN MUST be incremented.

5.2  Topology Information Base  The Topology Information Base stores topological ANSN MUST also be incremented if any
   locally advertised attached networks are added or removed.

5.2.2.  ANSN History Set

   A node records a set of "ANSN History Tuples", recording information
   describing
   about the network beyond freshness of the nodes neighborhood (i.e. beyond topology information received from each
   other node:

       (AH_orig_addr, AH_seq_number, AH_time)

   AH_orig_addr is the
   Neighborhood Information Base originator address of a received TC message;

   AH_seq_number is the highest ANSN in any TC message received which
      originated from AH_orig_addr;

   AH_time is the time at which this ANSN History Tuple expires and
      *MUST* be removed.

   In a node, the set of ANSN History Tuples is denoted the node).

5.2.1 "ANSN
   History Set".

5.2.3.  Topology Set

   Each node in the network maintains topology information about the
   network.

   For each destination
   network in the network, at least one form of "Topology Tuple" Tuples":

       (T_dest_iface_addr, T_last_iface_addr, T_seq, T_seq_number, T_time)

   is recorded.

   T_dest_iface_addr is the an OLSRv2 interface address of a destination
      node, which may be reached in one hop from the node with the
      OLSRv2 interface address T_last_iface_addr;

   T_last_iface_addr is, conversely, an OLSRv2 interface address of a
      node which is the last hop towards
      T_dest_iface_addr.  Typically, T_last_iface_addr is on a MPR path towards the node with OLSRv2
      interface address T_dest_iface_addr.  Typically, the node with
      OLSRv2 interface address T_last_iface_addr is a MPR of the node
      with OLSRv2 interface address T_dest_iface_addr;

   T_seq

   T_seq_number is a sequence number, and the highest received ANSN associated with the
      information contained in this Topology Tuple;

   T_time specifies the time at which this tuple Topology Tuple expires and
      *MUST* be removed.

   In a node, the set of Topology Tuples are is denoted the "Topology Set".

5.2.2

5.2.4.  Attached Network Set

   Each node in the network maintains information about attached
   networks.

   For each attached network, at least one
   networks in the form of "Attached Network Tuple" Tuples":

       (AN_net_addr, AN_prefix_lenght, AN_gw_addr, AN_seq_no, AN_gw_iface_addr, AN_seq_number, AN_time)

   is recorded.

   AN_net_addr is the network address (prefix) (including prefix length) of a an
      attached network, which may be reached via the node with the
      OLSRv2 interface address
      AN_gw_addr;

   AN_prefix_length is the length of the prefix of the network address
      AN_net_addr;

   AN_gw_addr AN_gw_iface_addr;

   AN_gw_iface_addr is the address of an OLSRv2 interface of a node
      which can act as gateway to the network identified by the AD_net_addr/
      AD_prefix_length;

   AN_seq_no AN_net_addr;

   AN_seq_number is a sequence number, and; the highest received ANSN associated with the
      information contained in this Attached Network Tuple;

   AN_time specifies the time at which this tuple Attached Network Tuple
      expires and *MUST* be removed.

   In a node, the set of Topology Attached Network Tuples are is denoted the "Topology
   "Attached Network Set".

5.2.3

5.2.5.  Routing Set

   A node records a set of "Routing Tuples":

      (R_dest_iface_addr, R_next_iface_addr, R_dist, R_iface_addr) Tuples" describing the next hop and distance of the selected path
   to each destination in the network for which a route is known.

   R_dest_iface_addr known:

      (R_dest_addr, R_next_iface_addr, R_dist, R_local_iface_addr)

   R_dest_addr is the interface address of the destination, either the address of
      an OLSRv2 interface of a destination node; node, or the network address
      of an attached network;

   R_next_iface_addr is the OLSRv2 interface address of the "next hop"
      on the selected path towards R_dest_iface_addr; to the destination;

   R_dist is the number of hops on the selected path to R_dest_iface_addr;

   R_iface_addr the destination;

   R_local_iface_addr is the address of the local interface over which a
      packet MUST be sent to reach R_next_iface_addr. the destination.

   In a node, the set of Routing Tuples is denoted the "Routing Set".

6.  OLSRv2 Control Message Structures

   Nodes using OLSRv2 exchange information through messages.  One or
   more messages sent by a node at the same time are combined into a
   packet.  These messages may have originated at the sending node, or
   have originated at another node and forwarded by the sending node.
   Messages with different originators may be combined in the same
   packet.

   The packet and message format used by OLSRv2 is defined in [4]. [3].
   However this specification contains some options which are not used
   by OLSRv2.  In particular (using the syntactical elements defined in
   the packet format specification):

   o  All OLSRv2 packets packets, not limited to those defined in this document,
      include a <packet-header>.

   o  All OLSRv2 packets, not limited to those defined in this document,
      have the pseqnum bit of <packet-semantics> cleared ('0'), i.e.
      they include a packet sequence number.

   o  OLSRv2 packets MAY include packet TLVs, however OLSRv2 itself does
      not specify any packet TLVs.

   o  All OLSRv2 messages, not limited to those defined in this
      document, include a full <msg-header> and hence have bits 0 the noorig
      and 1 nohops bits of <msg-semantics> cleared. cleared ('0').

   o  All OLSRv2 message defined in this document have the typedep bit,
      and all remaining reserved bits of <msg-semantics> cleared. cleared ('0').

   Other options defined in [4] [3] may be freely used, in particular any
   other values of <packet-semantics> or <tlv-semantics> consistent with
   its specification.  An
   implementation of OLSRv2 MAY take full advantage of the features of
   the message specification in [4] allowing decisions relating to
   whether a message should be forwarded and/or processed to be taken
   parsing only the message header (plus, if a message is to be
   processed but may be fragmented, only the first octets of the message
   body).

   OLSRv2 messages are sent using UDP, see Appendix C.

   The remainder of this section defines, within the framework of [4], [3],
   message types and TLVs specific to OLSRv2.

6.1

6.1.  General OLSRv2 Message TLVs

   This document specifies two message TLVs, which can be applied to any
   OLSRv2 control message, messages, VALIDITY_TIME and INTERVAL_TIME, detailed in
   this section.

6.1.1 INTERVAL_TIME.

6.1.1.  VALIDITY_TIME TLV

   All OLSRv2 messages specified in this specification document MUST include a
   VALIDITY_TIME TLV, specifying for how long a node may, upon receiving period following receipt of a
   message, after which the receiving node MUST consider the message
   content to no longer be valid. valid (unless repeated in a later message).
   The validity time of a message MAY be specified to depend on the
   distance from the originator (i.e. the <hop-count> field in the
   message header as defined in [4]). [3]).  Thus, the a VALIDITY_TIME TLV contains TLV's value
   field MAY contain a sequence of pairs (time, hop-limit) hop limit) in increasing hop-limit order, followed by
   hop limit order; it MUST contain a final default value.

   This is an extended, and compatible, version of the VALIDITY_TIME TLV
   defined in [4].

   Thus, an instance of a VALIDITY_TIME TLV could MAY have the following
   value: value
   field:

     <t_1><hl_1><t_2><hl_2> ... <t_i><hl_i> ....  <t_n><hl_n><t_default>

   Which would mean that the message, message carrying this VALIDITY_TIME TLV, TLV
   would have the following validity times:

   o  <t_1> in the interval from 0 (exclusive) to <hl_1> (inclusive)
      hops away from the originator;

   o  <t_i> in the interval from <hl_(i-1)> (exclusive) to <hl_i>
      (inclusive) hops away from the originator; and

   o  <t_default> in the interval from <hl_n> (exclusive) to 255> 255
      (inclusive) hops away from the originator.

   The VALIDITY_TIME message TLV specification is given in Table 2.

   VALIDITY_TIME message TLV specification overview

   +----------------+--------+-------------------+---------------------+ 1.

   +----------------+------+-------------------+-----------------------+
   |      Name      | Type |       Length      | Value                 |
   +----------------+--------+-------------------+---------------------+
   +----------------+------+-------------------+-----------------------+
   |  VALIDITY_TIME |  TBD |  (2*n+1) * 8 bits | {<time><hoplimit>}* {<time><hop_limit>}*  |
   |                |      |                   | <t_default>           |
   +----------------+--------+-------------------+---------------------+
   +----------------+------+-------------------+-----------------------+

                                  Table 2 1

   where <n> n is the number of (time, hop_limit) pairs in the TLV, TLV (i.e. is
   equal to (<length>-1)/2, where <length> is the length of the TLV
   value field) and where <time> and <t_default> are represented as
   specified in section
   Section 16.

6.1.2  INTERVAL_TIME TLV [3].

6.2.  HELLO Messages

   A HELLO message in OLSRv2 messages of a given type MAY include is generated as specified in [4].

   Additionally, an INTERVAL_TIME OLSRv2 node:

   o  MUST include TLV(s) with Type == MPR associated with all OLSRv2
      interface addresses included in the HELLO message
   TLV, specifying with a TLV with
      Type == LINK_STATUS and Value == SYMMETRIC if that address is also
      included in the interval at which messages node's MPR Set (if there is more than one copy of
      the address, this type are being
   generated by applies to the specific copy of the address to
      which the originator node.

   The INTERVAL_TIME message TLV specification is given associated);

   o  MUST NOT include any TLVs with Type == MPR associated with any
      other addresses;

   o  MAY include a message TLV with Type == WILLINGNESS, indicating the
      node's willingness to be selected as MPR.

6.2.1.  HELLO Message OLSRv2 Message TLVs

   In a HELLO message, a node MAY include a WILLINGNESS message TLV as
   specified in Table 3.

   INTERVAL_TIME TLV specification overview

   +----------------+--------+-------------------+---------------------+ 2.

   +----------------+------+-------------------+-----------------------+
   |      Name      | Type |       Length      | Value                 |
   +----------------+--------+-------------------+---------------------+
   +----------------+------+-------------------+-----------------------+
   |  INTERVAL_TIME   WILLINGNESS  |  TBD |       8 bits      | <time> The node's            |
   |                |      |                   | willingness to be     |
   |                |      |                   | selected as MPR, any  |
   |                |      |                   | unused bits (based on |
   |                |      |                   |
   +----------------+--------+-------------------+---------------------+

                                  Table 3

   where <time> is the time between two successive emissions of messages
   of the type, represented maximum           |
   |                |      |                   | willingness value     |
   |                |      |                   | WILL_ALWAYS) are      |
   |                |      |                   | RESERVED and SHOULD   |
   |                |      |                   | be set to zero.       |
   +----------------+------+-------------------+-----------------------+

                                  Table 2

   A node's willingness to be selected as specified in section Section 16.

6.2  Local Interface Blocks

   The first address block, plus following MPR ranges from WILL_NEVER
   (indicating that a node MUST NOT be selected as MPR by any node) to
   WILL_ALWAYS (indicating that a node MUST always be selected as MPR).

   If a node does not advertise a Willingness TLV block in HELLO messages, the
   node MUST be assumed to have a willingness of WILL_DEFAULT.

6.2.2.  HELLO or Message OLSRv2 Address Block TLVs

   In a HELLO message, a node MAY include MPR address block TLV(s) as
   specified in Table 3.

   +----------------+------+-------------------+-----------------------+
   |      Name      | Type |       Length      | Value                 |
   +----------------+------+-------------------+-----------------------+
   |       MPR      |  TBD |       0 bits      | No value, i.e.        |
   |                |      |                   | novalue bit ([3]) set |
   +----------------+------+-------------------+-----------------------+

                                  Table 3

6.3.  TC Messages

   A TC message is known MUST contain:

   o  a message TLV with Type == CONT_SEQ_NUM, as specified in [3];

   o  a Local Interface Block.  A Local Interface Block
   is not distinguished message TLV with Type == VALIDITY_TIME, as specified in any way other than by being the
      Section 6.1.1;

   o  a first address block in containing all of the node's OLSRv2
      interface addresses.  This is similar to the message.

   A Local Interface Block contains the
      specified in [4], however these addresses of all of the
   interfaces of the originating node that support OLSRv2 and
   participate MUST be included in the MANET, using the standard <address-block> syntax
   from [4].  In a TC message this is sufficient;
      same order in all copies of a HELLO given TC message,
   those addresses, if any, which correspond to interfaces other than
   that on regardless of
      which the HELLO message interface it is sent must have a corresponding
   OTHER_IF TLV.  In this case (only) this OTHER_IF TLV SHALL NOT have a
   <value> field.

   Note that a Local Interface Block may include more than one address
   for each interface, transmitted on, and hence in a HELLO message may contain more
   than one address without an no OTHER_IF TLV.

6.3  HELLO Messages

   A HELLO message MUST contain:

   o  a message TLV VALIDITY_TIME Section 6.1.1 address
      block TLV(s) are required;

   o  one or more  additional address blocks block(s) containing all addresses in the
      Advertised Address Set and Attached Network Set, the latter (only)
      with associated GATEWAY address block TLVs

   The first (mandatory) address block is a Local Interface Block, TLV(s), as specified in
      Section 6.2.  Other (optional) address blocks contain
   1-hop neighbors' interface addresses. 6.4, both with associated PREFIX_LENGTH TLV(s), as
      specified in [3], as necessary.

   A HELLO TC message MAY optionally contain:

   o  a message TLV INTERVAL_TIME as specified in Section 6.1.2

   o  a message TLV WILLINGNESS, INTERVAL_TIME, as specified in Section 6.3.1

6.3.1  HELLO [4].

6.4.  TC Message: Message OLSRv2 Address Block TLVs

   In a HELLO TC message, a node MAY include a message TLV GATEWAY address block TLV(s) as
   specified in Table 4.

   VALIDITY_TIME message TLV specification overview

   +----------------+--------+-------------------+---------------------+

   +----------------+------+-------------------+-----------------------+
   |      Name      | Type |       Length      | Value                 |
   +----------------+--------+-------------------+---------------------+
   +----------------+------+-------------------+-----------------------+
   |   WILLINGNESS     GATEWAY    |  TBD |       8       0 bits      | <The node's         |
   |                |        |                   | willingness to be No value, i.e.        |
   |                |      |                   | selected as MPR> novalue bit ([3]) set |
   +----------------+--------+-------------------+---------------------+
   +----------------+------+-------------------+-----------------------+

                                  Table 4

   A

7.  HELLO Message Generation

   An OLSRv2 HELLO message is composed as defined in [4], with the
   following TLVs added:

   o  a message TLV with Type == WILLINGNESS and Value == the node's
      willingness to be selected act as MPR ranges from WILL_NEVER
   (indicating that a node MUST NOT an MPR, MAY be included in the message;

   o  for each symmetric 1-hop neighbor OLSRv2 interface address which
      is included in the HELLO message with an associated TLV with Type
      == LINK_STATUS and is selected as an MPR (i.e. is present in the
      MPR Set), an address TLV with Type == MPR by any node) to
   WILL_ALWAYS (indicating that a node MUST always be included, this
      SHOULD be associated with the same copy of the address as the TLV
      with Type == LINK_STATUS;

   o  for each 1-hop neighbor OLSRv2 interface address which is included
      in the HELLO message but is not selected as MPR.

   If a node does an MPR (i.e. is not advertise a Willingness
      present in the MPR Set), an address TLV with Type == MPR MUST NOT
      be included.

7.1.  HELLO Message: Transmission

   Messages are retransmitted in the packet/message format specified by
   [3] with the ALL-MANET-NEIGHBORS address as destination IP address
   and with TTL (IPv4) or hop limit (IPv6) equal to 1.

8.  HELLO Message Processing

   Subsequent to the processing of HELLO messages, as specified in [4],
   the node MUST be assumed to have a MUST:

   1.  Determine the willingness of WILL_DEFAULT.

6.3.2  HELLO Message: Address Blocks TLVs the originating node to be an MPR
       by:

       *  if the HELLO message address block contains a message TLV specification overview

   +----------------+--------+-------------------+---------------------+
   |      Name      | with Type  |       Length      | Value               |
   +----------------+--------+-------------------+---------------------+
   |   LINK_STATUS  |   TBD  |       8 bits      | One ==
          WILLINGNESS then the willingness is the value of HEARD,       |
   |                |        |                   | SYMMETRIC, LOST.    |
   |                |        |                   |                     |
   |       MPR      |   TBD  |       0 that TLV,
          ignoring the reserved bits      | No value, i.e.      |
   |                |        |                   | novalue bit (see    |
   |                |        |                   | [4]) set            |
   |                |        |                   |                     |
   |    OTHER_IF    |   TBD  |    0 or 8 bits    | In a in that field;

       *  otherwise the willingness is WILL_DEFAULT.

   2.  Update each Link Tuple whose L_neighbor_iface_addr is present in
       the Local          |
   |                |        |                   | Interface Block     |
   |                |        |                   | none, otherwise     |
   |                |        |                   | either of SYMMETRIC |
   |                |        |                   | or LOST             |
   +----------------+--------+-------------------+---------------------+

                                  Table 5

6.4  TC Messages

   A TC message MUST contain:

   o  a message TLV VALIDITY_TIME the HELLO message, with:

       *  L_willingness = the willingness of the originating node;

   3.  Update its MPR Selector Set, according to Section 6.1.1

   o 8.1.

8.1.  Populating the MPR Selector Set

   On receiving a message TLV CONTENT_SEQUENCE_NUMBER [4]

   o  one or more address blocks with associated address block TLVs.

   The first (mandatory) address block is HELLO message, a Local Interface Block, as
   specified in Section 6.2.  Other (optional) address blocks contain
   1-hop neighbors' interface addresses and/or host or network addresses
   for which this node may act as a gateway.  In the latter case they
   may use PREFIX_LENGTH TLV(s) as specified in [4].

   A TC message MAY optionally contain:

   o MUST:

   1.  If a message node finds one of its own interface addresses with an
       associated TLV INTERVAL_TIME as specified with Type == MPR in Section 6.1.2

7.  HELLO Message Generation

   An OLSRv2 the HELLO message is composed of a set of message TLVs,
   describing general properties of (indicating
       that the message and originator node has selected the receiving node emitting as an
       MPR), the HELLO, and a set MPR Selector Set MUST be updated as follows:

       1.  For each address, henceforth neighbor address, in the Local
           Interface Block of address blocks (with associated TLV sets),
   describing the links and their associated properties.

   OLSRv2 received HELLO messages are generated and transmitted per interface,
   i.e. different HELLO messages are generated and transmitted per
   OLSRv2 interface of message, where the
           neighbor address is present as an N_neighbor_iface_addr in a node.

   OLSRv2 HELLO messages are generated and transmitted periodically,
           Symmetric Neighbor Tuple with N_STATUS == SYMMETRIC:

           1.  If there exists no MPR Selector Tuple with:

               -  MS_neighbor_iface_addr == neighbor address

               then a default interval between two consecutive HELLO emissions on
   the same interface new MPR Selector Tuple is created with:

               -  MS_neighbor_iface_addr = neighbor address

           2.  The MPR Selector Tuple (new or otherwise) with:

               -  MS_neighbor_iface_addr == neighbor address

               is then modified as follows:

               -  MS_time = current time + validity time

   2.  Otherwise if a node finds one of HELLO_INTERVAL.

   This section specifies its own interface addresses with
       an associated TLV with Type == LINK_STATUS and Value == SYMMETRIC
       in the requirements, which HELLO message
   generation MUST fulfill.  An example algorithm (indicating, since there is proposed in
   Appendix B.1.

   For each OLSRv2 interface a no TLV with Type
       == MPR, that originator node has de-selected the receiving node
       as an MPR), the MPR Selector Set MUST generate a HELLO message with a be updated as follows:

       1.  All MPR Selector Tuples whose N_neighbor_iface_addr is in the
           Local Interface Block as of the first address block, HELLO message are removed.

   MPR Selector Tuples are also removed upon expiration of MS_time, or
   upon symmetric link breakage as specified described in Section 6.2, followed by address blocks 8.2.

8.2.  Symmetric Neighborhood and address TLVs according 2-Hop Neighborhood Changes

   A node MUST also perform the following:

   1.  If a Link Tuple with L_STATUS == SYMMETRIC is removed, or its
       L_STATUS changes from SYMMETRIC to
   Table 6.

   +---------------------------+---------------------------------------+
   | The set HEARD or LOST, and if that
       Link Tuple's L_neighbor_iface_addr is an MS_iface_addr of neighbor       | TLV(s) (Type = Value)                 |
   | interfaces which are ...  |                                       |
   +---------------------------+---------------------------------------+
   | HEARD, but not an MPR
       Selector Tuple, then that MPR Selector Tuple MUST be removed.

   2.  If any of:

       *  a Link Tuple is added with L_STATUS == SYMMETRIC, OR;

       *  a Link Tuple with L_STATUS == SYMMETRIC  | LINK_STATUS=HEARD                     |
   | over the interface over   |                                       |
   | which the HELLO message   |                                       |
   | is being transmitted      |                                       |
   |                           |                                       |
   | removed, or its
          L_STATUS changes from SYMMETRIC over the        | LINK_STATUS=SYMMETRIC                 |
   | interface over which the  |                                       |
   | HELLO message to HEARD or LOST, or vice
          versa, OR;

       *  a 2-Hop Neighbor Tuple is being    |                                       |
   | transmitted               |                                       |
   |                           |                                       |
   | LOST over the interface   | LINK_STATUS=LOST                      |
   | over which added or removed, OR;

       *  the HELLO      |                                       |
   | message Neighbor Address Association Set is being          |                                       |
   | transmitted               |                                       |
   |                           |                                       |
   | Not SYMMETRIC over changed such that the    | OTHER_IF=SYMMETRIC                    |
   | interface over
          subset of any NA_neighbor_iface_addr_list consisting of those
          addresses which are the  |                                       |
   | HELLO message is being    |                                       |
   | transmitted, but          |                                       |
   | L_neighbor_iface_addr of a Link Tuple
          with L_STATUS == SYMMETRIC over one is changed, including the cases of
          removal or     |                                       |
   | more other interfaces addition of  |                                       |
   | the node                  |                                       |
   |                           |                                       |
   | Not SYMMETRIC over a Neighbor Address Association Tuple
          containing any    | OTHER_IF=LOST                         |
   | interface or LOST over    |                                       |
   | the interface over which  |                                       |
   | such addresses;

       then the MPR Set MUST be recalculated.

   An additional HELLO message is      |                                       |
   | being transmitted, but    |                                       |
   | previously reported as    |                                       |
   | OTHER_IF=SYMMETRIC MAY be sent when the MPR Set changes, in
   addition to the cases specified in [4], and    |                                       |
   | still HEARD or LOST over  |                                       |
   | subject to the same
   constraints.

9.  TC Message Generation

   A node with one or more interfaces of |                                       |
   | the node other than the   |                                       |
   | interface over OLSRv2 interfaces, and with a non-empty
   Advertised Neighbor Set or which the  |                                       |
   | HELLO message is being    |                                       |
   | transmitted               |                                       |
   |                           |                                       |
   | Selected acts as MPR for the   | MPR                                   |
   | interface over a gateway to an associated
   network which the  |                                       |
   | HELLO message is          |                                       |
   | transmitted               |                                       |
   +---------------------------+---------------------------------------+

                                  Table 6

   In order that an address can to be reported as OTHER_IF=LOST by a node
   with more than one interface participating advertised in the MANET, such a MUST generate TC
   messages.  A node
   MAY maintain with an Other Interface Set of addresses for each interface.
   The Other Interface empty Advertised Neighbor Set for an interface is updated when a HELLO
   message is to be transmitted over that interface, and used to
   determine which addresses are reported is
   not acting as OTHER_IF=LOST in that such a gateway SHOULD also generate "empty" TC messages
   for a period A_HOLD_TIME after it last generated a non-empty TC
   message.  The Other Interface Set of addresses is updated  TC messages (non-empty and used as
   follows: empty) are generated according
   to the following:

   1.  Each address that  the HELLO TC message is to include with MUST contain a
       corresponding message TLV with Type=LINK_STATUS Type ==
       CONT_SEQ_NUM and Value=SYMMETRIC is
       removed Value == ANSN from the set. Advertised Neighbor Set;

   2.  Each address that  the HELLO TC message is to include with MUST contain a
       corresponding message TLV with Type=OTHER_IF Type ==
       VALIDITY_TIME and Value=SYMMETRIC is added
       to the set if not already present.

   3.  Each other address in the set (not included Value == T_HOLD_TIME as specified in
       Section 6.1.1;

   3.  the HELLO TC message
       with MAY contain a corresponding message TLV with Type=OTHER_IF Type ==
       INTERVAL_TIME and Value=SYMMETRIC)

       1.  Is removed if Value == TC_INTERVAL as specified in [4];

   4.  the HELLO TC message is to include it with a
           corresponding TLV with Type=LINK_STATUS MUST contain the addresses of all of its OLSRv2
       interfaces in its first address block, note that the TC message
       generated on all OLSRv2 interfaces MUST be identical (including
       having identical message sequence number) and Value=LOST.

       2.  Is removed if it is hence these
       addresses are not HEARD ordered or LOST over an interface other
           than otherwise identified according to
       the interface over on which the HELLO TC message is transmitted;

   5.  the TC message MUST contain, in address blocks other than its
       first:

       1.  A_neighbor_iface_addr from each Advertised Neighbor Tuple;

       2.  the addresses of all associated hosts and networks for which
           this node is to be
           transmitted.

       3.  Otherwise act as a gateway and which is included to be
           advertised in the HELLO message MANET, each associated with a TLV with
           Type=OTHER_IF and Value=LOST.  (Note that Type
           == GATEWAY.

   6.  the address may
           also have TC message MAY be fragmented, only by its originator.  It
       SHOULD be fragmented only if necessary; if the TC message is
       fragmented, a corresponding FRAGMENTATION TLV with Type=LINK_STATUS MUST be included, and
           Value=HEARD if appropriate.)

7.1  HELLO each
       fragment SHOULD be indicated as "partially or wholly self
       contained" in it, and MUST indicate that the content sequence
       number (ANSN) is message type specific.

9.1.  TC Message: Transmission

   Messages

   TC messages are retransmitted in the packet/message format specified by
   [4] with the All-OLSRv2-Multicast address as destination IP address generated and transmitted periodically on all OLSRv2
   interfaces, with a TTL=1.

8.  HELLO Message Processing

   Upon receiving a HELLO message, a node will update its local link
   information base according to default interval between two consecutive TC
   emissions by the specification given same node of TC_INTERVAL.  TC messages MAY be
   generated in this
   section.

   For the purpose response to a change of this section, please notice contents (a change in ANSN, due
   to a change in the following:

   o Advertised Neighbor Set or the "validity time" advertised locally
   attached networks) but a node must respect a minimum interval of
   TC_MIN_INTERVAL between generated TC messages.

   TC messages SHOULD be generated with a message is calculated from hop limit [3] greater
   than or equal to the VALIDITY-
      TIME TLV expected network diameter (by default with a hop
   limit of 255).

   TC messages are transmitted with the message as specified in Section 6.1.1;

   o  the "Source Address" is the source ALL-MANET-NEIGHBORS multicast
   address as indicated by the
      source interface of the destination IP datagram containing address and are forwarded according to the message;

   o
   specification in Section 4.4.

10.  TC Message Processing

   When according to Section 4.3 a HELLO TC message MUST neither be forwarded nor is to be recorded in processed
   according to its type, this means that processing is carried out
   according to Section 10.1 and Section 10.2.  The timing of this
   processing depends on whether the
      Processing TC message is a fragment, and Forwarding Repositories; if so
   whether it is "partially or wholly self-contained":

   o  if the address blocks considered exclude the Local Interface Block,
      unless explicitly specified;

   o  a HELLO message is only valid when, for each address listed in not a fragment, then first Section 10.1 and then
      Section 10.2 are carried out when the
      address blocks:

      * message is received;

   o  if the address message is associated with a TLV with Type=Link Status OR a
         TLV with Type=Other Interface Status OR both, the latter either fragment which is "partially or wholly self-
      contained", then processing according to Section 10.1 is carried
      out when the TLV with Type=Link Status message is received, and processing according to
      Section 10.2 is carried out when all matching fragments have been
      received and all processing according to Section 10.1 has Value=HEARD, been
      carried out;

   o  if the message is a fragment which is not "partially or wholly
      self-contained", then processing according to Section 10.1 is
      carried out when all matching fragments have been received, and
      processing according to Section 10.2 is carried out when all
      matching fragments have been received and all processing according
      to Section 10.1 has been carried out.

   For all processing purposes, "ANSN" is defined as being the value of
   the message TLV with Type=Link Status has Value=LOST and Type == CONT_SEQ_NUM in the TLV with
         Type=Other Interface Status TC message.  If a TC
   message has Value=SYMMETRIC, AND

      * no such TLV then the processing of Section 10.1 and
   Section 10.2 MUST NOT be carried out.  (Note that if the address message is associated with a TLV with Type=MPR, then
   fragment it
         MUST also will have already been discarded according to
   Section 4.3.)  If more than one TC message is processed according to
   Section 10.2 then these must have the same ANSN to be associated with a TLV with Type=Link Status and
         Value=SYMMETRIC.

      Invalid HELLO messages are not processed.

8.1  Populating recognized as
   fragments of the Link Set

   Upon receiving a HELLO message, a node SHOULD update its Link Set
   with same message.

10.1.  Single TC Message Processing

   For the information contained purpose of this section, note the following:

   o  "validity time" is calculated from the VALIDITY_TIME message TLV
      in the HELLO.  Thus, for TC message according to the Local
   Interface Block (see specification in Section 6.2) 6.1.1;

   o  "originator address" refers to the Neighbor Address Association
   Set is updated as specified by Section 13.  For each address, listed originator address in the subsequent HELLO TC
      message address blocks (see Section 6): header;

   o  comparisons of sequence numbers are carried out as specified in
      Section 15.

   The TC message is processed as follows:

   1.  the ANSN History Set is updated according to Section 10.1.1; if
       the TC message is indicated as discarded in that processing then
       the following steps are not carried out;

   2.  the Topology Set is updated according to Section 10.1.2;

   3.  the Attached Network Set is updated according to Section 10.1.3.

10.1.1.  Populating the ANSN History Set

   The node MUST update its ANSN History Set as follows:

   1.  if there exists no link tuple is an ANSN History Tuple with:

       *  L_neighbor_iface_addr  AH_orig_addr == Source Address originator address; AND

       *  AH_seq_number > ANSN

       then the TC message MUST be discarded;

   2.  otherwise create a new tuple is created with

       *  L_neighbor_iface_addr = Source Address; ANSN History Tuple with:

       *  L_local_iface_addr  AH_orig_addr = Address of the interface which
          received the HELLO message; originator address;

       *  L_SYM_time  AH_seq_number = current time - 1 (expired); ANSN;

       *  L_time  AH_time = current time + validity time.

   2.  The tuple (existing or new)

       possibly replacing an existing ANSN History Tuple with L_neighbor_iface_addr == Source
       Address is then modified the same
       AH_orig_addr.

10.1.2.  Populating the Topology Set

   The node SHOULD update its Topology Set as follows:

   1.  if the node finds the address of the interface, which
           received the HELLO message,  for each address, henceforth local address, in one of the first address blocks
           included
       block in message, then the tuple is modified as follows: TC message:

       1.  if  for each address, henceforth advertised address, in an
           address block other than the occurrence of L_local_iface_addr first in the HELLO
               message is:

               - TC message, and
           which does not have an associated with a TLV with (Type Type == "LINK_STATUS",
                  Value GATEWAY:

           1.  if there is a Topology Tuple with:

               T_dest_iface_addr == LOST) advertised address; AND
               T_last_iface_addr == local address

               then

               -  L_SYM_time update this Topology Tuple to have:

               T_seq_number = ANSN;

               T_time = current time - 1 (i.e., expired) + validity time

           2.  else if the occurrence of L_local_iface_addr in the HELLO
               message:

               -  is associated with:

                  o  a TLV with (Type == "LINK_STATUS", Value ==
                     SYMMETRIC);

                  OR;

                  o  otherwise create a TLV with (Type == "LINK_STATUS", Value == HEARD);

               then

               -  L_SYM_time new Topology Tuple with:

               T_dest_iface_addr = current time + validity time,

               -  L_time advertised address;

               T_last_iface_addr = L_SYM_time + L_HOLD_TIME.

       2.  L_ASYM_time local address;

               T_seq_number = ANSN;

               T_time = current time + validity time;

       3.  L_time = max(L_time, L_ASYM_time)

   3.  Additionally, the willingness field is updated as follows:

          If a TLV with Type=="WILLINGNESS" is present in the message
          TLVs, then:

          +  L_willingness = Value of the TLV

          otherwise:

          +  L_willingness = WILL_DEFAULT

   The rule for setting L_time is the following: a link losing its
   symmetry SHOULD still be advertised in HELLOs (with the remaining
   status as defined by Table 1) during at least the duration of the
   "validity time".  This  allows neighbors to detect the link breakage.
   Thus, the Local Link Set must maintain information, also about LOST
   links, until the link would otherwise expire.

8.2 time.

10.1.3.  Populating the 2-Hop Neighbor Attached Network Set

   Upon receiving a HELLO message from a symmetric neighbor interface, a

   The node SHOULD update its 2-hop Neighbor Set.

   If the Source Address is the L_local_iface_addr from a link tuple
   included in the Link Set with L_STATUS equal to SYMMETRIC (in other
   words: if the Source Address is a symmetric neighbor interface) then
   the 2-hop Neighbor Attached Network Set SHOULD be updated as follows:

   1.  for each address, henceforth gateway address, in the first
       address (henceforth: 2-hop neighbor address), listed block in the HELLO TC message:

       1.  if the 2-hop neighbor address is  for each address, henceforth network address, in an interface address of the
           receiving node silently discard the 2-hop neighbor address
           (in
           block other words: a node is not its own 2-hop neighbor).

       2.  else if than the 2-hop neighbor address first in the TC message, and which has a
           an associated TLV with:

           +  (Type=LINK_STATUS, Value with Type == SYMMETRIC); OR

           +  (Type=OTHER_IF, Value=SYMMETRIC);

           a 2-hop tuple GATEWAY:

           1.  if there is created a Attached Network Tuple with:

           +  N2_local_iface_addr    =

               AN_net_addr == network address; AND

               AN_gw_iface_addr == gateway address of the interface over
              which the HELLO message was received;

           +  N2_neighbor_iface_addr

               then update this Attached Network Tuple to have:

               AN_seq_number = source address of the message; ANSN;

               AN_time = current time +  N2_2hop_iface_addr validity time

           2.  otherwise create a new Attached Network Tuple with:

               AN_net_addr = 2-hop neighbor network address;

           +  N2_time
               AN_gw_iface_addr = gateway address

               AN_seq_number = ANSN;

               AN_time = current time + validity time.

           This tuple may replace an older similar tuple with the same
           N2_local_iface_addr, N2_neighbor_iface_addr and
           N2_2hop_iface_addr values.

       3.  else if time

10.2.  Completed TC Message Processing

   The TC message(s) are processed as follows:

   1.  the 2-hop neighbor address has a TLV with:

           +  (Type == LINK_STATUS, Value == LOST); OR

           +  (Type == OTHER_IF, Value == LOST),

           then any 2-hop tuple with:

           +  N2_local_iface_addr equal Topology Set is updated according to Section 10.2.1;

   2.  the address of the interface
              over which the HELLO message was received; AND

           +  N2_neighbor_iface_addr equal Attached Network Set is updated according to Section 10.2.2.

10.2.1.  Purging the source Topology Set

   The Topology Set MUST be updated as follows:

   1.  for each address, henceforth local address, in the first address
       block of any of the
              message; TC messages, all Topology Tuples with:

       T_last_iface_addr == local address; AND

           +  and N2_2hop_iface_addr equal to the 2-hop neighbour
              address

       T_seq_number < ANSN

       MUST be deleted.

8.3  Populating removed.

10.2.2.  Purging the MPR Selector Attached Network Set

   Upon receiving a HELLO message, if a node finds one of its own
   interface addresses, listed with an MPR TLV (indicating that the
   originator node has selected one of the receiving node's interfaces
   as MPR), the MPR Selector

   The Attached Network Set SHOULD MUST be updated as follows:

   For

   1.  for each address address, henceforth local address, in the Local Interface Block first address
       block of any of the received
   message:

   1.  If there exists no MPR Selector tuple TC messages, all Attached Network Tuples
       with:

       *  MS_if_addr

       AN_gw_iface_addr == that address

       then a new tuple is created with:

       *  MS_if_addr   =  that address

   2.  The tuple (new or otherwise) with:

       *  MS_if_addr   == that address

       is then modified as follows:

       *  MS_time       =  current time + validity time. local address; AND

       AN_seq_number < ANSN

       MUST be removed.

11.  Populating the MPR Selector tuples are removed upon expiration Set

   Each node MUST select, from among its symmetric 1-hop neighbors, a
   subset of MS_time, or upon
   link breakage nodes as described in Section 8.4.

8.4  Neighborhood MPRs.  This subset MUST be selected such that a
   message transmitted by the node, and 2-Hop Neighborhood Changes

   A change retransmitted by all its MPRs,
   will be received by all of its symmetric strict 2-hop neighbors.

   Each node selects its MPR Set individually, utilizing the information
   in the neighborhood is detected when:

   o  Link Loss: Symmetric Neighbor Set, the L_SYM_time field of a link tuple expires (either
      due to time out, or 2-Hop Neighbor Set and the
   Neighborhood Address Association Set. Initially these sets will be
   empty, as a result of processing a TLV (Type ==
      LINK_STATUS, Value == LOST)).

   o  Link Acquisition: a new link tuple is inserted in will be the Link MPR Set. A node SHOULD recalculate its MPR Set
      with a non expired L_SYM_time or
   when a tuple with expired L_SYM_time
      is modified so that L_SYM_time becomes non-expired.  This relevant change is
      considered as a link acquisition if there was previously no such
      link tuple.

   o  Neighbor Loss: all links made to a neighbor node have have been lost.

   A change in the 2-hop neighborhood is detected when a Symmetric Neighbor Set, the
   2-Hop Neighbor
   Tuple expires Set or is deleted according to section Section 8.2.

   The following processing occurs when changes in the neighborhood or Neighborhood Address Association Set.

   More specifically, a node MUST calculate MPRs per interface, the 2-hop neighborhood are detected:

   o  In case
   union of link loss, all 2-Hop Neighbor Tuples with

      *  N2_local_iface_addr == interface address the MPR Sets of each interface make up the node where MPR Set for the
         link was lost

      *  N2_neighbor_iface_addr == interface address
   node.  All OLSRv2 interfaces of nodes selected as MPRs with which the neighbor
   node has a symmetric link MUST be deleted.

   o  In case of neighbor loss, all added to the MPR Selector tuples associated with
      that Set. Also
   symmetric 1-hop neighbor are deleted.  More precisely:

      *  all MPR selector tuples nodes with MS_iface_addr == interface address
         of willingness WILL_NEVER (as
   recorded in the neighbor Link Set) MUST NOT be deleted, along with any interface
         addresses associated considered as MPRs.

   MPRs are used to flood control messages from a node into the network
   while reducing the number of retransmissions that will occur in a
   region.  Thus, the Neighbor Address Association Set.

   o  The concept of MPR Set MUST be re-calculated when a link acquisition or loss is detected, or when an optimization of a change in classical
   flooding mechanism.  While it is not essential that the 2-hop neighborhood MPR Set is
      detected.

   o  An additional HELLO message MAY
   minimal, it is essential that all symmetric strict 2-hop neighbors
   can be sent when reached through the selected MPR nodes.  A node MUST select an
   MPR Set or the
      neighborhood changes.

   Additionally, proper update of the sets describing local topology
   should be made when a Neighbor Association Address Tuple has a list
   of addresses which such that any strict 2-hop neighbor is modified.

9.  TC Message Generation

   TC messages are, in OLSRv2, transmitted with the purpose of
   populating "covered" by at least
   one MPR node.  A node MAY select additional MPRs beyond the Topology Set, minimum
   set.  Keeping the Attached Network MPR Set and small ensures that the
   Neighborhood Address Association Set:

   o  Topology Discovery: ensure that information overhead of OLSRv2
   is present in each
      node describing all destinations and kept at a sufficient subset minimum.

   Appendix A contains an example heuristic for selecting MPRs.

12.  Populating Derived Sets

   The Relay Set and the Advertised Neighbor Set of links
      in order to provide least-hop paths to all destinations.

   o  Multiple Interface Declaration: ensure that nodes, up OLSRv2 are denoted
   derived sets, since updates to two hops
      away from the originator, these sets are aware not directly a function
   of message exchanges, but rather are derived from updates to other
   sets, in particular the interface configuration
      of MPR Selector Set.

12.1.  Populating the originator node.

   Thus, nodes with Relay Set

   The Relay Set contains the set of neighbor addresses, for which a non-empty
   node is supposed to relay broadcast traffic.  This set SHOULD at
   least contain all addresses in the MPR Selector Set. This set MAY
   contain additional symmetric 1-hop neighbor addresses.

12.2.  Populating the Advertised Neighbor Set, or which are
   specifically reporting an empty Set

   The Advertised Neighbor Set (for a period contains the set of T_HOLD_TIME following reporting a non-empty Advertised Neighbor
   Set) or with more than one OLSRv2 interface
   addresses of those 1-hop neighbors to which supports OLSRv2 and
   participates a node advertises a
   symmetric link in the MANET, MUST generate TC messages, according to messages.  This set SHOULD at least contain all
   of the following:

   1.  The node includes, in its first address block OLSRv2 interface addresses of the TC message,
       a Local Interface Block as specified nodes in Section 6.2

   2.  If the node has a non-empty Advertised Neighbor MPR Selector
   Set or is
       specifically reporting (i.e. all addresses associated with an empty Advertised Neighbor Set, or it
       has a one or more attached non-OLSRv2 networks, to which it
       wishes to advertise routes to MPR Selector node through
   the network, it furthermore:

       1.  includes a message TLV (Type = CONTENT_SEQ_NUMBER TLV, Value
           = Neighborhood Address Association Set, that is, appearing in the
   same NA_neighbor_iface_addr_list as any MS_neighbor_iface_addr).
   This set MAY also contain OLSRv2 interface addresses of other
   symmetric 1-hop neighbors.

   Whenever an address is removed from the Advertised Neighbor Set Sequence Number);

       2.  includes Set, the
   ANSN MUST be incremented.  Whenever an address blocks, containing its is added to the
   Advertised Neighbor Set, the ANSN MUST be incremented.

13.  Routing Table Calculation

   The Routing Set (if non-empty);

       3.  includes address blocks is updated when a change (an entry appearing or
   disappearing, or changing between SYMMETRIC and PREFIX_LENGTH TLVs, describing
           attached non-OLSRv2 networks;

       4.  sets the TTL of LOST) is detected in:

   o  the message to Link Set, OR;

   o  the network diameter.

   3.  Otherwise, Neighbor Address Association Set, OR;

   o  the node:

       1.  sets 2-Hop Neighbor Set, OR;

   o  the TTL of Topology Set, OR;

   o  the message Attached Network Set.

   Note that some changes to 2.

   OLSRv2 TC messages are generated and transmitted periodically, with these sets do not necessitate a
   default interval between two consecutive TC emissions by change to
   the same
   node of TC_INTERVAL.

9.1  TC Message: Transmission

   Messages are retransmitted Routing Set, in particular changes to the packet/message format specified by
   [4] Link Set which do not
   involve Link Tuples with L_STATUS == SYMMETRIC (either before or
   after the All-OLSRv2-Multicast address as destination IP address
   and is forwarded according change), similar changes to the specification Neighbor Address
   Association Set. A node MAY avoid updating the Routing Set in section
   Section 4.4.  If fragmentation is necessary, a FRAGMENTATION TLV MUST
   be included, and each fragment SHOULD be flagged as partially such
   cases.

   Updates to the Routing Set does not generate or
   wholly self contained as specified in [4].

10.  TC Message Processing

   Upon receiving a TC message, a node MUST update its topology
   information base according trigger any messages
   to be transmitted.  The state of the specification given Routing Set SHOULD, however, be
   reflected in this
   section.

   For the purpose of this section, note IP routing table by adding and removing entries from
   the following:

   o routing table as appropriate.

   To construct the "validity time" Routing Set of node X, a message shortest path algorithm is calculated from the
      VALIDITY_TIME message TLV according to
   run on the specification in
      Section 16; directed graph containing

   o  the "originator address" refers to the address, contained in the
      "originator address" field arcs X -> Y where there exists a Link Tuple with Y as
      L_neighbor_iface_addr and L_STATUS == SYMMETRIC (i.e.  Y is a
      symmetric 1-hop neighbor of the OLSRv2 message header specified
      in [4]; X), AND;

   o  the ASSN of the node, originating the TC message, arcs Y -> Z where Y is recovered added as above and the value of the CONTENT_SEQ_NO message TLV in the TC message, if
      any.

10.1  Checking Freshness & Validity of a TC message

   In order to be able Link Tuple with
      Y as L_neighbor_iface_addr has L_willingness not equal to ensure that only valid
      WILL_NEVER, and fresh information there exists a 2-Hop Neighbor Tuple with Y as
      N2_neighbor_iface_addr and Z as N2_2hop_iface_addr (i.e.  Z is recorded in a
      symmetric 2-hop neighbor of Z through Y, which does not have
      willingness WILL_NEVER), AND;

   o  the arcs U -> V, where there exists a Topology Set, each node maintains Tuple with U as
      T_last_iface_addr and V as T_dest_iface_addr (i.e. this is an ASSN History
   Set, recording the highest ASSN received from each node
      advertised link in the
   network, in the form of a "ASSN History Tuples":

       (AS_Address, AS_seq, AS_time)

   AS_Address is the originator address of a received TC message;

   AS_seq network).

   The graph is the highest received ASSN seen in complemented with:

   o  arcs Y -> W where there exists a TC message from
      AS_Address;

   AS_time is the time at which this tuple expires Link Tuple with Y as
      L_neighbor_iface_addr and L_STATUS == SYMMETRIC and MUST be removed.

   Upon receiving a TC message, a node MUST check if the TC message is
   fresh Neighborhood
      Address Association Tuple with Y and valid as follows:

   1.  If the TC message has more than one address block W both contained in
      NA_neighbor_iface_addr_list (i.e. not just
       a Local Interface Block)  Y and does not contain a message-TLV W are both addresses of
       type CONTENT_SEQ_NO. then
      the message MUST be discarded;

   2.  otherwise, if same symmetric 1-hop neighbor), AND;

   o  arcs U -> T where there exists an Attached Network Tuple with U as
      AN_net_addr and T as AN_gw_iface_addr (i.e.  U is a gateway to
      network T).

   The following procedure is given as an example for (re-)calculating
   the ASSN History Routing Set contains using a tuple where:

       *  AS_Address == Originator Address variation of the TC message; AND
       *  AS_seq > the ASSN recovered from the TC message,

       then the TC message MUST be discarded;

   3.  otherwise Dijkstra's algorithm.  Thus:

   1.  All Routing Tuples are removed.

   2.  For each Link Tuple with L_STATUS == SYMMETRIC, a tuple new Routing
       Tuple is inserted in the ASSN History Set added with:

       *  AS_Address  R_dest_addr = Originator Address in L_neighbor_iface_addr of the message; Link Tuple;

       *  AS_seq  R_next_iface_addr = The ASSN, extracted from L_neighbor_iface_addr of the message; Link Tuple;

       *  AS_time  R_dist = current time + AS_HOLD_TIME.

       possibly replacing an existing tuple with the same AS_Address.

10.2  Updating 1;

       *  R_local_iface_addr = L_local_iface_addr of the Topology Set

   A node SHOULD update its Topology Set as follows:

   1. Link Tuple.

   3.  For each address, LocAddr, from the Local Interface Block in the
       TC message:

       1.  For each advertised neighbor address, listed in an address
           block other than the Local Interface Block in the TC message, Neighbor Address Association Tuple, for which does NOT have an associated PREFIX_LENGTH TLV:

           1.  if there exists a tuple two
       addresses A1 and A2 are in the Topology Set NA_neighbor_iface_addr_list where:

               T_dest_iface_addr == advertised neighbor address; AND

               T_last_iface_addr

       *  there is a Routing Tuple with:

          +  R_dest_addr == LocAddr.

               then the tuple A1

       *  and there is updated as follows:

               T_time = current time no Routing Tuple with:

          + validity time

               T_seq = ASSN

           2.  Otherwise,  R_dest_addr == A2

       then a new topology tuple Routing Tuple is created added with:

               T_dest_iface_addr = advertised neighbor address, AND

               T_last_iface_addr

       *  R_dest_addr = LocAddr; AND

               T_seq A2;

       *  R_next_iface_addr = ASSN.

10.3  Purging Old Entries from the Topology Set

   Old entries from the Topology Set MUST be purged as follows:

   1.  For each address, LocAddr, from R_next_iface_addr of the Local Interface Block Routing Tuple in
          which R_dest_addr == A1;

       *  R_dist = 1;

       *  R_local_iface_addr = R_local_iface_addr of the
       TC message:

       1.  all tuples Routing Tuple
          in the Topology Set where:

           T_last_iface_addr which R_dest_addr == LocAddr AND

           T_seq < ASSN A1.

   4.  The following procedure, which adds Routing Tuples for
       destination nodes h+1 hops away, MUST be removed.

10.4  Updating the Attached Networks Set

   A node SHOULD update its Attached Networks Set as follows:

   1.  For executed for each address, LocAddr, from the Local Interface Block in the
       TC message:

       1.  For value
       of h, starting with h=2 and incrementing by 1 for each advertised neighbor address, listed in an address
           block other than the Local Interface Block iteration.
       The execution MUST stop if no new Routing Tuples are added in the TC message,
           which does have an associated PREFIX_LENGTH TLV:
       iteration.

       1.  For each Topology Tuple, if there exists a tuple in the Attached Networks Set
               where:

               AN_net_addr

           +  T_dest_iface_addr is not equal to R_dest_addr of any
              Routing Tuple, AND;

           +  T_last_iface_addr is equal to R_dest_addr of a Routing
              Tuple whose R_dist == advertised neighbor address; AND

               AN_prefix_length h;

           then a new Routing Tuple MUST be added, with:

           +  R_dest_addr = T_dest_iface_addr;

           +  R_next_iface_addr = R_next_iface_addr of the Routing Tuple
              whose R_dest_addr == T_last_iface_addr;

           +  R_dist = h+1;

           +  R_local_iface_addr = R_local_iface_addr of the prefix length as recoveredf from Routing
              Tuple whose R_dest_addr == T_last_iface_addr.

           Several Topology Tuples may be used to select a next hop
           R_next_iface_addr for reaching the PREFIX_LENGTH TLV; AND

               AN_gw_addr address R_dest_addr.  When
           h == LocAddr.

               then 1, ties should be broken such that nodes with highest
           willingness are preferred, and between nodes of equal
           willingness, MPR selectors are preferred over non-MPR
           selectors.

       2.  After the tuple above iteration has completed, if h == 1, for each
           2-Hop Neighbor Tuple where:

           +  N2_2hop_iface_addr is updated as follows:

               AN_time = current time not equal to R_dest_addr of any
              Routing Tuple, AND;

           + validity time

               AN_seq = ASSN

           2.  Otherwise,  N2_neighbor_iface_addr has a new topology tuple willingness (i.e. the
              L_willingness of the Link Tuple of which
              L_neighbor_iface_addr == N2_neighbor_iface_addr) which is created
              not equal to WILL_NEVER;

           a Routing Tuple is added with:

               AN_net_addr == advertised neighbor address; AND

               AN_prefix_length ==

           +  R_dest_addr = N2_2hop_iface_addr of the prefix length as recoveredf from 2-Hop Neighbor
              Tuple;
           +  R_next_iface_addr = R_next_iface_addr of the PREFIX_LENGTH TLV; AND

               AN_gw_addr Routing Tuple
              in which R_dest_addr == LocAddr.

               AN_time N2_neighbor_iface_addr;

           +  R_dist = current time 2;

           + validity time

               AN_seq  R_local_iface_addr = ASSN

10.5  Purging Old Entries from R_local_iface_addr of the Routing
              Tuple in which R_dest_addr == N2_neighbor_iface_addr.

   5.  For each Attached Network Set

   TBD

10.6  Processing Unfragmented TC Messages

   If an unfragmented TC message, i.e. a TC message without a
   FRAGMENTATION message TLV, Tuple, if

       *  AN_net_addr is received, it MUST be processed as
   follows:

   1.  Verify freshness and validity not equal to R_dest_addr of the TC message (see
       Section 10.1).  If the message is not discarded, then continue;

   2.  Update the Topology Set (see Section 10.2);

   3.  Purge old entries from the Topology Set (see Section 10.3);

   4.  Update the Attached Networks Set (see Section 10.4;

   5.  Purge old entries from the Attached Networks Set (see
       Section 10.5);

   6.  Update the Neighborhood Address Association Set (see Section 13).

10.7  Processing Partially or Wholly Self-Contained Fragmented TC
      Messagess

   If a TC message contains a FRAGMENTATION message TLV which indicates
   that the fragment any Routing Tuple,
          AND;

       *  AN_gw_iface_addr is a partially or wholly self-contained message,
   then the following processing SHOULD be carried out immediately upon
   receipt equal to R_dest_addr of each received fragment (if not a Routing Tuple;

       then it a new Routing Tuple MUST be carried out
   for each fragment once all fragments have been received):

   1.  Verify freshness and validity added, with:

       *  R_dest_addr = AN_net_addr;

       *  R_next_iface_addr = R_next_iface_addr of the TC message (see
       Section 10.1).  If the message is not discarded, then continue;
   2.  Update the Topology Set (see Section 10.2);

   3.  Update Routing Tuple
          whose R_dest_addr == AN_gw_iface_addr;

       *  R_dist = R_dist of the Neighborhood Address Association Set (see Section 13).

   4.  Update Routing Tuple whose R_dest_addr ==
          AN_gw_iface_addr;

       *  R_local_iface_addr = R_local_iface_addr of the Routing Tuple
          whose R_dest_addr == AN_gw_iface_addr.

       If more than one Attached Networks Set (see Section 10.4;

   Once all fragments have been received, Network Tuple has the following processing same AN_net_addr,
       then more than one Routing Tuple MUST NOT be carried out once:

   1.  Purge old entries from the Topology Set (see Section 10.3);

   2.  Purge old entries from the Attached Networks Set (see
       Section 10.5);

11.  Populating added, and the MPR Set

   Each node added
       Routing Tuple MUST select, from among its one-hop neighbors, a subset of
   nodes as MPRs. have minimum R_dist.

14.  Proposed Values for Constants

   This subset MUST be selected such that a message
   transmitted by section list the node, and retransmitted by all its MPR nodes, will
   be received by all nodes 2 hops away.

   Each node selects its MPR Set individually, utilizing values for the information constants used in the Link Set, 2-Hop Neighbor Set and
   description of the protocol.

14.1.  Neighborhood Address
   Association Set. Initially these sets will be empty, Discovery Constants

   The constants HELLO_INTERVAL, REFRESH_INTERVAL, HELLO_MIN_INTERVAL,
   H_HOLD_TIME, L_HOLD_TIME, N_HOLD_TIME and C are used as will be in [4].

14.2.  Message Intervals

   o  TC_INTERVAL = 5 seconds

   o  TC_MIN_INTERVAL = TC_INTERVAL/4

14.3.  Holding Times

   o  T_HOLD_TIME = 3 x TC_INTERVAL

   o  A_HOLD_TIME = T_HOLD_TIME

   o  P_HOLD_TIME = 30 seconds

   o  FG_HOLD_TIME = 30 seconds

   o  RX_HOLD_TIME = 30 seconds

   o  FW_HOLD_TIME = 30 seconds

14.4.  Willingness

   o  WILL_NEVER = 0

   o  WILL_DEFAULT = 3

   o  WILL_ALWAYS = 7

15.  Sequence Numbers

   Sequence numbers are used in OLSRv2 with the
   MPR Set. A node SHOULD recalculate its MPR Set when purpose of discarding
   "old" information, i.e. messages received out of order.  However with
   a relevant change limited number of bits for representing sequence numbers, wrap-
   around (that the sequence number is made incremented from the maximum
   possible value to zero) will occur.  To prevent this from interfering
   with the Link Set, 2-Hop Neighbor Set or Neighborhood Address
   Association Set.

   More specifically, a node operation of OLSRv2, the following MUST calculate MPRs per interface, be observed when
   determining the
   union ordering of sequence numbers.

   The term MAXVALUE designates in the MPR Sets of each interface make up following one more than the MPR Set
   largest possible value for the
   node.

   MPRs are used to flood control messages from a node into the network
   while reducing the number of retransmissions that will occur in sequence number.  For a
   region.  Thus, the concept of MPR is an optimization of a classical
   flooding mechanism.  While it is not essential that the MPR Set 16 bit sequence
   number (as are those defined in this specification) MAXVALUE is
   minimal, it
   65536.

   The sequence number S1 is essential that all strict 2-hop neighbors can said to be
   reached through "greater than" the selected MPR nodes.  A node MUST select an MPR
   Set such that any strict 2-hop neighbor sequence
   number S2 if:

   o  S1 > S2 AND S1 - S2 <= MAXVALUE/2 OR

   o  S2 > S1 AND S2 - S1 > MAXVALUE/2

   Thus when comparing two messages, it is covered by at least one
   MPR node.  A node MAY select additional MPRs beyond the minimum set.
   Keeping the MPR Set small ensures that possible - even in the overhead
   presence of OLSRv2 is kept
   at a minimum.

   Appendix A wrap-around - to determine which message contains an example heuristic the
   most recent information.

16.  IANA Considerations

16.1.  Multicast Addresses

   A well-known multicast address, ALL-MANET-NEIGHBORS, must be
   registered and defined for selecting MPRs.

12.  Populating Derived Sets

   The Relay Set both IPv6 and IPv4.  The addressing scope
   is link-local, i.e. this address is similar to the Advertised Neighbor Set all nodes/routers
   multicast address of IPv6 in that it targets all OLSRv2 are denoted
   derived sets, since updates capable nodes
   adjacent to these sets are not directly a function the originator of an IP datagram.

16.2.  Message Types

     OLSRv2 defines one message exchanges, but rather are derived type, which must be allocated from updates to other
   sets, in particular the MPR Selector Set.

12.1  Populating the Relay Set

   The Relay Set contains the set of neighbor addresses, for
                "Assigned Message Types" repository of [3]

   +--------------------+-------+--------------------------------------+
   |      Mnemonic      | Value | Description                          |
   +--------------------+-------+--------------------------------------+
   |         TC         |  TBD  | Topology Control (global signaling)  |
   +--------------------+-------+--------------------------------------+

                                  Table 5

16.3.  TLV Types

   OLSRv2 defines one Message TLV type, which a
   node is supposed to relay broadcast traffic.  This set SHOULD at
   least contain must be allocated from the addresses
              "Assigned message TLV Types" repository of the MPR Selector set (i.e. all
   addresses, associated with [3]

   +--------------------+-------+--------------------------------------+
   |      Mnemonic      | Value | Description                          |
   +--------------------+-------+--------------------------------------+
   |     WILLINGNESS    |  TBD  | Specifies a MPR selector through the Neighborhood
   Address Association Set).  This set MAY contain additional neighbor
   addresses.

12.2  Populating the Advertised Neighbor Set

   The Advertised Neighbor Set contains the set of neighbor addresses, node's willingness to which    |
   |                    |       | act as a node advertises links through TC messages.  This set
   SHOULD at least contain relay and to partake in     |
   |                    |       | network formation                    |
   +--------------------+-------+--------------------------------------+

                                  Table 6
    OLSRv2 defines one Address Block TLV type, which must be allocated
       from the addresses "Assigned address block TLV Types" repository of the [3]

   +--------------------+-------+--------------------------------------+
   |      Mnemonic      | Value | Description                          |
   +--------------------+-------+--------------------------------------+
   |         MPR Selector Set (i.e.
   all addresses, associated with        |  TBD  | Specifies that a MPR selector through the
   Neighborhood Address Association Set).  This set MAY contain
   additional neighbor addresses.

   Each time an address is removed from the Advertised Neighbor Set, the
   ASSN MUST be incremented.  When an given address is added    |
   |                    |       | selected as MPR                      |
   +--------------------+-------+--------------------------------------+

                                  Table 7

17.  References

17.1.  Normative References

   [1]  Clausen, T. and P. Jacquet, "The Optimized Link State Routing
        Protocol", RFC 3626, October 2003.

   [2]  Bradner, S., "Key words for use in RFCs to the Advertised
   Neighbor Set, the ASSN MUST be incremented.

13.  Populating the Indicate Requirement
        Levels", RFC 2119, BCP 14, March 1997.

   [3]  Clausen, T., Dean, J., Dearlove, C., and C. Adjih, "Generalized
        MANET Packet/Message Format", work in
        progress draft-ietf-manet-packetbb-01.txt, June 2006.

   [4]  Clausen, T., Dean, J., and C. Dearlove, "MANET Neighborhood Address Association Set

   All OLSRv2 messages containing
        Discovery Protocol (NHDP)", work in
        progress draft-ietf-manet-nhdp-00.txt, June 2006.

17.2.  Informative References

   [5]  Atkins, D., Stallings, W., and P. Zimmermann, "PGP Message
        Exchange Formats", RFC 1991, August 1996.

   [6]  ETSI, "ETSI STC-RES10 Committee.  Radio equipment and systems:
        HIPERLAN type 1, functional specifications ETS 300-652",
        June 1996.

   [7]  Jacquet, P., Minet, P., Muhlethaler, P., and N. Rivierre,
        "Increasing reliability in cable free radio LANs: Low level
        forwarding in HIPERLAN.", 1996.

   [8]  Qayuum, A., Viennot, L., and A. Laouiti, "Multipoint relaying:
        An efficient technique for flooding in mobile wireless
        networks.", 2001.

Appendix A.  Example Heuristic for Calculating MPRs

   The following specifies a Local Interface Block (including
   HELLO proposed heuristic for selection of MPRs.

   In graph theory terms, MPR computation is a "set cover" problem,
   which is a difficult optimization problem, but for which an easy and TC messages) SHOULD be used to update
   efficient heuristics exist: the Neighborhood
   Address Association Set as follows:

   1.  If there so-called "Greedy Heuristic", a
   variant of which is described here.  In simple terms, MPR computation
   constructs an MPR Set that enables a Neighborhood Address Association Tuple, node to reach any of
       whose addresses symmetric
   2-hop neighbors by relaying through an MPR node.

   There are in the Local Interface Block being processed,
       then discard several peripheral issues that tuple.

   2.  A tuple is added to the Neighborhood Address Association Set,
       where:

       *  NA_neighbor_addr_list = all addresses from the Local Interface
          Block;

       *  NA_time = current time + NA_HOLD_TIME.

14.  Routing Table Calculation algorithm needs to
   address.  The Routing Set first one is updated when a change (an entry appearing/
   disappearing) that some nodes have some willingness
   WILL_NEVER.  The second one is detected in:

   o  the Link Set,

   o  the Neighbor Address Association Set,

   o  the 2-hop Neighbor Set,

   o  the Topology Set,

   Updates to the Routing Set does not generate or trigger any messages
   to be transmitted. that some nodes may have several
   interfaces.

   The state of the Routing Set SHOULD, however, algorithm hence can be
   reflected summarized by:

   o  All 1-hop neighbor nodes with willingness equal to WILL_NEVER MUST
      ignored in the IP routing table by adding and removing entries from
   the routing table following algorithm: they are not considered as appropriate.

   To construct the Routing Set of node X, a shortest path algorithm
      1-hop neighbors (hence not used as MPRs).

   o  Because link sensing is
   run on the directed graph containing performed by interface, the arcs X -> Y where Y local network
      topology is any
   symmetric neighbor best described in terms of X (with Link Type equal to SYM), links: hence the arcs Y ->
   Z where Y algorithm
      is a considering 1-hop neighbor node with willingness different of WILL_NEVER OLSRv2 interfaces, and there exists an entry in the 2-hop Neighbor Set with Y as
   N2_neighbor_iface_addr and Z as N2_2hop_iface_addr, and the arcs U ->
   V, where there exists an entry
      neighbor OLSRv2 interfaces (and their addresses).  Additionally,
      asymmetric links are ignored.  This is reflected in the Topology Set with V as
   T_dest_iface_addr and U as T_last_iface_addr.  The graph
      definitions below.

   o  MPR computation is
   complemented with performed on each interface of the arcs W0 -> W1 where W0 and W1 node: on
      each interface I, the node MUST select some neighbor interfaces,
      so that all 2-hop neighbor interfaces are two addresses reached.

   From now on, MPR calculation will be described for one interface I on
   the node, and the following terminology will be used in describing
   the heuristics:

   neighbor interface (of I) - An OLSRv2 interface of a 1-hop neighbor
      to which there exist a symmetric link using interface I.

   N  - the set of such neighbor interfaces

   2-hop neighbor interface (of I) An interface of a same symmetric strict
      2-hop neighbor (in which can be reached from a neighbor address association
   tuple).

   The following procedure interface for
      I.

   N2 - the set of such 2-hop neighbor interfaces

   D(y): - the degree of a 1-hop neighbor interface y (where y is given a
      member of N), is defined as an example for (re-)calculating the Routing number of symmetric neighbor
      interfaces of node y which are in N2

   MPR Set (with a breadth-first algorithm):

   1.  All - the tuples from set of the Routing Set are removed.

   2. neighbor interfaces selected as MPRs.

   The new routing tuples are added starting with the symmetric
       neighbors (h=1) proposed heuristic selects iteratively some interfaces from N as the destinations.  Thus, for each tuple
   MPRs in the
       Link order to cover 2-hop neighbor interfaces from N2, as follows:

   1.  Start with an MPR Set where:

       *  L_STATUS           == SYMMETRIC (L_STATUS made of all members of N with L_willingness
       equal to WILL_ALWAYS

   2.  Calculate D(y), where y is calculated as
          indicated in Table 1) a new routing tuple is recorded member of N, for all interfaces in
       N.

   3.  Add to the Routing MPR Set with:

       *  R_dest_iface_addr  = L_neighbor_iface_addr, of those interfaces in N, which are the *only*
       nodes to provide reachability to an interface in N2.  For
       example, if interface B in N2 can be reached only through a
       symmetric link tuple;

       *  R_next_iface_addr  = L_neighbor_iface_addr, of to interface A in N, then add interface B to the link tuple;
       *  R_dist             = 1;

       *  R_iface_addr       = L_local_iface_addr of
       MPR Set. Remove the link tuple.

   3.  for each neighbor address association tuple, for interfaces from N2 which two
       addresses A1 and A2 exist in I_neighbor_iface_addr_list where:

       *  there exists are now covered by a routing tuple with:

          +  R_dest_iface_addr == A1

       *
       interface in the MPR Set.

   4.  While there is no routing tuple with:

          +  R_dest_iface_addr == A2

       then a tuple exist interfaces in N2 which are not covered by at
       least one interface in the Routing Set is created with:

       *  R_dest_iface_addr = A2;

       *  R_next_iface_addr = R_next_iface_addr of MPR Set:

       1.  For each interface in N, calculate the route tuple of
          A1;

       *  R_dist            = R_dist of reachability, i.e.,
           the route tuple of A1 (e.g. 1);

       *  R_iface_addr      = R_iface_addr number of interfaces in N2 which are not yet covered by
           at least one node in the route tuple of A1.

   4.  for each symmetric strict 2-hop MPR Set, and which are reachable
           through this neighbor where interface;

       2.  Select as an MPR the
       N2_neighbor_iface_addr has a willingness different from
       WILL_NEVER a tuple in interface with highest L_willingness
           among the Routing Set is created with:

       *  R_dest_iface_addr = N2_2hop_iface_addr interfaces in N with non-zero reachability.  In
           case of multiple choice select the 2-hop neighbor;

       *  R_next_iface_addr = interface which provides
           reachability to the R_next_iface_addr maximum number of the route tuple
          with:

          +  R_dest_iface_addr == N2_neighbor_iface_addr interfaces in N2.  In
           case of multiple interfaces providing the 2-hop
             tuple;

       *  R_dist            = 2;

       *  R_iface_addr      = the R_iface_addr of the route tuple with:

          +  R_dest_iface_addr == N2_neighbor_iface_addr same amount of
           reachability, select the 2-hop
             tuple;

   5.  The new route tuples for interface as MPR whose D(y) is
           greater.  Remove the destination nodes h+1 hops away interfaces from N2 which are
       recorded now covered
           by an interface in the routing table.  The following procedure MUST be
       executed MPR Set.

   Other algorithms, as well as improvements over this algorithm, are
   possible.  For example:

   o  Assume that in a multiple interface scenario there exists more
      than one link between nodes 'a' and 'b'.  If node 'a' has selected
      node 'b' as MPR for each value one of h, starting its interfaces, then node 'b' can be
      selected as MPR with h=2 and incrementing minimal performance loss by 1 any other
      interfaces on node 'a'.

   o  In a multiple interface scenario MPRs are selected for each iteration.
      interface of the selecting node, providing full coverage of all
      2-hop nodes accessible through that interface.  The execution will stop if no new tuple overall MPR
      Set is recorded in an iteration.

       1.  For each topology tuple in then the Topology Set, if its
           T_dest_iface_addr does not correspond to R_dest_iface_addr union of
           any route tuple in the Routing Set AND its T_last_iface_addr
           corresponds these sets.  These sets do not however
      have to R_dest_iface_addr of be selected independently, if a route tuple whose
           R_dist node is equal to h, then a new route tuple MUST selected as an MPR
      for one interface it may be recorded automatically added to the MPR
      selection for other interfaces.

Appendix B.  Heuristics for Generating Control Traffic

   A node creates HELLO messages and TC messages as specified in
   Section 7 and Section 9, the Routing Set (if it does not already exist) where:

           +  R_dest_iface_addr = T_dest_iface_addr;

           +  R_next_iface_addr = R_next_iface_addr former being a modification of the route tuple
              where:

              -  R_dest_iface_addr == T_last_iface_addr

           +  R_dist           = h+1; and

           +  R_iface_addr     = R_iface_addr
   specification in [4].  The heuristics for creation of HELLO messages
   in [4] remain applicable, with the division of the route tuple where:

              -  R_dest_iface_addr address TLVs with
   Type == T_last_iface_addr.

       2.  Several topology tuples may be used to select a next hop
           R_next_iface_addr LINK_STATUS and Value == SYMMETRIC into separate ranges with
   and without an associated TLV with Type == MPR.  The heuristics for reaching
   collection of addresses are also generally applicable to TC messages,
   excepting that the node R_dest_iface_addr.
           When h==1, ties first address block is not sorted as the Local
   Interface Block of a HELLO message is, and that other addresses
   recorded in TC messages are divided into those with and without a TLV
   with Type == GATEWAY.  These should be broken such ordered so that nodes with highest
           willingness the range of
   addresses without that TLV is continuous (and it is suggested that
   the range without is also continuous).

Appendix C.  Protocol and MPR selectors Port Number

   Packets in OLSRv2 are preferred as next hop.

15.  Proposed Values communicated using UDP.  Port 698 has been
   assigned by IANA for Constants exclusive usage by the OLSR (v1 and v2)
   protocol.

Appendix D.  Packet and Message Layout

   This section list specifies the values for translation from the constants used abstract descriptions
   of packets employed in the
   description protocol specification, and the bit-layout
   packets actually exchanged between the nodes.

Appendix D.1.  OLSRv2 Packet Format

   The basic layout of an OLSRv2 packet is as described in [3].  However
   the protocol.

15.1  Message Intervals

   o  HELLO_INTERVAL        = 2 seconds

   o  REFRESH_INTERVAL      = 2 seconds

   o  TC_INTERVAL           = 5 seconds

15.2  Holding Times

   o  L_HOLD_TIME           = 3 x HELLO_INTERVAL

   o  N2_HOLD_TIME          = following points should be noted.

   OLSRv2 uses only packets with a packet header including a packet
   sequence number, either with or without a packet TLV block.  Thus all
   OLSRv2 packets have the layout of either

      0                   1                   2                   3 x REFRESH_INTERVAL

   o  NA_HOLD_TIME          =
      0 1 2 3 x TC_INTERVAL

   o  T_HOLD_TIME           = 4 5 6 7 8 9 0 1 2 3 x TC_INTERVAL

   o  RX_HOLD_TIME          = 30 seconds

   o  FW_HOLD_TIME          = 30 seconds

   o  P_HOLD_TIME           = 30 seconds

   o  FG_HOLD_TIME          = 30 seconds

15.3  Willingness

   o  WILL_NEVER            = 4 5 6 7 8 9 0

   o  WILL_LOW              = 1

   o  WILL_DEFAULT          = 2 3

   o  WILL_HIGH             = 4 5 6

   o  WILL_ALWAYS           = 7

15.4  Time

   o  C                 = 0.0625 seconds (1/16 second)

16.  Representing Time

   OLSRv2 specifies several TLVs, where time, in seconds, is to be
   represented via an 8 bit field.

   Of these 9 0 1
     +-+-+-+-+-+-+-+-+-+-+-+-+-+-+-+-+-+-+-+-+-+-+-+-+-+-+-+-+-+-+-+-+
     |0 0 0 0 0 0 0 0| Reserved  |0|0|    Packet Sequence Number     |
     +-+-+-+-+-+-+-+-+-+-+-+-+-+-+-+-+-+-+-+-+-+-+-+-+-+-+-+-+-+-+-+-+
     |                                                               |
     |                       Message + Padding                       |
     |                                                               |
     +-+-+-+-+-+-+-+-+-+-+-+-+-+-+-+-+-+-+-+-+-+-+-+-+-+-+-+-+-+-+-+-+
     |                                                               |
     :                              ...                              :
     |                                                               |
     +-+-+-+-+-+-+-+-+-+-+-+-+-+-+-+-+-+-+-+-+-+-+-+-+-+-+-+-+-+-+-+-+
     |                                                               |
     |                       Message + Padding                       |
     |                                                               |
     +-+-+-+-+-+-+-+-+-+-+-+-+-+-+-+-+-+-+-+-+-+-+-+-+-+-+-+-+-+-+-+-+

   or
      0                   1                   2                   3
      0 1 2 3 4 5 6 7 8 bits, the highest four bits represent the mantissa (a) and
   the four lowest bits represent the exponent (b), yielding that:

   o  time = C*(1+a/16)* 2^b  [in seconds]

   where a is the integer represented by the four highest bits of the
   time field and b the integer represented by the four lowest bits of
   the time field.  The proposed value of the scaling factor C is
   specified in Section 15.  All nodes in the network MUST use the same
   value of C.

17.  IANA Considerations

17.1  Multicast Addresses

   A well-known multicast address, All-OLSRv2-Multicast, must be
   registered and defined for both IPv6 and IPv4.  The addressing scope
   is link-local, i.e. this address is similar to the all nodes/routers
   multicast address of IPv6 in that it targets all OLSRv2 capable nodes
   adjacent to the originator of an IP datagram.

17.2  Message Types

   OLSRv2 defines two message types, which must be allocated from the
   "Assigned Message Types" repository of [4]

   +--------------------+--------+-------------------------------------+
   |      Mnemonic      |  Value 9 0 1 2 3 4 5 6 7 8 9 0 1 2 3 4 5 6 7 8 9 0 1
     +-+-+-+-+-+-+-+-+-+-+-+-+-+-+-+-+-+-+-+-+-+-+-+-+-+-+-+-+-+-+-+-+
     |0 0 0 0 0 0 0 0| Reserved  |1|0|    Packet Sequence Number     | Description
     +-+-+-+-+-+-+-+-+-+-+-+-+-+-+-+-+-+-+-+-+-+-+-+-+-+-+-+-+-+-+-+-+
     |
   +--------------------+--------+-------------------------------------+                                                               |       HELLOv2
     |   TBD                       Packet TLV Block                        | Local Signaling
     |                                                               |
     |                               +-+-+-+-+-+-+-+-+-+-+-+-+-+-+-+-+
     |                               |            Padding            |        TCv2
     +-+-+-+-+-+-+-+-+-+-+-+-+-+-+-+-+-+-+-+-+-+-+-+-+-+-+-+-+-+-+-+-+
     |   TBD                                                               | Global Signaling
     |
   +--------------------+--------+-------------------------------------+

                                  Table 7

17.3  TLV Types

   OLSRv2 defines three                       Message TLV types, which must be allocated from
   the "Assigned message TLV Types" repository of [4]

   +--------------------+--------+-------------------------------------+
   |      Mnemonic      |  Value + Padding                       | Description
     |
   +--------------------+--------+-------------------------------------+                                                               |    VALIDITY_TIME
     +-+-+-+-+-+-+-+-+-+-+-+-+-+-+-+-+-+-+-+-+-+-+-+-+-+-+-+-+-+-+-+-+
     |   TBD                                                               | The time (in seconds) from receipt
     :                              ...                              :
     |                                                               |
     +-+-+-+-+-+-+-+-+-+-+-+-+-+-+-+-+-+-+-+-+-+-+-+-+-+-+-+-+-+-+-+-+
     |                                                               | of the message during which the
     |                       Message + Padding                       |
     |                                                               | information contained in
     +-+-+-+-+-+-+-+-+-+-+-+-+-+-+-+-+-+-+-+-+-+-+-+-+-+-+-+-+-+-+-+-+

   The reserved bits marked Resv SHOULD be cleared ('0').  The octets
   indicated as Padding are optional and MAY be omitted; if not omitted
   they SHOULD be used to pad to a 32 bit boundary and MUST all be zero.

   OLSRv2 uses only messages with a complete message header.  Thus all
   OLSRv2 messages, plus padding if any, have the following layout.

      0                   1                   2                   3
      0 1 2 3 4 5 6 7 8 9 0 1 2 3 4 5 6 7 8 9 0 1 2 3 4 5 6 7 8 9 0 1
     +-+-+-+-+-+-+-+-+-+-+-+-+-+-+-+-+-+-+-+-+-+-+-+-+-+-+-+-+-+-+-+-+
     | Message Type  |  Resv   |N|0|0|         Message Size          |
     +-+-+-+-+-+-+-+-+-+-+-+-+-+-+-+-+-+-+-+-+-+-+-+-+-+-+-+-+-+-+-+-+
     | is to be valid                      |                      Originator Address                       |
     +-+-+-+-+-+-+-+-+-+-+-+-+-+-+-+-+-+-+-+-+-+-+-+-+-+-+-+-+-+-+-+-+
     |   Hop Limit   |   Hop Count   |    Message Sequence Number    |    INTERVAL_TIME
     +-+-+-+-+-+-+-+-+-+-+-+-+-+-+-+-+-+-+-+-+-+-+-+-+-+-+-+-+-+-+-+-+
     |   TBD                                                               | The time (in seconds) between two
     |                         Message Body                          |
     |                                                               | successive transmissions of
     |                               +-+-+-+-+-+-+-+-+-+-+-+-+-+-+-+-+
     |                               |            Padding            |
     +-+-+-+-+-+-+-+-+-+-+-+-+-+-+-+-+-+-+-+-+-+-+-+-+-+-+-+-+-+-+-+-+
   The reserved bits marked Resv SHOULD be cleared ('0').  In standard
   OLSRv2 messages of a given type            |
   |                    |        |                                     |
   |     WILLINGNESS    |   TBD  | Specifies a node's willingness      |
   |                    |        | [0-7] to act as a relay (HELLO and to      |
   |                    |        | partake in network formation        |
   +--------------------+--------+-------------------------------------+
                                  Table 8

   OLSRv2 defines three Address Block TLV types, which must TC) the type dependent sequence number bit
   marked N SHOULD also be allocated
   from cleared ('0').

   The layouts of the "Assigned message body, address block, TLV block and TLV Types" repository of [4]

   +--------------------+--------+-------------------------------------+
   |      Mnemonic      |  Value | Description                         |
   +--------------------+--------+-------------------------------------+
   |      OTHER_IF      |   TBD  | Specifies that an are
   as in [3], allowing all options.  Standard (HELLO and TC) messages
   contain a first address is        |
   |                    |        | associated to an block which contains local interface address
   information, all other    |
   |                    |        | than the one where address blocks contain neighbor interface
   address information (or for a TC message address information for
   which it is a gateway) specific to the message type.

   An example HELLO message, using IPv4 (four octet) addresses is as
   follows.  The overall message length is   |
   |                    |        | transmitted, 56 octets (it does not need
   padding).  The message has a hop limit of 1 and may specify a hop count of 0, as
   sent by its    |
   |                    |        | status (verified bidirectional or   |
   |                    |        | lost)                               |
   |                    |        |                                     |
   |     LINK_STATUS    |   TBD  | Specifies originator.

   The message has a given link's status     |
   |                    |        | (asymmetric, verified               |
   |                    |        | bidirectional, lost)                |
   |                    |        |                                     |
   |         MPR        |   TBD  | Specifies that message TLV block with content length 12 octets
   containing three message TLVs.  These TLVs represent message validity
   time, message interval time and willingness.  Each uses a given TLV with
   semantics value 4, indicating no start and stop indexes are included,
   and each has a value length of 1 octet.

   The first address block contains a 1 local interface address, with
   head length 4.  This is   |
   |                    |        | selected as MPR                     |
   +--------------------+--------+-------------------------------------+

                                  Table 9

18.  References

   [1]  Clausen, T., "The Optimized Link State Routing Protocol",
        RFC 3626, October 2003.

   [2]  Bradner, S., "Key words for use in RFCs equal to Indicate Requirement
        Levels", RFC 2119, BCP 14, March 1997.

   [3]  Atkins, D., Stallings, W., and P. Zimmermann, "PGP Message
        Exchange Formats", RFC 1991, August 1996.

   [4]  Clausen, T., Dean, J., and C. Dearlove, "Generalized MANET
        Packet/Message Format", Work In
        Progress draft-ietf-manet-packetbb-00.txt, February 2006.

   [5]  ETSI, "ETSI STC-RES10 Committee.  Radio equipment and systems:
        HIPERLAN type 1, functional specifications ETS 300-652",
        June 1996.

   [6]  Jacquet, P., Minet, P., Muhlethaler, P., the address length, thus no tail or
   mid sections of the address are included.  This address block has no
   TLVs (the TLV block content length is 0 octets).

   The second, and N. Rivierre,
        "Increasing reliability in cable free radio LANs: Low level
        forwarding in HIPERLAN.", 1996.

   [7]  Qayuum, A., Viennot, L., last, address block reports 4 neighbor interface
   addresses, with address head length 3 octets, and A. Laouiti, "Multipoint relaying:
        An efficient technique for flooding in mobile wireless
        networks.", 2001.

Authors' Addresses

   Thomas Heide Clausen
   LIX, Ecole Polytechnique, France

   Phone: +33 6 6058 9349
   Email: T.Clausen@computer.org
   URI:   http://www.lix.polytechnique.fr/Labo/Thomas.Clausen/

   Christopher M. Dearlove
   BAE Systems Advanced Technology Centre

   Phone: +44 1245 242194
   Email: chris.dearlove@baesystems.com

   The OLSRv2 Design Team
   MANET Working Group

Appendix A.  Example Heuristic for Calculating MPRs no tail octet (zero
   tail length).  Thus each mid address section is of length one octet.
   The following specifies a proposed heuristic for selection address TLV block (content length 11 octets) includes
   two TLVs.

   The first of MPRs.

   In graph theory terms, MPR computation is a "set cover" problem,
   which is a difficult optimization problem, but for which an easy and
   efficient heuristics exist: these TLVs reports the so-called "Greedy Heuristic", a
   variant link status of which is described here.  In simple terms, MPR computation
   constructs an MPR Set that enables all four neighbors
   in a node to reach any 2-hop
   interfaces by relaying through an MPR node.

   There single multivalue TLV, the first two addresses are several peripheral issues that HEARD, the algorithm need to
   address.
   last two addresses are SYMMETRIC.  The first one TLV semantics value of 12
   indicates, in addition to that this is a multivalue TLV, that some nodes have some willingness
   WILL_NEVER. no
   start index and stop index are included, hence values for all
   addresses are included.  The second TLV value length of 4 octets indicates
   one is that some nodes may have several
   interfaces. octet per value per address.

   The algorithm hence need to be precised in the following way:

   o  All neighbor nodes with willingness equal to WILL_NEVER MUST
      ignored in second of these TLV indicates that the following algorithm: they are not considered as
      neighbors (hence not used as MPRs), nor as 2-hop neighbors (hence
      no attempt to cover them is made).

   o  Because link sensing is performed by interface, the local network
      topology, is best described in terms of links: hence the algorithm last address (start index
   3, stop index 3) is considering neighbor interfaces, and 2-hop neighbor interfaces
      (and their addresses).  Additionally, asymmetric links are
      ignored. an MPR.  This is reflected in the definitions below.

   o  MPR computation is performed on each interface of the node: on
      each interface I, the node MUST select some neighbor interfaces,
      so that all 2-hop interfaces are reached.

   From now on, MPR calculation will be described for one interface I on
   the node, and the following terminology will be used in describing
   the heuristics:

   neighbor interface (of I) - An interface of a neighbor to which there
      exist a symmetrical link on interface I.

   N  - the set of such neighbor interfaces

   2-hop neighbor interface (of I) An interface of a symmetric strict
      2-hop neighbor and which can be reached from a neighbor interface
      for I.

   N2 - the set of such 2-hop neighbor interfaces

   D(y): - the degree of a 1-hop neighbor interface y (where y is a
      member of N), is defined as the number of symmetric neighbor
      interfaces of node y which are in N2

   MPR Set - the set of the neighbor interfaces selected as MPRs.

   The proposed heuristic selects iteratively some interfaces from N TLV has no value, or value length,
   fields, as
   MPRs in order to cover 2-hop neighbor interfaces from N2, as follows:

   1.  Start with an MPR Set made of all members of N with N_willingness
       equal to WILL_ALWAYS

   2.  Calculate D(y), where y is a member of N, for all interfaces in
       N.

   3.  Add to the MPR Set those interfaces in N, which are the *only*
       nodes to provide reachability to an interface in N2.  For
       example, if interface B in N2 can be reached only through a
       symmetric link to interface A in N, then add interface B to the
       MPR Set. Remove the interfaces from N2 which are now covered by a
       interface in the MPR Set.

   4.  While there exist interfaces in N2 which are not covered by at
       least one interface in the MPR Set:

       1.  For each interface in N, calculate the reachability, i.e.,
           the number of interfaces in N2 which are not yet covered by
           at least one node in the MPR Set, and which are reachable
           through this neighbor interface;

       2.  Select as an MPR the interface with highest N_willingness
           among the interfaces in N with non-zero reachability.  In
           case of multiple choice select the interface which provides
           reachability to the maximum number of interfaces in N2.  In
           case of multiple interfaces providing the same amount of
           reachability, select the interface as MPR whose D(y) is
           greater.  Remove the interfaces from N2 which are now covered
           by an interface in the MPR Set.

   Other algorithms, as well as improvements over this algorithm, are
   possible.  For example:

   o  Some 2-hop neighbors may have several interfaces.  The described
      algorithm attempts to reach every such interface of the nodes.
      However, whenever information that several 2-hop interfaces are,
      in fact, interfaces of the same 2-hop neighbor, is available, it
      can be used: only one of the interfaces of the 2-hop neighbor
      needs to be covered.  This information is provided in the
      Neighborhood Address Association Set.

   o  Assume that in a multiple interface scenario there exists more
      than one link between nodes 'a' and 'b'.  If node 'a' has selected
      node 'b' as MPR for one of its interfaces, then node 'b' can be
      selected as MPR with minimal performance loss by any other
      interfaces on node 'a'.

   o  In a multiple interface scenario MPRs are selected for each
      interface of the selecting node, providing full coverage of all
      2-hop nodes accessible through that interface.  The overall MPR
      Set is then the union of these sets.  These sets do not however
      have to be selected independently, if a node is selected as an MPR
      for one interface it may be automatically added to the MPR
      selection for other interfaces.

Appendix B.  Example Algorithms for Generating Control Traffic

   The proposed generation of the control messages proceeds in four
   steps.  HELLO messages and TC messages both essentially consist of a
   list of advertised addresses of neighbors (some part of the
   topology).

   Hence, a first step is to collect the set of relevant addresses which
   are to be advertised.  Because there are a number of TLVs which can
   be associated with each address (including mandatory ones), this step
   results in a list of addresses, each associated with a certain number
   of TLVs.

   The second step is then to regroup the addresses which share exactly
   the same TLVs (same Type and same Value), into an address block which
   will be associated with a list of TLVs.

   The third step is to pack the message header and message TLVs into a
   sequence of octets.

   The fourth step consists of packing every address block obtained in
   the second step by finding the longest common prefix of the addresses
   in the address block (the head), then, packing the list of the tails
   of the addresses into a sequence of octets, followed by the TLVs of
   the address block.

   This generation method can be used for TC generation and HELLO
   generation: in each case, all what need to be specified is the
   message headers, message TLVs, and the list of each address with its
   associated TLVs.

   The Local Interface Block MUST include all of the participating
   interface addresses of the node (including the one of chosen as the
   node's originator address and included in the message header).

Appendix B.1  Example Algorithm for Generating HELLO messages

   This section proposes an algorithm for generating HELLO messages.
   Periodically, on each interface I, the node generates a HELLO message
   specific to that interface, as follows:

   1.  First, the list of the links of the interface is collected.  It
       is the list of the Link Tuples where:

       *  L_local_iface_addr == address of the interface

       Each corresponding address L_neighbor_iface_addr is then
       advertised with the following TLVs:

       *  Type="LINK-STATUS", Value=L_STATUS, the status of the link
          (see Section 5.1.1);

       *  Type="OTHER_IF", if and only if as specified in Section 7);

       *  Type="MPR", if and only of the address L_neighbor_iface_addr
          is an interface address in the MPR Set.

   2.  Second, if the node has more than one interface, for each address
       which was not previously advertised and for which there exists a
       Link Tuple on another interface where:

       *  L_local_iface_addr is different from address of the interface
          I; AND

       *  L_STATUS == SYMMETRIC

       the corresponding address L_neighbor_iface_addr is advertised
       with the following TLV:

       *  Type="OTHER_IF", Value=SYMMETRIC.

   3.  Third, if the node has more than one interface, for each
       interface address which is to be reported as LOST as specified in
       Section 7) the interface address is advertised with the following
       TLV:

       *  Type="OTHER_IF", Value=LOST.

   4.  Then a HELLO message is generated using the previous method, with
       the specified headers and TLVs:

       *  a message TLV with Type="VALIDITY_TIME" and Value=encoding of
          L_HOLD_TIME, SHALL be added

       *  a message TLV with Type="INTERVAL_TIME" and Value=encoding of
          HELLO_INTERVAL, SHOULD be added

       *  a message TLV with Type="WILLINGNESS" and Value=the
          willingness of the node.  This SHOULD NOT be included if this
          value is WILL_DEFAULT, it SHALL be included otherwise.

Appendix B.2  Example Algorithm for Generating TC messages

   Periodically, the node generates TC messages, broadcast on all the
   interfaces of the node, as follows:

   1.  Each A_iface_addr in the Advertised Neighbor Set, SHALL be
       included in the TC message.

   2.  The TC message is generated with the proper headers, and (except
       where the Advertised Neighbor Set is empty and the TC message is
       not specifically reporting this, see Section 9) including the
       message TLV, Type="CONTENT_SEQUENCE_NUMBER", Value=the current
       ASSN of the node.

Appendix C.  Protocol and Port Number

   Packets in OLSRv2 are communicated using UDP.  Port 698 has been
   assigned by IANA for exclusive usage by the OLSR (v1 and v2)
   protocol.

Appendix D.  Packet and Message Layout

   This section specifies the translation from the abstract descriptions
   of packets employed in the protocol specification, and the bit-layout
   packets actually exchanged between the nodes.

Appendix D.1  OLSRv2 Packet Format

   The basic layout of an OLSRv2 packet is as described in [4].  However
   the following points should be noted.

   OLSRv2 uses only packets with a packet header.  Thus all OLSRv2
   packets have the following layout.

      0                   1                   2                   3

      0 1 2 3 4 5 6 7 8 9 0 1 2 3 4 5 6 7 8 9 0 1 2 3 4 5 6 7 8 9 0 1

     +-+-+-+-+-+-+-+-+-+-+-+-+-+-+-+-+-+-+-+-+-+-+-+-+-+-+-+-+-+-+-+-+

     |0 0 0 0 0 0 0 0|   Reserved    |    Packet Sequence Number     |

     +-+-+-+-+-+-+-+-+-+-+-+-+-+-+-+-+-+-+-+-+-+-+-+-+-+-+-+-+-+-+-+-+

     |                                                               |

     |                            Message                            |

     |                                                               |

     +-+-+-+-+-+-+-+-+-+-+-+-+-+-+-+-+-+-+-+-+-+-+-+-+-+-+-+-+-+-+-+-+

     |                                                               |

     :                              ...                              :

     |                                                               |

     +-+-+-+-+-+-+-+-+-+-+-+-+-+-+-+-+-+-+-+-+-+-+-+-+-+-+-+-+-+-+-+-+

     |                                                               |

     |                            Message                            |

     |                                                               |
     +-+-+-+-+-+-+-+-+-+-+-+-+-+-+-+-+-+-+-+-+-+-+-+-+-+-+-+-+-+-+-+-+

   All reserved bits are also unset (zero).

   OLSRv2 uses only packets with a complete message header.  Thus all
   OLSRv2 messages have the following layout.

      0                   1                   2                   3

      0 1 2 3 4 5 6 7 8 9 0 1 2 3 4 5 6 7 8 9 0 1 2 3 4 5 6 7 8 9 0 1

     +-+-+-+-+-+-+-+-+-+-+-+-+-+-+-+-+-+-+-+-+-+-+-+-+-+-+-+-+-+-+-+-+

     |  Message Type | Resv  |U|N|0|0|         Message Size          |

     +-+-+-+-+-+-+-+-+-+-+-+-+-+-+-+-+-+-+-+-+-+-+-+-+-+-+-+-+-+-+-+-+

     |                      Originator Address                       |

     +-+-+-+-+-+-+-+-+-+-+-+-+-+-+-+-+-+-+-+-+-+-+-+-+-+-+-+-+-+-+-+-+

     |  Time To Live |   Hop Count   |    Message Sequence Number    |

     +-+-+-+-+-+-+-+-+-+-+-+-+-+-+-+-+-+-+-+-+-+-+-+-+-+-+-+-+-+-+-+-+

     |                                                               |

     |                    Message Body + Padding                     |

     |                                                               |

     +-+-+-+-+-+-+-+-+-+-+-+-+-+-+-+-+-+-+-+-+-+-+-+-+-+-+-+-+-+-+-+-+

   In standard OLSRv2 messages (HELLO and TC) the U and N bits are also
   unset(zero).  In all OLSRv2 messages the reserved bits marked Resv
   above are also unset (zero).

   The layouts of the message body, address block, TLV block and TLV are
   as in [4], allowing all options.  Standard (HELLO and TC) messages
   contain a first address block which contains local interface
   addresses, all other address blocks contain information specific to
   the message type.  Except by being first, the local interface address
   block is not distinguished in any way.

   An example HELLO message, using IPv4 (four octet) addresses is as
   follows.  The overall message length is 56 octets (it does not need
   padding).  The message has a TTL of 1 and a hop count of 0, as sent
   by its originator.

   The message has a message TLV block with content length 12 octets
   containing three message TLVs.  These TLVs represent message validity
   time, message interval time and willingness.  Each uses a TLV with
   semantics value 4, indicating no start and stop indexes are included,
   and each has a value length of 1 octet.

   The first address block contains a single local interface address,
   with head length 4; thus although 1 tail is indicated, no tail octets
   are included.  This address block has no TLVs (TLV block content
   length 0 octets).

   The second, and last, address block reports 4 neighbour interface
   addresses, with address head length 3 octets.  The following TLV
   block (content length 11 octets) includes two TLVs.

   The first of these TLVs reports the link status of all four
   neighbours in a single multivalue TLV, the first two addresses are
   HEARD, the last two addresses are SYMMETRIC.  The TLV semantics value
   of 12 indicates, in addition to that this is a multivalue TLV, that
   no start index and stop index are included, since values for all
   addresses are included.  The TLV value length of 4 octets indicates
   one octet per value per address.

   The second of these TLV indicates that the last address (start index
   3, stop index 3) is an MPR.  This TLV has no value, or value length,
   fields, as indicated by its semantics octet being indicated by its semantics octet being equal to 1.

      0                   1                   2                   3
      0 1 2 3 4 5 6 7 8 9 0 1 2 3 4 5 6 7 8 9 0 1 2 3 4 5 6 7 8 9 0 1
     +-+-+-+-+-+-+-+-+-+-+-+-+-+-+-+-+-+-+-+-+-+-+-+-+-+-+-+-+-+-+-+-+
     |     HELLO     |0 0 0 0 0 0 0 0|0 0 0 0 0 0 0 0 0 0 1 1 1 0 0 0|
     +-+-+-+-+-+-+-+-+-+-+-+-+-+-+-+-+-+-+-+-+-+-+-+-+-+-+-+-+-+-+-+-+
     |                      Originator Address                       |
     +-+-+-+-+-+-+-+-+-+-+-+-+-+-+-+-+-+-+-+-+-+-+-+-+-+-+-+-+-+-+-+-+
     |0 0 0 0 0 0 0 1|0 0 0 0 0 0 0 0|    Message Sequence Number    |
     +-+-+-+-+-+-+-+-+-+-+-+-+-+-+-+-+-+-+-+-+-+-+-+-+-+-+-+-+-+-+-+-+
     |0 0 0 0 0 0 0 0 0 0 0 0 1 1 0 0| VALIDITY-TIME VALIDITY_TIME |0 0 0 0 0 1 0 0|
     +-+-+-+-+-+-+-+-+-+-+-+-+-+-+-+-+-+-+-+-+-+-+-+-+-+-+-+-+-+-+-+-+
     |0 0 0 0 0 0 0 1|     Value     | INTERVAL-TIME INTERVAL_TIME |0 0 0 0 0 1 0 0|
     +-+-+-+-+-+-+-+-+-+-+-+-+-+-+-+-+-+-+-+-+-+-+-+-+-+-+-+-+-+-+-+-+
     |0 0 0 0 0 0 0 1|     Value     |  WILLINGNESS  |0 0 0 0 0 1 0 0|
     +-+-+-+-+-+-+-+-+-+-+-+-+-+-+-+-+-+-+-+-+-+-+-+-+-+-+-+-+-+-+-+-+
     |0 0 0 0 0 0 0 1|     Value     |0 0 0 0 0 0 0 1|0 0 0 0 0 1 0 0|     Head      |
     +-+-+-+-+-+-+-+-+-+-+-+-+-+-+-+-+-+-+-+-+-+-+-+-+-+-+-+-+-+-+-+-+
     |                             Head (cont)                              |
     +-+-+-+-+-+-+-+-+-+-+-+-+-+-+-+-+-+-+-+-+-+-+-+-+-+-+-+-+-+-+-+-+
     |0 0 0 0 0 0 0 1|

     +-+-+-+-+-+-+-+-+-+-+-+-+-+-+-+-+-+-+-+-+-+-+-+-+-+-+-+-+-+-+-+-+

     |0 0 0 0 0 0 0 0 0 0 0|0 0 0 0 0 1 0 0|0 0 0 0 0 0 1 1|     Head      |
     +-+-+-+-+-+-+-+-+-+-+-+-+-+-+-+-+-+-+-+-+-+-+-+-+-+-+-+-+-+-+-+-+
     |                     Head (cont)          |0 0 0 0 0 1 0 0|     Tail                      |      Mid      |
     +-+-+-+-+-+-+-+-+-+-+-+-+-+-+-+-+-+-+-+-+-+-+-+-+-+-+-+-+-+-+-+-+
     |     Tail      Mid      |     Tail      Mid      |     Tail      Mid      |0 0 0 0 0 0 0 0|
     +-+-+-+-+-+-+-+-+-+-+-+-+-+-+-+-+-+-+-+-+-+-+-+-+-+-+-+-+-+-+-+-+
     |0 0 0 0 1 0 1 1|  LINK-STATUS  LINK_STATUS  |0 0 0 0 1 1 0 0|0 0 0 0 0 1 0 0|
     +-+-+-+-+-+-+-+-+-+-+-+-+-+-+-+-+-+-+-+-+-+-+-+-+-+-+-+-+-+-+-+-+
     |     HEARD     |     HEARD     |   SYMMETRIC   |   SYMMETRIC   |
     +-+-+-+-+-+-+-+-+-+-+-+-+-+-+-+-+-+-+-+-+-+-+-+-+-+-+-+-+-+-+-+-+
     |      MPR      |0 0 0 0 0 0 0 1|0 0 0 0 0 0 1 1|0 0 0 0 0 0 1 1|
     +-+-+-+-+-+-+-+-+-+-+-+-+-+-+-+-+-+-+-+-+-+-+-+-+-+-+-+-+-+-+-+-+

   An example TC message, using IPv4 (four octet) addresses, is as
   follows.  The overall message length is 67 64 octets, the final octet is it also does not
   need padding.

   The message has a message TLV block with content length 13 octets
   containing three TLVs.  The first TLV is a content sequence number
   TLV used to carry the 2 octet ANSN.  The semantics value is 4
   indicating that no index fields are included.  The other two TLVs are
   validity and interval times as for the HELLO message above.

   The message has three address blocks.  The first address block
   contains 3 local interface addresses (with common head length 2
   octets) and has a TLV block with content length 4 octets containing a
   single TLV with semantics value 1, indicating that the TLV has no
   value field, or length thereof.  This TLV indicates that the second
   and third of these addresses (indexes 1 to 2) are for other
   interfaces than the one on which this TC message is transmitted. 0 octets.

   The other two address blocks contain neighbour neighbor interface addresses,
   with head lengths 2 and 4 respectively. addresses.

   The first of these, with contains 3
   addresses, addresses and has an empty TLV block (content
   length 0 octets).  The
   second, which second contains 1 address, has address.  The head octet
   (hex 82) indicates a head length of two octets and the presence of a
   tail octet.  The tail octet (hex 82) indicates a tail length of two
   octets, all zero bits and not included.  The following TLV block
   (content length 4 6 octets) with a single TLV includes two TLVs, the first (semantics
   value 4 indicating no indexes are needed) indicating indicates that this is a network the address
   has a netmask, with the given
   prefix length (itself with given by the value (of length 1 octet). octet) of
   16.  Thus this address is Head.0.0/16.  The second TLV indicates that
   the originating node is a gateway to this network, the TLV semantics
   value of 5 indicates that neither indexes nor value are needed.

      0                   1                   2                   3
      0 1 2 3 4 5 6 7 8 9 0 1 2 3 4 5 6 7 8 9 0 1 2 3 4 5 6 7 8 9 0 1
     +-+-+-+-+-+-+-+-+-+-+-+-+-+-+-+-+-+-+-+-+-+-+-+-+-+-+-+-+-+-+-+-+
     |      TC       |0 0 0 0 0 0 0 0|0 0 0 0 0 0 0 0 0 1 0 0 0 0 1 1| 0 0|
     +-+-+-+-+-+-+-+-+-+-+-+-+-+-+-+-+-+-+-+-+-+-+-+-+-+-+-+-+-+-+-+-+
     |                      Originator Address                       |
     +-+-+-+-+-+-+-+-+-+-+-+-+-+-+-+-+-+-+-+-+-+-+-+-+-+-+-+-+-+-+-+-+
     | Time to Live   Hop Limit   |   Hop Count   |    Message Sequence Number    |
     +-+-+-+-+-+-+-+-+-+-+-+-+-+-+-+-+-+-+-+-+-+-+-+-+-+-+-+-+-+-+-+-+
     |0 0 0 0 0 0 0 0 0 0 0 0 1 1 0 1| CONT_SEQ_NUM  |0 0 0 0 0 1 0 0|
     +-+-+-+-+-+-+-+-+-+-+-+-+-+-+-+-+-+-+-+-+-+-+-+-+-+-+-+-+-+-+-+-+
     |0 0 0 0 0 0 1 0|         Value (ASSN) (ANSN)          | VALIDITY_TIME |
     +-+-+-+-+-+-+-+-+-+-+-+-+-+-+-+-+-+-+-+-+-+-+-+-+-+-+-+-+-+-+-+-+
     |0 0 0 0 0 1 0 0|0 0 0 0 0 0 0 1|     Value     | INTERVAL_TIME |
     +-+-+-+-+-+-+-+-+-+-+-+-+-+-+-+-+-+-+-+-+-+-+-+-+-+-+-+-+-+-+-+-+
     |0 0 0 0 0 1 0 0|0 0 0 0 0 0 0 1|     Value     |0 0 0 0 0 0 1 0| 1|
     +-+-+-+-+-+-+-+-+-+-+-+-+-+-+-+-+-+-+-+-+-+-+-+-+-+-+-+-+-+-+-+-+

     |             Head
     |0 0 0 0 0 0 1 1|     Tail 0|             Head              |      Mid      |
     +-+-+-+-+-+-+-+-+-+-+-+-+-+-+-+-+-+-+-+-+-+-+-+-+-+-+-+-+-+-+-+-+
     |  Tail  Mid (cont)   |             Tail              Mid              |     Tail      Mid      |
     +-+-+-+-+-+-+-+-+-+-+-+-+-+-+-+-+-+-+-+-+-+-+-+-+-+-+-+-+-+-+-+-+
     |  Tail  Mid (cont)   |0 0 0 0 0 0 0 0 0 0 0 0 0 1 0 0|   OTHER_IF    |

     +-+-+-+-+-+-+-+-+-+-+-+-+-+-+-+-+-+-+-+-+-+-+-+-+-+-+-+-+-+-+-+-+

     |0 0 0 0 0 0 0 1|0 0 0 0 0 0 0 1|0 0 0 0 0 0 1 0|0 0 0 0 0 0 1 0| 1|
     +-+-+-+-+-+-+-+-+-+-+-+-+-+-+-+-+-+-+-+-+-+-+-+-+-+-+-+-+-+-+-+-+

     |             Head
     |0 0 0 0 0 0 1 1|     Tail 0|             Head              |      Mid      |
     +-+-+-+-+-+-+-+-+-+-+-+-+-+-+-+-+-+-+-+-+-+-+-+-+-+-+-+-+-+-+-+-+
     |  Tail  Mid (cont)   |             Tail              Mid              |     Tail      Mid      |
     +-+-+-+-+-+-+-+-+-+-+-+-+-+-+-+-+-+-+-+-+-+-+-+-+-+-+-+-+-+-+-+-+
     |  Tail  Mid (cont)   |0 0 0 0 0 0 0 0 0 0 0 0 0 0 0 0|0 0 0 0 0 1 0 0| 0 1|
     +-+-+-+-+-+-+-+-+-+-+-+-+-+-+-+-+-+-+-+-+-+-+-+-+-+-+-+-+-+-+-+-+

     |
     |1 0 0 0 0 0 1 0|             Head                              |

     +-+-+-+-+-+-+-+-+-+-+-+-+-+-+-+-+-+-+-+-+-+-+-+-+-+-+-+-+-+-+-+-+

     |0              |1 0 0 0 0 0 1 0|
     +-+-+-+-+-+-+-+-+-+-+-+-+-+-+-+-+-+-+-+-+-+-+-+-+-+-+-+-+-+-+-+-+
     |0 0 0 0 1|0 0 0 0 0 0 0 0 0 0 1 1 0| PREFIX_LENGTH |0 0 0 0 0 1 0 0| PREFIX-LENGTH |
     +-+-+-+-+-+-+-+-+-+-+-+-+-+-+-+-+-+-+-+-+-+-+-+-+-+-+-+-+-+-+-+-+
     |0 0 0 0 0 1 0 0|0 0 1|0 0 0 1 0 0 0 1|Value (Length) 0|    GATEWAY    |0 0 0 0 0 1 0 0 0| 1|
     +-+-+-+-+-+-+-+-+-+-+-+-+-+-+-+-+-+-+-+-+-+-+-+-+-+-+-+-+-+-+-+-+

Appendix E.  Node Configuration

   OLSRv2 does not make make any assumption about node addresses, other than
   that each node is assumed to have at least one a unique and routable
   IP address for each interface that it has which participates in the
   MANET.

   When applicable, a recommended way of connecting an OLSRv2 network to
   an existing IP routing domain is to assign an IP prefix (under the
   authority of the nodes/gateways connecting the MANET with the routing
   domain) exclusively to the OLSRv2 area, and to configure the gateways
   statically to advertise routes to that IP sequence to nodes in the
   existing routing domain.

Appendix F.  Jitter

   In a wireless network, simultaneous packet transmission by nearby
   nodes is undesirable as, depending on the medium access control and
   other lower layer mechanisms, the interference between these messages
   may cause at best increased delay, and at worst complete packet loss
   by both nodes.  This is often particularly true when using a
   broadcast mechanism, such as is used by OLSRv2 packets.

   The problems of simultaneous packet transmission in OLSRv2 are
   increased by the following features of the protocol:

   o  If two nodes send packets containing regularly scheduled messages
      of the same type at the same time, then if, as is typical, they
      are using the same message interval, further transmissions of
      these messages will also be at the same time, and will also
      interfere.  This node synchronization could even result in
      complete operational failure of these nodes.

   o  OLSRv2 allows nodes to respond to changes in their circumstances,
      usually changes in the neighborhood, with immediate messages of
      appropriate types.  Nearby nodes will have overlapping
      neighborhoods, and may respond to the same change in
      circumstances.  For example a single link failure can result in a
      node having to change its MPR Set, and then two or more of its
      neighbors having changed MPR status responding simultaneously with
      revised TC messages, whose packets may interfere.

   o  When a node sends such a responsive message, there is no apparent
      reason why it should not restart its message schedule of the
      appropriate type of message.  This results in nodes responding to
      the same change not just sending single simultaneous packets, but
      becoming synchronized.

   o  Nodes also forward messages they receive from other nodes.  Two
      nearby nodes will thus commonly receive and forward the same
      message.  The consequent packet transmissions can easily interfere
      with each other.

   Because interference can easily occur, is self-reinforcing, and is
   anything from undesirable to catastrophic, mechanisms to minimize it,
   and to break synchronization of nodes, SHOULD be used in OLSRv2.
   These all make a deliberate adjustment to the timing, known as
   "jitter".  Three cases exist:

   o  When a node generates a control message periodically, it would
      normally wait for a delay equal to MESSAGE_INTERVAL (e.g.
      HELLO_INTERVAL for HELLO messages or TC_INTERVAL for TC messages)
      between two transmissions of messages of that type.  This delay
      SHOULD be mitigated by subtracting a jitter time, so that the
      delay between consecutive transmissions of a messages of the same
      type SHOULD be equal to MESSAGE_INTERVAL - jitter, where jitter is
      a random value whose generation is discussed below.  Note that
      this is a deliberately asymmetric process.  It ensures that the
      message interval does not exceed MESSAGE_INTERVAL (which leaves
      MESSAGE_INTERVAL an appropriate value for reporting in an
      INTERVAL_TIME message TLV) and also allows different nodes to
      become completely desynchronized as each interval is based on the
      previous actual transmission time, not on a fixed clock of period
      MESSAGE_INTERVAL.

   o  When a node responds to an externally triggered change in
      circumstances, it SHOULD delay the transmission of a message in
      response by a random jitter time.  It MAY restart its schedule of
      messages of the appropriate type based on that new time.  If such
      a message is delayed due to the need to respect the appropriate
      MESSAGE_MIN_INTERVAL (e.g.  HELLO_MIN_INTERVAL for HELLO messages
      or TC_MIN_INTERVAL for TC messages) then the node MAY reduce this
      minimum interval by a jitter time as the normal message interval
      is reduced (thus allowing MESSAGE_MIN_INTERVAL to equal
      MESSAGE_INTERVAL even when using jitter).

   o  When a node forwards a message, it SHOULD delay the message
      retransmission by a random jitter time.

   In the first and second cases above, the maximum jitter time may be
   specified by a parameter MAXJITTER.  It is necessary only that this
   be significantly less than each MESSAGE_INTERVAL, and less than each
   MESSAGE_MIN_INTERVAL.  Normally the actual value of the jitter
   (reduction in message interval or delay of responsive message) SHOULD
   be uniformly generated in the interval 0 <= jitter <= MAXJITTER,
   however this may be modified as indicated below.

   In the third case above, a message SHOULD be delayed by a jitter
   value which is significantly less than the originating node's message
   interval.  This MAY be available in an INTERVAL_TIME message TLV in
   the message to be forwarded.  If not so available, a node MAY
   estimate an acceptable maximum jitter by any assumption about other means available to
   it, which may be by use of its own MAXJITTER parameter for as long as
   this works.  In a network in which this is likely to be unsuccessful,
   nodes SHOULD include an INTERVAL_TIME message TLV in messages which
   are to be forwarded.

   In all cases, as well as constraints imposed by message intervals and
   message minimum intervals, the maximum jitter delay SHOULD only be as
   large as is required to achieve the required objective of minimizing
   interference due to synchronization.  This is because all jitter, and
   forwarding jitter in particular, is undesirable for otherwise ideal
   functioning of the network.

   Because of differing parameters, or due to responsive messages with a
   small minimum message interval, a node addresses, other than
   that each may receive a message from an
   originating node is assumed while still waiting to forward an earlier message of
   the same type originating from the same node.  The forwarding node
   SHOULD NOT allow forwarding jitter delay to reorder these messages.
   A node MAY discard the earlier message, transmitting the later
   message no later than the earlier message was due to be
   retransmitted, if, and only if, it can guarantee that this will not
   have at least one any adverse effect.

   OLSRv2 messages are transmitted in potentially multi-message packets.
   Whilst a unique packet is a hop by hop construct and routable
   IP address for each interface that it has is the messages in
   it which participates are forwarded, if a number of messages are received in the
   MANET.

   When applicable,
   same packet, they SHOULD (if their maximum jitter delays are
   compatible) be permitted to be forwarded in the same new packet.
   This may be accomplished by generating the same random delay for all
   messages received in a recommended way of connecting an OLSRv2 network single packet.  Furthermore, the opportunity
   to
   an existing IP routing domain is combine messages to assign an IP prefix (under be forwarded from different sources, and
   locally generated messages in a single packet SHOULD be allowed even
   when this means adjusting (forwards or backwards) the strictly
   uniformly generated random jitter times, however these SHOULD NOT be
   allowed to exceed their maximum value, nor allow a message interval
   to be exceeded, nor compromise the
   authority purpose of jitter.  (It is for
   this reason that messages in the nodes/gateways connecting the MANET with the routing
   domain) exclusively to same packet should be given the OLSRv2 area, and same
   random jitter, as giving them independent jitter values but then, for
   example, allowing all to configure be sent with the gateways
   statically to advertise routes to that IP sequence to nodes in earliest would reduce the
   existing routing domain.
   mean jitter delay.)

Appendix F. G.  Security Considerations

   Currently, OLSRv2 does not specify any special security measures.  As
   a proactive routing protocol, OLSRv2 makes a target for various
   attacks.  The various possible vulnerabilities are discussed in this
   section.

Appendix F.1 G.1.  Confidentiality

   Being a proactive protocol, OLSRv2 periodically diffuses topological
   information.  Hence, if used in an unprotected wireless network, the
   network topology is revealed to anyone who listens to OLSRv2 control
   messages.

   In situations where the confidentiality of the network topology is of
   importance, regular cryptographic techniques, such as exchange of
   OLSRv2 control traffic messages encrypted by PGP [3] [5] or encrypted by
   some shared secret key, can be applied to ensure that control traffic
   can be read and interpreted by only those authorized to do so.

Appendix F.2 G.2.  Integrity

   In OLSRv2, each node is injecting topological information into the
   network through transmitting HELLO messages and, for some nodes, TC
   messages.  If some nodes for some reason, malicious or malfunction,
   inject invalid control traffic, network integrity may be compromised.
   Therefore, message authentication is recommended.

   Different such situations may occur, for instance:

   1.  a node generates TC messages, advertising links to non-neighbor
       nodes;

   2.  a node generates TC messages, pretending to be another node;

   3.  a node generates HELLO messages, advertising non-neighbor nodes;

   4.  a node generates HELLO messages, pretending to be another node;

   5.  a node forwards altered control messages;

   6.  a node does not forward control messages;

   7.  a node does not select multipoint relays correctly;

   8.  a node forwards broadcast control messages unaltered, but does
       not forward unicast data traffic;
   9.  a node "replays" previously recorded control traffic from another
       node.

   Authentication of the originator node for control messages (for
   situations 2, 4 and 5) and on the individual links announced in the
   control messages (for situations 1 and 3) may be used as a
   countermeasure.  However to prevent nodes from repeating old (and
   correctly authenticated) information (situation 9) temporal
   information is required, allowing a node to positively identify such
   delayed messages.

   In general, digital signatures and other required security
   information may be transmitted as a separate OLSRv2 message type,
   thereby allowing that "secured" and "unsecured" nodes can coexist in
   the same network, if desired, or signatures and security information
   may be transmitted within the OLSRv2 HELLO and TC messages, using the
   TLV mechanism.

   Specifically, the authenticity of entire OLSRv2 control messages can
   be established through employing IPsec authentication headers,
   whereas authenticity of individual links (situations 1 and 3) require
   additional security information to be distributed.

   An important consideration is, that all control messages in OLSRv2
   are transmitted either to all nodes in the neighborhood (HELLO
   messages) or broadcast to all nodes in the network (TC messages).

   For example, a control message in OLSRv2 is always a point-to-
   multipoint transmission.  It is therefore important that the
   authentication mechanism employed permits that any receiving node can
   validate the authenticity of a message.  As an analogy, given a block
   of text, signed by a PGP private key, then anyone with the
   corresponding public key can verify the authenticity of the text.

Appendix F.3 G.3.  Interaction with External Routing Domains

   OLSRv2 does, through the use of TC messages, provide a basic
   mechanism for injecting external routing information to the OLSRv2
   domain.  Appendix E also specifies that routing information can be
   extracted from the topology table or the routing table of OLSRv2 and,
   potentially, injected into an external domain if the routing protocol
   governing that domain permits.

   Other than as described in Appendix E, when operating nodes,
   connecting OLSRv2 to an external routing domain, care MUST be taken
   not to allow potentially insecure and untrustworthy information to be
   injected from the OLSRv2 domain to external routing domains.  Care
   MUST be taken to validate the correctness of information prior to it
   being injected as to avoid polluting routing tables with invalid
   information.

   A recommended way of extending connectivity from an existing routing
   domain to an OLSRv2 routed MANET is to assign an IP prefix (under the
   authority of the nodes/gateways connecting the MANET with the exiting
   routing domain) exclusively to the OLSRv2 MANET area, and to
   configure the gateways statically to advertise routes to that IP
   sequence to nodes in the existing routing domain.

Appendix F.4 G.4.  Node Identity

   OLSRv2 does not make any assumption about node addresses, other than
   that each node is assumed to have at least one a unique and routable
   IP address for each interface that it has which participates in the
   MANET.

Appendix G. H.  Flow and Congestion Control

   TBD

Appendix H.  Sequence Numbers

   Sequence numbers are used in OLSR with the purpose of discarding
   "old" information, i.e., messages received out of order.  However
   with a limited number of bits for representing sequence numbers,
   wrap-around (that the sequence number is incremented from the maximum
   possible value to zero) will occur.  To prevent this from interfering
   with the operation of OLSRv2, the following MUST be observed.

   The term MAXVALUE designates in the following the largest possible
   value for a sequence number.

   The sequence number S1 is said to be "greater than" the sequence
   number S2 if:

   o  S1 > S2 AND S1 - S2 <= MAXVALUE/2 OR

   o  S2 > S1 AND S2 - S1 > MAXVALUE/2

   Thus when comparing two messages, it is possible - even in the
   presence of wrap-around - to determine which message contains the
   most recent information.

Appendix I.  Contributors

   This specification is the result of the joint efforts of the
   following contributers contributors -- listed alphabetically.

   o  Cedric Adjih, INRIA, France, <Cedric.Adjih@inria.fr>

   o  Emmanuel Baccelli, Hitachi Labs Europe, France,
      <Emmanuel.Baccelli@inria.fr>

   o  Thomas Heide Clausen, PCRI, France<T.Clausen@computer.org>

   o  Justin Dean, NRL, USA<jdean@itd.nrl.navy.mil>

   o  Christopher Dearlove, BAE Systems, UK,
      <Chris.Dearlove@baesystems.com>

   o  Satoh Hiroki, Hitachi SDL, Japan, <h-satoh@sdl.hitachi.co.jp>

   o  Philippe Jacquet, INRIA, France, <Philippe.Jacquet@inria.fr>

   o  Monden Kazuya, Hitachi SDL, Japan, <monden@sdl.hitachi.co.jp>

   o  Kenichi Mase, Niigata University, Japan, <mase@ie.niigata-u.ac.jp>

   o  Ryuji Wakikawa, KEIO University, Japan, <ryuji@sfc.wide.ad.jp>

Appendix J.  Acknowledgements

   The authors would like to acknowledge the team behind OLSRv1,
   specified in RFC3626, including Anis Laouiti, Pascale Minet, Laurent
   Viennot (all at INRIA, France), and Amir Qayuum (Center for Advanced
   Research in Engineering) Engineering, Pakistan) for their contributions.

   The authors would like to gratefully acknowledge the following people
   for intense technical discussions, early reviews and comments on the
   specification and its components: Kenichi Mase (Niigata University), Li Li (CRC), Louise Lamont (CRC),
   Joe Macker (NRL), Alan Cullen (BAE Systems), Philippe Jacquet
   (INRIA), Khaldoun Al Agha (LRI), Richard Ogier (?), (SRI), Song-Yean Cho
   (Samsung Software Center), Shubhranshu Singh (Samsung AIT) and the
   entire IETF MANET working group.

Authors' Addresses

   Thomas Heide Clausen
   LIX, Ecole Polytechnique, France

   Phone: +33 6 6058 9349
   Email: T.Clausen@computer.org
   URI:   http://www.lix.polytechnique.fr/Labo/Thomas.Clausen/

   Christopher M. Dearlove
   BAE Systems Advanced Technology Centre

   Phone: +44 1245 242194
   Email: chris.dearlove@baesystems.com
   URI:   http://www.baesystems.com/ocs/sharedservices/atc/

   Philippe Jacquet
   Project Hipercom, INRIA

   Phone: +33 1 3963 5263
   Email: philippe.jacquet@inria.fr
   URI:   http://hipercom.inria.fr/test/Jacquet.htm

   The OLSRv2 Design Team
   MANET Working Group

Intellectual Property Statement

   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.

Disclaimer of Validity

   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 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.

Copyright Statement

   Copyright (C) The Internet Society (2006).  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.

Acknowledgment

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