WO2008007805A2

WO2008007805A2 - Signaling-transmission managing method and relay node used therefor

Info

Publication number: WO2008007805A2
Application number: PCT/JP2007/064130
Authority: WO
Inventors: Takako Hori; Hong Cheng; Toyoki Ue; Shivanajay Marwaha
Original assignee: Panasonic Corporation
Priority date: 2006-07-11
Filing date: 2007-07-11
Publication date: 2008-01-17
Also published as: WO2008007805A3

Abstract

There is disclosed a technology to provide a signaling-transmission managing method or the like which is capable of continuing a link of a network, even if the link goes down, reducing burdens on a terminal end node, inhibiting wasting of a network resource and reducing burdens on a node between the terminal end node and a node where signaling gathers, and according to the technology, an upstream relay node transmits a first message indicating that the link has been disconnected to a first relay node, the first relay node transmits a second message indicating that information of a state stored in a correction node should be updated to the correction node, the correction node transmits a third message to update the states of relay nodes along a path from the correction node itself to a data reception node, and the relay node which has received a fourth message from the upstream relay node to the data reception node for a correction request of the correction node and the third message transmits a message to update the states of the relay nodes in a case where it is judged based on the messages that the node itself is a new correction node.

Description

DESCRIPTION

SIGNALING-TRANSMISSION MANAGING METHOD AND RELAY NODE USED THEREFOR

TECHNICAL FIELD

The present invention relates to a method of managing transmission of signaling messages to be transmitted in a communication network, and a relay node for use in the method.

BACKGROUND ART

As a new signaling protocol, Next Step in Signaling (NSIS) has been standardized by the NSIS working group (see Non-Patent Document 1 below) of IETF. The NSIS is expected to be especially effective in QoS resource reservation. According to the existing internet draft, the NSIS signaling message for QoS management passes through the same route as that of a data packet. The signaling message (hereinafter referred to also as the signaling) in an upstream direction (from a data reception side to a data transmission side) traces a router (QNE) which recognizes the same NSIS QoS in a downstream direction (from the data transmission side to the data reception side) . Signaling message and information on path in the same session is identified by a session ID, and signaling message and information on path in the same flow is identified by a flow ID (see Non-Patent Document 2 below) . The session ID does not change in a continuing period of a communication session. However, since the flow ID of the same session consists of header information (IP addresses, port numbers and the like of a transmitter and a destination) of the data packet, the flow ID changes with, for example, address change of a terminal end node. In a case where the QoS resource reservation is made on the path in the NSIS, it is assumed that information of the flow ID is used as filter information as information for identifying the QoS reservation for the data packet having any kind of the header information. For example, in Non-Patent Document 6 below, a method is disclosed in which a filter list is defined separately from the flow ID, so that a plurality of filter information can be stored in this list. For example, when data is downloaded using a file transfer protocol (FTP) , a client uses a plurality of ports, for example, ports 10001, 10002 and 10003 once. When the QoS resource is reserved in this data route, one of the plurality of port numbers, for example, only 10001 is adopted as the flow ID, and the filter information including the port numbers (10001, 10002 and 10003) can all be stored in the filter list.

On the other hand, in a large-scaled communication network, a data path of the session changes with a network management regulation such as load balancing. Here, FIG. 19 shows a conventional communication network. As shown in FIG. 19, when a data packet (hereinafter referred to simply as the packet) is transmitted from a data transmission node 20 to a data reception node 21, the packets to be transmitted to the same destination are output from different interfaces of router at random by the load balancing or the like, depending on situations of the communication network. Therefore, the packet flow is split into several packet flows at a part of an end- to-end data path (corresponding to QNE2 (a load balance initiator: LB-I) in FIG. 19). The existing signaling proposal having concepts of the session ID and the flow ID cannot be adapted to the above-mentioned case. The load balancing includes a method of distributing the packets based on the header information of the packets. In this case, the packet having the same header information is surely output from the same interface. However, even in a case where the packets have the same source and destination IP addresses, when the packets have different port numbers, the packets might be output from different interfaces. In this case, if the signaling message employs the filter list and a plurality of filter information are stored in the filter list, a path where the signaling message passes might be different from a path where the packet having, in a header thereof, the same information as the filter information stored in the filter list of this signaling message passes. The existing signaling proposal cannot cope with the above-mentioned case.

The existing signaling proposal does not support the signaling in a plurality of paths generated by the load balancing. Therefore, when the signaling message is sent, the signaling message passes through only one of the split paths at the path splitting part. Therefore, state management is not executed on the other paths. The existing signaling technology does not provide any method of appropriately managing the signaling at the path splitting part or a path coupling part halfway in the end-to-end path. As shown in, for example, FIG. 20, when the signaling message (here a query message) from a data transmission node 20' is duplicated in QNE2 (LB-I), a terminal-end node (a data reception node 21') receives the query message several times. As a result, QNE8, QNE9 disposed along a common path cannot appropriately respond to or support any mobility. The signaling load increases at a common path part.

In a technique "route pinning", the packet is subjected to compulsory processing in order to transmit the packet through a determined router. The packet correctly passes through the same route by this technique, and this also applies to the signaling. However, this method hampers network optimization.

The signaling message is duplicated and transmitted through all possible paths in order to incorporate (install, set) reserved states of all data paths generated by the load balancing. In this case, the session IDs and the flow IDs of all the paths are the same, because these IDs are end-to-end assigned. Therefore, it is considered that the path splitting nodes (CRN) are caused by change of the route. As a result, the state of the only one route is maintained. Non- Patent Document 3 below discloses a method of setting a "no replace" flag in order to maintain a plurality of states of one session. If this flag is set in the signaling message, the state incorporated beforehand is maintained, and a new state is added.

[Non-Patent Document 1] R. Hancock et al,"Next Steps in Signaling: Framework", RFC4080, June 2005

[Non-Patent Document 2] H. Schulzrinne et al, "GIMPS : General Internet Messaging Protocol for Signaling", draft-ietf-nsis-ntlp-07.txt, July 2005

[Non-Patent Document 3] J. Manner, Georgios Karagiannis, Andrew McDonald and Sven Van den Bosch"NSLP for Quality-of-Service signaling", draft-ietf-nsis-qos- nslp-07.txt, July 2005

[Non-Patent Document 4] T. Sanda, T. Ue and H.Cheng, "Path type support for NSIS signaling", draft- sanda-nsis-path-type-02. txt, February 2005

[Non-Patent Document 5] T. Sanda and T. Ue, "Pre CRN discovery from proxy on candidate new path", draft- sanda-nsis-mobility-qos-proxy-01. txt, February 2004

[Non-Patent Document 6] H. Cheng, Q. Huang, T. Sanda and T Ue, "NSIS Flow ID and packet classification issues" , draft-cheng-nsis-flowid-issues-01.txt, July 2005 [Non-Patent Document 7] Braden,R., "Resource reservation Protocol (RSVP) -Versionl Functional Specification", RFC2205, September 1997

However, in this case, if actual route change occurs, the CRN of the load balancing path cannot identify the path (the state) to be opened, because the session IDs and flow IDs of all the paths are the same. That is, in FIG. 19, since the route passing through a QNE6 is discontinued, the route is changed to a route passing through a QNElO, but the route passing through the QNE6 is left. Here, to identify a plurality of signaling paths of the same session, Non-Patent Document 4 proposes a pass type ID. The signaling messages of the same session are managed by use of the pass type ID, respectively. However, in the load balancing, the signaling is split at the split of the path halfway in the network. Moreover, the terminal end node receives all the signaling messages. Therefore, to reserve receiver initialization, the terminal end node has to send a plurality of reserve message for one session, and burdens are imposed on the node. Moreover, since a large number of signaling messages flow, a network resource is wasted. The QNEs (the nodes) disposed between the terminal end node and the QNE (the node) on which the split (duplicated) signaling messages are collected (converged) have to separately have the states of the paths, respectively, and burdens are imposed on the nodes Moreover, FIG. 21 shows a case where the QNE2 (LB-

I) performs the load balancing, depending on the header information, the signaling message (regarded as the RESERVE message) sent from a data transmission node 30 has a filter list and three pieces of filter information (filter 1, filter 2 and filter 3) are stored in the list. In the QNE2 (LB-I) , data having the same header information as that of the filter 1 is sent to a path passing through QNE4 and QNE6, and data having the same header information as those of the filter 2 and filter 3 is sent to a path passing through QNE3, QNE5. Here, if the RESERVE message is duplicated in the QNE (LB-I) and QoS reservations are equally made to the paths, an extra QoS resource is given to the path through which the data having the same header as that of the filter 1 passes, and a QoS resource to be given to the path through which the data having the same header information as that of the filter 2 and filter 3 passes falls short.

Furthermore, as shown in FIG. 12, in a case where a link and a router are down halfway in the route where the load balancing is performed and a change occurs in a data route, when the QoS reservation is not appropriately or quickly updated, the data packet passing through the changed route during the update cannot be given QoS guarantee .

DISCLOSURE OF THE INVENTION

The present invention has been developed in view of the above problem, and an object thereof is to provide a signaling-transmission managing method which is capable of reducing burdens on a terminal end node, inhibiting wasting of a network resource and reducing burdens on a node between the terminal end node and a node where signaling gathers and which is further capable of continuing a link halfway in a network, even if the link goes down (shutdown) for some reason, reducing burdens on a terminal end node, inhibiting wasting of a network resource and reducing burdens on a node between the terminal end node and a node where signaling gathers, and to provide a relay node for use in the method.

To achieve the above object, according to the present invention, there is provided a signaling- transmission managing method in a communication network including a data transmission node, a data reception node and a plurality of relay nodes which relay signaling to be transmitted from the data transmission node to the data reception node, in which, in a case where the communication network has a predetermined communicating situation, a first relay node positioned at a split point of paths of the communication network among the relay nodes duplicates global signaling received from the data transmission node to obtain duplicated signaling, includes predetermined information in the duplicated signaling and transmits the duplicated signaling to the plurality of split paths, respectively, and in which, in a case where a second relay node which receives the duplicated signaling among the relay nodes judges, based on the predetermined information included in the received duplicated signaling, that the node itself is a correction node to integrate and correct the duplicated signaling into one global signaling, the second relay node corrects the integrated duplicated signaling into the one global signaling to transmit the global signaling, the method comprising: a step in which, in a case where one or more links among a plurality of links connected to the communication network are disconnected and it is recognized that an upstream relay node which has been connected to the disconnected link and which is positioned on an upstream side where the duplicated signaling flows has been disconnected, the upstream relay node transmits a first message indicating that the link has been disconnected to the first relay node; a step in which the first relay node that has received the first message transmits to the correction node a second message indicating that information of a state stored in the correction node should be updated, when the link is disconnected; a step in which the correction node that has received the second message transmits to the data reception node a third message to update the state of the relay node existing on a route from the correction node to the data reception node based on the second message; and a step in which the relay node that has received a fourth message transmitted from the upstream relay node to the data reception node along a new route and indicating a request for correction of the correction node and the third message judges, based on the received third and fourth messages, whether or not the relay node itself is a new correction node and in which the relay node itself transmits a message to update the information of the state stored in the relay nodes, in a case where it is judged that the relay node is the new correction node. According to this constitution, the new correction node can easily be found. Moreover, according to the present invention, there is provided a signaling-transmission managing method in a communication network including a data transmission node, a data reception node and a plurality of relay nodes which relay signaling to be transmitted from the data transmission node to the data reception node, in which the signaling includes a list including information to indicate whether it is QoS reservation corresponding to a data packet having what kind of header information, in a case where the QoS reservation is made to a QoS corresponding node along a data route among the relay nodes, and in which a first relay node positioned at a split point of paths of the communication network among the relay nodes duplicates the signaling including the list to obtain duplicated signaling based on a load balancing policy set beforehand, includes predetermined information in the duplicated signaling and transmits the duplicated signaling to the plurality of split paths, respectively, and in which, in a case where a second relay node to receive the duplicated signaling among the relay nodes judges, based on the predetermined information included in the received duplicated signaling, that the relay node itself is a correction node to integrate and correct the duplicated signaling into one global signaling, the second relay node corrects the integrated duplicated signaling into the one global signaling to transmit the global signaling, the method comprising: a step in which, in a case where one or more links among a plurality of links connected to the communication network are disconnected and it is recognized that an upstream relay node which has been connected to the disconnected link and which is positioned on an upstream side where the duplicated signaling flows has been disconnected, the upstream relay node transmits a first message indicating that the link has been disconnected to the first relay node; a step in which the first relay node that has received the first message transmits to the correction node a second message indicating that information of a state stored in the correction node should be updated, when the link is disconnected; a step in which the correction node that has received the second message transmits to the data reception node a third message to update the state of the relay node existing on a route from the correction node to the data reception node based on the second message; and a step in which the relay node that has received a fourth message transmitted from the upstream relay node to the data reception node along a new route and indicating a request for correction of the correction node and the third message judges, based on the received third and fourth messages, whether or not the relay node itself is a new correction node and in which the relay node itself transmits a message to update the information of the state stored in the relay nodes, in a case where it is judged that the relay node is the new correction node. According to this constitution, the new correction node can easily be found.

Furthermore, according to the present invention, there is provided a signaling-transmission managing method in a communication network including a data transmission node, a data reception node and a plurality of relay nodes which relay signaling to be transmitted from the data transmission node to the data reception node, in which, in a case where the communication network has a predetermined communicating situation, a first relay node positioned at a split point of paths of the communication network among the relay nodes duplicates global signaling received from the data transmission node to obtain duplicated signaling, includes predetermined information in the duplicated signaling and transmits the duplicated signaling to the plurality of split paths, respectively, and in which, in a case where a second relay node which receives the duplicated signaling among the relay nodes judges, based on the predetermined information included in the received duplicated signaling, that the relay node itself is a correction node to integrate and correct the duplicated signaling into the one global signaling, the second relay node corrects the integrated duplicated signaling into the one global signaling to transmit the global signaling, the method comprising: a step in which, in a case where disconnection of a link is caused by the correction node in an upstream link connected to the correction node among a plurality of links connected to the communication network, an upstream relay node that is positioned on an upstream side of the correction node and that has been connected to the correction node transmits to the first relay node a first message indicating that the link connected to the correction node is not used; a step in which the first relay node that has received the first message transmits to the relay node positioned on an upstream side of the first relay node itself a third message indicating a request for transmission of a second message indicating that the correction node is changed based on the received first message; a step in which the relay node that has received the third message and that is positioned on the upstream side of the first relay node transmits the second message to the data reception node; and a step in which the relay node that is positioned on a downstream side of the first relay node and that has received a fourth message transmitted from the relay node connected to the first relay node to the data reception node along a new route and indicating a request for correction of the correction node and the second message judges, based on the received second and fourth messages, whether or not the relay node itself is a new correction node and in which the relay node itself transmits a message to update information of a state stored in the relay nodes, in a case where it is judged that the relay node is the new correction node. According to this constitution, the new correction node can easily be found.

In addition, according to the present invention, there is provided a signaling-transmission managing method in a communication network including a data transmission node, a data reception node and a plurality of relay nodes which relay signaling to be transmitted from the data transmission node to the data reception node, in which the signaling includes a list including information to indicate whether it is QoS reservation corresponding to a data packet having what kind of header information, in a case where the QoS reservation is made to a QoS corresponding node along a data route among the relay nodes and in which a first relay node positioned at a split point of paths of the communication network among the relay nodes duplicates the signaling including the list to obtain duplicated signaling based on a load balancing policy set beforehand, includes predetermined information in the duplicated signaling and transmits the duplicated signaling to the plurality of split paths, respectively, and in which, in a case where a second relay node to receive the duplicated signaling among the relay nodes judges, based on the predetermined information included in the received duplicated signaling, that the relay node itself is a correction node to integrate and correct the duplicated signaling into one global signaling, the second relay node corrects the integrated duplicated signaling into the one global signaling to transmit the global signaling, the method comprising: a step in which an upstream relay node that is positioned on an upstream side of the correction node and that has been connected to the correction node transmits to the first relay node a first message indicating that a link connected to the correction node is not used, in a case where disconnection of the link is caused by the correction node in an upstream link connected to the correction node among a plurality of links connected to the communication network; a step in which the first relay node that has received the first message transmits to the relay node positioned on an upstream side of the first relay node itself a third message indicating a request for transmission of a second message indicating that the correction node is changed based on the received first message; a step in which the relay node that has received the third message and that is positioned on the upstream side of the first relay node transmits the second message to the data reception node; and a step in which the relay node that is positioned on a downstream side of the first relay node and that has received a fourth message transmitted from the relay node connected to the first relay node to the data reception node along a new route and indicating a request for correction of the correction node and the second message judges, based on the received second and fourth messages, whether or not the relay node itself is a new correction node and in which the relay node itself transmits a message to update information of a state stored in the relay nodes, in a case where it is judged that the relay node is the new correction node. According to this constitution, the new correction node can easily be found.

Moreover, according to the present invention, there is provided a signaling-transmission managing method in a communication network including a data transmission node, a data reception node and a plurality of relay nodes which relay signaling to be transmitted from the data transmission node to the data reception node, in which, in a case where the communication network has a predetermined communicating situation, a first relay node positioned at a split point of paths of the communication network among the relay nodes duplicates global signaling received from the data transmission node to obtain duplicated signaling, includes predetermined information in the duplicated signaling and transmits the duplicated signaling to the plurality of split paths, respectively, and in which, in a case where a second relay node which receives the duplicated signaling among the relay nodes judges, based on the predetermined information included in the received duplicated signaling, that the relay node itself is a correction node to integrate and correct the duplicated signaling into the one global signaling, the second relay node corrects the integrated duplicated signaling into the one global signaling to transmit the global signaling, the method comprising: a step in which, in a case where disconnection of a link is caused by the correction node in a plurality of links on an upstream side connected to the correction node among a plurality of links connected to the communication network, upstream relay nodes that are positioned on an upstream side of the correction node and that have been connected to the correction node transmit to the first relay node first messages indicating that the links from the upstream relay nodes themselves to the correction node have been disconnected; a step in which the first relay node to receive the first messages transmits to the relay node positioned on an upstream side of the first relay node itself a third message indicating a request for transmission of a second message indicating that the correction node is changed, in a case where the first relay node receives the plurality of first messages; a step in which the relay node that has received the third message and that is positioned on the upstream side of the first relay node transmits the second message to the data reception node; and a step in which the relay node that is positioned on a downstream side of the first relay node and that has received a fourth message transmitted from the relay node connected to the first relay node to the data reception node along a new route and indicating a request for correction of the correction node and the second message judges, based on the received second and fourth messages, whether or not the relay node itself is a new correction node and in which the relay node itself transmits a message to update information of a state stored in the relay nodes, in a case where it is judged that the relay node is the new correction node. According to this constitution, the new correction node can easily be found.

Furthermore, according to the present invention, there is provided a signaling-transmission managing method in a communication network including a data transmission node, a data reception node and a plurality of relay nodes which relay signaling to be transmitted from the data transmission node to the data reception node, in which the signaling includes a list including information to indicate whether it is QoS reservation corresponding to a data packet having what kind of header information, in a case where the QoS reservation is made to a QoS corresponding node along a data route among the relay nodes and in which a first relay node positioned at a split point of paths of the communication network among the relay nodes duplicates the signaling including the list to obtain duplicated signaling based on a load balancing policy set beforehand, includes predetermined information in the duplicated signaling and transmits the duplicated signaling to the plurality of split paths, respectively, and in which, in a case where a second relay node to receive the duplicated signaling among the relay nodes judges, based on the predetermined information included in the received duplicated signaling, that the relay node itself is a correction node to integrate and correct the duplicated signaling into one global signaling, the second relay node corrects the integrated duplicated signaling into the one global signaling to transmit the global signaling, the method comprising: a step in which, in a case where disconnection of a link is caused by the correction node in a plurality of links on an upstream side connected to the correction node among a plurality of links connected to the communication network, upstream relay nodes that are positioned on an upstream side of the correction node and that have been connected to the correction node transmit to the first relay node first messages indicating that the links from the upstream relay nodes themselves to the correction node have been disconnected; a step in which the first relay node to receive the first messages transmits to the relay node positioned on an upstream side of the first relay node itself a third message indicating a request for transmission of a second message indicating that the correction node is changed, in a case where the first relay node receives the plurality of first messages; a step in which the relay node that has received the third message and that is positioned on the upstream side of the first relay node transmits the second message to the data reception node; and a step in which the relay node that is positioned on a downstream side of the first relay node and that has received a fourth message transmitted from the relay node connected to the first relay node to the data reception node along a new route and indicating a request for correction of the correction node and the second message judges, based on the received second and fourth messages, whether or not the relay node itself is a new correction node and in which the relay node itself transmits a message to update information of a state stored in the relay nodes, in a case where it is judged that the relay- node is the new correction node. According to this constitution, the new correction node can easily be found. In addition, in the signaling-transmission managing method of the present invention, it is a preferable configuration of the present invention that the second message includes information indicating that duplication is prohibited. According to this constitution, unnecessary processing can be avoided. Moreover, according to the present invention, there are provided a relay node for use in a signaling- transmission managing method in a communication network including a data transmission node, a data reception node and the plurality of relay nodes which relay signaling to be transmitted from the data transmission node to the data reception node, the relay node comprising: reception means for receiving duplicated signaling obtained by- duplicating global signaling to be transmitted from the data transmission node to the data reception node by another relay node; judgment means for judging whether or not the relay node itself is a correction node which integrates and corrects the duplicated signaling into the one global signaling based on predetermined information included in the received duplicated signaling; correction means for correcting the duplicated signaling into the one global signaling, in a case where it is judged that the relay node is the correction node; and transmission means for transmitting the one corrected global signaling, wherein, in a case where one or more links among a plurality of links connected to the communication network are disconnected and an upstream relay node which has been connected to the disconnected link and which is positioned on an upstream side where the duplicated signaling flows recognizes that the links have been disconnected, the reception means receives a second message indicating that information of a state stored in the correction node should be updated and transmitted, when the links are disconnected, by a first relay node which has received a first message indicating that the link has been disconnected and transmitted by the upstream relay node, the correction means generates a third message to update the state of the relay node existing along a route from the relay node itself to the data reception node based on the received second message, and the transmission means transmits the generated third message to the data reception node. According to this constitution, a new correction node can easily be found.

Furthermore, according to the present invention, there are provided a relay node for use in a signaling- transmission managing method in a communication network including a data transmission node, a data reception node and the plurality of relay nodes which relay signaling to be transmitted from the data transmission node to the data reception node, the relay node comprising: reception means for receiving duplicated signaling obtained by duplicating, by another relay node, global signaling transmitted from the data transmission node to the data reception node and including a list including information to indicate whether it is QoS reservation corresponding to a data packet having what kind of header information, in a case where the QoS reservation is made to a QoS corresponding node along a data route among the relay nodes; judgment means for judging whether or not the relay node itself is a correction node which integrates and corrects the duplicated signaling into the one global signaling based on predetermined information included in the received duplicated signaling; correction means for correcting the duplicated signaling into the one global signaling, in a case where it is judged that the relay node is the correction node; and transmission means for transmitting the one corrected global signaling, wherein, in a case where one or more links among a plurality of links connected to the communication network are disconnected and an upstream relay node which has been connected to the disconnected link and which is positioned on an upstream side where the duplicated signaling flows recognizes that the links have been disconnected, the reception means receives a second message indicating that information of a state stored in the correction node should be updated and transmitted, when the links are disconnected, by a first relay node which has received a first message indicating that the link has been disconnected and transmitted by the upstream relay node, the correction means generates a third message to update the state of the relay node existing along a route from the relay node itself to the data reception node based on the received second message, and the transmission means transmits the generated third message to the data reception node. According to this constitution, a new correction node can easily be found. In addition, according to the present invention, there are provided a relay node for use in a signaling- transmission managing method in a communication network including a data transmission node, a data reception node and the plurality of relay nodes which relay signaling to be transmitted from the data transmission node to the data reception node, the relay node comprising: reception means for receiving global signaling to be transmitted from the data transmission node to the data reception node; duplication means for, in a case where the relay node itself is positioned at a split point of paths of the communication network and the communication network has a predetermined communicating situation, duplicating the global signaling received by the reception means to obtain duplicated signaling as many as paths to be split; and transmission means for including predetermined information in the duplicated signaling and transmitting the duplicated signaling to the plurality of split paths, respectively, wherein, in a case where disconnection of a link is caused by a correction node in an upstream link connected to the correction node among a plurality of links connected to the communication network, the reception means receives a first message transmitted by an upstream relay node which is positioned on an upstream side of the correction node and which has been connected to the correction node and indicating that the link connected to the correction node is not used, the duplication means generates a third message indicating a request for transmission of a second message indicating that the correction node is changed based on the received first message, and the transmission means transmits the generated third message to the relay node which is positioned on an upstream side of the relay node itself. According to this constitution, the new correction node can easily be found.

Moreover, according to the present invention, there are provided a relay node for use in a signaling- transmission managing method in a communication network including a data transmission node, a data reception node and the plurality of relay nodes which relay signaling to be transmitted from the data transmission node to the data reception node, the relay node comprising: reception means for receiving global signaling transmitted from the data transmission node to the data reception node and including a list including information to indicate whether it is reservation corresponding to a data packet having what kind of header information, in a case where the QoS reservation is made to a QoS corresponding node along a data route among the relay nodes; duplication means for duplicating the global signaling to obtain duplicated signaling based on a load balancing policy set beforehand, in a case where the relay node is positioned at a split point of paths of the communication network; and transmission means for including predetermined information in the duplicated signaling and transmitting the duplicated signaling to the plurality of split paths, respectively, wherein, in a case where disconnection of a link is caused by a correction node in an upstream link connected to the correction node among a plurality of links connected to the communication network, the reception means receives a first message transmitted by an upstream relay node which is positioned on an upstream side of the correction node and which has been connected to the correction node and indicating that the link connected to the correction node is not used, the duplication means generates a third message indicating a request for transmission of a second message indicating that the correction node is changed based on the received first message, and the transmission means transmits the generated third message to the relay node which is positioned on an upstream side of the relay node itself. According to this constitution, the new correction node can easily be found.

Furthermore, according to the present invention, there are provided a relay node for use in a signaling- transmission managing method in a communication network including a data transmission node, a data reception node and the plurality of relay nodes which relay signaling to be transmitted from the data transmission node to the data reception node, the relay node comprising: reception means for receiving global signaling to be transmitted from the data transmission node to the data reception node; duplication means for, in a case where the relay node itself is positioned at a split point of paths of the communication network and the communication network has a predetermined communicating situation, duplicating the global signaling received by the reception means to obtain duplicated signaling as many as paths to be split; and transmission means for including predetermined information in the duplicated signaling and transmitting the duplicated signaling to the plurality of split paths, respectively, wherein, in a case where disconnection of a link is caused by a correction node in a plurality of links on an upstream side connected to the correction node among a plurality of links connected to the communication network, the reception means receives first messages transmitted by upstream relay nodes which are positioned on an upstream side of the correction node and which have been connected to the correction node and indicating that the links from the upstream relay nodes themselves to the correction node have been disconnected, the duplication means generates a third message indicating a request for transmission of a second message indicating that the correction node is changed, in a case where it is judged that the plurality of first messages are received, and the transmission means transmits the generation third message to the relay node positioned on the upstream side of the relay node itself. According to this constitution, the new correction node can easily be found. In addition, according to the present invention, there are provided a relay node for use in a signaling- transmission managing method in a communication network including a data transmission node, a data reception node and the plurality of relay nodes which relay signaling to be transmitted from the data transmission node to the data reception node, the relay node comprising: reception means for receiving global signaling transmitted from the data transmission node to the data reception node and including a list including information to indicate whether it is QoS reservation corresponding to a data packet having what kind of header information, in a case where the QoS reservation is made to a QoS corresponding node along a data route among the relay nodes; duplication means for duplicating the global signaling to obtain duplicated signaling based on a load balancing policy set beforehand, in a case where the relay node is positioned at a split point of paths of the communication network; and transmission means for including predetermined information in the duplicated signaling and transmitting the duplicated signaling to the plurality of split paths, respectively, wherein, in a case where disconnection of a link is caused by a correction node in a plurality of links on an upstream side connected to the correction node among a plurality of links connected to the communication network, the reception means receives first messages transmitted by upstream relay nodes which are positioned on an upstream side of the correction node and which have been connected to the correction node and indicating that the links from the upstream relay nodes themselves to the correction node have been disconnected, the duplication means generates a third message indicating a request for transmission of a second message indicating that the correction node is changed, in a case where it is judged that the plurality of first messages are received, and the transmission means transmits the generation third message to the relay node positioned on the upstream side of the relay node itself. According to^v this constitution, the new correction node can easily be found.

Moreover, in the relay node of the present invention, it is a preferable configuration of the present invention that the second message includes information indicating that duplication is prohibited. According to this constitution, unnecessary processing can be avoided.

The signaling-transmission managing method and the relay nodes for use in the method of the present invention have the above-mentioned constitutions, can reduce burdens on a terminal end node, can inhibit wasting of a network resource, and can reduce burdens on a node between the terminal end node and a node on which signaling gathers. Furthermore, even if the link halfway in the network goes down (shutdown) for some reason, the link can be continued to reduce the burdens on the terminal end node, inhibit the wasting of the network resource and reduce the burdens on the node between the terminal end node and the node on which signaling gathers,

BRIEF DESCRIPTION OF THE DRAWINGS

FIG. 1 is a schematic diagram showing a constitution of a communication network according to a first embodiment of the present invention;

FIG. 2 is a constitution diagram showing a constitution of a relay node according to the first embodiment of the present invention;

FIG. 3 is a constitution diagram showing a constitution of another relay node according to the first embodiment of the present invention;

FIG. 4 is a sequence chart showing a sequence of transmission management of signaling according to the first embodiment of the present invention;

FIG. 5 is a schematic diagram of a constitution of a communication network showing a signaling-transmission managing method according to a second embodiment of the present invention;

FIG. 6 is a diagram of conventional message communication showing the signaling-transmission managing method according to the second embodiment of the present invention;

FIG. 7 is an explanatory view of a transmission method of a reserve message in the signaling-transmission managing method according to the second embodiment of the present invention;

FIG. 8 is a schematic diagram showing a constitution of a communication network according to a third embodiment of the present invention;

FIG. 9 is an explanatory view of a first path detecting method of a load-balanced flow according to the third embodiment of the present invention; FIG. 10 is an explanatory view showing a second path detecting method of the load-balanced flow according to the third embodiment of the present invention;

FIG. 11 is an explanatory view showing a third path detecting method of the load-balanced flow according to the third embodiment of the present invention;

FIG. 12 is a schematic diagram showing a constitution of a communication network according to a fourth embodiment of the present invention; FIG. 13 is a sequence chart showing a sequence of finding a new correction node according to the fourth embodiment of the present invention;

FIG. 14 is a sequence chart showing a processing sequence of an LB-T state update message by a signaling processable node (QNE7) according to the fourth embodiment of the present invention;

FIG. 15 is a sequence chart showing a sequence of a signaling message according to a fifth embodiment of the present invention; FIG. 16 is a flow chart showing one example of a message processing logic at a signaling processable node according to the fifth embodiment of the present invention;

FIG. 17 is a schematic diagram showing a constitution of a communication network according to a sixth embodiment of the present invention; FIG. 18 is a schematic diagram showing a constitution of a communication network according to a seventh embodiment of the present invention;

FIG. 19 is a diagram showing a conventional communication network;

FIG. 20 is a diagram showing a flow of signaling in the conventional communication network; and

FIG. 21 is a diagram showing a flow of signaling including a filter list in the conventional communication network.

BEST MODE FOR CARRYING OUT THE INVENTION <First Embodiment> A first embodiment of the present invention will hereinafter be described with reference to FIGS. 1 to 4. FIG. 1 is a schematic diagram showing a constitution of a communication network according to a first embodiment of the present invention. FIG. 2 is a constitution diagram showing a constitution of a relay node according to the first embodiment of the present invention. FIG. 3 is a constitution diagram showing a constitution of another relay node according to the first embodiment of the present invention. FIG. 4 is a sequence chart showing a sequence of transmission management of signaling according to the first embodiment of the present invention. First, the constitution of the communication network according to the first embodiment of the present invention will be described with reference to FIG. 1. The communication network includes a data transmission node 100, a data reception node 101 and QNEl to QNE9 (relay nodes) . In the following, it is assumed that signaling messages at split points of data paths (hereinafter referred to simply as the paths) of the communication network is duplicated by road balancing of the communication network. It is to be noted that actual duplication of the signaling messages at the split points of the paths may be performed for another cause such as network management or policy.

As shown in FIG. 1, the data transmission node 100 transmits a signaling message (hereinafter referred to simply as the signaling) 102 (here, also referred to as the "global signaling message (an original message transmitted from the data transmission node 100)") to the data reception node 101. When the transmitted signaling 102 reaches the QNE2 (LB-I), the QNE2 as a split point of the path duplicates the signaling 102 by the load balancing, and transmits the duplicated signaling a and signaling b to paths of the QNE3 and the QNE4. The duplicated signaling a, b include information (e.g., a path type ID) to identify the paths through which the signaling messages pass. The duplicated signaling also include information (e.g., information of split bits or the like) indicating that "this signaling is the signaling due to the load balancing", and information (e.g., information of a "full value" and a "divided value" described later) by which the QNE (a signaling processable node) that has received the duplicated signaling judges whether or not the node itself is a correction node described later.

The information of the "full value" and the "divided value" determines whether or not the QNE is a node having a function of merging all duplicated signaling into one signaling. Assuming that the path is split into "N" paths at a path split point, original signaling is duplicated into "N" signaling messages. These signaling messages have a divided value "ai", and a full value "S" is represented by the following equation

(D :

[Equation 1]

The above-mentioned "divided value" and "full value" may have any form as long as the above equation is satisfied. For example, the values may have forms of an integer and a total integer, forms of different bits or the like of a field and the whole field. The QNE2 generates a local Qspec (indicating a degree at which QoS is desired) value to be applied to the load balancing. Information of the generated local Qspec value (represented by a rate or the like) and information of a Qspec value of the whole communication network may be included in the duplicated signaling. It is to be noted that the above "full value" and "divided value" may be represented based on the generated local Qspec value. It is to be noted that any form of the above information may be incorporated in the signaling. For example, the "divided value" and the "full value" may be separate parameters or one parameter.

Here, when one of the duplicated signaling, for example, the signaling a reaches the QNE3, the QNE3 checks information indicating that "this signaling is the signaling due to the load balancing". At this time, when the QNE3 has states of the same session ID and flow ID and different path type IDs, the QNE3 adds the "divided value" held (accumulated) in a predetermined storage region at a time when the signaling was received to the "divided value" included in the signaling a, and checks whether or not the total is the "full value". In this example, the QNE3 does not have states of the same session ID and flow ID and different path type IDs. ^• Therefore, the divided values are not totaled to form the full value, and the message is forwarded without any change, after the QNE3 stores the "divided value" to the predetermined storage region. The same processing is performed in QNE5, 7, 8, 9 and the data reception node 101.

Next, when the signaling b reaches the QNE4 and QNE6, processing similar to the above processing is performed. When the signaling b reaches the QNE7 (load balance terminator: LB-T), the QNE7 has already processed the signaling a. Therefore, when the signaling b reaches, the QNE7 has states of the same session ID and flow ID and the different path type IDs. The QNE7 adds the

"divided value" held (accumulated) in the predetermined storage region at the time when the signaling b was received to the "divided value" included in the signaling b, and checks whether or not the total is the "full value". When the total is the "full value", the QNE7 recognizes that the node itself is the LB-T (the correction node) , and the QNE7 corrects the signaling b into the global signaling message (the signaling) 102. Specifically, all the information added by the QNE2 (the LB-I) is removed from the signaling b. If the local Q spec value is included, the value is removed. At this time, the QNE7 sends corrected signaling 102' to the data reception node 101. It is to be noted that the QNE7 stores mapping information of the divided state and the whole state in the predetermined storage region. When the QNE8 and the QNE9 receive the signaling 102', the QNE8, 9 replace all the states of the same session ID and flow ID with new states.

Next, a constitution of the relay node according to the first embodiment of the present invention will be described with reference to FIGS. 1 and 2. The relay node shown in FIG. 2 is the relay node corresponding to the QNE2 of FIG. 1 described above. That is, the relay node is positioned at the split point of the communication network, and duplicates the signaling transmitted from the data transmission node by the load balancing or the like. Here, it is assumed that such a relay node is a first relay node 200. As shown in FIG. 2, the first relay node 200 includes reception means 201, transmission means 202, duplication means 203, numeric value generation means 204, identification information generating means 205 and storage means 206.

The reception means 201 receives the signaling, a data packet and the like transmitted between the data transmission node 100 and the data reception node 101. The transmission means 202 transmits the received data packet to another QNE, or transmits the signaling duplicated by the duplication means 203 described later to another QNE. In this case, the transmission means 202 includes information generated by the numeric value generation means 204 and the identification information generating means 205 described later. The above information indicates that "this signaling is the signaling due to the load balancing", and the above mentioned local Qspec value is included in the duplicated signaling. When the reception means 201 receives the signaling from the data transmission node 100 to the data reception node 101, the duplication means 203 duplicates the signaling for each of the split paths created by the load balancing or the like, because the first relay node 200 is positioned at the split point. For example, when there are N paths split from the first relay node 200, N signaling messages are duplicated.

The numeric value generation means 204 generates information such as the above "full value" and "divided value" for the QNE that has received the duplicated signaling to judge whether or not the node itself is the above correction node. The identification information generating means 205 generates information such as the above path type ID to identify the path through which the duplicated signaling passes. In the storage means 206, a control program for controlling an operation of the first relay node 200 and information necessary for the operation are stored. Also in the storage means 206, information and the like generated when the first relay node 200 performs processing are stored. Next, a constitution of another relay node according to the first embodiment of the present invention will be described with reference to FIGS. 1 and 3. The relay node shown in FIG. 3 corresponds to the relay nodes QNEl, QNE3 to QNE9 of FIG. 1 described above. Here, such relay nodes will be described as a second relay node 300. As shown in FIG. 3, the second relay node 300 includes reception means 301, transmission means 302, judgment means 303, correction means 304, and storage means 305. It is to be noted that here the relay node is divided into the first relay node 200 and the second relay node 300, but the present invention may be carried out as one relay node integrally having functions of the relay nodes. In consequence, arrangement of the relay nodes does not have to be considered.

The reception means 301 receives the signaling, the data packet and the like to be transmitted between the data transmission node 100 and the data reception node 101. The transmission means 302 transmits the received data packet to another QNE, or transmits the signaling 102' corrected by the correction means 304 described later to the data reception node 101. The judgment means 303 judges whether or not the second relay node 300 itself is a correction node based on predetermined information included in the received duplicated signaling. Specifically, as described above, the previously received "divided value" of the signaling due to the load balancing, held (stored) in the storage means 305 or the like, is added to the divided value included in the received duplicated signaling. When the total equals the "full value", it is judged that the node is the correction node. In a case where the judgment means 303 judges that the second relay node 300 is the correction node, the correction means 304 corrects the duplicated signaling into one global signaling. Specifically, the signaling b is corrected into the signaling 102'. That is, the correction means 304 removes the information added by the first relay node 200 from the signaling b to obtain the signaling 102'. On the other hand, in a case where it is judged that the second relay node 300 is not the correction node, the transmission means 302 transfers the received signaling as it is, and the correction means 304 adds (updates) the divided value included in the signaling to the divided value previously stored in the storage means 305. In the storage means 305, a control program for controlling an operation of the second relay node 300 and information necessary for the operation are stored. Also in the storage means 305, information (e.g., the above information of the divided value) and the like generated in a case where the second relay node 300 performs processing are stored. Next, a sequence of signaling-transmission management according to the first embodiment of the present invention will be described with reference to FIG. 4. First, the data transmission node 100 transmits the signaling 102 to the QNEl in a direction toward the data reception node 101 (step S401) . Subsequently, the QNE2 which has received the signaling 102 from the QNEl and which is positioned at the split point of the paths of the communication network duplicates the signaling 102 into signaling a and signaling b by the load balancing (step S402). Moreover, the duplicated signaling a and signaling b are allowed to include information (e.g., the path type ID) with which paths where the signaling messages pass are identified, the information (the information of the split bits or the like) indicating that "this signaling is the signaling due to the load balancing", the information (e.g., the information of the "full value" and the "divided value") by which the QNE that has received the duplicated signaling judges whether or not the node itself is the correction node, the information (represented by a rate or the like) of the generated local Qspec value, and the information of the Qspec value of the whole communication network (step S403) . The signaling a and the signaling b are transmitted to the QNE3 and the QNE4 (step S404) respectively. Subsequently, for example, on first receiving the signaling a, the QNE7 which receives the signaling a and the signaling b duplicated by the QNE2 and transmitted via the QNE3, 5 and QNE4, 6 checks the information included in the signaling a. When it is judged that the signaling is the signaling due to the load balancing, it is checked whether or not an divided value of the load balancing from another path (e.g. associated with the same session ID and flow ID, but different path type ID) exist in the storage means 305 of the node itself. If any, the stored divided value is added to the divided value included in the signaling a (step S405) . Moreover, the QNE7 judges whether or not the added value forms the full value (step S406) . In this case, since the received signaling a is the first received signaling, no divided value is stored in the storage means 305 of the QNE7. Moreover, since the divided value of the signaling a is not the full value, the QNE7 transmits the signaling a to the QNE8 as it is (step S407) . It is to be noted that at this time, the QNE7 stores the divided value included in the signaling a in the storage means 305 of the node itself.

Subsequently, on receiving the signaling b transmitted via the QNE4, 6, the QNE7 checks the information included in the signaling b. When it is judged that the signaling is the signaling due to the load balancing, it is checked whether or not an divided value of the load balancing from another path (e.g. associated with the same session ID and flow ID, but different path type ID) exist in the storage means 305 of the node itself. In this case, since the divided value of the previously received signaling a is stored, the divided value stored in the storage means 305 is added to the divided value of the signaling b (step S408). Moreover, it is judged whether or not the added value forms the full value (step S409) . In this case, since the full value is reached, the QNE7 removes the information added by the QNE2 from the signaling b, and transmits the resulted signaling 102' toward the data reception node 101 (step S410) .

It is to be noted that the QNE3 to 6 between the QNE2 and the QNE7 perform processing similar to that of the QNE7. The QNE8, 9 between the QNE7 and the data reception node 101 update a state owing to the signaling 102'. In consequence, unlike a conventional technology, the data reception node 101 does not have to transmit a reserve message for reservation for each of the received signaling, and burdens are reduced. Here, only the case where the QNE7 first receives the signaling a has been described, but the operation could be carried in the same manner when signaling b is first received. <Second Embodiment> Next, a signaling-transmission managing method according to a second embodiment of the present invention will be described with reference to FIGS. 5 to 7. FIG. 5 is a schematic diagram of a constitution of a communication network showing the signaling-transmission managing method according to the second embodiment of the present invention. FIG. 6 is a diagram of conventional signaling transmission showing the signaling-transmission managing method according to the second embodiment of the present invention. FIG. 7 is an explanatory view of a transmission method of a reserve message in the signaling-transmission managing method according to the second embodiment of the present invention.

In the signaling-transmission managing method according to the second embodiment of the present invention, it is assumed that the data reception node 101 of the first embodiment is a mobile node (MN) . It is to be noted that a constitution of a QNE group including a data transmission node and relay nodes is basically similar to that of the first embodiment, and therefore description of the constitution is omitted. The signaling-transmission managing method according to the second embodiment will hereinafter be described with reference to FIGS. 5 to 7.

First, when an MN 501 determines to move to another sub-net, the MN 501 requests a QNE13 (hereinafter referred to also as the proxy) to fine a CRN and make a QoS reservation in advance as disclosed in Non-Patent Document 5. As shown in FIG. 5, when a data transmission node 500 transmits an UCRN_DISCOVER message (hereinafter referred to also as the signaling) disclosed in Non- Patent Document 5, a QNE2 duplicates the signaling, and a QNEIl corrects the signaling as described in the first embodiment. The QNEs compare not only a session ID and a flow ID but also information of a distinguishable path (information such as a path type ID) . The QNE only carry out crossover node actions when all IDs of the QNE agree with IDs included in the signaling. It is to be noted that, unlike a conventional technology, if the QNE (the QNEIl) to correct the duplicated signaling cannot be discovered, a plurality of signaling messages flow into the proxy even after movement of the MN 501 as shown in FIG. 6. In consequence, the proxy has to transmit a plurality of reserve messages, and burdens are imposed.

The proxy obtains two CRNs (a QNE3 and a QNE4) by load balancing. The CRN holds the information of the path type ID. When the proxy transmits the reserve message to the data transmission node 500, the proxy transmits only one reserve message (a global reserve message) including (the information of) the CRN having the path type ID as shown in FIG. 7. When the reserve message reaches the QNEIl, the reserve message is duplicated so that all load-balanced paths can transmit signaling for QoS reservation. In FIG. 7, the message is duplicated into two messages. The duplicated reserve messages include a local Qspec, the path type ID and the (information of the) CRN. The local Qspec and the path type ID are obtained from information accumulated beforehand in the QNEIl (in the same manner as in the first embodiment) .

When the duplicated reserve message reaches the CRN (the QNE3 and the QNE4), the CRN performs processing as disclosed in Non-Patent Document 4. In the processing, an old path is removed, and the reserve message is corrected in order to update a path between the CRN (the QNE3 and the QNE4) and the data transmission node 500. When all the duplicated reserve messages reach LB-I (the QNE2), the LB-I corrects (merges) the reserve messages, generates the global reserve message, and transmits the generated global reserve message to the data transmission node 500.

It is to be noted that, to support mobility, the pass type ID of each of the split paths needs to be unique to the session. In consequence, in the mobility, the CRNs of the individual paths are appropriately found. Since the paths may split several times between an LB-I (the QNE2) and an LB-T (the QNE7 or the QNEIl), the path type ID needs to be used so as to guarantee uniqueness of the path type ID. For example, the path is split into three paths at a first split point, and path type IDs of Nl, N2 and N3 are assigned to these split three paths. At another split point, the path having the path type ID: Nl is further split into other two paths, and path type IDs of Ml, M2 are assigned to the split two paths. Another path having the path type ID: N2 is split into three paths at another split point, and path type IDs of Kl, K2 and K3 are assigned to the split three paths. It is to be noted that Nl, N2, N3, Ml, M2, Kl, K2 and K3 need to be all different. There are many methods of identifying (varying) Nl, N2, N3, Ml, M2, Kl, K2 and K3. For example, the path type ID is represented by an integer of 16 bits, and Nl, N2 and N3 are generated by setting 001, 010 and 100 to first three bits of 16 bits, respectively. At this time, Ml, M2 are generated by setting 01, 10 to the next two bits, respectively. That is, Ml and M2 have 00101, 00110 in first five bits, respectively. Similarly, when the integer is similarly assigned to another path, Kl, K2 and K3 have 010001, 010010 and 010100 in first six bits, respectively. In consequence, all the path type IDs are different. It is to be noted that another form of the path type identification may be used as long as the types can mutually be identified. <Third Embodiment> A third embodiment of the present invention will hereinafter be described with reference to FIG. 8. FIG. 8 is a schematic diagram showing a constitution of a communication network according to the third embodiment of the present invention. First, the constitution of the communication network according to the third embodiment of the present invention will be described with reference to FIG. 8. The communication network includes a data transmission node 800, a data reception node 801, and a QNEl to a QNE9 (relay nodes) . In the following, it is assumed that a signaling message is duplicated at a split point of data paths (hereinafter referred to simply as the paths) of the communication network owing to load balancing and that the load balancing is performed depending on header information.

Moreover, it is assumed that the signaling message (e.g., the reserve message) sent from the data transmission node 800 has a filter list and that the list has a plurality of filter information (a filter 1, a filter 2, and a filter 3 in FIG. 8) . In the QNE2 (an LB- I), data having the same header information as that of the filter 1 is sent to a path passing through the QNE4, the QNE6, and data having the same header information as those of the filter 2 and filter 3 is sent to a path passing through the QNE3, the QNE5.

As shown in FIG. 8, the data transmission node 800 transmits a signaling message (hereinafter referred to simply as the signaling) 802 (here referred to also as a "global signaling message") to the data reception node 801. When the transmitted signaling 802 reaches the QNE2 (the LB-I), the QNE2 as a path split point duplicates the signaling 802 by the load .balancing and transmits duplicated signaling a and signaling b to the paths of the QNE3 and the QNE4. In this case, contents of the filter list of the global signaling message are checked against the load balancing policy, e.g., which interface transmits data of the header as indicated by the filter information. Moreover, the filter list is reconstructed according to the load balancing policy. That is, filter information 2 and 3 are included in the filter list of the duplicated signaling a, and filter information 1 is included in the filter list of the duplicated signaling b. Moreover, when each filter list is reconstructed, a local Qspec may be generated, and attached to each duplicated signaling message. The duplicated signaling a, b include information (e.g., a path type ID) by which the path where each signaling passes is identified. The signaling also includes information (e.g., information of split bits or the like) indicating that "this signaling is the signaling due to the load balancing", and information (e.g., information of a "full value" and a "divided value") by which the QNE (a signaling processable node) that has received the duplicated signaling judges whether or not the node itself is a correction node.

The information of the "full value" and the "divided value" determines whether or not the QNE is a node having a function of merging all duplicated signaling into one signaling in the same manner as in the first embodiment. In addition to the examples of these "divided value" and "full value" described in the first embodiment, the values can be realized by using the filter information. That is, the filter information included in the original global signaling message is the "full value", and the filter information included in the reconstructed filter list is the "divided value". A example of how such information is used will hereinafter be described. Here, when one duplicated signaling, for example, the signaling a reaches the QNE3, the QNE3 checks the information indicating that "this signaling is the signaling due to the load balancing". At this time, when the QNE3 has states such as the same session ID and flow ID and different path type IDs, the QNE3 adds the

"divided value" held (accumulated) in a predetermined storage region at a time when the signaling was received, that is, the number of the filters included in the filter list to the "divided value" included in the signaling a, that is, the number of the filters included in the filter list, and checks whether or not the total agrees with the "full value", that is, the number of the filters included in the filter list of the global signaling message. In this example, the QNE3 does not have information such as the same session ID and flow ID and the different path type IDs. Therefore, the divided values are not totaled to form the full value. Processing to transfer the signaling a without any change is performed in the QNE5, 7, 8, 9 and the data reception node 801.

Next, when the signaling b reaches the QNE4 and the QNE6, processing similar to the above processing is performed. When the signaling b reaches the QNE7 (an LB- T), the QNE7 has already processed the signaling a. Therefore, it has stored divided value information of the same session ID and flow ID and the different path type IDs as that of signaling b. The QNE7 adds the "divided value" held (accumulated) in the predetermined storage region at the time when the signaling b was received, that is, the number of the filters included in the filter list, to the "divided value" included in the signaling b, that is, the number of the filters included in the filter list of signaling b, and checks whether or not the total is the "full value", that is, the number of the filters included in the filter list of the global signaling message. When the total is the "full value", the QNE7 recognizes that itself is the LB-T (the correction node) , and the QNE7 corrects the signaling b into a global signaling message (the signaling) 802'.

Specifically, all the information added by the QNE2 (the LB-I) is removed from the signaling b. If the local Qspec value is included, the value is removed. The filter list is reconstructed into the original form to be held by the global signaling message (in a case where the filter list of the original global signaling message is included in the signaling b, this list may be used, instead of reconstructing the list) . At this time, the QNE7 stores mapping information of the divided state and the whole state in the predetermined storage region. When the QNE8 and the QNE9 receive the signaling 802', the QNE8, 9 replace all the states of the same session ID and flow ID with new states. It is to be noted that, in a case where the LB-I uses the filter list of the global signaling message as it is in the duplicated signaling message instead of reconstructing the filter list and appropriately distributes QoS resources to the paths, only the local Qspec may be prepared. In this case, the number of the filter lists cannot be used as the "full value" and the "divided value". Therefore, other elements may be used as the "full value" and the "divided value".

It is to be noted that, in a case where the signaling message has the filter list, there is a method in which the LB-I does not divide the filter list or the Qspec but the data transmission node divides the filter list and the Qspec beforehand. In this case, the data transmission node needs to know the path through which the load-balanced flow passes. In consequence, the signaling can be merged into one signaling which passes through the same path. Three types of methods are considered as a method of detecting the path of the load- balanced flow.

In a first method, as shown in FIG. 9, first a data transmission node 900 transmits signaling messages (e.g., QUERY messages) to a data reception node 901 by use of flow IDs corresponding to three flows, and a QNE transmits to the data transmission node 900 signaling (e.g., a response message) indicating that the QNE (a QNE2 in FIG. 9) has detected that the signaling path is split. For example, in FIG. 9, in a case where the Ql is first transmitted from the data transmission node 900, an intermediate node does not have any state with respect to the same session ID, and no split is detected. In addition, the information of the flow ID and the session ID of Ql is stored in the QNEl, 2, 4, 6, 7, 8 and 9. Subsequently, when Q2 is sent from the data transmission node 900, the QNE2 compares the information with information stored at a time when Ql was sent to detect the split. Moreover, information indicating that "Q2 is transmitted in a direction different from that of Ql" is transmitted to the data transmission node 900. Subsequently, in a case where Q3 is sent, the QNE2 similarly transmits information indicating that "Q3 is transmitted in a direction different from that of Ql" to the data transmission node 900. In a case where intermediate QNEl to 9 do not transmit this signaling, the data transmission node 900 interprets that all flows pass through the same path.

Moreover, when there is a plurality of splits, for example, in FIG. 9, Ql, Q2 and Q3 are sent in this order from the data transmission node 900, and the QNE2 sends the messages in different directions, respectively. In this case, the QNE2 which has detected the split of Q3 transmits information indicating that "Q3 is also transmitted in a direction different from that of Ql and Q2" to the data transmission node 900. The signaling transmitted by the data transmission node 900 includes the total number of the transmitted signaling messages and an index number. In a case where the data reception node 901 detects that all signaling messages (QUERY messages) have been received, signaling (a response message) indicating this is transmitted to the data transmission node 900. In consequence, the data transmission node 900 can decide if the same route to the data reception node 901 is used by the different flows. In a second method, as shown in FIG. 10, a data transmission node 1000 transmits the signaling by use of the flow IDs corresponding to three flows in the same manner as in the first method. These signaling messages include other flow ID information as a payload. For example, the signaling corresponding to the flow IDl includes flow ID2 and flow ID3 information. The QNE (a QNE2 in FIG. 10) which has detected that another flow splits from the flow of the present signaling based on this flow ID information deletes the flow ID information corresponding to the split flow from the payload. On receiving all signaling messages (query messages) , a data reception node 1001 can detect from the flow ID information included in the payload whether or not two arbitrary flows pass through the same path or different paths. Detected flow split information is transmitted toward the data transmission node 1000 (a RESPONSE message) . To detect that the data reception node 1001 receives all signaling messages, the signaling transmitted by the data transmission node 1000 includes the total number of the transmitted signaling messages and the index number in the same manner as in the first method. In consequence, the data transmission node 1000 can judge a combination of the flows of the same session which passes through the same route to the data reception node 1001.

In a third method, as shown in FIG. 11, a data transmission node 1100 transmits signaling to a data reception node 1101 only once by use of any one flow ID. This signaling includes all flow ID information as a payload. A QNE (a QNE2 in FIG. 11) which has detected split of flows based on this flow ID information generates new signaling for each split flow. The payload of the new signaling includes flow ID information of the flow which passes through the path.

In FIG. 11, the QNE2 transmits the signaling including the flow ID information corresponding to filter 2 and filter 3 as the payload to a path of QNE3 and QNE5, and transmits the signaling including the flow ID information corresponding to a filter 1 as the payload to a path of QNE4 to QNE6. The splitting QNE includes the total number of the split and index number into the generated signaling message. In consequence, the data reception node 1101, and in turn the data transmission node 1100 can judge if a combination of the flows of the same session passes through the same route to the data reception node 1101.

An example of traffic in one direction has been described above, but the present invention can be carried out for traffic in two directions. In this case, the LB- T and the LB-I have mutual functions, and the functions need to be performed in a reverse direction. <Fourth Embodiment> In a fourth embodiment, a method of solving problems which might occur in the above first and subsequent embodiments will be described with reference to FIG. 12. An example of the first embodiment will hereinafter be described. As shown in FIG. 12, a QNE2 (an LB-I) duplicates signaling 102, and duplicated signaling a and signaling b are transmitted to paths of a QNE3 and a QNE4. It is to be noted that the signaling a and the signaling b include information of a "split bit", information of a "full value" and a "divided value" and the like. However, when the LB-T is found, excessive elements included in the signaling are removed from the signaling in a QNE7. Moreover, the QNE7 transmits an update message toward a data reception node 101 in order to remove information on split in a signaling processable node (a QNE8, a QNE9) .

As shown in FIG. 12, for example, when a link 1200 connecting the QNE3 to a QNE5 goes down in the QNE3 as one QNE, a new path (e.g., a path passing through the QNE3, a QNElO and the QNE8) is formed, and the link is connected to the data reception node 101 via the QNE8. In this case, to create resource reservation along the new path, local correction signaling is made by the node (the QNE3) which comes in contact with the shutdown link. Since the new path is a part of a load balanced branch, the signaling has the information of the "split bit", the information of the "full value" and the "divided value" and the like. However, as shown in FIG. 12, the new LB-T is the QNE8, and the QNE8 is disposed ahead of the QNE7, and does not have the information of the "split bit" or the like. In this case, the QNE8 cannot recognize that the node itself is a new LB-T, and cannot execute necessary processing such as converging of the signaling in different paths or updating of an original path. To solve the above problem, processing will hereinafter be described with reference to FIG. 13. As shown in FIG. 13, when the link 1200 between the QNE3 and the QNE5 is not connectable, a message cannot be sent through the link 1200. The non-connectable state is caused by, for example, shutdown of the link, congestion of the link, a management problem of a network and the like.

In a usual soft state based signaling function, periodic soft state signaling is sent to a signaling path in order to guarantee of the liveliness of the path (the link). When the link 1200 goes down, as shown in FIG. 13, soft-state signaling (hereinafter referred to simply as the soft-state) cannot pass (step S1301) . In this case, as shown in a step S1303 of FIG. 13, a soft-state timer of the QNE5 as the signaling processable node in an adjacent downstream direction times out. The QNE5 transmits state update (LB-T state update) of the LB-T down in a downstream direction (e.g., toward a data reception node 101) by use of time-out as a trigger. This LB-T state update includes the following information (e.g., besides other signaling information such as a session ID and a flow ID) . The information includes a path type ID, a LB-T reactivation flag and information of a full value and a divided value.

The path type ID is a path type ID used in a route between the QNE3 and the QNE5 before the link 1200 goes down. The "full value" is a value set by the LB-I (a QNE2), and the "divided value" is a value of the route between the QNE3 and the QNE5 before the link 1200 goes down. FIG. 14 shows a processing sequence of a LB-T state update message by a QNE7 as a signaling processable node. The signaling processable node receives the LB-T state update message (LB-T State Update) shown in step S1305 of FIG. 13 (step S1401), and the node then checks whether the node itself is the LB-T (step S1403) . This check is performed by checking, for example, the "LB-T reactivation flag" and a signaling state of the node itself.

When the current node is not the LB-T, the node transfers the LB-T state update message in the downstream direction (step S1405) . When the current node is the LB-T, for example, the QNE7, a LB-T state is updated (step S1407) . The corresponding "divided value" is removed from, for example, stored state information, and the "split bit" or the like is set. It is obvious for any person skilled in the art that the LB-T executes a special operation based on a signaling scheme and a local policy. Specifically, it is obvious for any person skilled in the art that a local Qspec needs to be updated. After processing the LB-T state update message, the LB-T (the QNE7) transmits a LB reactivation message toward the data reception node 101 (step S1409) . This message updates all signaling processable nodes, for example, the QNE8 in order to support finding of the next LB-T. The information included in the LB reactivation message is information of a split bit, a path type ID, a full value and a divided value.

The above-mentioned "path type ID" is a "path type ID" updated by the QNE7. For example, it (path type ID) is obtained by setting a corresponding bit of the "path type ID" of the received LB-T state update message. The "full value" is a "full value" set by the LB-I (the QNE2), and the "divided value" is obtained by subtracting the "divided value" of the received LB-T state update message from the "full value".

It is to be noted that it is obvious for any person skilled in the art that the LB reactivation message includes information necessary for a normal signaling operation, such as information of a session ID, a flow ID and the like. The signaling aware node, for example, the QNE8 receives the LB reactivation message (LB Re-activation) (step S1307), and updates its state. Specifically, the "split bit" is set, the "full value" and the "divided value" are installed, and the "path type ID" is updated. Moreover, the QNE8 forwards the message toward the data reception node 101 (step S1309) .

At this time, when an upstream node of the link 1200, for example, the QNE3 transmits a local repair message, for example, a LB repair request (LB Repair Request) toward the data reception node 101 (step S1311) , the request is transmitted via a new node, for example, a QNElO.

The information included in the LB repair request is information of a pass type ID, a "split bit", a full value and a divided value. In this case, the "path type ID" is a "path type ID" used in a route passing through the QNE3 and the QNE5 before the link 1200 goes down. The "full value" is a "full value" set by the LB-I (the QNE2), and the "divided value" is a "divided value" of a route passing through the QNE3 and the QNE5 before the link 1200 goes down. It is obvious for any person skilled in the art that the LB repair request could be a local repair signaling message sent by the QNE3, for example, the signaling message in the solution described in the first and second embodiments.

On receiving the LB repair request, the signaling aware node, for example, the QNElO executes regular processing as described in the solution of the first embodiment. Specifically, a corresponding state is generated by the "path type ID", and the "full value" and the "divided value" are stored together with the "split bit".

When it is decided that the node, for example, the QNElO is not the LB-T after executing the processing, for example, the total of the "divided values" is not equal to the "full value", the LB repair request (LB Repair Request) is forwarded in the downstream direction (step S1313) .

When it is decided that the node, for example, the QNE8 itself is the LB-T after the processing, for example, the total of the "divided values" is equal to the "full value", the state is updated, and it is indicated that the node itself is the LB-T. Subsequently, the QNE8 generates and transmits a LB repair response (LB Repair Response) to the node which has made a local repair request, for example, the QNE3 (step S1315). Moreover, an intermediate node, for example, the QNElO transfers the LB repair response (LB Repair Response) to the QNE3

(step S1317) .

It is obvious for any person skilled in the art that the LB repair response could be a local repair signaling message of the existing scheme, for example, the signaling message in the solution described in the first and second embodiments.

Moreover, the new LB-T, for example, the QNE8 generates a LB deactivation message (LB De-activation) , and transmits the message to the data reception node 101

(step S1319) . This message removes, from the QNE along the path, state information of load balancing, for example, the "split bit", the "full value", the "divided value", the corresponding "path type ID" and the like.

Other information such as the local Qspec is removed and replaced with a global Qspec.

The new LB-T, for example, the QNE8 transmits a LB update message (LB Update) in another direction via an old path of the QNE7 and the QNE5 in order to remove the corresponding state information from these QNEs (steps

S1321, S1323) .

Next, the relay node (e.g., the QNE7) according to the fourth embodiment of the present invention will be described. Since the relay node (the QNE7 ) according to the fourth embodiment is basically constituted in the same manner as in the relay node (the QNE7) according to the first embodiment, the relay node (the QNE7) according to the fourth embodiment will be described here with reference to FIG. 3. The QNE7 includes reception means 301, transmission means 302, judgment means 303, correction means 304, and storage means 305.

In a case where it is judged that the node itself is the correction node and then one or more links of a plurality of links connected to a communication network are disconnected, the reception means 301 receives a first message (LB-T Sate Update) transmitted by a relay node (e.g., the QNE5) on a downstream side where duplicated signaling of the disconnected link flows. The first message is indicating that a state of the correction node should be updated. Moreover, the correction means 304 processes the first message received by the reception means 301 to generate a second message (LB Re-activation) . The transmission means 302 transmits the second message generated by the correction means 304 to the data reception node 101.

Next, the relay node (the relay node (e.g., the QNE8) constituting a new correction node) according to the fourth embodiment of the present invention will be described. Since the relay node (the QNE8) according to the fourth embodiment has a basic constitution similar to that of the relay node (the QNE7) according to the first embodiment, the relay node (the QNE8) according to the fourth embodiment will be described here with reference to FIG. 3. The QNE8 includes reception means 301, transmission means 302, judgment means 303, correction means 304 and storage means 305.

The reception means 301 receives the second message (LB Re-activation) transmitted by the QNE7. The reception means 301 receives a third message (LB Repair Request) transmitted from the QNE3. When the "divided value" included in the received second message and the "divided value" included in the third message are combined to form the "full value", the judgment means 303 judges that the node itself is the new correction node. In a case where it is judged that the node itself is the new correction node, the correction means 304 generates the above-mentioned LB Update and LB De-activation. The transmission means 302 transmits the generated message.

Next, the relay node (e.g., the QNE5) according to the fourth embodiment of the present invention will be described. Since the relay node (the QNE5) according to the fourth embodiment has a basic constitution similar to that of the relay node (the QNE7) according to the first embodiment, the relay node (the QNE5) according to the fourth embodiment will be described here with reference to FIG. 3. It is to be noted that the QNE5 will be described here as one typical example, but even other relay node can similarly be described. The QNE5 includes reception means 301, transmission means 302, judgment means 303, correction means 304, and storage means 305. The judgment means 303 judges whether or not the link connected to the node itself is disconnected. It is to be noted that specifically the judgment means 303 recognizes whether or not the link is disconnected based on the reception of the signaling (Soft-State) to guarantee the liveliness of the link. As described above, if the above-mentioned Soft-State does not present even after the elapse of a predetermined time, it is judged that the link has been disconnected. In a case where it is judged that the connected link has been disconnected, the correction means 304 generates a first message indicating that the state of the correction node (the QNE7) should be updated. The transmission means 302 transmits the generated first message to the data reception node 101.

It is to be noted that it is preferable that the relay node arranged in this communication network have all the above-mentioned functions of the relay nodes. In this case, the arrangement of the relay nodes does not have to be considered. In the fourth embodiment, the first embodiment has been described as one example, but this embodiment is similarly considered even when applied to the second and third embodiments. <Fifth Embodiment>

In a fifth embodiment, a solution different from that of the fourth embodiment will be described with reference to FIG. 12 used in the description of the fourth embodiment. It is to be noted that the first embodiment will hereinafter be described as one example. As shown in FIG. 12, a case where a link 1200 between a QNE3 and a QNE5 goes down (is disconnected) and cannot be used at a certain time will be considered. When the link 1200 goes down, a new route needs to be constructed. In a case where a network supports a local repair signaling message, for example, a scheme (see Non-Patent Document 7) similar to a resource reservation protocol (RSVP) , the QNE3 in an upstream direction waits for a period, and transmits a signaling message toward a data reception node 101 in order to find a new path and reserve the corresponding resource. However, with the above load balancing scheme, the local repair signaling message transmitted by the QNE3 is not sufficient in locating the proper paths.

For example, as shown in FIG. 12, the selected path directly passes through a QNElO toward a QNE8. Moreover, the path does not pass through a QNE7 as a correction node before disconnection of the link 1200. In this case, since the QNE3 is disposed along a split path of a session, the signaling message transmitted from the QNE3 includes a split bit, a path type ID, a full value and a divided value in addition to usual signaling information. However, the QNE8 as the new correction node locates beyond the QNE7 as the previous correction node. Therefore, the QNE8 does not have information such as the split bit or the path type ID. The local repair signaling message transmitted from the QNE3 via the QNElO causes an unexpected behavior such as deletion of a state installed in other path. To solve this problem, the previous correction node QNE7 or QNE5 checks (judges) the presence of Soft State (whether or not Soft State can be checked in a predetermined time) . When a periodic new message of Soft State does not arrive, a LB-T in a downstream direction, for example, the QNE7 starts redetection of a common path in finding the correction node. This is performed by the QNE7 which transmits an update message to the data reception node 101 in order to set, for example, the split bit, the corresponding path type ID, the full value, the divided value and the like. The QNE8 as the node along the common path prepares for reception of the local repair signaling message by use of this information, and an appropriate action is carried out.

The above solving method is performed owing to timeout of Soft State, and therefore has a defect. The timeout of Soft State is corrected based on characteristic of local connection. Therefore, control of local repair does not necessarily become successful. The timeout of Soft State is usually 30 seconds on a regular link. Needless to say, the local repair signaling message may have already reached the QNE8 before the QNE7 detects the timeout of Soft State. On the other hand, when a timer for local repair is set too short, an unnecessary overhead on the network is imposed. Another problem concerning the above solution is that the local repair is started voluntarily by the QNE3. This does not necessarily influence actual decision of the LB-I, for example, the QNE2. After a link event such as the link 1200 goes down, the LB-I might decide that traffic is moved to a path which does not include any QNE3. In this case, the local repair signaling message started by the QNE3 is troublesome, and affects correct processing of load balancing.

Moreover, details of the fifth embodiment will be described with reference to FIG. 15. FIG. 15 shows a sequence of a signaling message applied to the fifth embodiment. When a link event such as shutdown, congestion or the like of the link occurs in the link 1200, the QNE3 is notified through data plane indication (step S1501) . In the network, this data plane indication may be of a different form such as an Internet Control Message Protocol (ICMP) message, Transmission timeout, routing protocol indication or the like. It is obvious for any person skilled in the art that the form of the data plane indication does not influence essence of the present invention. The data plane indication is usually used as a trigger of traffic rerouting in the network. After the node in the upstream direction, for example, the QNE3 or the LB-I (the QNE2), receives the data plane indication, the traffic is changed to another route. Therefore, this is one of the fastest methods to send path coupled control signaling to a substitute path to be selected.

When the QNE3 receives the data plane indication, it needs to be determined whether or not the path to be selected can be used by the node itself. For example, when the QNE3 has only one link such as the only link

1200 in the downstream direction, it is considered that local route change (repair) cannot be performed with the current relay node. Moreover, whether or not to perform the local repair in the QNE3 depends on a network policy. For example, when the LB-I (the QNE2) decides that the traffic is changed to a different path, the QNE3 transmits an instruction (LB Reroute Notification) toward a neighboring LB-I (the QNE2) in the upstream direction (step S1503) . In the LB reroute notification, the QNE3 indicates a property of the link event, whether or not the local repair is possible, and information of load balancing control. Moreover, the QNE3 may indicate information of a signaling state at this time, for example, the split bit, the path type ID, the full value, the divided value and the like. The LB reroute notification notifies the

LB-I (the QNE2) of the information of the signaling state of the load balancing control. When the LB-I (the QNE2) stores the state for each route (stateful) , some part of the information in the signaling message is optional, and relevant information can be obtained from LB-I' s stored state about the route. However, when the state for each route is not stored in the LB-I (stateless) or an only part of the state is stored (reduced state) , the information in the signaling message is necessary. When the state information is included, the LB reroute notification can be sent to the LB-I through the link capable of using the LB reroute notification, for example, a path faster than a usual signaling path. This especially serves, when an off path signaling scheme is used in a reverse direction or signaling optimization is usable.

One example of the LB reroute notification message will hereinafter be described.

LB Reroute Notification: = Session ID; Flow ID;

Link .event type; [Local repair option] ; [Path type ID] ; [Split bit] ; [Full value] ; and r [Divided value] .

Here, the "link event type" indicates the events occurred on the link in the downstream direction, for example, an event on the link 1200. Examples of this event include link shutdown and link congestion. This allows the node that decides the load balancing to determine data traffic routing.

The "local repair option" indicates possibility of using local repair selection in the current node, for example, the QNE3. Examples of the option include "local repair possible" and "local repair impossible".

The remaining elements of the information are obtained from a signaling state of the link 1200 stored in the QNE3. Examples of the remaining elements of the information include the "session ID", the "flow ID", the "path type ID", the "full value" and the "divided value".

It is obvious for any person skilled in the art that additional information can be included in the LB reroute notification message. Some of optional fields of the message may be omitted based on a network arrangement configuration.

The LB reroute notification is processed by the signaling aware node of the path until the notification reaches the LB-I. In a hierarchical load balancing support network, traffic change might occur in a LB-I other than a LB-I closest to the QNE3 in the upstream direction. This depends on, for example, network policy decision. In this case, the LB reroute notification is forwarded until the appropriate LB-I is identified.

FIG. 16 shows one example of a message processing logic at the signaling aware node. When the node receives the LB reroute notification (step S1601), the node checks whether or not the "local repair option" is indicated in the message (step S1603) . When the LB reroute notification does not indicate the local repair at the node in the downstream direction, the current node checks the possibility of local repair in the network polity and the local state. That is, it is checked whether or not the local repair can be supported (step S1605) . When the local node cannot support the local repair owing to the state of the node or the network policy, the corresponding LB reroute notification is updated, and transferred to the node in the upstream direction (step S1611). The update of the LB reroute notification depends on the state information of the local node. For example, when the current node is also the LB-I, the corresponding "path type ID", "full value" and "divided value" and the like need to be updated. In a case where the node sees in the step S1603 that the LB reroute notification indicates the local repair, it is checked whether or not the local node is the LB-I (step S1607) . When this node is the LB-I, transfer of the LB reroute notification is completed, and further processing is executed by the LB-I (step S1609) . If not, the LB reroute notification is updated and forwarded toward the node in the upstream direction (step S1611) . In a case where it is judged in the step S1605 that the local node can perform the local repair, it is checked whether or not the local node is the LB-I (step S1607) . When the local node is the LB-I, the transfer of the message ends at the node, and appropriate processing is performed by the LB-I (step S1609) . If not, the node adds the "local repair option" to an appropriate field such as the message to update the message, and transfers the message in the upstream direction (step S1611) . The "local repair option" includes, for example, an address of the local node for the LB-I in the upstream direction to locate a place of the local repair.

It is obvious to any person skilled in the art that only main processing steps have been described above in the message processing logic. When the present invention is applied, other processing of the signaling message might be performed. Examples of this processing include processing of a usual signaling message such as signaling state update. These additional processing steps do not influence the essence of the processing of the present invention. When the appropriate LB-I is identified and the corresponding LB reroute notification is received, as shown in FIG. 15, the LB-I such as the QNE2 transmits a LB update request toward the LB-T in a downstream direction (the data reception node 101) through a path different from a path through which the LB reroute notification has been received (step S1505) . The path for use in the transmission of the LB update request before reaching the LB-I is based on the information of the local node and the network control policy. For example, the LB-I can select the path to support the best QoS or the path which does not have any delay.

The LB update request includes information of the path to which a problem link such as the link 1200 belongs. Examples of the information include the "path type ID" and the "divided value" of the path where the link 1200 exists. One example of a format of the LB update request will hereinafter be described. LB Update Request: = Session ID;

Flow ID; [Path type ID] ;

[Divided value] ; [Full value]; and [Update option] .

Here, the "path type ID" and the "divided value" are the information of the path where the link 1200 exists. The "update option" indicates a processing instruction from the LB-I to the LB-T. Examples of the instruction include a "reroute type" and a "local repair type". Use of these types will be described later.

This LB update request is directed to the LB-T such as the LB-T that joins the new path and the original path where the link 1200 exists. Therefore, usual nodes such as the QNE4 and the QNE6 only transfer the message without performing any detailed processing (steps S1507, S1509) . It is determined whether or not the LB-T is an appropriate LB-T to process the message based on the

"path type ID" included in the message. In a case where the current stored state includes a value of the "path type ID" of the message, it is meant that the LB-T is a LB-T to process the message. If not, the LB-T transfers the message in the downstream direction toward the data reception node until the message reaches a correct LB-T such as the QNE7.

When the LB-T such as the QNE7 receives the LB update request, execution of necessary processing for the path in the downstream direction is started. For example, as shown in FIG. 15, the QNE7 transmits LB-T update toward the data reception node 101 in the downstream direction (step S1511) . The LB-T update sets the "split bit" to all of the intermediate nodes disposed halfway toward the data reception node 101. The message sets the corresponding "full value", "divided value" and "path type ID" to each node. The LB-T sets the value, depending on the "update option" set to the LB update request .

For example, when the "update option" is of the "reroute type", the LB-T does not use the "full value", "divided value" and "path type ID" stored in the LB-T update. The last LB-T disposed along the path sends signaling to the data reception node 101 with the "divided value" equal to the "full value" and the "path type ID" including IDs of all the paths. However, when the "update option" is of the "local repair type", the divided value is obtained by subtracting the "divided value" of the LB update request from the "divided value" stored in the LB-T. Moreover, the "path type ID" is obtained by subtracting the "path type ID" of the LB update request from the "path type ID" stored in the LB-T, One example of a format of the LB-T update will hereinafter be described.

LB-T Update: = Session ID; Flow ID;

[Split bit set] ; [Path type ID] [Full value] ; and [Divided value] .

When all signaling aware nodes to set split bits along the path recognize a "split bit set" of the LB-T update, the corresponding state information will be stored and forwarded to the data reception node 101. In consequence, these nodes prepare for local repair message processing. The LB-T such as the QNE7 transmits LB Update Ack to the LB-I such as the QNE2 which has transmitted the LB update request (step S1515) . This message notifies the LB-I of an update status. Since this LB Update Ack is directly returned to the QNE2, a reliable path which can quickly transmit the message can be used.

Moreover, when the received LB update request includes an update option of the "local repair type", the LB-T such as the QNE7 transmits old path update toward a path indicated by the "path type ID" of the LB update request (step S1519) . This message is transferred by the node in the upstream direction before reaching the link 1200. In this case the reservation over the old path could be removed faster.

After receiving LB Update Ack, the LB-I such as the QNE2 transmits local repair indication toward the node indicated by the "local repair option" of the LB reroute notification (step S1517). This is directly transmitted to the node or through a usual hop-by-hop style. One example of a format of the local repair indication will hereinafter be described. Local Repair Indication: = Session ID;

Flow ID; [Path type ID] ; [Divided value] ; and [Repair option] . The "repair option" includes policy information from the LB-I such as the QNE2, and is concerned with potential local repair. For example, the LB-I can indicate that the local repair should be performed after certain delay. When a local repair node such as the QNE3 receives this local repair indication, the local repair is executed. As shown in FIG. 15, the local repair is transmitted by the QNE3 (step S1523) toward the node in the downstream direction (step S1525). A signaling aware node such as the QNElO at the path installs a corresponding state according to information in the local repair signaling. The local repair includes information on an original path (a path before disconnected) . One example of a format of the local repair will hereinafter be described. Local Repair: = Session ID;

Flow ID; [Path type ID] ; [Split bit] ; [Full value]; and [Divided value] . Here, the "path type ID", the "full value" and the

"divided value" were used for the original path where the link 1200 was on.

The local repair is processed by the signaling aware node according to a usual processing procedure. For example, when the QNE8 as a path convergence node receives the message, the state is updated according to the message. For example, the "divided value" of the message is added to the stored "divided value", or the stored "path type ID" and the "path type ID" of the message are integrated. Subsequently, the QNE8 compares the total of new "divided values" with the full value. If these values are equal, it is meant that the new LB-T has been determined. If not, the node transfers the message with the updated "divided value" and "path type ID" toward the data reception node 101.

When the LB-T is recognized as the last LB-T, for example, the QNE8, it transmits the LB-T update with an option of "split bit release" toward the data reception node 101 (step S1527). This message removes the "split bit", "full value", "divided value" and "path type ID" from a state of the signaling aware node on the path between the LB-T and the data reception node 101.

As an option, the LB-T such as the QNE8 transmits local repair response toward a local repair node such as the QNE3 (step S1529) . This message is transferred to the QNE3 as acknowledgment of local repair processing (step S1531) .

Next, the relay node (the QNE7) according to the fifth embodiment will be described. The relay node (the QNE7) according to the fifth embodiment has a basic constitution similar to that of the relay node (the QNE7) according to the first embodiment. Here, the relay node (the QNE7) according to the fifth embodiment will be described with reference to FIG. 3. The QNE7 includes reception means 301, transmission means 302, judgment means 303, correction means 304, and storage means 305. In a case where one or more links of a plurality of links connected to a communication network are disconnected and an upstream relay node connected to the disconnected link (the link 1200) detects the disconnection, the reception means 301 receives a second message (LB Update Request) transmitted by a first relay node (the QNE2) that has received a first message (LB Reroute Notification) indicating the link has been disconnected. The second message is indicating that information of a state stored in the correction node should be updated. The correction means 304 generates a third message (LB-T Update) to update the state of the relay node existing along a path from the node itself (the QNE7 ) to the data reception node 101 based on the received second message. The transmission means 302 transmits the generated third message toward the data reception node 101.

It is to be noted that the first embodiment has been described as one example in the fifth embodiment, but even the second and third embodiments are similarly described.

A sixth embodiment will be described with reference to FIG. 17. FIG. 17 is a schematic diagram showing a constitution of a communication network according to the sixth embodiment. As shown in FIG. 17, a link event such as disconnection of a link might occur at a path convergence node (a correction node) . In this case, both of links 1713 and 1715 cannot be used. Therefore, LB update request described in the fifth embodiment does not reach a LB-T such as a QNE7.

There are two methods to solve this problem. In a first method, a node in an upstream direction of the LB-T stores state information. This is achieved by setting a flag indicating that "the node in a downstream direction is the LB-T" during path finding processing of a signaling scheme. For example, when the QNE7 as the LB-T is found during the signaling, nodes such as a QNE5 and a QNE6 in an upstream direction of both paths are notified so as to set the corresponding flag. In this method, when the QNE5 notices the link event by the QNE7, it is seen that the same LB-T cannot be reached even via other path.

Therefore, a special "upper level LB-I" flag is included in LB reroute notification. In consequence, when the LB update request is transmitted, an appropriate LB-I such as a QNEl is selected by signaling processing. Moreover, a "path type ID" of the other path connected to the same LB-T, for example, the QNE7 may be included so that the LB-I such as a QNE2 which has received the LB reroute notification notices that the indicated path should be avoided in selecting the path. Therefore, the QNE2 further transmits the message to the LB-I such as the QNEl in the upstream direction. Moreover, instead of transmitting the LB update request from the QNE2, the message is transmitted from the QNEl via a QNEIl through paths 1701, 1703.

In a second method of this problem solving method, relevant information is stored in the LB-I. For example, when the QNE7 goes down, it is noticed that it is difficult to access both of the links 1713 and 1715. Therefore, both of the QNE5 and the QNEβ receive data plane indication. This is a trigger for two separate LB reroute notifications directed to the LB-I such as the QNE2. The LB-I such as the QNE2 stores a state of the transmitted LB reroute notification. Therefore, when the second LB reroute notification reaches the QNE2, it is seen that there is not any path capable of transmitting the LB update request from this LB-I (the QNE2 ) . Therefore, the QNE2 updates the LB reroute notification by adding, for example, a "divided value" and a "path type ID", and transmits the notification to the LB-I such as the QNEl in the upstream direction. At this time, the QNEl transmits the LB update request based on new information of a path to be selected (e.g., the paths 1701, 1703 passing through the QNEIl) .

Next, the relay node (the QNE2) in the above first method according to the sixth embodiment will be described. Since a basic constitution of the relay node (the QNE2) according to the sixth embodiment is similar to that of the relay node (the QNE2) according to the first embodiment, the relay node (the QNE2 ) in the above first method according to the sixth embodiment will be described here with reference to FIG. 2. The QNE2 includes reception means 201, transmission means 202, duplication means 203, numeric value generation means 204, identification information generating means 205, and storage means 206.

In a case where disconnection of a link is caused by a link (the links 1713, 1715) connected to the correction node on an upstream side among a plurality of links connected to the communication network, the reception means 201 receives a first message (LB Reroute Notification) transmitted by an upstream relay node (e.g., the QNE5) positioned on the upstream side of the correction node and connected to the correction node. The first message is indicating that the link connected to the correction node cannot be used. The duplication means 203 generates a third message for request of transmission of a second message (LB Update Request) indicating that the correction node is changed, based on received the first message. The transmission means 202 transmits the generated third message to the relay node (the QNEl) positioned on the upstream side of the relay node itself.

Next, the relay node (the QNE2) in the above second method according to the sixth embodiment will be described. Since a basic constitution of the relay node (the QNE2) according to the sixth embodiment is similar to that of the relay node (the QNE2) according to the first embodiment, the relay node (the QNE2) in the above second method according to the sixth embodiment will be described here with reference to FIG. 2. The QNE2 includes reception means 201, transmission means 202, duplication means 203, numeric value generation means 204, identification information generating means 205 and storage means 206.

In a case where the link disconnection is caused by the link (the links 1713, 1715) connected to the correction node on the upstream side among a plurality of links connected to the communication network, the reception means 201 receives a first message (LB Reroute Notification) transmitted by upstream relay nodes (e.g., the QNE5 and the QNE6) positioned on the upstream side of the correction node and connected to the correction node. The first message is indicating that the link of the node itself connected to the correction node has been disconnected. In a case where it is judged that a plurality of the first message have been received, the duplication means 203 generates a third message for request of transmission of a second message (LB Update Request) indicating that the correction node is changed. The transmission means 202 transmits the generated third message to the relay node (the QNEl) positioned on the upstream side of the relay node itself.

It is to be noted that the first embodiment has been described as one example in the sixth embodiment, but even the second and third embodiments are similarly described. <Seventh Embodiment>

A seventh embodiment will be described with reference to FIG. 18. FIG. 18 is a schematic diagram showing a constitution of a communication network according to the seventh embodiment. As shown in FIG. 18, between a LB-I such as a QNE2 and a LB-T (a QNE7), a pair of nested load balancing paths (e.g., a QNE4 as the LB-I and a QNE6 as the LB-T) could additionally exist. In this case, a nested LB-I such as the QNE4 is to duplicate a signaling message such as LB update request from the QNE2. This means that a plurality of LB update requests is transmitted from the QNE4 in a downstream direction. This is a cause for unnecessary signaling message processing, and increases complexity of the processing logic of a signaling aware node.

To avoid the above adverse effect, when the LB-I such as the QNE2 transmits the LB update request, a

"multiplex forbidden" flag is used in a header of the signaling. When an intermediate LB-I such as the QNE4 confirms the "multiplex forbidden" flag, the message is not duplicated. Therefore, only one copy of the LB update request is transmitted toward the LB-T such as the QNE7 in the downstream direction.

Next, the relay node (the QNE2) according to the seventh embodiment will be described. In a network constituted as shown in FIG. 18, when duplication means of the relay node (the QNE2 ) transmits the LB update request, the above "multiplex forbidden" flag is added to the header of the signaling. In consequence, the unnecessary signaling message processing and the complexity of the processing logic of the signaling aware node can be avoided. <Eighth Embodiment>

In the above embodiments descriptions, the duplicated signaling from the LB-I are transmitted to the data receiving node, before the proper LB-T is discovered and proper state established. In certain network configurations, some signaling aware nodes are placed outside the domain that supports the load balancing. In such a configuration, these signaling aware nodes could simply ignore the duplicated signaling received. Alternatively, if the LB-I arranges the duplicated signaling in such a manner that the original signaling information is attached as well, these signaling aware node could interpret the original signaling information and act accordingly, without processing any additional information inserted by the LB-I, for example the local Qspec. With such arrangement, unnecessary signaling processing on the nodes could be avoided.

Each functional block used in the explanations of each embodiment of the present embodiment, described above, can be realized as a large scale integration (LSI) that is typically an integrated circuit. Each functional block can be individually formed into a single chip. Alternatively, some or all of the functional blocks can be included and formed into a single chip. Although referred to here as the LSI, depending on differences in integration, the integrated circuit can be referred to as the integrated circuit (IC), a system LSI, a super LSI, or an ultra LSI. The method of forming the integrated circuit is not limited to LSI and can be actualized by a dedicated circuit or a general-purpose processor. A field programmable gate array (FPGA) that can be programmed after LSI manufacturing or a reconfigurable processor of which connections and settings of the circuit cells within the LSI can be reconfigured can be used. Furthermore, if a technology for forming the integrated circuit that can replace LSI is introduced as a result of the advancement of semiconductor technology or a different derivative technology, the integration of the functional blocks can naturally be performed using the technology. For example, the application of biotechnology is a possibility.

INDUSTRIAL APPLICABILITY

A signaling-transmission managing method and relay nodes for use in the method according to the present invention are capable of reducing burdens on a terminal end node, inhibiting wasting of a network resource and reducing burdens on a node between the terminal end node and a node where signaling gathers, and further capable of continuing a link halfway in a network, even if the link goes down (shutdown) for some reason, reducing burdens on the terminal end node, inhibiting wasting of a network resource and reducing burdens on a node between the terminal end node and a node where signaling gathers. Therefore, the present invention is useful for a transmission managing method of signaling to be transmitted in a communication network, relay nodes for use in the method and the like.

Claims

1. A signaling-transmission managing method in a communication network including a data transmission node, a data reception node and a plurality of relay nodes which relay signaling to be transmitted from the data transmission node to the data reception node, in which, in a case where the communication network has a predetermined communicating situation, a first relay node positioned at a split point of paths of the communication network among the relay nodes duplicates global signaling received from the data transmission node to obtain duplicated signaling, includes predetermined information in the duplicated signaling and transmits the duplicated signaling to the plurality of split paths, respectively, and in which, in a case where a second relay node which receives the duplicated signaling among the relay nodes judges, based on the predetermined information included in the received duplicated signaling, that the node itself is a correction node to integrate and correct the duplicated signaling into one global signaling, the second relay node corrects the integrated duplicated signaling into the one global signaling to transmit the global signaling, the method comprising: a step in which, in a case where one or more links among a plurality of links connected to the communication network are disconnected and it is recognized that an upstream relay node which has been connected to the disconnected link and which is positioned on an upstream side where the duplicated signaling flows has been disconnected, the upstream relay node transmits a first message indicating that the link has been disconnected to the first relay node; a step in which the first relay node that has received the first message transmits to the correction node a second message indicating that information of a state stored in the correction node should be updated, when the link is disconnected; a step in which the correction node that has received the second message transmits to the data reception node a third message to^' update the state of the relay node existing on a route from the correction node to the data reception node based on the second message; and a step in which the relay node that has received a fourth message transmitted from the upstream relay node to the data reception node along a new route and indicating a request for correction of the correction node and the third message judges, based on the received third and fourth messages, whether or not the relay node itself is a new correction node and in which the relay node itself transmits a message to update the information of the state stored in the relay nodes, in a case where it is judged that the relay node is the new correction node .

2. A signaling-transmission managing method in a communication network including a data transmission node, a data reception node and a plurality of relay nodes which relay signaling to be transmitted from the data transmission node to the data reception node, in which the signaling includes a list including information to indicate whether it is QoS reservation corresponding to a data packet having what kind of header information, in a case where the QoS reservation is made to a QoS corresponding node along a data route among the relay nodes, in which a first relay node positioned at a split point of paths of the communication network among the relay nodes duplicates the signaling including the list to obtain duplicated signaling based on a load balancing policy set beforehand, includes predetermined information in the duplicated signaling and transmits the duplicated signaling to the plurality of split paths, respectively, and in which, in a case where a second relay node to receive the duplicated signaling among the relay nodes judges, based on the predetermined information included in the received duplicated signaling, that the relay node itself is a correction node to integrate and correct the duplicated signaling into one global signaling, the second relay node corrects the integrated duplicated signaling into the one global signaling to transmit the global signaling, the method comprising: a step in which, in a case where one or more links among a plurality of links connected to the communication network are disconnected and it is recognized that an upstream relay node which has been connected to the disconnected link and which is positioned on an upstream side where the duplicated signaling flows has been disconnected, the upstream relay node transmits a first message indicating that the link has been disconnected to the first relay node; a step in which the first relay node that has received the first message transmits to the correction node a second message indicating that information of a state stored in the correction node should be updated, when the link is disconnected; a step in which the correction node that has received the second message transmits to the data reception node a third message to update the state of the relay node existing on a route from the correction node to the data reception node based on the second message; and a step in which the relay node that has received a fourth message transmitted from the upstream relay node to the data reception node along a new route and indicating a request for correction of the correction node and the third message judges, based on the received third and fourth messages, whether or not the relay node itself is a new correction node and in which the relay- node itself transmits a message to update the information of the state stored in the relay nodes, in a case where it is judged that the relay node is the new correction node .

3. A signaling-transmission managing method in a communication network including a data transmission node, a data reception node and a plurality of relay nodes which relay signaling to be transmitted from the data transmission node to the data reception node, in which, in a case where the communication network has a predetermined communicating situation, a first relay node positioned at a split point of paths of the communication network among the relay nodes duplicates global signaling received from the data transmission node to obtain duplicated signaling, includes predetermined information in the duplicated signaling and transmits the duplicated signaling to the plurality of split paths, respectively, and in which, in a case where a second relay node which receives the duplicated signaling among the relay nodes judges, based on the predetermined information included in the received duplicated signaling, that the relay node itself is a correction node to integrate and correct the duplicated signaling into the one global signaling, the second relay node corrects the integrated duplicated signaling into the one global signaling to transmit the global signaling, the method comprising: a step in which, in a case where disconnection of a link is caused by the correction node in an upstream link connected to the correction node among a plurality of links connected to the communication network, an upstream relay node that is positioned on an upstream side of the correction node and that has been connected to the correction node transmits to the first relay node a first message indicating that the link connected to the correction node is not used; a step in which the first relay node that has received the first message transmits to the relay node positioned on an upstream side of the first relay node itself a third message indicating a request for transmission of a second message indicating that the correction node is changed based on the received first message; a step in which the relay node that has received the third message and that is positioned on the upstream side of the first relay node transmits the second message to the data reception node; and a step in which the relay node that is positioned on a downstream side of the first relay node and that has received a fourth message transmitted from the relay node connected to the first relay node to the data reception node along a new route and indicating a request for correction of the correction node and the second message judges, based on the received second and fourth messages, whether or not the relay node itself is a new correction node and in which the relay node itself transmits a message to update information of a state stored in the relay nodes, in a case where it is judged that the relay node is the new correction node.

4. A signaling-transmission managing method in a communication network including a data transmission node, a data reception node and a plurality of relay nodes which relay signaling to be transmitted from the data transmission node to the data reception node, in which the signaling includes a list including information to indicate whether it is QoS reservation corresponding to a data packet having what kind of header information, in a case where the QoS reservation is made to a QoS corresponding node along a data route among the relay nodes, in which a first relay node positioned at a split point of paths of the communication network among the relay nodes duplicates the signaling including the list to obtain duplicated signaling based on a load balancing policy set beforehand, includes predetermined information in the duplicated signaling and transmits the duplicated signaling to the plurality of split paths, respectively, and in which, in a case where a second relay node to receive the duplicated signaling among the relay nodes judges, based on the predetermined information included in the received duplicated signaling, that the relay node itself is a correction node to integrate and correct the duplicated signaling into one global signaling, the second relay node corrects the integrated duplicated signaling into the one global signaling to transmit the global signaling, the method comprising: a step in which, in a case where disconnection of a link is caused by the correction node in an upstream link connected to the correction node among a plurality of links connected to the communication network, an upstream relay node that is positioned on an upstream side of the correction node and that has been connected to the correction node transmits to the first relay node a first message indicating that a link connected to the correction node is not used; a step in which the first relay node that has received the first message transmits to the relay node positioned on an upstream side of the first relay node itself a third message indicating a request for transmission of a second message indicating that the correction node is changed based on the received first message; a step in which the relay node that has received the third message and that is positioned on the upstream side of the first relay node transmits the second message to the data reception node; and a step in which the relay node that is positioned on a downstream side of the first relay node and that has received a fourth message transmitted from the relay node connected to the first relay node to the data reception node along a new route and indicating a request for correction of the correction node and the second message judges, based on the received second and fourth messages, whether or not the relay node itself is a new correction node and in which the relay node itself transmits a message to update information of a state stored in the relay nodes, in a case where it is judged that the relay- node is the new correction node.

5. A signaling-transmission managing method in a communication network including a data transmission node, a data reception node and a plurality of relay nodes which relay signaling to be transmitted from the data transmission node to the data reception node, in which, in a case where the communication network has a predetermined communicating situation, a first relay node positioned at a split point of paths of the communication network among the relay nodes duplicates global signaling received from the data transmission node to obtain duplicated signaling, includes predetermined information in the duplicated signaling and transmits the duplicated signaling to the plurality of split paths, respectively, and in which, in a case where a second relay node which receives the duplicated signaling among the relay nodes judges, based on the predetermined information included in the received duplicated signaling, that the relay node itself is a correction node to integrate and correct the duplicated signaling into the one global signaling, the second relay node corrects the integrated duplicated signaling into the one global signaling to transmit the global signaling, the method comprising: a step in which, in a case where disconnection of a link is caused by the correction node in a plurality of links on an upstream side connected to the correction node among a plurality of links connected to the communication network, upstream relay nodes that are positioned on an upstream side of the correction node and that have been connected to the correction node transmit to the first relay node first messages indicating that the links from the upstream relay nodes themselves to the correction node have been disconnected; a step in which the first relay node to receive the first messages transmits to the relay node positioned on an upstream side of the first relay node itself a third message indicating a request for transmission of a second message indicating that the correction node is changed, in a case where the first relay node receives the plurality of first messages; a step in which the relay node that has received the third message and that is positioned on the upstream side of the first relay node transmits the second message to the data reception node; and a step in which the relay node that is positioned on a downstream side of the first relay node and that has received a fourth message transmitted from the relay node connected to the first relay node to the data reception node along a new route and indicating a request for correction of the correction node and the second message judges, based on the received second and fourth messages, whether or not the relay node itself is a new correction node and in which the relay node itself transmits a message to update information of a state stored in the relay nodes, in a case where it is judged that the relay node is the new correction node.

6. A signaling-transmission managing method in a communication network including a data transmission node, a data reception node and a plurality of relay nodes which relay signaling to be transmitted from the data transmission node to the data reception node, in which the signaling includes a list including information to indicate whether it is QoS reservation corresponding to a data packet having what kind of header information, in a case where the QoS reservation is made to a QoS corresponding node along a data route among the relay nodes, in which a first relay node positioned at a split point of paths of the communication network among the relay nodes duplicates the signaling including the list to obtain duplicated signaling based on a load balancing policy set beforehand, includes predetermined information in the duplicated signaling and transmits the duplicated signaling to the plurality of split paths, respectively, and in which, in a case where a second relay node to receive the duplicated signaling among the relay nodes judges, based on the predetermined information included in the received duplicated signaling, that the relay node itself is a correction node to integrate and correct the duplicated signaling into one global signaling, the second relay node corrects the integrated duplicated signaling into the one global signaling to transmit the global signaling, the method comprising: a step in which, in a case where disconnection of a link is caused by the correction node in a plurality of links on an upstream side connected to the correction node among a plurality of links connected to the communication network, upstream relay nodes that are positioned on an upstream side of the correction node and that have been connected to the correction node transmit to the first relay node first messages indicating that the links from the upstream relay nodes themselves to the correction node have been disconnected; a step in which the first relay node to receive the first messages transmits to the relay node positioned on an upstream side of the first relay node itself a third message indicating a request for transmission of a second message indicating that the correction node is changed, in a case where the first relay node receives the plurality of first messages; a step in which the relay node that has received the third message and that is positioned on the upstream side of the first relay node transmits the second message to the data reception node; and a step in which the relay node that is positioned on a downstream side of the first relay node and that has received a fourth message transmitted from the relay node connected to the first relay node to the data reception node along a new route and indicating a request for correction of the correction node and the second message judges, based on the received second and fourth messages, whether or not the relay node itself is a new correction node and in which the relay node itself transmits a message to update information of a state stored in the relay nodes, in a case where it is judged that the relay node is the new correction node.

7. The signaling-transmission managing method according to claim 1, wherein the second message includes information indicating that duplication is prohibited.

8. A relay node for use in a signaling-transmission managing method in a communication network including a data transmission node, a data reception node and the plurality of relay nodes which relay signaling to be transmitted from the data transmission node to the data reception node, the relay node comprising: reception means for receiving duplicated signaling obtained by duplicating global signaling to be transmitted from the data transmission node to the data reception node by another relay node; judgment means for judging whether or not the relay node itself is a correction node which integrates and corrects the duplicated signaling into the one global signaling based on predetermined information included in the received duplicated signaling; correction means for correcting the duplicated signaling into the one global signaling, in a case where it is judged that the relay node is the correction node; and transmission means for transmitting the one corrected global signaling, wherein, in a case where one or more links among a plurality of links connected to the communication network are disconnected and an upstream relay node which has been connected to the disconnected link and which is positioned on an upstream side where the duplicated signaling flows recognizes that the links have been disconnected, the reception means receives a second message indicating that information of a state stored in the correction node should be updated and transmitted, when the links are disconnected, by a first relay node which has received a first message indicating that the link has been disconnected and transmitted by the upstream relay node, the correction means generates a third message to update the state of the relay node existing along a route from the relay node itself to the data reception node based on the received second message, and the transmission means transmits the generated third message to the data reception node.

9. A relay node for use in a signaling-transmission managing method in a communication network including a data transmission node, a data reception node and the plurality of relay nodes which relay signaling to be transmitted from the data transmission node to the data reception node, the relay node comprising: reception means for receiving duplicated signaling obtained by duplicating, by another relay node, global signaling transmitted from the data transmission node to the data reception node and including a list including information to indicate whether it is QoS reservation corresponding to a data packet having what kind of header information, in a case where the QoS reservation is made to a QoS corresponding node along a data route among the relay nodes; judgment means for judging whether or not the relay node itself is a correction node which integrates and corrects the duplicated signaling into the one global signaling based on predetermined information included in the received duplicated signaling; correction means for correcting the duplicated signaling into the one global signaling, in a case where it is judged that the relay node is the correction node; and transmission means for transmitting the one corrected global signaling, wherein, in a case where one or more links among a plurality of links connected to the communication network are disconnected and an upstream relay node which has been connected to the disconnected link and which is positioned on an upstream side where the duplicated signaling flows recognizes that the links have been disconnected, the reception means receives a second message indicating that information of a state stored in the correction node should be updated and transmitted, when the links are disconnected, by a first relay node which has received a first message indicating that the link has been disconnected and transmitted by the upstream relay node, the correction means generates a third message to update the state of the relay node existing along a route from the relay node itself to the data reception node based on the received second message, and the transmission means transmits the generated third message to the data reception node.

10. A relay node for use in a signaling-transmission managing method in a communication network including a data transmission node, a data reception node and the plurality of relay nodes which relay signaling to be transmitted from the data transmission node to the data reception node, the relay node comprising: reception means for receiving global signaling to be transmitted from the data transmission node to the data reception node; duplication means for, in a case where the relay node itself is positioned at a split point of paths of the communication network and the communication network has a predetermined communicating situation, duplicating the global signaling received by the reception means to obtain duplicated signaling as many as paths to be split; in and transmission means for including predetermined information in the duplicated signaling and transmitting the duplicated signaling to the plurality of split paths, respectively, wherein, in a case where disconnection of a link is caused by a correction node in an upstream link connected to the correction node among a plurality of links connected to the communication network, the reception means receives a first message transmitted by an upstream relay node which is positioned on an upstream side of the correction node and which has been connected to the correction node and indicating that the link connected to the correction node is not used, the duplication means generates a third message indicating a request for transmission of a second message indicating that the correction node is changed based on the received first message, and the transmission means transmits the generated third message to the relay node which is positioned on an upstream side of the relay node itself.

11. A relay node for use in a signaling-transmission managing method in a communication network including a data transmission node, a data reception node and the plurality of relay nodes which relay signaling to be transmitted from the data transmission node to the data reception node, the relay node comprising: reception means for receiving global signaling transmitted from the data transmission node to the data reception node and including a list including information to indicate whether it is QoS reservation corresponding to a data packet having what kind of header information, in a case where the QoS reservation is made to a QoS corresponding node along a data route among the relay nodes; duplication means for duplicating the global signaling to obtain duplicated signaling based on a load balancing policy set beforehand, in a case where the relay node is positioned at a split point of paths of the communication network; and transmission means for including predetermined information in the duplicated signaling and transmitting the duplicated signaling to the plurality of split paths, respectively, wherein, in a case where disconnection of a link is caused by a correction node in an upstream link connected to the correction node among a plurality of links connected to the communication network, the reception means receives a first message transmitted by an upstream relay node which is positioned on an upstream side of the correction node and which has been connected to the correction node and indicating that the link connected to the correction node is not used, the duplication means generates a third message indicating a request for transmission of a second message indicating that the correction node is changed based on the received first message, and the transmission means transmits the generated third message to the relay node which is positioned on an upstream side of the relay node itself.

12. A relay node for use in a signaling-transmission managing method in a communication network including a data transmission node, a data reception node and the plurality of relay nodes which relay signaling to be transmitted from the data transmission node to the data reception node, the relay node comprising: reception means for receiving global signaling to be transmitted from the data transmission node to the data reception node; duplication means for, in a case where the relay node itself is positioned at a split point of paths of the communication network and the communication network has a predetermined communicating situation, duplicating the global signaling received by the reception means to obtain duplicated signaling as many as paths to be split; and transmission means for including predetermined information in the duplicated signaling and transmitting the duplicated signaling to the plurality of split paths, respectively, wherein, in a case where disconnection of a link is caused by a correction node in a plurality of links on an upstream side connected to the correction node among a plurality of links connected to the communication network, the reception means receives first messages transmitted by upstream relay nodes which are positioned on an upstream side of the correction node and which have been connected to the correction node and indicating that the links from the upstream relay nodes themselves to the correction node have been disconnected, the duplication means generates a third message indicating a request for transmission of a second message indicating that the correction node is changed, in a case where it is judged that the plurality of first messages are received, and the transmission means transmits the generation third message to the relay node positioned on the upstream side of the relay node itself.

13. A relay node for use in a signaling-transmission managing method in a communication network including a data transmission node, a data reception node and the plurality of relay nodes which relay signaling to be transmitted from the data transmission node to the data reception node, the relay node comprising: reception means for receiving global signaling transmitted from the data transmission node to the data reception node and including a list including information to indicate whether it is QoS reservation corresponding to a data packet having what kind of header information, in a case where the QoS reservation is made to a QoS corresponding node along a data route among the relay nodes; duplication means for duplicating the global signaling to obtain duplicated signaling based on a load balancing policy set beforehand, in a case where the relay node is positioned at a split point of paths of the communication network; and transmission means for including predetermined information in the duplicated signaling and transmitting the duplicated signaling to the plurality of split paths, respectively, wherein, in a case where disconnection of a link is caused by a correction node in a plurality of links on an upstream side connected to the correction node among a plurality of links connected to the communication network, the reception means receives first messages transmitted by upstream relay nodes which are positioned on an upstream side of the correction node and which have been connected to the correction node and indicating that the links from the upstream relay nodes themselves to the correction node have been disconnected, the duplication means generates a third message indicating a request for transmission of a second message indicating that the correction node is changed, in a case where it is judged that the plurality of first messages are received, and the transmission means transmits the generation third message to the relay node positioned on the upstream side of the relay node itself.

14. The relay node according to claim 8, wherein the second message includes information indicating that duplication is prohibited.