Case 1:16-cv-02690-AT Document 121-17 Filed 08/05/16 Page 1 of 3001
`Case 1:16-cv—02690-AT Document 121-17 Filed 08/05/16 Page 1 of 3001
`
`E-7
`
`E-7
`
`

`

`The ‘314 Patent – Claims
`
`
`Leiner Reference
`
`Case 1:16-cv-02690-AT Document 121-17 Filed 08/05/16 Page 2 of 3001
`
`Exhibit B11 – Invalidity Chart for Brownrigg Family based on Leiner Reference
`“The basic job of the network management algorithms is to allow data packets to be
`least possible number of additional
`routed through the network in an efficient and reliable manner. This entails two basic
`second nodes, the path to the first
`node through the most robust
`tasks. The first is the establishment of routes through the network, and the second is
`the forwarding of packets along those routes.” Leiner at 12.
`additional second nodes, the path to
` “Because flooding techniques do not require a priori knowledge of the network
`the first node through the second
`nodes with the least amount of traffic,
`connectivity, they are easily used for disseminating network management and control
`and the path to the first node through
`information which is used to determine that connectivity.” Leiner at 13.
`the fastest second nodes.
` “Point‐to‐point routing methods typically involve the association of a route (a
`sequence of links) with a source‐destination pair. One method of doing point‐to‐point
`routing is to explicitly associate information in each node with a source‐ destination
`pair (connection). Typically such techniques involve a route establishment phase that
`occurs when the ‘connection’ is first recognized, and then the information stored at
`each node is used to perform the actual routing of the packets. Forwarding of packets
`then simply involves looking up the appropriate forwarding information based on the
`connection identifier (which is carried in the packet). If topology changes occur, a
`new route establishment (or re‐establishment) phase would occur to assure that the
`correct information is stored at all the nodes in the intended route.” Leiner at 13.
` “Thus we see that all three routing methods have a place in packet radio networks. In
`relatively static networks, it is often most efficient to have the nodes determine their
`connectivity, and then determine relatively fixed routes (which would then be
`modified if connectivity changed due to mobility, etc.). For more dynamic networks,
`where connectivity is constantly changing, higher channel efficiency can be achieved
`by reducing the connection setups and the associated overhead. Finally, in the most
`dynamic networks, where network delays preclude tracking of connectivity on
`any but the most local basis, flooding techniques would appear to be a reasonable
`26
`
`
`
`

`

`The ‘314 Patent – Claims
`
`
`Leiner Reference
`
`Case 1:16-cv-02690-AT Document 121-17 Filed 08/05/16 Page 3 of 3001
`
`Exhibit B11 – Invalidity Chart for Brownrigg Family based on Leiner Reference
`approach.” Leiner at 13.
`
` “HOW SHOULD THE INFORMATION THAT EACH NODE REQUIRES TO ROUTE
`PACKETS BE DISSEMINATED TO THOSE NODES? For any type of routing method
`(with the exception of the most simple flooding methods), the local connectivity
`information must be processed and made available to the nodes so that they may
`route the packets. Note that this is somewhat independent of the type of routing being
`used. However, it does depend on the method for determining link connectivity and in
`particular, where the resulting connectivity information resides.
` A popular method for doing routing in networks where functional distribution is not
`needed (e.g., for survivability) is to use a centralized routing server. (This, in fact, was
`the method used in the early DARPA packet radio network [3].) This technique has
`each node send its local connectivity information to a central location. At this location,
`routes are determined and the information required by each node to process and
`forward packets (such as the next node along the route) is sent to the individual
`network nodes on either a request basis or as a background operation which
`constantly updates tables in the nodes.” Leiner at 13.
` “Use of a centralized routing server has several advantages over more distributed
`techniques. Because the server has all the connectivity information available (albeit
`not necessarily current), it can be quite efficient in the computation of routes. This
`can be a significant advantage in packet radio situations where both connectivity and
`congestion are more visible globally and where some nodes are typically collocated
`with mobile users as opposed to being located in some predetermined location. The
`centralized techniques can generally be extended to a small number of servers for
`load‐sharing and/or backup, thus overcoming some of the problems of size and
`robustness inherent in a centralized method.” Leiner at 13.
`27
`
`
`
`

`

`The ‘314 Patent – Claims
`
`
`Leiner Reference
`
`Case 1:16-cv-02690-AT Document 121-17 Filed 08/05/16 Page 4 of 3001
`
`Exhibit B11 – Invalidity Chart for Brownrigg Family based on Leiner Reference
` “One method for distributing the routing process is to provide enough information to
`each node so that each node can simply compute for itself the best total route and
`then take action locally that is commensurate with that global optimum. For example,
`based on the computed best total route, a node may determine which is the best node
`to forward the packet. At the next node, the route may be recomputed or the entire
`route (or portion) could be included in the packet. (The latter is considerably less
`robust in the face of changing topology.) This form of distributed routing can be
`accomplished by having each node transmit its local connectivity information
`explicitly to every other node. Typically a form of flooding is used to disseminate the
`information.” Leiner at 13‐14.
` “This method is quite robust (except for errors in tables or transmissions) and, in fact,
`is the (new) algorithm used in the Arpanet [21] and is planned for use in the gateways
`of the DARPA Internet system [22], [23]. However, if the network has a relatively high
`rate of topology changes, the amount of traffic on the network could be very high, as
`every substantial topology change can produce a number of packets roughly equal to
`the number of nodes in the network times the number of nodes directly affected by
`the change. Thus this method of routing is well‐suited to a network like the Arpanet
`or a packet radio network consisting of fixed locations where topology changes are
`infrequent.” Leiner at 14.
` “Another interesting routing structure occurs when packet radio networks are
`hierarchically organized. If the network is assumed to consist of clusters of packet
`radios that are interconnected, the topology between clusters is likely to change at a
`slower rate than that between radios, and therefore hierarchical techniques may be
`applicable. We see this applied to packet radio in [24] and [6].” Leiner at 14.
`
`28
`
`
`
`

`

`The ‘314 Patent – Claims
`
`
`Leiner Reference
`
`Case 1:16-cv-02690-AT Document 121-17 Filed 08/05/16 Page 5 of 3001
`
`Exhibit B11 – Invalidity Chart for Brownrigg Family based on Leiner Reference
`
`
`10. A client node in a network
`“Fig. 1 shows a typical packet radio network structure [8]. A packet radio unit consists
`including a server node having a
`of a radio, antenna, and digital controller. The radio provides connectivity to a
`server radio modem and a server
`number of neighboring radios, but typically is not in direct connectivity with all
`controller which implements a server
`radios in the network. Thus the controller needs to provide for store‐and‐forward
`process that includes controlling the
`operation, relaying packets to accomplish connectivity between the originating and
`server node to receive and transmit
`destination users.” Leiner at 6.
`data packets via said server node to
` “Thus there are many design choices that must be made in the development of a
`other nodes in the network, the client
`node comprising:
`packet radio network. There is usually no single correct choice, and the decisions are
`a client node radio modem; and a
`dependent on the environment that the network must work in, the requirements for
`client node controller; said client node
`performance and other functionalities, and the cost and other limitations. In addition,
`controller implementing a process
`as new hardware and software technologies become available, the parameters
`including receiving and transmitting
`governing the decisions change and often result in different selections.” Leiner at 7.
`data packets via said client modem;
` “1) Gateways: Gateways can perform many functions but, as far as addressing is
`concerned, they are packet translation devices that interpret addresses at the internet
`level and impose headers (addresses) appropriate both to the local networks to
`which they are attached as well as other networks. They are host‐level devices and to
`work correctly must have some relationship with not only the other gateways of the
`internet but the network‐attached hosts themselves. Gateways may have an
`additional role in highly mobile networks such as packet radio where topological
`partitioning may occur dynamically. Under these circumstances, the gateways,
`normally internet devices, may take on a role of intranetwork addressing and routing.
`Specifically, the internet may become the trajectory over which an intranet packet
`gets delivered when a single network temporarily divides [22), Whenever gateways
`play important roles such as this in mobile packet radio networks, the following issue
`arises: SHOULD ADDRESSING AND ROUTING BE NETWORK‐ OR GATEWAY‐BASED?
`29
`
`
`
`

`

`The ‘314 Patent – Claims
`
`
`Leiner Reference
`
`Case 1:16-cv-02690-AT Document 121-17 Filed 08/05/16 Page 6 of 3001
`
`Exhibit B11 – Invalidity Chart for Brownrigg Family based on Leiner Reference
`Network‐based addressing means that each network has a unique name and address
`of which all relevant gateways are aware. In this case, all points within a single
`network share some portion of their address in common. In contrast, if gateway‐
`based addressing is used, then internet packets are routed from gateway to gateway
`and each gateway attached to a network must have some means to route packets to
`destinations within that network. Furthermore, in this case, hosts must have a means
`to bind themselves dynamically to at least one gateway. Gateway‐based routing, while
`somewhat less intuitive, provides a solution to the problem of what to do when a
`single network becomes partitioned.” Leiner at 17.
`
`“In the section above on data link control, the tradeoff between the various link
`parameters was discussed. In addition, there must be an interaction between network
`level routing algorithms (discussed below) and the control of the link parameters
`[20). If link connectivity is lost, the network must determine whether it should try
`harder on that link (by, for example, increasing power or coding gain) or it should
`attempt to find a different route, thereby possibly suffering some delay and lost
`packets while the new route is determined.” Leiner at 12.
` “One method for distributing the routing process is to provide enough information to
`each node so that each node can simply compute for itself the best total route and
`then take action locally that is commensurate with that global optimum. For example,
`based on the computed best total route, a node may determine which is the best node
`to forward the packet. At the next node, the route may be recomputed or the entire
`route (or portion) could be included in the packet. (The latter is considerably less
`robust in the face of changing topology.) This form of distributed routing can be
`accomplished by having each node transmit its local connectivity information
`explicitly to every other node. Typically a form of flooding is used to disseminate the
`information.” Leiner at 13‐14.
`30
`
`selecting a radio transmission path to
`said server node that is one of a direct
`link to said server node and an
`indirect link to said server node
`through at least one other client node;
`
`
`
`

`

`Leiner Reference
`
`
`
`Case 1:16-cv-02690-AT Document 121-17 Filed 08/05/16 Page 7 of 3001
`
`Exhibit B11 – Invalidity Chart for Brownrigg Family based on Leiner Reference
`“Radio connectivity must be determined by the two ends of the radio link (i.e., the two
`implementing a process requesting
`updated radio transmission path data
`packet radio units which are connected). The information from each node can be
`collected at a central location where connectivity is then determined, or it can be
`from said server node, and in
`determined by the nodes themselves through a cooperative mechanism, such as
`response thereto, implementing by
`exchange of the number of transmitted and received packets. In either case, a decision
`the server node changes to upgrade
`must be made as to the nature of the information that will be used to determine the
`the selected transmission path to an
`existence of a link.” Leiner at 11.
`optimized transmission path.
`
` “The basic job of the network management algorithms is to allow data packets to be
`routed through the network in an efficient and reliable manner. This entails two basic
`tasks. The first is the establishment of routes through the network, and the second is
`the forwarding of packets along those routes.” Leiner at 12.
` “Because flooding techniques do not require a priori knowledge of the network
`connectivity, they are easily used for disseminating network management and control
`information which is used to determine that connectivity.” Leiner at 13.
` “Point‐to‐point routing methods typically involve the association of a route (a
`sequence of links) with a source‐destination pair. One method of doing point‐to‐point
`routing is to explicitly associate information in each node with a source‐ destination
`pair (connection). Typically such techniques involve a route establishment phase that
`occurs when the ‘connection’ is first recognized, and then the information stored at
`each node is used to perform the actual routing of the packets. Forwarding of packets
`then simply involves looking up the appropriate forwarding information based on the
`connection identifier (which is carried in the packet). If topology changes occur, a
`new route establishment (or re‐establishment) phase would occur to assure that the
`correct information is stored at all the nodes in the intended route.” Leiner at 13.
`31
`
`The ‘314 Patent – Claims
`
`
`
`
`

`

`The ‘314 Patent – Claims
`
`
`Leiner Reference
`
`Case 1:16-cv-02690-AT Document 121-17 Filed 08/05/16 Page 8 of 3001
`
`Exhibit B11 – Invalidity Chart for Brownrigg Family based on Leiner Reference
` “Thus we see that all three routing methods have a place in packet radio networks. In
`relatively static networks, it is often most efficient to have the nodes determine their
`connectivity, and then determine relatively fixed routes (which would then be
`modified if connectivity changed due to mobility, etc.). For more dynamic networks,
`where connectivity is constantly changing, higher channel efficiency can be achieved
`by reducing the connection setups and the associated overhead. Finally, in the most
`dynamic networks, where network delays preclude tracking of connectivity on
`any but the most local basis, flooding techniques would appear to be a reasonable
`approach.” Leiner at 13.
`
` “HOW SHOULD THE INFORMATION THAT EACH NODE REQUIRES TO ROUTE
`PACKETS BE DISSEMINATED TO THOSE NODES? For any type of routing method
`(with the exception of the most simple flooding methods), the local connectivity
`information must be processed and made available to the nodes so that they may
`route the packets. Note that this is somewhat independent of the type of routing being
`used. However, it does depend on the method for determining link connectivity and in
`particular, where the resulting connectivity information resides.
` A popular method for doing routing in networks where functional distribution is not
`needed (e.g., for survivability) is to use a centralized routing server. (This, in fact, was
`the method used in the early DARPA packet radio network [3].) This technique has
`each node send its local connectivity information to a central location. At this location,
`routes are determined and the information required by each node to process and
`forward packets (such as the next node along the route) is sent to the individual
`network nodes on either a request basis or as a background operation which
`constantly updates tables in the nodes.” Leiner at 13.
`
`32
`
`
`
`

`

`The ‘314 Patent – Claims
`
`
`Leiner Reference
`
`Case 1:16-cv-02690-AT Document 121-17 Filed 08/05/16 Page 9 of 3001
`
`Exhibit B11 – Invalidity Chart for Brownrigg Family based on Leiner Reference
`“Use of a centralized routing server has several advantages over more distributed
`techniques. Because the server has all the connectivity information available (albeit
`not necessarily current), it can be quite efficient in the computation of routes. This
`can be a significant advantage in packet radio situations where both connectivity and
`congestion are more visible globally and where some nodes are typically collocated
`with mobile users as opposed to being located in some predetermined location. The
`centralized techniques can generally be extended to a small number of servers for
`load‐sharing and/or backup, thus overcoming some of the problems of size and
`robustness inherent in a centralized method.” Leiner at 13.
` “One method for distributing the routing process is to provide enough information to
`each node so that each node can simply compute for itself the best total route and
`then take action locally that is commensurate with that global optimum. For example,
`based on the computed best total route, a node may determine which is the best node
`to forward the packet. At the next node, the route may be recomputed or the entire
`route (or portion) could be included in the packet. (The latter is considerably less
`robust in the face of changing topology.) This form of distributed routing can be
`accomplished by having each node transmit its local connectivity information
`explicitly to every other node. Typically a form of flooding is used to disseminate the
`information.” Leiner at 13‐14.
` “This method is quite robust (except for errors in tables or transmissions) and, in fact,
`is the (new) algorithm used in the Arpanet [21] and is planned for use in the gateways
`of the DARPA Internet system [22], [23]. However, if the network has a relatively high
`rate of topology changes, the amount of traffic on the network could be very high, as
`every substantial topology change can produce a number of packets roughly equal to
`the number of nodes in the network times the number of nodes directly affected by
`the change. Thus this method of routing is well‐suited to a network like the Arpanet
`33
`
`
`
`

`

`The ‘314 Patent – Claims
`
`
`Leiner Reference
`
`Case 1:16-cv-02690-AT Document 121-17 Filed 08/05/16 Page 10 of 3001
`
`Exhibit B11 – Invalidity Chart for Brownrigg Family based on Leiner Reference
`or a packet radio network consisting of fixed locations where topology changes are
`infrequent.” Leiner at 14.
` “Another interesting routing structure occurs when packet radio networks are
`hierarchically organized. If the network is assumed to consist of clusters of packet
`radios that are interconnected, the topology between clusters is likely to change at a
`slower rate than that between radios, and therefore hierarchical techniques may be
`applicable. We see this applied to packet radio in [24] and [6].” Leiner at 14.
`
`
`“Fig. 1 shows a typical packet radio network structure [8]. A packet radio unit consists
`of a radio, antenna, and digital controller. The radio provides connectivity to a
`number of neighboring radios, but typically is not in direct connectivity with all
`radios in the network. Thus the controller needs to provide for store‐and‐forward
`operation, relaying packets to accomplish connectivity between the originating and
`destination users.” Leiner at 6.
` “Thus there are many design choices that must be made in the development of a
`packet radio network. There is usually no single correct choice, and the decisions are
`dependent on the environment that the network must work in, the requirements for
`performance and other functionalities, and the cost and other limitations. In addition,
`as new hardware and software technologies become available, the parameters
`governing the decisions change and often result in different selections.” Leiner at 7.
` “1) Gateways: Gateways can perform many functions but, as far as addressing is
`concerned, they are packet translation devices that interpret addresses at the internet
`level and impose headers (addresses) appropriate both to the local networks to
`which they are attached as well as other networks. They are host‐level devices and to
`34
`
`
`12. A first node providing a gateway
`between two networks, where at least
`one of the two networks is a wireless
`network, said first node comprising:
`a radio modem capable of
`communicating with a first network
`that operates in part, by wireless
`communication;
`a network interface to communicating
`with a second network;
`a digital controller coupled to said
`radio modem and to said network
`interface, said digital controller
`communicating with said first
`network via said radio modem and
`communicating with said second
`network via said network interface,
`said digital controller passing data
`
`
`

`

`The ‘314 Patent – Claims
`
`
`Leiner Reference
`
`Case 1:16-cv-02690-AT Document 121-17 Filed 08/05/16 Page 11 of 3001
`
`Exhibit B11 – Invalidity Chart for Brownrigg Family based on Leiner Reference
`work correctly must have some relationship with not only the other gateways of the
`packets received from said first
`internet but the network‐attached hosts themselves. Gateways may have an
`network that are destined for said
`second network to said second
`additional role in highly mobile networks such as packet radio where topological
`partitioning may occur dynamically. Under these circumstances, the gateways,
`network, and passing data packets
`normally internet devices, may take on a role of intranetwork addressing and routing.
`received from said second network
`Specifically, the internet may become the trajectory over which an intranet packet
`that are destined for said first
`gets delivered when a single network temporarily divides [22), Whenever gateways
`network to said first network,
`play important roles such as this in mobile packet radio networks, the following issue
`arises: SHOULD ADDRESSING AND ROUTING BE NETWORK‐ OR GATEWAY‐BASED?
`Network‐based addressing means that each network has a unique name and address
`of which all relevant gateways are aware. In this case, all points within a single
`network share some portion of their address in common. In contrast, if gateway‐
`based addressing is used, then internet packets are routed from gateway to gateway
`and each gateway attached to a network must have some means to route packets to
`destinations within that network. Furthermore, in this case, hosts must have a means
`to bind themselves dynamically to at least one gateway. Gateway‐based routing, while
`somewhat less intuitive, provides a solution to the problem of what to do when a
`single network becomes partitioned.” Leiner at 17.
`
`“Radio connectivity must be determined by the two ends of the radio link (i.e., the two
`packet radio units which are connected). The information from each node can be
`collected at a central location where connectivity is then determined, or it can be
`determined by the nodes themselves through a cooperative mechanism, such as
`exchange of the number of transmitted and received packets. In either case, a decision
`must be made as to the nature of the information that will be used to determine the
`existence of a link.” Leiner at 11.
`
`35
`
`said digital controller maintaining a
`map of data packet transmission
`paths to a plurality of second nodes of
`said first network, where a
`transmission path of a second node of
`said first network to said first node
`can be through one or more of other
`second node of said first network;
`
`
`

`

`The ‘314 Patent – Claims
`
`
`Leiner Reference
`
`Case 1:16-cv-02690-AT Document 121-17 Filed 08/05/16 Page 12 of 3001
`
`Exhibit B11 – Invalidity Chart for Brownrigg Family based on Leiner Reference
`
`wherein said digital controller
`“Radio connectivity must be determined by the two ends of the radio link (i.e., the two
`changes the transmission paths of
`packet radio units which are connected). The information from each node can be
`each of the second nodes to optimize
`collected at a central location where connectivity is then determined, or it can be
`the transmission paths including
`determined by the nodes themselves through a cooperative mechanism, such as
`changing each transmission path from
`exchange of the number of transmitted and received packets. In either case, a decision
`on of the plurality of said second
`must be made as to the nature of the information that will be used to determine the
`nodes to the first node so that the
`existence of a link.” Leiner at 11.
`path to the first node is chosen from
`
` “The basic job of the network management algorithms is to allow data packets to be
`the group consisting essentially of the
`path to the first node through the
`routed through the network in an efficient and reliable manner. This entails two basic
`least possible number of additional
`tasks. The first is the establishment of routes through the network, and the second is
`second nodes, the path to the first
`the forwarding of packets along those routes.” Leiner at 12.
`node through the most robust
` “Because flooding techniques do not require a priori knowledge of the network
`additional second nodes, the path to
`the first node through the second
`connectivity, they are easily used for disseminating network management and control
`nodes with the least amount of traffic,
`information which is used to determine that connectivity.” Leiner at 13.
`and the path to the first node through
` “Point‐to‐point routing methods typically involve the association of a route (a
`the fastest second nodes.
`sequence of links) with a source‐destination pair. One method of doing point‐to‐point
`routing is to explicitly associate information in each node with a source‐ destination
`pair (connection). Typically such techniques involve a route establishment phase that
`occurs when the ‘connection’ is first recognized, and then the information stored at
`each node is used to perform the actual routing of the packets. Forwarding of packets
`then simply involves looking up the appropriate forwarding information based on the
`connection identifier (which is carried in the packet). If topology changes occur, a
`new route establishment (or re‐establishment) phase would occur to assure that the
`correct information is stored at all the nodes in the intended route.” Leiner at 13.
`36
`
`
`
`

`

`The ‘314 Patent – Claims
`
`
`Leiner Reference
`
`Case 1:16-cv-02690-AT Document 121-17 Filed 08/05/16 Page 13 of 3001
`
`Exhibit B11 – Invalidity Chart for Brownrigg Family based on Leiner Reference
` “Thus we see that all three routing methods have a place in packet radio networks. In
`relatively static networks, it is often most efficient to have the nodes determine their
`connectivity, and then determine relatively fixed routes (which would then be
`modified if connectivity changed due to mobility, etc.). For more dynamic networks,
`where connectivity is constantly changing, higher channel efficiency can be achieved
`by reducing the connection setups and the associated overhead. Finally, in the most
`dynamic networks, where network delays preclude tracking of connectivity on
`any but the most local basis, flooding techniques would appear to be a reasonable
`approach.” Leiner at 13.
`
` “HOW SHOULD THE INFORMATION THAT EACH NODE REQUIRES TO ROUTE
`PACKETS BE DISSEMINATED TO THOSE NODES? For any type of routing method
`(with the exception of the most simple flooding methods), the local connectivity
`information must be processed and made available to the nodes so that they may
`route the packets. Note that this is somewhat independent of the type of routing being
`used. However, it does depend on the method for determining link connectivity and in
`particular, where the resulting connectivity information resides.
` A popular method for doing routing in networks where functional distribution is not
`needed (e.g., for survivability) is to use a centralized routing server. (This, in fact, was
`the method used in the early DARPA packet radio network [3].) This technique has
`each node send its local connectivity information to a central location. At this location,
`routes are determined and the information required by each node to process and
`forward packets (such as the next node along the route) is sent to the individual
`network nodes on either a request basis or as a background operation which
`constantly updates tables in the nodes.” Leiner at 13.
`
`37
`
`
`
`

`

`The ‘314 Patent – Claims
`
`
`Leiner Reference
`
`Case 1:16-cv-02690-AT Document 121-17 Filed 08/05/16 Page 14 of 3001
`
`Exhibit B11 – Invalidity Chart for Brownrigg Family based on Leiner Reference
`“Use of a centralized routing server has several advantages over more distributed
`techniques. Because the server has all the connectivity information available (albeit
`not necessarily current), it can be quite efficient in the computation of routes. This
`can be a significant advantage in packet radio situations where both connectivity and
`congestion are more visible globally and where some nodes are typically collocated
`with mobile users as opposed to being located in some predetermined location. The
`centralized techniques can generally be extended to a small number of servers for
`load‐sharing and/or backup, thus overcoming some of the problems of size and
`robustness inherent in a centralized method.” Leiner at 13.
` “One method for distributing the routing process is to provide enough information to
`each node so that each node can simply compute for itself the best total route and
`then take action locally that is commensurate with that global optimum. For example,
`based on the computed best total route, a node may determine which is the best node
`to forward the packet. At the next node, the route may be recomputed or the entire
`route (or portion) could be included in the packet. (The latter is considerably less
`robust in the face of changing topology.) This form of distributed routing can be
`accomplished by having each node transmit its local connectivity information
`explicitly to every other node. Typically a form of flooding is used to disseminate the
`information.” Leiner at 13‐14.
` “This method is quite robust (except for errors in tables or transmissions) and, in fact,
`is the (new) algorithm used in the Arpanet [21] and is planned for use in the gateways
`of the DARPA Internet system [22], [23]. However, if the network has a relatively high
`rate of topology changes, the amount of traffic on the network could be very high, as
`every substantial topology change can produce a number of packets roughly equal to
`the number of nodes in the network times the number of nodes directly affected by
`the change. Thus this method of routing is well‐suited to a network like the Arpanet
`38
`
`
`
`

`

`The ‘314 Patent – Claims
`
`
`Leiner Reference
`
`Case 1:16-cv-02690-AT Document 121-17 Filed 08/05/16 Page 15 of 3001
`
`Exhibit B11 – Invalidity Chart for Brownrigg Family based on Leiner Reference
`or a packet radio network consisting of fixed locations where topology changes are
`infrequent.” Leiner at 14.
` “Another interesting routing structure occurs when packet radio networks are
`hierarchically organized. If the network is assumed to consist of clusters of packet
`radios that are interconnected, the topology between clusters is likely to change at a
`slower rate than that between radios, and therefore hierarchical techniques may be
`applicable. We see this applied to packet radio in [24] and [6].” Leiner at 14.
`
`
`“1) Gateways: Gateways can perform many functions but, as far as addressing is
`concerned, they are packet translation devices that interpret addresses at the internet
`level and impose h