JP5884658B2 - COMMUNICATION METHOD, COMMUNICATION DEVICE, AND COMMUNICATION PROGRAM - Google Patents

Description

  The present invention relates to communication performed between a plurality of communication devices included in a network.

  As the number of communication devices included in the network increases, the number of paths stored in the router also increases. Therefore, in recent years, an ID-Locator separation technique has attracted attention as a technique for preventing an increase in route information held by a router. As one of ID-Locator separation technologies, LISP (Locator Identifier Separation Protocol) is known.

  FIG. 1 shows an example of a network to which LISP is applied. A network in which LISP is used includes an access network (A1, A2), a core network 3, and a location management server 10, and the core network and the access network are connected by a router. Hereinafter, a router belonging to both the core network and the access network is referred to as an “edge node”. In the example of FIG. 1, the access network A1 and the core network 3 are connected by an edge node 2a, and the access network A2 and the core network 3 are connected by an edge node 2b. The location management server 10 acquires ID-LOC correspondence information from each edge node and holds it in the location information management table 15. Here, the ID-LOC correspondence information is a combination of the address (host ID) assigned to the communication device in the access network including the edge node and the address of the edge node. For example, when the host ID of the communication device 1a is IP1, the host ID of the communication device 1b is IP2, the address of the edge node 2a is LOC1, and the address of the edge node 2b is LOC2, the location information management table 15 shown in FIG. Is done. Communication between communication devices belonging to different access networks is performed using ID-LOC correspondence information.

  For example, a case where the communication device 1a transmits the packet 6 addressed to the communication device 1b will be described. The edge node 2a requests the location management server 10 for ID-LOC correspondence information including the host ID (IP2) of the destination communication device 1b (LOC request message). The location management server 10 refers to the location information management table 15 to identify that the address associated with IP2 is LOC2. Further, the location management server 10 uses the LOC response message to notify the communication device 1a that the address assigned to the edge node of the access network to which IP2 belongs is LOC2. The communication device 1a stores the notified ID-LOC correspondence information in the LOC cache table 5. The communication device 1a generates an encapsulated packet 7 by encapsulating the packet 6 using a header (outer header) whose destination address is the address of the edge node notified from the location management server. The encapsulated packet 7 is routed to the edge node 2b by the outer header. The edge node 2b decapsulates the encapsulated packet 7 and transfers the packet 6 to the communication device 1b. Thereafter, when receiving the user IP packet addressed to the communication device 1b from the communication device 1a, the edge node 2a transfers the IP packet using the information stored in the LOC cache table 5.

  Incidentally, in a network to which LISP is applied, a communication device may move between a plurality of access networks. When the destination communication device moves to another access network, the packet is not transferred to the destination communication device unless the LOC cache table 5 is updated in the edge node accessed by the transmission source communication device. Therefore, a method has been proposed in which each edge node acquires and stores address information from a received packet in order to change the packet transfer path as the communication apparatus moves. Here, the address information includes the source host ID of the received packet, the address of the edge node accessed by the source communication device, the destination host ID, and the address of the edge node accessed by the destination communication device. In this method, when the combination of the source host ID, the destination host ID, and the address of the edge node accessed by the destination communication device satisfies a predetermined condition, the destination communication is transmitted to the edge node accessed by the source communication device. ID-LOC correspondence information about the device is notified.

  Furthermore, a communication system is also devised that includes a correspondence table that associates a packet source host ID, a destination host ID, and an edge node accessed by a source communication device. In this system, when the edge node accessed by the transmission destination communication device is changed, the mapping information notification device refers to the correspondence table and notifies the edge node accessed by the transmission source communication device of the mapping change. To do.

  As a related technique, there has been devised a method for notifying a compatible node identifier to a server in charge of a mobile node when each mobile node moves. In this method, when the destination address of the packet is designated by the compatible node identifier, the mobile node accesses the server in charge of the destination node and acquires the destination address. An error recovery method is also known in which transfer error information associated with context information relating to a first call session between a first calling terminal and a call system resource is generated. In this method, when transfer of a call relating to the first calling terminal fails, transfer error information is transmitted to the first calling terminal.

JP 2011-78030 A JP 2011-55236 A JP 2001-230803 A JP 2011-511599 A

Locator / ID Separation Protocol (LISP) --- draft-ietf-lisp-ms-16.txt (March 2012) cisco, Locator / ID SeparationProtocol (LISP) VM Mobility Solution white paper, 2011

  In the method in which each edge node uses the address information acquired from the received packet to notify the edge node accessible to the packet transmission source communication device of the movement of the communication device, the stored address information is old information. There may be. Then, since the edge node holding the address information will notify the movement of the communication device to the edge node communicating with the communication device that has completed communication, the useless control packet Will be sent and received. Similarly, even in a system in which the mapping information notification device holds a correspondence table, useless control packets are transmitted and received if information about communication devices that have completed communication exists in the correspondence table.

  An object of the present invention is to efficiently perform communication even when a communication apparatus moves between access networks.

  In a method according to an embodiment, the first transfer device receives a user packet addressed to the second communication device from the first communication device. Then, the first transfer device sends the user packet to the second transfer device set as the end point of the first tunnel communication that has been used for transferring the user packet addressed to the second communication device. Is encapsulated and sent. If the second transfer device does not hold a route to the second communication device, the second transfer device notifies the first transfer device of a failure to transfer the user packet. The first transfer device requests information on a transfer device belonging to a network in which the second communication device is located, from a location management server that manages the location of the second communication device. The first transfer device transfers the user packet to the third transfer device by the second tunnel communication with the third transfer device notified from the location management server as an end point. The third transfer device transfers the user packet to the second communication device.

  Even if the communication device moves between access networks, communication is performed efficiently.

It is a figure which shows the example of the network to which LISP is applied. It is a figure which shows the example of a structure of a transfer apparatus. It is a figure which shows the example of a format of a packet and an encapsulation packet. It is a figure which shows the example of the information element contained in a control packet. It is a figure which shows the example of the hardware constitutions of a transfer apparatus. It is a figure which shows the example of a structure of a position management server. It is a figure which shows the example of the hardware constitutions of a position management server. It is a figure explaining the example of the communication method concerning 1st Embodiment. It is a flowchart explaining the example of the process which the transfer apparatus which operate | moves as an ingress edge node performs. It is a flowchart explaining the example of the process which the transfer apparatus which operate | moves as an egress edge node performs. It is a flowchart explaining the example of the process which a position management server performs. It is a figure explaining the example of the communication method concerning 2nd Embodiment. It is a figure which shows the example of the format of the encapsulation packet produced | generated by 2nd Embodiment. It is a figure explaining the example of the communication method concerning 2nd Embodiment. It is a flowchart explaining the example of the process which the transfer apparatus which operate | moves as an ingress edge node performs. It is a flowchart explaining the example of the process which the transfer apparatus which operate | moves as an egress edge node performs. It is a figure which shows the example of the format of the encapsulation packet produced | generated by 3rd Embodiment. It is a figure which shows the example of the format of the error notification produced | generated in 3rd Embodiment. It is a flowchart explaining the example of the process which the transfer apparatus which operate | moves as an egress edge node performs. 10 is a flowchart illustrating an example of processing performed when a transfer device receives an error notification. It is a figure which shows an example when a 3rd party attack is performed.

<Device configuration>
FIG. 2 shows an example of the configuration of the transfer device 20. The transfer device 20 includes a reception unit 21, a type determination unit 22, an encapsulation unit 23, a decapsulation unit 24, a transfer control unit 25, an error notification generation unit 26, a transmission unit 27, a position information processing unit 30, and a LOC information processing unit 40. The storage unit 50 is provided. The position information processing unit 30 includes a learning unit 31 and a position information notification generation unit 32. The LOC information processing unit 40 includes a LOC request unit 41 and a LOC response processing unit 42. The storage unit 50 stores a route table 51 and a transfer device table 52.

  In the route table 51, information on the route to the communication device 1 included in the access network is recorded. On the other hand, when the tunnel communication of the encapsulated packet 7 is performed in order to transmit the packet 6, the transfer device table 52 records information that associates the transfer device 20 that is the end point of the tunnel communication with the destination communication device 1. ing. Hereinafter, the transfer device 20 serving as an end point of tunnel communication may be referred to as an “exit edge node”. The egress edge node is the transfer device 20 that can be accessed by the communication device 1 that is the destination of the packet 6. In addition, the transfer device 20 that is the starting point of the tunnel communication may be described as an “entrance edge node”. The ingress edge node is a transfer device 20 that can be accessed by the communication device 1 that is the source of the packet 6.

  The receiving unit 21 receives a packet transmitted or transferred to the transfer device 20. For example, the receiving unit 21 receives a packet from the communication device 1 included in the access network or another transfer device 20 included in the core network 3. The receiving unit 21 also receives an LOC response message transmitted from the location management server 10 and an error notification transmitted from the transfer device 20 included in the core network 3. The receiving unit 21 outputs the received packet to the type determining unit 22.

  The type determination unit 22 determines the type of the input packet and distributes it. For example, the type determination unit 22 outputs a data packet received from the access network to the transfer control unit 25, and outputs the encapsulated packet 7 received from the core network 3 to the decapsulation unit 24. Further, the type determination unit 22 outputs a LOC response message to the LOC response processing unit 42 and outputs an error notification to the LOC request unit 41. Further, among the encapsulated packets 7 received from the core network 3, the type determination unit 22 outputs a packet to be transferred to another transfer device 20 by tunnel communication to the transfer control unit 25.

  The encapsulation unit 23 encapsulates the packet input from the transfer control unit 25. An example of the format of the packet 6 and the encapsulated packet 7 is shown in FIG. Packet 6 includes a data field and a header. The header of the packet 6 includes a source address field and a destination address field. For example, when the address of the communication device 1a is IP1 and the address of the communication device 1b is IP2, the packet 6 transmitted from the communication device 1a to the communication device 1b is expressed as a packet 6a in FIG. . In the following description, the header of the packet 6 may be referred to as an “inner header” in order to make the distinction from the outer header easier to understand.

  The encapsulation unit 23 generates an encapsulated packet 7 by adding an outer header to the packet 6 and encapsulating the packet 6. The outer header includes an egress edge node address and an ingress edge node address. For example, it is assumed that the communication device 1a belongs to the same network as the transfer device 20a, and the communication device 1b belongs to the same network as the transfer device 20b. Further, it is assumed that an address LOC1 is assigned to the transfer device 20a and an address LOC2 is assigned to the transfer device 20b. Then, the encapsulating unit 23 encapsulates the packet 6a to generate an encapsulated packet 7a. Note that the encapsulating unit 23 can appropriately access the transfer device table 52 in order to acquire the address of the egress edge node. The encapsulation unit 23 outputs the encapsulated packet 7 to the transfer control unit 25.

  The decapsulating unit 24 of the egress edge node generates the packet 6 by removing the outer header from the encapsulated packet 7 input from the type determining unit 22. The decapsulation unit 24 outputs the packet 6 to the transfer control unit 25. At this time, the decapsulation unit 24 outputs the information included in the outer header removed to generate the packet 6 in association with the packet 6 to the transfer control unit 25.

  The transfer control unit 25 determines the destination of the packet 6 and the encapsulated packet 7 using the route table 51. Furthermore, the transfer control unit 25 outputs the packet 6 to the encapsulation unit 23 when the destination address of the packet 6 is an address assigned to the communication device 1 in the access network connected to the other transfer device 20. Then, the encapsulation unit 23 is requested to perform encapsulation. The route table 51 records a route to the communication device 1 included in the access network to which the transfer device 20 is connected and a route to another transfer device 20 in the core network 3. When the packet 6 is input, the transfer control unit 25 refers to the transfer control unit 25 using the destination address in the inner header of the input packet 6 as a key, and determines the transfer destination. On the other hand, when the encapsulated packet 7 is input, the transfer control unit 25 determines a transfer destination according to the egress edge node. The transfer control unit 25 transfers the packet 6 or the encapsulated packet 7 whose transmission destination has been determined via the transmission unit 27. Further, the transfer control unit 25 also performs a process of deleting the route information about the communication device 1 whose communication period has exceeded a predetermined threshold from the route table 51.

  When the transfer control unit 25 cannot determine the transfer destination for the packet 6 input from the decapsulation unit 24, the transfer control unit 25 notifies the packet 6 that has failed to determine the transfer destination and the packet 6 that has failed to determine the transfer destination. The information is output to the error notification generator 26. Therefore, the error notification generation unit 26 acquires the addresses of the ingress edge nodes for the packet 6 that failed to determine the transfer destination and the encapsulated packet 7 that includes the packet 6 that failed to determine the transfer destination. The error notification generation unit 26 generates an error notification addressed to the ingress edge node. The error notification includes the destination address (destination host ID) of the decapsulated packet 6 and information indicating that the destination is unknown. An example of information elements included in the error notification is shown in FIG. It should be noted that the error notification may include information that can uniquely identify the destination communication device 1 of the packet 6 instead of the destination address of the packet 6. Further, the entire error packet 6 may be included in the error notification.

  The learning unit 31 extracts a transmission source address from the packet generated by the communication device 1 included in the access network, and registers it as route information in the route table. The position information notification generation unit 32 associates the transmission source address extracted by the learning unit 31 with the address assigned to the transfer device 20 and uses it as position information. The location information notification generation unit 32 generates a packet (location information notification) used to notify the location management server 10 of location information.

  The LOC request unit 41 generates a control message (LOC request message) for inquiring the location management server 10 about the egress edge node. The LOC request message includes information indicating that the address of the communication device 1 that is the destination of the packet 6 and the address of the transfer device 20 that is the egress edge node are requested. An example of the LOC request message is shown in FIG. In the example of FIG. 4B, setting to Type = LOC request indicates that the control message requests the address of the egress edge node associated with the destination communication device 1. The destination of the LOC request message is the location management server 10, and the transmission source is the transfer device 20 serving as an ingress edge node. In the example of FIG. 4B, the address of the communication device 1 that is the transmission source of the packet 6 is also included in the LOC request message, but this information is optional and can be omitted. The inquiry about the egress edge node is included in the access network in which the communication device 1 that is the destination of the packet 6 is connected to another transfer device 20, and information on the egress edge node of the tunnel communication is recorded in the transfer device table 52. If not done. In addition, when the encapsulated packet 7 is transmitted to the egress edge node included in the transfer device table 52 and an error notification is received from the egress edge node, the egress edge node is also inquired.

  The LOC response processing unit 42 processes the LOC response message received from the location management server 10. FIG. 4C shows an example of information elements included in the LOC response message. The LOC response message includes a destination address, a source address, a Type, information for identifying the destination communication device 1 of the packet 6, and an address of the egress edge node. In the LOC response message, the transmission source is the location management server 10 and the destination is the ingress edge node that transmitted the LOC request message. “Type = LOC response” is information indicating that this is a control message for notifying the address of the egress edge node in response to the LOC request message. The LOC response processing unit 42 extracts a combination of the destination communication device 1 and the address of the egress edge node from the LOC response message, and records it in the transfer device table 52.

  FIG. 5 shows an example of the hardware configuration of the transfer device 20. The transfer device 20 includes a processor 61, a switch circuit 62, a storage device 63, a bus 64, and interface circuits 65 (65a to 65d). The transfer device 20 may be realized by a router, for example. In the example of FIG. 5, the transfer device 20 includes four interface circuits 65, which are interface circuits 65a to 65d, but the number of interface circuits 65 included in the transfer device 20 is arbitrary.

  The processor 61 can be any processing circuit including a Central Processing Unit (CPU). The processor 61 operates as the type determination unit 22, the encapsulation unit 23, the decapsulation unit 24, the error notification generation unit 26, the position information processing unit 30, and the LOC information processing unit 40. The processor 61 can execute a program stored in the storage device 63, for example. The transfer control unit 25 is realized by the processor 61 and the switch circuit 62. The storage device 63 operates as the storage unit 50. Furthermore, the storage device 63 also appropriately stores data obtained by the operation of the processor 61 and data used for the processing of the processor 61. The bus 64 connects the processor 61, the switch circuit 62, the storage device 63, and the interface circuit 65 so that data can be exchanged between them. The receiving unit 21 and the transmitting unit 27 are realized by the interface circuit 65 (65a to 65d).

  FIG. 6 shows an example of the configuration of the location management server 10. The location management server 10 includes a reception unit 11, a transmission unit 12, a location information processing unit 13, a LOC request processing unit 14, and a location information management table 15. The receiving unit 11 receives the LOC request message and the position information notification transmitted from the transfer device 20. The receiving unit 11 outputs a position information notification to the position information processing unit 13. The position information processing unit 13 records information included in the position information notification input from the receiving unit 11 in the position information management table 15. In the location information management table 15, the address of the transfer device 20 that is the transmission source of the location information notification and the address of the communication device 1 included in the access network to which the transfer device 20 belongs are recorded in association with each other. The receiving unit 11 outputs the LOC request message to the LOC request processing unit 14. The LOC request processing unit 14 acquires the address of the destination communication device 1 of the packet 6 from the LOC request message, and searches the location information management table 15 using the acquired address as a key. The LOC request processing unit 14 generates a LOC response message using the address obtained by the search as the address of the egress edge node. The LOC request processing unit 14 outputs a LOC response message to the transmission unit 12. The transmission unit 12 transmits the LOC response message to the ingress edge node that made the inquiry about the egress edge node.

  FIG. 7 shows an example of the hardware configuration of the location management server 10. The location management server 10 includes a processor 71, a memory 72, an input device 73, an output device 74, a bus 75, an external storage device 76, a medium drive device 77, and a network connection device 79. The location management server 10 may be realized by a computer, for example.

  The processor 71 can be any processing circuit including a Central Processing Unit (CPU). The processor 71 operates as the position information processing unit 13 and the LOC request processing unit 14. The processor 71 can execute a program stored in the external storage device 76, for example. The memory 72 stores the position information management table 15. Further, the memory 72 appropriately stores data obtained by the operation of the processor 71 and data used for processing of the processor 71. The network connection device 79 performs processing for communication with the transfer device 20 and operates as the reception unit 11 and the transmission unit 12.

  The input device 73 is realized as, for example, a button, a keyboard, or a mouse, and the output device 74 is realized as a display or the like. The bus 75 connects the processor 71, the memory 72, the input device 73, the output device 74, the external storage device 76, the medium drive device 77, and the network connection device 79 so as to exchange data with each other. The external storage device 76 stores programs, data, and the like, and provides the stored information to the processor 71 as appropriate. The medium driving device 77 can output the data of the memory 72 and the external storage device 76 to the portable storage medium 78 and can read programs, data, and the like from the portable storage medium 78. Here, the portable storage medium 78 may be any portable storage medium including a floppy disk, a Magneto-Optical (MO) disk, a Compact Disc Recordable (CD-R), and a Digital Versatile Disk Recordable (DVD-R). It can be a medium.

<First Embodiment>
FIG. 8 is a diagram illustrating an example of a communication method according to the first embodiment. Hereinafter, an example of a communication method performed when the destination communication device 1 of the packet 6 moves between access networks will be described with reference to FIG. In the following description, in order to make it easy to understand which operation is performed by the transfer apparatuses 20a to 20c, the same alphabet as the alphabet included in the number of the transfer apparatus 20 performing the operation is described at the end of the reference number. There shall be. For example, the learning unit 31 included in the transfer device 20b is described as “learning unit 31b”, but the learning unit 31 included in the transfer device 20a is described as “learning unit 31a”. Further, the following description is an example, and it is assumed that the procedure may be modified depending on the implementation, for example, the order of the procedures (6) and (7) is changed.

As shown in FIG. 8A, at the start of communication, it is assumed that the transmission source communication device 1a is located in the access network A1 and the destination communication device 1b is located in the access network A2. Further, it is assumed that the transfer device 20a connects the access network A1 and the core network 3, and the transfer device 20b connects the access network A2 and the core network 3. Similarly, the transfer device 20c connects the access network A3 and the core network 3. In the following description, the addresses are as follows.
Address of transfer device 20a: LOC1
Address of transfer device 20b: LOC2
Address of transfer device 20c: LOC3
Address of communication device 1a: IP1
Address of communication device 1b: IP2

  (1) The learning unit 31b of the transfer device 20b specifies the source address (IP2) included in the header of the packet transmitted by the communication device 1b. The learning unit 31b notifies the obtained information to the position information notification generation unit 32b and appropriately records the information in the route table 51b. The location information notification generation unit 32b transmits a location information notification in which the address (LOC2) assigned to the transfer device 20b is associated with the address of the communication device 1b to the location management server 10 via the transmission unit 27b.

  (2) Upon receiving the location information notification transmitted from the transfer device 20b, the location information processing unit 13 of the location management server 10 records LOC2 and IP2 in association with each other in the location information management table 15. Here, it is assumed that the location management server 10 holds a location information management table 15α shown in FIG.

  (3) The communication device 1a transmits the packet 6 (6a in FIG. 3) addressed to the communication device 1b. The destination address of the packet 6 is the address (IP2) of the communication device 1b, and the transmission source address is the address (IP1) of the communication device 1a.

(4) The transfer device 20a receives the packet 6a at the receiving unit 21a. The type determination unit 22a outputs the packet 6a to the transfer control unit 25a. Since the communication device 1b that is the destination of the packet 6a is not included in the access network A1, the transfer control unit 25a determines to perform tunnel communication, and outputs the packet 6a to the encapsulation unit 23a. The encapsulation unit 23a refers to the transfer device table 52a and tries to identify the egress edge node. However, when the packet 6a addressed to the communication device 1b arrives first, the egress edge node associated with the communication device 1b is not recorded in the transfer device table 52a. Therefore, the encapsulation unit 23a requests the LOC request unit 41a to acquire information on the egress edge node. In response to the request from the encapsulation unit 23a, the LOC request unit 41a transmits a LOC request message including the following information elements to the location management server 10.
Destination address: Location management server 10 address
Source address: LOC1
Type: LOC request
Destination communication device address: IP2
Source communication device address: IP1

  (5) The LOC request processing unit 14 of the location management server 10 that has received the LOC request message refers to the location information management table 15α and searches for the transfer device 20 to which the communication device 1b is connected. At this time, since the location information management table 15α is referenced, LOC2 is detected as the address of the egress edge node. Therefore, the LOC request processing unit 14 generates a LOC response message including the following information elements.

Destination address: LOC1
Source address: address of the location management server 10
Type: LOC response
Destination communication device address: IP2
Egress edge node address: LOC2
The LOC request processing unit 14 transmits the generated LOC response message to the transfer device 20a via the transmission unit 12.

  (6) The LOC response message received by the transfer device 20a is input to the LOC response processing unit 42a by the type determination unit 22a. The LOC response processing unit 42a extracts the information included in the LOC response message, thereby notifying the encapsulation unit 23a that LOC2 becomes an egress edge node of the encapsulated packet 7 including the packet addressed to IP2. . The encapsulation unit 23a generates the encapsulated packet 7 (7a in FIG. 3) by encapsulating the packet 6a using the information notified from the LOC response processing unit 42a. The encapsulation unit 23a outputs the encapsulated packet 7 to the transfer control unit 25a. The transfer control unit 25 a performs transfer control of the encapsulated packet 7.

  (7) The LOC response processing unit 42a updates the transfer device table 52a to the state of the transfer device table 52α shown in FIG. 8A based on the information included in the LOC response message.

  (8) The encapsulated packet 7 transmitted to the transfer device 20b in the procedure (6) is decapsulated in the decapsulation unit 24b of the transfer device 20b and returns to the state of the packet 6a. The transfer controller 25b transfers the packet 6a obtained by decapsulation to the communication device 1b with reference to the route table 51b. As a result, the communication device 1b can receive the packet 6a from the communication device 1a.

  (9) Next, as shown in FIG. 8B, it is assumed that the communication device 1b has moved from the access network A2 to the access network A3. When a certain period elapses after the communication device 1b moves, the transfer control unit 25b deletes the route to the communication device 1b from the route table 51b.

  (10) It is assumed that the data packet or control message generated by the communication device 1b is received by the transfer device 20c. Then, the learning unit 31c specifies the address (IP2) of the communication device 1b using the received packet. The learning unit 31c notifies the obtained information to the position information notification generation unit 32c and appropriately records the information in the route table 51c. The location information notification generation unit 32c transmits to the location management server 10 a location information notification in which the address (LOC3) assigned to the transfer device 20c is associated with the address of the communication device 1b.

  (11) Upon receiving the location information notification transmitted from the transfer device 20c, the location information processing unit 13 of the location management server 10 records LOC3 and IP2 in association with each other in the location information management table 15. For this reason, the position information management table 15α is updated to the position information management table 15β.

  (12) The transfer device 20a is not notified that the communication device 1b has moved. Therefore, when the transfer device 20a next receives the packet 6 addressed to the communication device 1b from the communication device 1a, the encapsulated packet 7 encapsulating the packet 6 is transferred to the transfer device 20b based on the record in the transfer device table 52α. Send.

(13) The decapsulation unit 24b of the transfer device 20b decapsulates the encapsulated packet 7 received from the transfer device 20a, and outputs it to the transfer control unit 25b together with the information in the outer header. The transfer control unit 25b refers to the route table 51b, but the route to the communication device 1b is not included in the route table 51b by the processing of the procedure (9). Therefore, the transfer control unit 25b determines that the transfer has failed, and outputs the packet obtained by decapsulation and the information in the outer header to the error notification generation unit 26b. Then, the error notification generation unit 26b generates an error notification including the following information.
Destination address: LOC1
Source address: LOC2
Type: Destination unknown error
Destination communication device address: IP2
The error notification generation unit 26b transmits the generated error notification to the transfer device 20a via the transmission unit 27b.

  (14) The type determination unit 22a of the transfer device 20a that has received the error notification outputs the error notification to the LOC request unit 41a. The LOC request unit 41a extracts the address of the destination communication device of the packet that failed to be transferred from the error notification. The LOC request unit 41a deletes the record of the egress edge node associated with the extracted address from the transfer device table 52. Here, the LOC request unit 41a deletes the record of the egress edge node (LOC2) associated with IP2 when specifying that the address of the destination communication device 1 of the packet that failed to be transferred is IP2 from the error notification . Accordingly, the transfer device table 52α is updated to the transfer device table 52β.

  (15) The LOC request unit 41a inquires of the location management server 10 about an egress edge node when performing the tunnel communication of the packet 6 addressed to the communication device 1b. The operation at this time is similar to the procedure (4).

(16) Receiving the LOC request message, the location management server 10 transmits a LOC response message to the transfer device 20a by the same processing as in the procedure (5). Here, the LOC response message transmitted to the transfer apparatus 20a includes the following information elements.
Destination address: LOC1
Source address: address of the location management server 10
Type: LOC response
Destination communication device address: IP2
Egress edge node address: LOC3

  (17) By processing the LOC response message in the same manner as in the procedure (6), the encapsulating unit 23a recognizes that the LOC3 is an egress edge node of the encapsulated packet 7 addressed to IP2. The encapsulation unit 23a generates the encapsulated packet 7 using the LOC3 as an egress edge node. As shown in FIG. 8C, the transfer control unit 25a performs transfer control of the encapsulated packet 7 generated by the encapsulation unit 23a.

  (18) The LOC response processing unit 42a updates the transfer device table 52β to the transfer device table 52γ based on the information included in the LOC response message received in step (17).

  (19) The encapsulated packet 7 transmitted to the transfer device 20c in the procedure (17) is processed in the transfer device 20c in the same manner as in the procedure (8). As a result, the packet obtained by decapsulating the encapsulated packet 7 is transmitted from the transfer device 20c to the communication device 1b. Therefore, the communication device 1b can receive the packet from the communication device 1a.

  FIG. 9 is a flowchart illustrating an example of processing performed by the transfer device 20 that operates as an ingress edge node. FIG. 9 is an example. For example, the order of steps S7 and S8 may be changed. In addition, the order of steps S9 and S10 may be changed.

  The receiving unit 21 receives the packet and outputs it to the type determining unit 22 (step S1). The type determination unit 22 determines whether the received packet is a data packet received from the access network (step S2). If the received packet is a data packet, the type determination unit 22 outputs the data packet to the transfer control unit 25 (Yes in step S2). When the packet is addressed to the communication device 1 included in an access network other than the connected access network, the transfer control unit 25 requests the encapsulation unit 23 to perform encapsulation. The encapsulation unit 23 searches the transfer device table 52 using the destination address as a key, and determines whether there is a corresponding entry (step S3). If there is a corresponding entry, the encapsulating unit 23 generates an encapsulated packet 7 with the egress edge node included in the extracted entry as the destination, and the transfer control unit 25 transfers the encapsulated packet 7 to the destination. (Step S4). On the other hand, if there is no corresponding entry in the transfer device table 52, the LOC request unit 41 transmits a LOC request message to the location management server 10 (step S5).

  If the received packet is not a data packet, the type determination unit 22 determines whether the received packet is a LOC response message (No in step S2, step S6). If the received packet is a LOC response message, the type determination unit 22 outputs the LOC response message to the LOC response processing unit 42. The LOC response processing unit 42 records the information included in the LOC response message in the transfer device table 52 (step S7). Further, the LOC response processing unit 42 notifies the information in the LOC response message to the encapsulation unit 23, and the encapsulation unit 23 encapsulates the data packet based on the notified information. The encapsulated packet is transmitted toward the egress edge node (step S8).

  On the other hand, if the received packet is not a LOC response message in step S6, the type determination unit 22 determines that an error notification has been received, and outputs the received packet to the LOC request unit 41 (No in step S6). The LOC request unit 41 acquires the information of the destination communication device 1 stored in the error notification, and deletes the entry corresponding to the destination communication device 1 from the transfer device table 52 (step S9). Further, the LOC request unit 41 requests the information of the egress edge node used for communication to the destination communication device 1 by transmitting a LOC request message to the location management server 10 (step S10).

  FIG. 10 is a flowchart illustrating an example of processing performed by the transfer device 20 that operates as an egress edge node. The receiving unit 21 receives a data packet from the core network 3 via tunnel communication and outputs it to the type determining unit 22 (step S21). The type determination unit 22 outputs the received packet to the decapsulation unit 24. The decapsulation unit 24 removes the outer header of the input packet, and outputs the decapsulated packet 6 and outer header information to the transfer control unit 25 (step S22). The transfer control unit 25 searches the route table 51 using the destination address of the input packet as a key (step S23). When the route to the destination address is found, the transfer control unit 25 transfers the packet using the route table 51 (step S24). On the other hand, when the route to the destination address is not found, the transfer control unit 25 outputs the decapsulated packet and the information in the outer header to the error notification generation unit 26. The error notification generation unit 26 generates an error notification addressed to the transmission source address in the outer header, and transmits the error notification via the transmission unit 27 (step S25).

  FIG. 11 is a flowchart for explaining an example of processing performed by the location management server 10. When receiving the LOC request message, the receiving unit 11 outputs the LOC request message to the LOC request processing unit 14 (step S31). The LOC request processing unit 14 searches the location information management table 15 for an entry associated with the address of the destination communication device 1 included in the LOC request message (step S32). When there is an entry corresponding to the destination address, the LOC response message storing the address of the egress edge node associated with the address of the destination communication device 1 is transmitted to the transfer device 20 that is the transmission source of the LOC request message ( Step S33). On the other hand, if there is no entry corresponding to the address of the destination communication device 1, it is determined that the destination is unknown and error processing is performed (step S34).

  Thus, the egress edge node sends an error notification to the ingress edge node when it receives the encapsulated packet 7 including the packet 6 that cannot be transferred. Furthermore, the ingress edge node inquires of the location management server 10 about the egress edge node in response to the error notification. Therefore, according to the first embodiment, for example, for the transfer device 20d communicating with the communication device 1c that has completed communication with the moved communication device 1b, an error notification or the movement of the communication device 1b is performed. Earlier notifications are not sent. For this reason, in 1st Embodiment, transmission / reception of a useless control packet can be prevented and efficient communication can be performed.

  Furthermore, according to the first embodiment, the relationship between the ingress edge node and the destination communication device 1 need not be stored in the egress edge node. When the egress edge node stores the relationship between the ingress edge node and the destination communication device 1, the forwarding device monitors the address information included in the encapsulated packet 7 received when no error has occurred. Therefore, the burden on the transfer device is large. On the other hand, in the first embodiment, the egress edge node does not memorize the relationship between the ingress edge node and the destination communication device 1, so that the egress edge node stores the address information in the encapsulated packet 7 and the like. The burden on the transfer device 20 is reduced.

<Second Embodiment>
FIG. 12 is a diagram illustrating an example of a communication method according to the second embodiment. In the second embodiment, a method of using the time elapsed after the transfer device table 52 is updated in order to further reduce control packets transmitted and received than in the first embodiment will be described.

  The transfer device table 52 used in the second embodiment records the address of the egress edge node and the comparison result in association with the address of the destination communication device 1, as in the transfer device table 52x of FIG. Is done. The comparison result is a result of comparing the time elapsed from the time when the information of the egress edge node regarding the destination communication device 1 is updated with a predetermined threshold (elapsed time threshold). When the time elapsed from the time when the information of the egress edge node is updated is shorter than the elapsed time threshold, the comparison result is “New”. On the other hand, when the time elapsed from the time when the information of the egress edge node is updated is equal to or greater than the elapsed time threshold, the comparison result is “Old”.

  The elapsed time threshold is used as an index for determining the reliability of the information of the egress edge node that is the processing target. The elapsed time threshold is a threshold that can be determined according to the implementation. For example, the elapsed time threshold may be set to a time that is predicted to be taken from the time when the communication device 1 moves to the time when the location information management table 15 is updated according to the movement of the communication device 1. is there. In this case, the elapsed time threshold value can be, for example, an average value of results of experiments in which the time from the movement of the communication device 1 to the change of the position information management table 15 is measured a plurality of times.

  When the time elapsed from the time when the information of the egress edge node related to the destination communication device 1 is updated is less than the elapsed time threshold, the information recorded in the transfer device table 52 indicates that the time indicated by the elapsed time threshold has elapsed. Newer than the information you have. For this reason, the possibility that the information about the communication device 1 for which the elapsed time threshold has not elapsed since the acquisition time has been changed in the location information management table 15 after being recorded in the transfer device table 52 is represented by the elapsed time threshold. Compared to the possibility that information acquired before time has been changed. Even if the information acquired by the ingress edge node from the location management server 10 is incorrect, if the location information management table 15 is not updated, the ingress edge node is notified of the wrong egress edge node information again by an inquiry. And the query is wasted. Therefore, in order to prevent the ingress edge node from querying the egress edge node information multiple times under the situation where the information in the location information management table 15 is not changed, in the second embodiment, the freshness of the information of the egress edge node is set. Accordingly, it is determined whether to send an error notification. That is, in the second embodiment, in the tunnel communication performed based on the information that has not passed the elapsed time threshold after being acquired from the location information management table 15, the egress edge node fails to transfer the packet. Also does not send an error notification to the ingress edge node.

  Hereinafter, a case where a packet is transmitted from the transmission source communication device 1a to the destination communication device 1b and the communication device 1b moves between access networks will be described with reference to FIG. Assume that the addresses and positions of the communication device 1a and the transfer devices 20a to 20c are the same as those described with reference to FIG. The address of the communication device 1b is the same as in the example of FIG. In the following description, the elapsed time threshold is Th1.

  (21) The registration of the location information of the communication device 1b, the inquiry of the information of the egress edge node, and the like are the same as the procedures (1) to (5) described with reference to FIG. FIG. 12A shows the procedure numbers (1) to (5) used in the description of FIG.

  (22) The LOC response processing unit 42a updates the transfer device table 52 based on the information included in the LOC response message. At this time, the LOC response processing unit 42a records the comparison result of the address of the egress edge node, the elapsed time from the time when the information is acquired, and the elapsed time threshold in association with the address of the destination communication device 1. When the transfer device table 52 is updated based on the LOC response message, the time elapsed from the time when the information acquired from the LOC response message is acquired is shorter than the time represented by the elapsed time threshold. Therefore, the comparison result is New. Here, it is assumed that the transfer device table 52x shown in FIG. 12A is obtained by the processing of the LOC response processing unit 42a.

  (23) The encapsulation unit 23a performs the encapsulation using the transfer device table 52x. FIG. 13 shows an example of the format of the encapsulated packet 8 generated in the second embodiment. The encapsulated packet 8 includes a usage period flag in addition to the information elements included in the encapsulated packet 7. The usage period flag = 0 indicates that the time elapsed from the time when the information used for encapsulation is acquired from the location management server 10 is shorter than the time represented by the elapsed time threshold. On the other hand, the usage period flag = 1 indicates that the time elapsed from the time when the information used for encapsulation is acquired from the location management server 10 is equal to or longer than the time represented by the elapsed time threshold. Here, it is recorded in the transfer device table 52x that the elapsed time threshold has not elapsed since the information used to encapsulate the packet 6 addressed to the communication device 1b is acquired. Therefore, the encapsulation unit 23a generates an encapsulated packet with a usage period flag = 0. That is, the encapsulation unit 23a generates the encapsulated packet 8a shown in FIG. When the encapsulating unit 23a outputs the encapsulated packet 8a to the transfer control unit 25a, the transfer control unit 25a performs transfer control of the encapsulated packet 8a.

  (24) The packet 6 obtained by decapsulation is transferred to the communication device 1b by the same process as the procedure (8) described with reference to FIG.

  (25) When the time elapsed from the time when the information of the egress edge node used to encapsulate the packet 6 addressed to the communication device 1b is acquired becomes equal to or greater than the elapsed time threshold Th1, the transfer control unit 25a changes the comparison result to Old. To do. That is, the transfer device table 52x is changed as shown in the transfer device table 52y.

  (26) Next, as shown in FIG. 12B, it is assumed that the communication device 1b has moved from the access network A2 to the access network A3.

  (27) The transfer device 20a is not notified that the communication device 1b has moved. Therefore, when the transfer device 20a next receives the packet 6 addressed to the communication device 1b from the communication device 1a, the transfer device 20a generates an encapsulated packet with a use period flag = 1 based on the record in the transfer device table 52y. That is, the encapsulation unit 23a generates the encapsulated packet 8b in FIG. The transfer device 20a transmits the encapsulated packet 8b to the transfer device 20b.

  (28) The decapsulation unit 24b of the transfer device 20b decapsulates the encapsulated packet 8b received from the transfer device 20a, and outputs it to the transfer control unit 25b together with the information in the outer header. At this time, since the communication device 1b has already moved, the transfer control unit 25b fails to transfer. The transfer control unit 25b outputs the packet obtained by decapsulation and the information in the outer header to the error notification generation unit 26b.

  The error notification generation unit 26b determines whether the value of the usage period flag in the outer header is 1. When the usage period flag = 1, the information on the egress edge node has already been changed in the location management server 10 because the time equal to or greater than the elapsed time threshold has elapsed since the information used for tunnel communication was acquired. there is a possibility. Therefore, when the usage period flag = 1, the error notification generation unit 26b generates an error notification. Here, since the usage period flag = 1, the error notification generation unit 26b generates an error notification addressed to the transfer device 20a. The generated error notification is transmitted to the transfer device 20a via the transmission unit 27b.

  (29) Upon receiving the error notification, the LOC request unit 41a of the transfer device 20a inquires of the location management server 10 about the egress edge node associated with the address of the destination communication device of the packet that failed to be transferred. The operation at this time is the same as the procedures (14) and (15) described with reference to FIG.

(30) Receiving the LOC request message, the location management server 10 transmits the LOC response message to the transfer device 20a by the same process as the procedure (16) described with reference to FIG. At this time, since the information regarding the communication device 1b in the position information management table 15 is not updated, a LOC response message including the following information elements is transmitted to the transfer device 20a.
Destination address: LOC1
Source address: address of the location management server 10
Type: LOC response
Destination communication device address: IP2
Egress edge node address: LOC2

  (31) The LOC response processing unit 42a updates the transfer device table 52 based on the information included in the LOC response message by performing the same processing as in the procedure (22). Here, it is assumed that the transfer device table 52u illustrated in FIG. 14A is obtained by the processing of the LOC response processing unit 42a.

  (32) Since the information is updated by the LOC response message as shown in the transfer device table 52u, the transfer device 20a transfers the encapsulated packet 8 including the packet 6 addressed to the communication device 1b to the transfer device 20b (LOC2) by tunnel communication. ). Since the encapsulated packet 8 transmitted at this time is transmitted immediately after the update of the transfer device table 52u, the usage period flag = 0.

  (33) Since the communication device 1b is not in the access network A2, the transfer device 20b cannot transfer the packet 6 obtained by decapsulating the encapsulated packet 8. The error notification generation unit 26b extracts the value of the usage period flag in the outer header. Here, the usage period flag = 0. That is, the time represented by the elapsed time threshold has not elapsed since the information used for the current tunnel communication is acquired. For this reason, the information on the exit edge node has not yet been updated in the location management server 10, and there is a possibility that correct information cannot be obtained even if the entrance edge node inquires the location management server 10 about the exit edge node. Therefore, the error notification generation unit 26b ends the process without generating an error notification.

  (34) It is assumed that a data packet, a control message, etc. generated by the communication device 1b are received by the transfer device 20c. Then, a location information notification in which the address (LOC3) of the transfer device 20c and the address of the communication device 1b are associated is transmitted to the location management server 10 by the same processing as the procedure (10) described with reference to FIG. .

  (35) Since the location information processing unit 13 of the location management server 10 records the LOC3 and IP2 in association with each other on the location information management table 15 based on the location information notification from the transfer device 20c, the location information management table 15β can get.

  (36) The communication device 1a determines that the packet transmission has failed because there is no reception response from the communication device 1b, and transmits again the packet 6 containing the same data as the previously transmitted packet. When the transfer device 20a receives the packet 6 retransmitted from the communication device 1a, it is assumed that the time represented by the elapsed time threshold has elapsed since the transfer device table 52 was updated in step (31). Then, the comparison result is set to Old as shown in the transfer device table 52w. The transfer device 20a generates the encapsulated packet 8 using the transfer device table 52w of FIG. 14B, and transmits it to the transfer device 20b. For this reason, the usage period flag of the outer header is set to 1.

  (37) Since the communication device 1b is not in the access network A2, the transfer device 20b cannot transfer the packet 6 obtained by decapsulating the encapsulated packet 8. The error notification generation unit 26b extracts the value of the usage period flag in the outer header. Here, the usage period flag = 1. Accordingly, the error notification generation unit 26b determines that the encapsulated packet 8 has been erroneously transmitted to the transfer device 20b because the information of the egress edge node used in the procedure (36) is old. Therefore, the error notification generation unit 26b generates an error notification and transmits it to the transfer device 20a via the transmission unit 27b as in the procedure (28).

  (38) Upon receiving the error notification, the LOC request unit 41a determines that the packet tunnel communication to the communication device 1b has failed. The LOC request unit 41a inquires of the location management server 10 about the egress edge node associated with the communication device 1b using the LOC request message.

  (39) The location management server 10 that has received the LOC request message transmits a LOC response message to the transfer device 20a by the same processing as in the procedure (30). At this time, since the location information management table 15 is as shown in FIG. 14B, a LOC response message including the following information elements is transmitted to the transfer device 20a.

Destination address: LOC1
Source address: address of the location management server 10
Type: LOC response
Destination communication device address: IP2
Egress edge node address: LOC3
(40) From the LOC response message, the encapsulation unit 23a recognizes that the LOC3 is an egress edge node for tunnel communication addressed to IP2. The encapsulating unit 23a generates the encapsulated packet 8 using the LOC3 as an egress edge node, and the transfer control unit 25a performs transfer control of the encapsulated packet 8 generated by the encapsulating unit 23a.

  (41) The LOC response processing unit 42a updates the transfer device table 52w to the transfer device table 52z based on the information included in the LOC response message.

  (42) The packet obtained by decapsulating the encapsulated packet 8 transmitted to the transfer device 20c is transmitted from the transfer device 20c to the communication device 1b. Therefore, the communication device 1b can receive the packet from the communication device 1a.

  FIG. 15 is a flowchart illustrating an example of processing performed by the transfer device 20 that operates as an ingress edge node. The processing in steps S41 to S43 is the same as that in steps S1 to S3 described with reference to FIG. When there is an entry corresponding to the transfer device table 52, the encapsulation unit 23 determines whether the comparison result included in the extracted entry is “New” (step S44). When the comparison result is “Old”, the encapsulating unit 23 generates the encapsulated packet 8 including the use period flag = 1 with the egress edge node included in the extracted entry as the destination (step S45). On the other hand, when the comparison result is “New”, the encapsulating unit 23 generates the encapsulated packet 8 including the use period flag = 0 with the egress edge node included in the extracted entry as the destination (step S46). ). In both steps S45 and S46, the generated encapsulated packet 8 is transmitted to the egress edge node. On the other hand, if there is no corresponding entry in the transfer device table 52, the LOC request unit 41 transmits a LOC request message to the location management server 10 (determined that there is none in step S43, step S47).

  The processing in steps S48 to S50 performed when the received packet is not a data packet is the same as steps S6 to S8 described with reference to FIG. However, in step S49, the LOC response processing unit 42 registers the comparison result as “New” in the transfer device table 52 in addition to the information of the egress edge node. When the time elapsed from the time when the LOC response message is received becomes equal to or greater than the elapsed time threshold, the transfer control unit 25 sets the comparison result in the transfer device table 52 to “Old” (steps S51 and S52). The processes in steps S53 and S54 performed when the received packet is an error notification are the same as steps S9 and S10 described in FIG.

  FIG. 16 is a flowchart illustrating an example of processing performed by the transfer device 20 that operates as an egress edge node. The processing of steps S61 to S64 is the same as steps S21 to S24 described with reference to FIG. On the other hand, when the route to the destination address is not found, the transfer control unit 25 outputs the decapsulated packet and the information in the outer header to the error notification generation unit 26. The error notification generator 26 determines whether the value of the usage period flag in the outer header is 1 (step S65). When the value of the usage period flag is 1, the error notification generation unit 26 generates an error notification addressed to the transmission source address in the outer header and transmits the error notification via the transmission unit 27 (Yes in Step S65, Step S66). ). If the value of the usage period flag is 0, the error notification generation unit 26 ends the process without generating an error notification (No in step S65).

  As described above, in the second embodiment, the error notification is not notified from the egress edge node unless the time represented by the elapsed time threshold has elapsed from the time when the information in the transfer device table 52 is updated. For this reason, it is possible to prevent multiple inquiries from the ingress edge node to the egress edge node before the information held by the location management server 10 is changed due to the movement of the communication device 1 or the like. . That is, according to the second embodiment, control that is transmitted and received between the ingress edge node and the location management server 10 before the information held by the location management server 10 is changed as the communication device 1 moves. The amount of packets is reduced. Furthermore, also in the second embodiment, the egress edge node sends an error notification to the ingress edge node when it receives the encapsulated packet 8 including the packet 6 that cannot be transferred. For this reason, the change of the egress edge node is not notified to the transfer apparatus 20 that does not transfer the packet addressed to the communication apparatus 1 that has moved. Therefore, in the second embodiment, as in the first embodiment, useless transmission / reception of control packets can be prevented and efficient communication can be performed.

  In the second embodiment, even when a packet is transmitted from the communication device 1a to the communication device 1b after the destination communication device 1b is turned off, transmission / reception of useless control packets can be suppressed. For example, it is assumed that the transfer device 20b deletes the route from the route table 51b to the communication device 1b one minute after the time when the communication device 1b is turned off. In addition, it is assumed that the transfer device 20b transmits a combination of the address of the communication device and the address of the transfer device 20b included in the access network A2 to the location management server 10 as location information notifications every 3 minutes. Then, even if the route of the communication device 1b is deleted from the transfer device 20b, the location information management table 15 of the location management server 10 registers that the exit edge node for communication to the communication device 1b is the transfer device 20b. A continuing case occurs. In this case, if the second embodiment is not used, the transfer device 20b mechanically sends an error notification to the transfer device 20a. Therefore, the transfer device 20a is requested several times before the location information management table 15 is updated. Will also send a LOC request message. In addition, the location management server 10 transmits a LOC response message to the transfer device 20a every time a LOC request message is transmitted. However, in the second embodiment, the freshness of the information of the egress edge node included in the outer header is indicated by the usage period flag, and is encapsulated by information in which the time represented by the elapsed time threshold has not elapsed since acquisition. No error notification is sent when an encrypted packet is transmitted. For this reason, in the second embodiment, it is possible to prevent unnecessary transmission / reception of the LOC request message and the LOC response message during the time lag until the location information management table 15 is updated.

<Third Embodiment>
In the third embodiment, by using an identification number, a communication method that can prevent an increase in the number of control packets transmitted and received in the network due to the attack even if an attack from a malicious third party is received. Will be explained. In the third embodiment, the ingress edge node generates an encapsulated packet including an identification number at the time of encapsulation. Note that the method of acquiring information of the egress edge node and the method of transmitting the data packet are the same as in the first embodiment.

  FIG. 17 shows an example of the format of the encapsulated packet 9 generated in the third embodiment. The outer header added when generating the encapsulated packet 9 includes an egress edge node address, an ingress edge node address, and an identification number. The identification number is a number used to determine whether the packet is transmitted from the ingress edge node to the egress edge node. It is assumed that the encapsulation unit 23 stores a combination of the destination communication device 1 and an identification number. For example, when the encapsulated packet 9a is generated, the encapsulating unit 23 stores that the packet having the identification number 1234 is transmitted to the communication device 1b (IP2). The operation when the transfer device 20b succeeds in transferring the packet after decapsulation is the same as that in the first embodiment.

  FIG. 18 is a diagram illustrating an example of a format of an error notification generated in the third embodiment. In the third embodiment, when the transfer device 20 operating as the egress edge node fails to transfer the packet after decapsulation, as shown in FIG. 18, the transfer device 20 generates an error notification including the identification number in the outer header. To the ingress edge node.

  When receiving the error notification, the type determination unit 22 of the transfer device 20 operating as the ingress edge node outputs an error notification to the LOC request unit 41. The LOC request unit 41 determines whether the combination of the identification number in error notification and the destination communication device 1 is stored in the encapsulation unit 23. When the combination of the identification number in error notification and the destination communication device 1 is stored in the encapsulation unit 23, the LOC request unit 41 determines that the error is with respect to the encapsulated packet transmitted by the transfer device 20, and Request the egress edge node from the management server 10. On the other hand, when the combination of the identification number in error notification and the destination communication device 1 is not stored in the encapsulation unit 23, the LOC request unit 41 determines that the attack is from a third party and does not transmit the LOC request message. .

  For example, it is assumed that the transfer device 20a receives an error notification after the encapsulated packet 9a is transmitted from the transfer device 20a to the transfer device 20b. At this time, if the combination of (identification number, destination communication device 1) is (1234, IP2), the LOC request unit 41 uses the LOC request message to specify the egress edge node corresponding to the destination communication device 1. Inquire. On the other hand, if the combination of (identification number, destination communication device 1) is (2234, IP2), the LOC request unit 41 ends the process without generating the LOC request message.

  FIG. 19 is a flowchart for explaining an example of processing performed by the transfer apparatus operating as an egress edge node. Steps S71 to S74 are the same as steps S21 to S24 described with reference to FIG. When the route to the destination address is not found, the transfer control unit 25 outputs the decapsulated packet and information in the outer header to the error notification generation unit 26. The error notification generator 26 determines whether identification information is included in the outer header (step S75). When the identification information is not included, the error notification generation unit 26 determines that an attack from a malicious third party has been received, and does not generate an error notification (No in step S75). When the identification information is included in the outer header, the error notification generation unit 26 generates an error notification including the identification information addressed to the transmission source address in the outer header, and transmits the error notification via the transmission unit 27 (step 27). Yes in S75, step S76).

  FIG. 20 is a flowchart for explaining an example of processing performed when the transfer apparatus operating as an ingress edge node receives an error notification. If the type determination unit 22 determines that an error notification has been received, the type determination unit 22 outputs the received packet to the LOC request unit 41 (step S81). The LOC request unit 41 compares the combination of the identification number stored in the error notification and the destination communication device 1 with the combination of the identification number stored in the encapsulation unit 23 and the destination communication device 1 (step S82). ). When the combination of the identification number in error notification and the destination communication device 1 matches the combination stored in the encapsulation unit 23, the LOC request unit 41 has received an error notification for the packet transmitted from the transfer device 20 Determine (determined to be the same in step S82). Therefore, the LOC request unit 41 deletes the entry corresponding to the destination communication device 1 stored in the error notification from the transfer control unit 25 (step S83). Further, the LOC request unit 41 requests the information of the egress edge node used for communication to the destination communication device 1 by transmitting a LOC request message to the location management server 10 (step S84). On the other hand, if the combination of the identification number in error notification and the destination communication device 1 does not match the combination stored in the encapsulation unit 23, the LOC request unit 41 is under attack from a malicious third party. And the process ends (determined as different in step S82).

  In the above description, the case where the freshness of the transfer device table 52 is not used for the method for determining whether to transmit an error notification has been described as an example. However, also in the third embodiment, the freshness of information included in the transfer device table 52 may be used, as in the second embodiment. In this case, as described in the second embodiment, the ingress edge node includes a transfer device table 52 that includes a comparison result between the time when the egress edge node information is acquired and the elapsed time threshold value. Further, the encapsulated packet transmitted from the ingress edge node to the egress edge node includes a use period flag as in the encapsulated packet 9b of FIG. The egress edge node does not generate an error notification when the usage period flag = 0 by the same method as in the second embodiment.

  Furthermore, when the freshness of the information included in the transfer device table 52 is used in the third embodiment, the ingress edge node compares the time elapsed from the time when the egress edge node information was acquired with the elapsed time threshold comparison result. May be expressed by the presence or absence of identification information. That is, when the time elapsed from the time when the information of the egress edge node is acquired is equal to or greater than the elapsed time threshold, the egress edge node transmits a packet including an identification number to the egress edge node as in the encapsulated packet 9a of FIG. To do. On the other hand, when the time elapsed from the time when the information of the egress edge node is acquired is shorter than the elapsed time threshold, the egress edge node transmits the encapsulated packet 9a that does not include the identification number to the egress edge node. The ingress edge node can enter an invalid value as the identification number when the time elapsed from the time when the information of the egress edge node is acquired is shorter than the elapsed time threshold. If the outer header of the received encapsulated packet includes a valid identification number, the egress edge node notifies the transfer failure by sending an error notification including the identification number to the ingress edge node. On the other hand, when a valid identification number is not included in the outer header of the encapsulated packet, the egress edge node does not generate an error notification to the ingress edge node.

FIG. 21 is a diagram illustrating an example when a third person 90 attack is performed. In the example of FIG. 21, it is assumed that the destination communication device 1b is located in the access network A2. While knowing that the communication device 1b does not exist in the access network A3, the third party 90 transmits an encapsulated packet including the following information to the transfer device 20c belonging to the access network A3.
Egress edge node address: LOC3
Ingress edge node address: LOC1
Inner header destination address: IP2
Identification information: 0

  Then, since LOC3 fails to transfer the packet after decapsulation, it transmits an error notification to the transfer device 20a. When the received error notification does not include the combination of the identification information stored in the encapsulation unit 23 and the communication device 1, the transfer device 20a increases the processing load on the transfer device 20a by the malicious third party 90. It can be determined that an attack was made. Therefore, when the transfer device 20a determines that the error notification resulting from the attack of the malicious third party 90 has been received, the transfer device 20a does not transmit the LOC request message to the location management server 10 as illustrated in FIG. Therefore, the transfer device 20a can avoid a load of generating and transmitting a LOC request message in vain. Further, if the LOC request message is not transmitted, the location management server 10 also does not transmit the LOC response message to the transfer device 20a. Therefore, the transfer device 20a can avoid the load caused by the processing of the LOC response message. . Furthermore, since the location management server 10 does not receive the LOC request message from the transfer device 20a, the processing load on the location management server 10 is also avoided.

<Others>
The embodiment is not limited to the above, and can be variously modified. Some examples are described below.

  In the above description, the case where the transfer device 20 can operate as both an ingress edge node and an egress edge node has been described as an example. However, depending on the implementation, a transfer device that operates as an egress edge node but does not operate as an ingress edge node may be generated. In this case, the transfer device 20 may not include the error notification generation unit 26. Similarly, a transfer device that operates as an ingress edge node but does not operate as an egress edge node may be generated.

  In addition, the information of the communication device 1 included in the access network may be acquired not only by the information performed by the learning unit 31 but also by manual setting from the operator or notification from the location management server 10 or the like.

  In the third embodiment, the encapsulating unit 23 may store the egress edge node in combination with the identification number. In this case, the LOC request unit 41 compares the combination of the identification number and the egress edge node stored in the error notification with the combination of the identification number stored in the encapsulation unit 23 and the destination communication device 1.

  The portable storage medium 78 read by the location management server 10 may store programs and data that are operated by the transfer device 20. In this case, the location management server 10 can read the program or data recorded in the portable storage medium 78 and transmit it to the transfer device 20 as appropriate. The transfer device 20 can store the program and data received from the location management server 10 in the storage device 63. Further, the processor 61 can execute a program recorded in the storage device 63, and uses data recorded in the storage device 63 as appropriate. In this case, the transfer device 20 operates as an ingress edge node or an egress edge node when the processor 61 executes the program.

DESCRIPTION OF SYMBOLS 1 Communication apparatus 2 Edge node 3 Core network 5 LOC cache table 6 Packet 7, 8, 9 Encapsulated packet 10 Location management server 11, 21 Reception unit 12, 27 Transmission unit 13 Location information processing unit 14 LOC request processing unit 15 Location information Management table 20 Transfer device 22 Type determination unit 23 Encapsulation unit 24 Decapsulation unit 25 Transfer control unit 26 Error notification generation unit 30 Location information processing unit 31 Learning unit 32 Location information notification generation unit 40 LOC information processing unit 41 LOC request unit 42 LOC response processing unit 50 storage unit 51 path table 52 transfer device table 61, 71 processor 62 switch circuit 63 storage device 64, 75 bus 65 interface circuit 72 memory 73 input device 74 output device 76 external storage device 77 medium drive device 7 Portable storage medium 79 network connection device 80 network

Claims (9)

When the first transfer device receives a user packet addressed to the second communication device from the first communication device, the first transfer device has been used for transferring the user packet addressed to the second communication device. The user packet is encapsulated and transmitted to the second transfer device set as the end point of
When the second transfer device does not hold a route to the second communication device, the second transfer device notifies the first transfer device of the transfer failure of the user packet,
The first transfer device requests information on a transfer device belonging to a network in which the second communication device is located to a location management server that manages the location of the second communication device,
Through the second tunnel communication with the third transfer device notified from the location management server as an end point, the first transfer device transfers the user packet to the third transfer device,
The communication method, wherein the third transfer device transfers the user packet to the second communication device. The second transfer device interrupts communication with the second communication device when a period during which communication with the second communication device is not performed exceeds a threshold. Delete the route to the second communication device from the route table that records the route to the communication device that can be transferred by the second transfer device without notifying the transfer device of
When the route to the second communication device is not included in the route table, the second transfer device notifies the first transfer device of the transfer failure of the user packet. The communication method according to claim 1. The third transfer device includes:
Using the packet transmitted from the second communication device, the second communication device is specified to be included in the network to which the third transfer device belongs,
A location information notification for notifying a combination of a first identifier for identifying the second communication device and a second identifier for identifying the third transfer device is transmitted to the location management server;
When the location management server receives the location information notification, the location management server changes the information on the transfer device belonging to the network in which the second communication device is located from the second transfer device to the third transfer device. The communication method according to claim 1 or 2. The first transfer device uses the location information of the second communication device notified from the location management server as the elapsed time threshold, the time elapsed from the time when the location information of the second communication device was notified. Along with the comparison result obtained by the comparison, it is recorded in the transfer device table,
Adding the comparison result to the user packet and sending it to the second transfer device which is the end point of the first tunnel communication;
When the comparison result indicates that the time elapsed from the time when the location information of the second communication device is notified is shorter than the elapsed time threshold, the second transfer device indicates that the transfer of the user packet has failed. The communication method according to claim 1, wherein the first transfer device is not notified. The first transfer device adds an identification number for identifying the user packet transferred by the first transfer device to the user packet, and transfers the end point of the tunnel communication starting from the first transfer device. To the endpoint device
When the error notification for notifying the transfer error of the user packet includes the identification number, the location management server is requested for information on the transfer device belonging to the network where the second communication device is located,
When the error notification for notifying the transfer error of the user packet does not include the identification number, the location management server is not requested to request information on the transfer device belonging to the network in which the second communication device is located. The communication method according to any one of claims 1 to 4, characterized in that: The first transfer device identifies a user packet transferred by the first transfer device when a time elapsed from the time when the position information of the second communication device is notified is equal to or greater than the elapsed time threshold. An identification number is added to the user packet and transmitted to an end point device which is a transfer device at the end of tunnel communication starting from the first transfer device,
The end device is
If the received packet does not include the identification number, the user packet transfer failure is not notified to the first transfer device,
5. The communication method according to claim 1, wherein, when the received packet includes the identification number, the first transfer apparatus is notified of the transfer failure of the user packet. A second transfer device communicable with a first transfer device belonging to the network in which the first communication device is located,
A receiving unit that receives a packet encapsulating a user packet addressed to a second communication device from the first communication device by tunnel communication starting from the first transfer device;
A route table that records routes to transferable communication devices;
A transfer control unit that performs control to delete a route to the second communication device from the route table when a period during which communication with the second communication device is not performed exceeds a threshold;
If the route to the second communication device is not included in the route table, an error notification notifying that an error has occurred in the transfer to the second communication device is sent to the first transfer device. An error notification generator to generate as
A transfer apparatus comprising: a transmission unit that transmits the error notification to the first transfer apparatus. The transfer control unit is configured to delete the route to the second communication device from the route table because a period during which communication with the second communication device is not performed exceeds the threshold. The transfer device according to claim 7, wherein the first transfer device is not notified that communication with the second communication device is interrupted. A second transfer device capable of communicating with the first transfer device belonging to the network in which the first communication device is located;
Receiving a packet encapsulating a user packet addressed to a second communication device from the first communication device by tunnel communication starting from the first transfer device;
When a period during which communication with the second communication device is not performed exceeds a threshold, the route to the second communication device is deleted from the route table in which the route to the transferable communication device is recorded. ,
If the route to the second communication device is not included in the route table, an error notification notifying that an error has occurred in the transfer to the second communication device is sent to the first transfer device. Produces as
A communication program for executing a process of transmitting the error notification to the first transfer apparatus.