JPH10257060A - Method for selecting and connecting route of communication node, atm node and network - Google Patents

Abstract

PROBLEM TO BE SOLVED: To eliminate unfairness caused by a fact that reception of an effect by a crank back processing in a PNNI protocol differs by a kind of connection. SOLUTION: In an ATM node used for a network to start communication by previously selecting and reserving a resource on a communication route, a signaling processing means (SG) 1103 to be operated according to the PNNI protocol and to simultaneously decide connection setting propriety on the route initially selected by a source node or the connection setting propriety on a selected substitutive route, a crank back processing part 1102 to execute a crank back processing of the PNNI when the connection is impossible to be set by the SG and a means 1101 to change a connection setting propriety deciding method executed by the SG according to presence or absence of crank back execution are provided. The unfairness is corrected by changing an algorithm of decision of the connection setting propriety based on whether a signaling processing is performed before or after the crank back when the signaling processing is performed in the ATM node.

Description

DETAILED DESCRIPTION OF THE INVENTION

[0001]

BACKGROUND OF THE INVENTION 1. Field of the Invention The present invention relates to a PNNI interface (Private Network Network).
The present invention relates to an ATM node having an rk interface, and particularly to an ATM node and a network route selection connection method which execute a signaling method defined by PNNI and also execute a crankback process in a signaling process for setting an ATM connection. .

[0002]

2. Description of the Related Art ATM (Asynchronous Transfer Mode);
Asynchronous transfer mode) was created as a basic method for realizing a high-speed broadband communication network (B-ISDN) suitable for multimedia communication. That is, a communication speed of 64 for a network for digital communication services
kbps ISDN (Integrate Services Digital Network
ork (integrated service digital network) was realized, and it became possible to provide not only voice but also facsimile and data communication, but this was considered only as a system based on the digital telephone network. Therefore, the speed at which communication is performed is based on 64 kbps, which is adapted to voice communication.

[0003] However, moving images and the like are to be transmitted, and furthermore, multimedia communications aiming at providing advanced services such as video on demand have to be realized. The concept of B-ISDN has emerged as a highly reliable communication service, and ATM has emerged from the study of B-ISDN.

In recent years, a large number of network products compatible with the ATM communication system have appeared on the market. In particular, the network products have been decided by the ATM-Forum (ATM Forum) which is a group that creates ATM implementation standards. ATM that conforms to the UNI (User Network Interface) standards UNI 3.0 and UNI 3.1
Many products having a switch and an ATM interface have appeared on the market and are being put to practical use, and the number of places where an in-house network is replaced with an ATM-LAN is gradually increasing.

[0005] As an interface between ATM switch nodes, the ATM Forum uses PNNI (Priva).
te Network Network Interf
ace) was determined in 1996, and implementation of this PNNI method has been started by many ATM switch vendors.

In the PNNI interface specification, a network connecting a plurality of ATM switches is recognized as a network having a virtual hierarchical structure. In this virtual network, the reduction of topology information to be stored is realized by recognizing a subnet composed of a plurality of switch nodes, which is called a peer group in the PNNI interface specification, as one logical node. I'm trying.

That is, in the PNNI interface specification, a plurality of ATs called "peer groups"
By recognizing a subnet composed of M switch nodes as one logical node, a network connecting a plurality of ATM switches is recognized as a network having a virtual hierarchical structure. By recognizing the topology (network form) in such a virtual hierarchical structure, in the PNNI system, each A in the network is recognized.
The topology information to be stored by the TM switch node is reduced.

FIG. 8 shows a method of recognizing a network topology in the PNNI system. In the example ATM network shown in FIG. 9, a plurality of ATM switch nodes 101, 102,.
5, 106, to 122, and these ATM switch nodes are connected by a link.

In the PNNI system, links between the ATM switch nodes and topology information of the logical ATM switch nodes are exchanged between these ATM switch nodes. In the example of the ATM network of FIG. 8, the above-mentioned 22 existing ATM switch nodes 101,
To 122 are divided into six ATM switch node groups (peer groups), and each peer group is assigned a logical AT.
They are recognized as M switch nodes 31 to 36.

Here, the existing ATM switch node 101,
A peer group consisting of 102, 103, 104 is a logical ATM switch node 31, a peer group consisting of existing ATM switch nodes 105, 106, 107 is a logical ATM switch node 32, and ATM switch nodes 108, 109, 110, 111. Is a logical ATM switch node 33, and a peer group consisting of existing ATM switch nodes 112, 113, 114 and 115 is a logical ATM switch node 34 and a peer group consisting of existing ATM switch nodes 116, 117 and 118. Are logical ATM switch nodes 35, existing ATM switch nodes 119, 120, 121, 12
The two peer groups are logical ATM switch nodes 36.

Therefore, in the PNNI, 22 real ATM switch nodes 10 shown in the lower part of FIG.
The network constituted by 1 to 122 is recognized as a virtual “lower-layer network”. Also, the logical network constituted by the six logical ATM switch nodes 31 to 36 shown in the upper part of FIG. 8 is recognized as a virtual “upper layer network”.

In an ATM network, several QOSs (Quos) are used to transfer a plurality of types of applications using the same data link protocol.
Arity of Service (quality of service) parameters are set, and these values can be requested to the network simultaneously with the connection setting request.

There are a plurality of types of QOS parameters such as an effective band, a transfer delay time, and a transfer delay fluctuation, and a routing path for optimizing these parameters is obtained.

The case where data is transmitted from the existing ATM switch node 101 to the existing ATM switch node 122 will be specifically described with reference to the drawings. In this case, the existing ATM switch node 101 on the network shown in FIG. 8, which is the calling node, behaves as follows. That is, P
In the case of the NNI system, the existing ATM switch node 101 recognizes the entire network as a topology as shown in FIG. 9 by using the above-described topology recognition system having a virtual hierarchical configuration.

FIG. 9 shows an example of the QO of the recognized network.
A case where “delay time” and “effective band” are given as S parameters is shown. That is, in FIG.
For example, between the switch node 101 which is the calling node and the adjacent switch node 102, "delay time"
Is "1 msec" and the "effective band" is "10 Mbps". Between the switch node 101 which is the calling node and the adjacent switch node 103, the "delay time" is "5 msec" and the "effective band" is "30 Mbps". "And the switch node 102 is adjacent to the switch node 1
04, the “delay time” is “2 msec” and the “effective band” is “20 Mbps”. Between the switch node 103 and the adjacent switch node 104, the “delay time” is “3 msec” and the “effective band”. Is "40 Mbps", between the switch node 104 and the adjacent switch node 32, the "delay time" is "10 msec", the "effective band" is "50 Mbps", and the switch node 32 and the adjacent switch node 3
Between the three, the “delay time” is “10 msec”, the “effective band” is “20 Mbps”, and so on.

Here, when setting up an ATM connection from the ATM switch node 101, first, in the existing ATM switch node 101 which has generated the call, the ATM switch shown in FIG.
The setting route of the ATM connection is searched by using the network of the topology shown in FIG. 9, and the connection setting route on the topology shown in FIG. 9 is determined.

That is, the existing ATM switch node 101 that has generated the call is connected to the other existing ATM switch nodes 102, 103, and 10 in the peer group to which it belongs.
4 and the logical AT in the virtual upper-layer network
Logical ATM switch node 3 which is an M switch node
2, 33, 34, 35, to find one route that can finally reach the logical ATM switch node 36, and from the QOS (Quality of Service) parameter on the route, the search is performed. It is determined whether or not the route is appropriate.

As a result, if the result is inappropriate, another route is searched to determine whether or not the route is appropriate. If the route is appropriate, this is set as a set route candidate, and the next operation, connection setting The process proceeds to the determination process.

The process for determining whether a connection can be set is performed as follows. The existing ATM switch node 101 transmits a packet in which path information called a DTL stack is written on the set path, and sequentially determines whether a connection can be set on each link. At this time, the logical ATM switch node in FIG. 9 uses the topology of the ATM switch node of a virtual lower hierarchy in each logical switch node to independently set the connection setting path in each peer group. And transfers a packet in which path information called a DTL stack is written on the selected connection setting path, and sequentially judges whether a connection can be set on each link.

Therefore, in the logical ATM switch node 32 in the topology shown in FIG. 9, the existing ATM switch node 10 in the network shown in FIG.
The connection setting path between the ATMs 5, 106, and 107 is determined. Similarly, the logical ATM in the topology shown in FIG.
In the switch node 36, the existing ATM switch nodes 119 and 12 in the network shown in FIG.
A connection setting route between 0, 121, and 122 will be determined.

As described above, in the PNNI interface specification, a plurality of A called "peer groups"
By recognizing a subnet composed of TM switch nodes as one logical node, a network connecting a plurality of ATM switches is recognized as a network having a virtual hierarchical structure. By recognizing the topology (network form), the topology information to be stored by each ATM switch node in the network is reduced.

In the PNNI system, a call setup process for setting up a connection is defined in several stages.

First, in a source node (that is, an ATM switch node accommodating a terminal for which a connection is to be set), a source routing process for selecting / determining a route of the entire connection is executed. Then, as a result, a stack in which path information called “DTL stack” is written is transferred onto the selected path, and processing for securing resources on each link is executed. If resources can be secured, communication will be performed between the calling and called terminals using the secured resources or the source.

Further, if the resource reservation fails, an alternative route selection process called "Crankback" is executed, whereby the "DTL stack" is returned to an appropriate node, and then a new route is selected. The selection / decision processing of the route of the appropriate connection is executed.

In the PNNI specification, such a crankback process is intended to improve the call blocking rate of the entire network.

A signaling procedure using a DTL stack in the PNNI protocol using a network whose topology is recognized as shown in FIG.
The crankback procedure executed at that time will be described.

[Signaling by DTL Stack] P
In NNI routing, called a DTL stack,
A signaling scheme using a stack storing path information selected at the calling node is shown. D
The procedure of the routing using the TL stack is as follows.

[Step 1]. Routing is executed at the calling node (the ATM switch node in which the calling terminal is accommodated), and a DTL stack in which the selected path is written is created.

[Step 2]. The created DTL stack is transferred on the selected route.

[Step 3]. When the transferred DTL stack reaches the peer group on the selected path, the peer group selects a path in the group based on the information written in the DTL stack.

[Step 4]. A part of the route information of the arriving DTL stack is rewritten to the route information newly selected in the peer group, and the partially rewritten new DT
Transfer the L stack on the selected path.

[Step 5]. The above procedure [Step 2]. ~ [Stage
Four]. Are sequentially performed, and when the DTL stack arrives at the called node (the ATM switch node accommodating the called terminal), the routing ends.

Thus, in the PNNI routing,
The DTL stack storing the selected route information is transferred to the selected route, and when the peer group on the route is reached, the peer group selects a route in the group and selects a part of the route information of the DTL stack. The route information is rewritten with the route information, and the rewritten DTL stack is transferred on the route according to the route information of the DTL stack, so that the DTL stack is finally sent to the destination.

A specific example will be described.

Referring to FIG. 10, consider the case where a call is made from the real node "A.1.1" to the real node "C.2.2" in the lower hierarchy which is the existing hierarchy.
All the real nodes “A.1.1” (nodes by the ATM switch) are regarded as logical nodes except for the peer group to which the own node belongs.
It is recognized that the logical topology is as shown in FIG.

Then, the route selection algorithm is executed on this logical network, and the calling node "A.1.
1 ”, the called node“ C. 2.2 "(FIG. 11
In (a), a route to the logical node C) is selected.

Here, the selected route selected by the calling node "A.1.1." Is the real node "A.1.
1 "→ real node" A. 1.2 ”→ real node“ A. 1.
3 "→ Peer group" A. 2 "(logical node) → peer group" A. 3 "(logical node) → peer group" B "
If (logical node) → peer group “C” (logical node), the DTL stack (packet in which the route information is written) created at that time is as shown in FIG. 11B.

The DTL stack created in this way is:
The data is transferred on the above-mentioned route selected by the calling node “A.1.1”, and the DTL stack is rewritten as follows in each peer group (see FIG. 12).

(1). As shown in FIG. 12 (b), the data transmitted from the peer group "A.1.3" [A. 1.1
A. 1.2 A. 1.3] [A. 1-A. 2-A.
3] The DTL stack “[1.]” having the content [ABC] arrives at the peer group “A.2”. Then, when the DTL stack “[1.]” arrives at the peer group “A.2”, a route in the peer group “A.2” is selected. Then, as a result, [A. 2.2-A.
2.1] is selected, the path information part in “A.1” of the DTL stack “[1.]” is newly selected as the [A. 2.2-A. 2.1] as shown in FIG. 12 (b).
2.2-A. 2.1] [A. 1-A. 2-A. 3] [A
-B-C]. And
[A. of the DTL stack "[1.]" 2.2-A. 2.
1] according to the information of the content “A.2.2” → “A.
2.1 is transmitted. 2.1 ", the DTL stack" [1. ]], The information [A.1-A.2-A.3] indicates that “A. 3 to the peer group of this DTL stack "[1. ]], But “[1. ] Is the peer group "A. 3 ”, the peer group“ A. 3 "is selected, and as a result," A. As a route in 3 ", [A.3.1-
A. 3.2-A. 3.3] is selected, DT
The path information part in “A.3” of the L stack “[1.]” is the newly selected [A. 3.1-A. 3.2-
A. 3.3] in FIG.
As shown in (c), [A. 3.1-A. 3.2-A.
3.3] [A. 1-A. 2-A. 3] A DTL stack “[2.]” having the content [ABC] is obtained. And
[A. of the DTL stack "[2.]" 3.1-A. 3.
2-A. 3.3] according to the information of the content “A.3.
1 "→" A. 3.2 "→" A. 3.3 ”is transmitted.
Then, [ABC] of the DTL stack “[2.]”
The DTL stack “[2.]”
Is transmitted to the peer group of “B”.

(2). Next, when the DTL stack “[2.]” arrives at the peer group “B”,
The route selection within the peer group "B" is performed. “B”
[B. 1.1-B. 1.3] is selected, the DTL stack “[2.]”, [A.
3.1 A. 3.2 A. 3.3] is newly selected as shown in FIG. 12C. 1.1-B. 1.3].

In this way, the DTL stack created at the calling node is transferred to the corresponding peer group according to the route information, and the peer group selects a route within the own group and rewrites the corresponding portion of the DTL stack. Processing. This secures resources at the node that has become the transfer route.

If resource reservation fails in the middle of the selected transfer route, an alternative route selection process (crankback) is performed, the above-mentioned DTL stack is returned to an appropriate node, and then a new connection is established. Execute route selection / determination processing.

The crankback algorithm is as follows.

[Crankback Algorithm] PNNI
In this method, after periodically rewriting the topology information, the topology information is notified to each node. Therefore, a change in topology information caused by a call generated during the rewrite cycle is not notified to other nodes until the next rewrite is completed. For this reason,
It is expected that a call loss will occur due to a mismatch in topology information.

To solve this problem, the PNNI employs an algorithm called "crankback", in which, when a connection requested by a call cannot be set, a path for which another connection can be set is sequentially searched.

More specifically, when a connection cannot be set on the route selected by the calling node, the process returns to the previous peer group (that is, the peer group in which the DTL stack has been rewritten) and the route is selected again. Is a method to redo.

In the routing on the logical network shown in FIG. 11, when crankback occurs,
FIG. 13 shows an example of the connection path and the DTL stack that are changed before and after that, and the execution procedure is shown below.

The calling node, which is the node accommodating the calling terminal, is called "A.1.1", and the node accommodating the called terminal is called the peer group "C". As shown in FIG. 13A, the contents of the DTL stack, which is information of the route selected by the calling node “A.1.1”, are [A. 1.1
A. 1.2 A. 1.3] [A. 1-A. 2-
A. 3] It is assumed that [A-B-C]. In this case, D
The TL stack is “A.1.1” → “A.1.2” →
It is transferred to follow the route of “A.1.3”, and further from “A.1.3” to the peer group “A.2”, and from “A.2” to “A.3”. Will be transferred.

1. Now, when the DTL stack is transferred on the route selected by the calling node “A.1.1” and arrives between the logical nodes “A.2” to “A.3”, the connection cannot be set. It is assumed that (Reject).

2. Then, it notifies that the connection could not be set between "A.2" and "A.3" while reversing the route ("Reject notification"). Here, since a new route cannot be selected in “A.2”, the call is backed up to the calling node “A.1.1”.

3. The node “A.1.1” receiving the “Reject notification” is a logical node “A.2” to
The route selection is executed assuming that there is no link between “A.3”. And a DTL storing the re-selected route information.
A new stack is created and transferred on the newly selected route.

As described above, in the crankback algorithm, when a connection cannot be set,
Returning to where appropriate, a new DTL stack will be created. Therefore, in order to execute the crankback, each peer group or node must store the DTL stack that has passed through its own peer group or its own node.

The effect of such a crankback process is to improve the call blocking rate of the entire network. However, it has been found that this effect is not imparted fairly to all connections occurring in the network.

For example, in a so-called multi-connection environment in which connections requesting a plurality of types of bandwidths occur, the effect of crankback is given mainly to connections having a smaller required bandwidth among the plurality of connections. . It is also known that the use of crankback may cause a call loss rate of a connection with a large required bandwidth to be degraded.

As a problem in a multi-connection environment, this results in promoting the previously pointed out problem of "unfairness of the call loss rate due to the difference in the type of connection".

Normally, the unfairness of the call loss rate due to the difference in the bandwidth required by the connection is determined by the relationship between the bandwidth of the link and the required bandwidth (so-called division loss), and is inevitable in a sense. .

However, the fact that the addition of such a new function called crankback promotes this unfairness means that connections occurring in the network cannot be treated equally. I have. From a user's point of view, it feels like the service is not provided fairly.

[0058]

In the PNNI system, a crankback process is performed if the resource allocation fails in order to improve the call blocking rate of the entire network when setting up a connection.
Not all connections that occur in the network are provided fairly.

For example, in a so-called multi-connection environment in which connections requesting a plurality of types of bandwidths occur, the effect of crankback is given mainly to the connection having a smaller required bandwidth among the plurality of connections. Therefore, for a connection with a large required bandwidth, the call blocking rate may be degraded due to crankback.

As a problem in a multi-connection environment, this has resulted in the fact that "unfairness of the call loss rate due to the difference in the type of connection" which has been pointed out in the past becomes apparent.

As described above, the currently defined crankback processing method can improve the call blocking rate of the entire network, but the effect is not imparted fairly in terms of individual connections. There was a problem. Therefore, establishment of technology to improve this inequity is expected.

Therefore, the object of the present invention is to
Even if the PNNI crankback algorithm is executed in a multi-connection environment, the effect of the crankback can be obtained so that all the connections can be enjoyed fairly and the loss probability of each connection occurring in the network can be uniformly improved. It is to provide an ATM node which implements a possible call setup algorithm.

A route selection and connection method in a network for starting communication by selecting and reserving resources on a communication route in advance, and performing an alternative route selection process when resources of the communication route selected first cannot be reserved. Is to provide.

[0064]

In order to achieve the above object, the present invention is configured as follows.

The first is a network for starting communication by selecting and reserving resources on a communication path in advance. If a resource on the communication path selected at the beginning cannot be reserved, a new network is substituted. In a network in which a selection process of an alternative route is executed, a communication node includes communication route selection means in a communication network for performing communication in accordance with a request of a call generated in the communication network, and communication on the selected route. Communication resource acquisition means for acquiring resources is provided. A communication path reselecting means for re-selecting a communication path when a communication resource cannot be obtained on the communication path selected by the resource obtaining means; and a communication path on the reselected communication path. Communication resource reacquisition means for acquiring resources is provided. Further, there is provided an algorithm changing means for changing a communication resource obtaining algorithm executed in the communication resource obtaining means and the communication resource reacquiring means.

Thus, the influence on the call generated in the communication network by the communication resource acquisition means and the influence on the call generated in the communication network by the communication resource reacquisition means can be made independent. Therefore, the influence on the call by the communication path selection / reselection means can be dispersed.

Second, the ATM node of the network includes a PNNI protocol executing means for executing the PNNI protocol, including a crankback means, and a path or crank selected first by the source node at the time of connection setting. Signaling processing means for deciding whether to set up a connection on the path selected by the alternative path selection processing after the execution of the back. Further, in the signaling means, whether to determine whether to set up a connection on the path initially selected by the source node or whether to set up or not to set up a connection on the path selected by the alternative path selection processing after execution of crankback is determined. Signaling processing recognizing means for recognizing whether to determine,
A connection setting feasibility determining algorithm changing means for changing an algorithm for determining whether a connection can be set, which is executed in the signaling processing recognized by the signaling processing recognizing means.

As a result, it is possible to disperse the influence of the call generated in the network by the signaling processing at the first source node and the influence of the signaling processing to the alternative path after the crankback processing.

Third, in the first or second configuration, the algorithm changing means or the connection setting is performed only for a call requesting an appropriate band among calls generated on the network. An algorithm change area determining means for executing the propriety determining algorithm changing means is provided.

Fourth, the communication node and the ATM node include the communication node having the first configuration or the second configuration and the ATM node.
In addition to the means possessed by the TM node, when the signaling processing recognizing means recognizes that the signaling processing being executed is determining whether or not to establish a connection on the path initially selected by the source node. If a bandwidth equal to or greater than the bandwidth required by the generated call exists in the link, connection setting is permitted, and the signaling processing recognition means causes the signaling processing being executed to be performed as an alternative route after the execution of the crankback processing. If it is recognized that the setting of the connection on the selected route has been determined, the connection setting is permitted if there is a certain or more specified bandwidth in the link. It has a connection setting possible / impossible judgment algorithm changing means.

As a result, a call generated in the network can receive the effects of the alternative route selection processing and the crankback processing without depending on the value of the required bandwidth.

Fifth, in addition to the means of the communication node and the ATM node of the first or second configuration,
If the signaling processing recognition means recognizes that the signaling processing being executed determines whether or not a connection can be set up on the path initially selected by the source node, the generated call requests. If a band equal to or larger than the band exists in the link, the connection setting is permitted, and
If it is recognized that the signaling process being executed determines whether or not a connection can be set on the route selected as an alternative route after the execution of the crankback process, the connection process occurring in the network is performed. ,
It has a connection setting availability determination algorithm changing means for permitting connection setting if a band higher than the band required by the call requesting the largest band exists in the link.

As a result, all calls generated in the network can receive the effects of the alternative route selection processing and the crankback processing without depending on the value of the required bandwidth.

Further, according to the present invention, the communication is started by selecting and reserving resources on the communication path in advance, and if the resources on the communication path selected first cannot be reserved, the alternative path selecting process is executed. In the route selection connection method in the network to be performed, a plurality of judgment algorithms having different judgment criteria are prepared, and each time the communication path is selected, it is specified whether or not resources on the selected communication path can be reserved. Determine using the determination algorithm, when the resource reservation is not possible in the determination, perform an alternative communication path selection process, to determine whether or not resources can be reserved on the selected communication path, The determination algorithm to be used is changed depending on whether or not the resource selection of the communication path is determined by the first selection.

As a result, a call generated in the network can receive the effects of the alternative route selection process and the crankback process without depending on the value of the required bandwidth.

[0076]

DETAILED DESCRIPTION OF THE PREFERRED EMBODIMENTS The present invention relates to a network configured by an ATM node implementing a PNNI protocol, in which a connection setting is made depending on whether a signaling process is a process after crankback or a process before crankback. Is changed so that the effect of the crankback processing is not unfair depending on the type of connection. Hereinafter, embodiments of the present invention will be described with reference to the drawings. Will be described in detail.

FIG. 1 shows an example of the topology of an ATM network to which the present invention is applied using ATM nodes.
In FIG. 1, reference numerals 101 to 116 denote ATM nodes (ATM switch nodes).
03 and 204 are peer groups. Peer group 20
1 belongs to 101 to 104 ATM nodes, 105 to 108 ATM nodes belong to peer group 202, 109 to 111 ATM nodes belong to peer group 203, and 113 to 113 ATM nodes belong to peer group 204. 116
ATM nodes belong.

In such an ATM network,
In the following embodiment, processing when setting a connection from the ATM node 101 to the ATM node 116 using the PNNI algorithm will be described, and the routing / signaling processing executed at that time and the ATM of the present invention will be described. The details of the crankback processing by the node will be described.

First, the ATM node 101 in FIG.
FIG. 2 shows an example of a virtual network topology recognized by the topology recognition method in the NI.

As shown in FIG. 2, the ATM node 101
Recognizes peer groups other than the one to which the ATM node 101 belongs in the network as logical ATM nodes (202, 203, 204 in FIG. 1). Then, using the topology of FIG.
1 is performed.

FIG. 3 shows an example of a flowchart of the PNNI routing / signaling process in the present embodiment.
FIG. 4 shows an example of an algorithm used in the ATM node of the present invention to determine whether a call can be set up.

The determination algorithm executed in the ATM node according to the present invention is characterized in that whether the call setting to be executed is judged as “judgment of connection setting on the path selected first at the source node” The algorithm is characterized in that the algorithm to be executed is changed depending on whether it is “determination of whether connection can be set on the selected path after crankback processing”. Details will be described later.

Using the network topology shown in FIGS. 1 and 2 and the flowchart shown in FIG. 3 and an example of the execution algorithm shown in FIG. 4, an ATM connection setting process using the PNNI algorithm when the ATM node of the present invention is used ( Routing / signaling processing).

The ATM connection setting process of the present invention
As shown in FIG. 3, when the calling node that accommodates the calling terminal receives a connection setting request from the calling terminal, the calling node selects a route, and stores information on the selected route in the DTL stack. The stored information is transferred to the selected path, and the peer group determines from the transferred path information of the DTL stack whether or not it is possible to follow the path in its own peer group and link (connect). If possible, select a route in the own peer group, change a part of the received information of the DTL stack to the information of the selected route, and change the DTL stack to a route according to the route information. If there is no link using the route of the route information of the DTL stack, it is checked whether or not there is an alternative route.
An alternative route in the own peer group is selected, a part of the information of the DTL stack is changed to the information of the selected route, and the DTL stack is transferred to a route according to the route information. If the link using the path of the stack path information cannot be established and there is no alternative path, the DTL stack is returned to the node at a position where the alternative path can be selected, and the path is selected again (crank). back).

However, when the alternative route is selected by performing the crankback, the algorithm used for determining whether the connection can be set is changed. Specifically, a comparison is made between the largest bandwidth requested by a call generated in the network and the effective bandwidth in a link or an ATM node on a selected route for which a connection is to be established. Then, it is determined whether the connection can be set.

The algorithm for determining whether or not a connection can be set when crankback is not performed is based on the required bandwidth (Ban) described in the transmitted DTL stack.
d. Req) is a link for which a connection is to be set up, or an effective bandwidth (Band. Rema) in an ATM node.
In), it is determined that the requested connection can be set, and conversely, if it is larger or equal, it is determined that the connection cannot be set.

Determination of connection setting availability, that is,
The determination as to whether linking (connection) is possible is shown in FIG.
Follow the flow of.

As described above, in the peer group, DT
In performing the connection setting based on the information of the L stack, the algorithm of the connection setting determination is changed depending on whether the process is performed after the crankback process or not. Since the determination condition and the determination condition after performing the crankback process can be made independent and the determination condition after performing the crankback process is equally imposed on all calls,
The effect of the crankback process will be fair regardless of the type of connection.

The above is the outline of the present invention. Next, the specific contents of the processing will be described in detail. First, the flow of FIG. 3 will be described. When a connection setting request (call setting request) is received, processing S701 is executed.

[Process S701 in FIG. 3] First, when a call setup request is sent from an ATM terminal accommodated in the ATM node "101", the ATM node "101" is described in the request information. AT that accommodates the destination terminal
A route search process for setting a connection to the N node (here, the ATM node "116") is executed (process S701 in FIG. 3).

The route [101-103-104-] is obtained by the routing process by the ATM node "101".
203-204] is selected, the ATM node "101", which is the source node, creates a DTL stack in which the above-described selected path is written, and sequentially transfers the DTL stack along the selected path. Will be. At this time, the ATM node "10" which is the source node
No. 1 "describes the created DTL stack in a connection setup request message (so-called Setup message) and transfers it.

In the following, this message will be described as “DTL stack (Setup message)”. When transferring the DTL stack (Setup message), the ATM node existing on the selected route compares the QOS information requested by the generated call with the resource status of its own ATM node and the adjacent link on the route. If the requested QOS information can be satisfied, it is determined that the requested connection can be set, and the next ATM node on the selected path is set to the DT.
Transfer the L stack (Setup message). The determination as to whether or not a connection can be set follows the algorithm in FIG.

In the present embodiment, a case is considered in which it is determined whether a connection can be set using bandwidth information as required QOS information.

[Process S710 in FIG. 3] The process in S710 is as follows. At each ATM node that has received the DTL stack (Setup message) after the above-described processing (processing in S701), it is determined whether the bandwidth actually requested can be set on the link between the own node and the link between the nodes. I do.

FIG. 4 shows an example of an algorithm specifically executed at each ATM node at this time. Each AT
In the M node, first, the sent connection setting request is sent by the DTL stack (Setup message) created as a result of the first routing process in the source node, or is created after the crankback process. DTL stack (Setup
(S801).

At this time, the type of the transmitted DTL stack (Setup message) is determined by the crankback data in the signaling message (SetUp message).
Information element (Crackback
This is possible by referring to the description of the Information Element.

Further, by referring to the cause field (Cause Field) described in the crankback information element, it is possible to know the reason why the crankback processing was executed. It is also possible to change the processing to be executed. However, in the present embodiment, it is assumed that the crankback process occurs only when the band requested by the call does not actually exist, and a method of changing the process based on the value of the Cause Field is omitted.

FIGS. 5 and 6 show a crankback information element (Crackback inf).
5 shows an example of an operation element and a cause field.

Each ATM node uses the above method to
DTL stack sent (Setup message)
Is created as a result of the first routing process in the source node (ATM node 101) or newly created after the crankback process (process S801 in FIG. 4).

As a result, if the transmitted DTL stack (Setup message) is created as a result of the first routing process in the source node, a normal call setting availability determination algorithm (S in FIG. 4)
803 ("Algorithm-2"). FIG.
In this example, it is assumed that the type of the requested connection is a CBR connection, and an example is shown in which it is determined whether the requested bandwidth can be set by the peak assignment process (process S803 in FIG. 4).

That is, the transmitted DTL stack (S
The required bandwidth (Band.Req in FIG. 4) described in the E.Tup message) is smaller than the effective bandwidth (Band.Remain in FIG. 4) in the link on the selected route or the ATM node where the connection is to be set. In that case, it is determined that the requested connection can be set,
On the other hand, if the values are larger / equal, it is determined that connection setting is impossible.

Also, the transmitted DTL stack (Se
If the “tup message” is created as a result of the routing process after the crankback process, the connection between the largest bandwidth requested by the call generated in the network and the connection is set. The link on the selected route or the effective bandwidth in the ATM node is compared to determine whether or not the connection can be set ("the algorithm of S802 in FIG. 4"
1 ").

Also in this setting availability determination processing, an example is shown in which the type of the requested connection is considered to be a CBR connection, and the availability of the connection setup is determined by peak assignment processing.

Specifically, the ATM node of this embodiment is provided with means for storing bandwidth information requested by a call occurring in the network, and the required bandwidth information ( From S804 in FIG. 4, it is determined whether connection setting is possible or not by using the largest bandwidth information (Band.C in FIG. 4). In other words, the sent DT
The above-mentioned maximum bandwidth (Band.C) required by the call generated in the network is requested regardless of the required bandwidth (Band.Req) described in the L stack (Setup message). Considering the band, this value
A link on a selected route for which a connection is to be set or an effective band (Band. Remai) in an ATM node.
If n is smaller than n), it is determined that the requested connection can be set. Conversely, if it is larger or equal, it is determined that connection cannot be set (process S8 in FIG. 4).
02).

If it is determined that a connection can be set within the peer group “201”, the ATM node “104”, which is a link to the next peer group “203” on the selected route. And ATM node "109"
It is determined whether or not the connection requested on the link between can be set. The decision algorithm at this time is
This can be realized by executing the method in the same manner as shown in FIG.

At this stage, as a result of the first routing processing in the source node, the processing related to the transmitted DTL stack (Setup message) is performed. -2 ".

[Crankback processing S771] If it is determined in the processing S710 of FIG. 3 that connection setting within the peer group "201" is impossible, crankback processing is executed (processing of FIG. 3). S77
1). Since the crankback processing at this time is based on the result of the first routing processing by the source node, the DTL stack (Set
The ATM node to which the "up message" is returned is always the source node "101".

Upon receiving the cranked-back DTL stack (SetUp message), the source node “101” executes a new route selection process, and the newly transferred DTL stack (Setup message) executes the crankback process. The latter is indicated by the crankback information element (Cra
nkback Information Element
t), and the created message (including the created DTL stack) is transferred on the selected route in the same manner as in the above-described method.

In the connection setting availability determination algorithm in the above-described processing S710, the transmitted DTL stack (Setup message) was created as a result of the first routing processing in the source node. -2 "has been selected. However, after this processing, the transmitted DTL stack (Setup message) is created after the crankback processing, so that the determination algorithm executed by each ATM node is “algorithm—
1 ".

[Process S702 in FIG. 3] Process S71 in FIG.
0, the ATM node "1" which is a link within the peer group "201" and to the peer group "203".
When it is determined that the requested connection can be set on the link between “04” and “109”, the DT created by the ATM node “101” as the source node
The L stack (Setup message) is transferred to the ATM node “109”, and the ATM node 109 newly executes routing / signaling processing in the peer group “203”. ATM node "1"
09 ”indicates that the next transfer destination peer group is“ 204 ”in the received DTL stack (in the Setup message).
Therefore, in order to connect from the own node “109” to the peer group “204”, the routing / connection in the peer group “203” to which the own node belongs is described.
A signaling process is performed (process S702 in FIG. 3).

In FIG. 3, the routing [109-110-
112-204] is selected.

At this time, the ATM node 109 checks the peer group 2 in the route information of the transmitted DTL stack.
01 for the newly created peer group “2”.
A DTL stack rewritten with the route information in the “03” is created, and the DTL stack (Setup message) is sequentially transferred along the newly selected route in the peer group “203”.

At this time, the link and the ATM node existing on the newly selected route compare the QOS information of the requested call with the resource status of the own link and the ATM node in the same manner as in the previous case. If the requested QOS information can be satisfied, it is determined that connection setup is possible, and the DTL is sent to the next ATM node on the selected route.
Transfer the stack (Setup message).

[Process S720 of FIG. 3] Next, the DTL stack (Set) sent from the ATM node "109"
Each ATM node that has received the "up message" determines whether the bandwidth actually requested can be set on the link between the own node and the node (step S720 in FIG. 3). At this time, the algorithm specifically executed in each ATM node is the algorithm shown in FIG. 4 as in the case described above. Therefore, each ATM node determines whether the transmitted DTL stack (Setup message) is created as a result of the first routing process in the source node or created after several crankback processes. I do. As a result,
DTL stack sent (Setup message)
Is created as a result of the first routing processing in the source node, the “algorithm-2” in FIG. 4 is executed.

That is, the transmitted DTL stack (S
bandwidth (Ban) described in the setup message)
d. Req) is a link for which a connection is to be set up, or an effective bandwidth (Band. Rema) in an ATM node.
in), it is determined that the requested connection can be set, and conversely, if it is greater / equal,
It is determined that connection cannot be set.

Further, the received DTL stack (Se
If the “tup message” is created as a result of the routing process after the crankback process, the connection between the largest bandwidth requested by the call generated in the network and the connection is set. The link on the selected route or the effective bandwidth in the ATM node is compared to determine whether the connection can be set ("Algorithm-1" in FIG. 4).

Specifically, the required bandwidth information stored in the means for storing the bandwidth information requested by the call generated in the network in the ATM node of the present embodiment (S804 in FIG. 4). , The connection setting is determined using the largest band information (“Band.C” in FIG. 4). In other words, the maximum required bandwidth of the above-mentioned call generated in the network is determined regardless of the required bandwidth (“Band.Req”) of the call described in the transmitted DTL stack (Setup message). If this value is smaller than the effective bandwidth in the link or the ATM node on the selected route for which connection is to be set, it is determined that the requested connection can be set, and conversely, it is larger. If they are equal, it is determined that the connection cannot be set (process S80 in FIG. 4).
2).

If it is determined that a connection can be set within the peer group “203”, the ATM node “112” which is a link to the next peer group “204” on the selected route. And ATM node "114"
To determine whether the requested connection can be set up on the link between. The determination algorithm in this case can also be realized by executing the same method as the method shown in FIG.

[Crankback Processing S772] If it is determined in the processing S720 in FIG. 3 that the connection cannot be set in the peer group “203”, the crankback processing is executed (the processing in FIG. 3). S7
72). In the crankback process executed at this time,
Make sure that the source node (ATM node "101") has DT
The L stack (Setup message) is not always returned, and the existence of an alternative route in the peer group “203” and between the peer groups “203” and “204” may be first confirmed (FIG. 3). Process S740).

For example, the AT in the peer group “203”
If it is determined that a connection cannot be set on the link between the M nodes “111” and “112” and the crankback process is started, the peer group “20” is set.
Since there is an alternative route [109-110-112-204] in "3", the DTL stack (Setup message) is returned to the ATM node "109", and the routing / signaling process in the peer group 203 is executed again. (Step S781 in FIG. 3).

On the other hand, for example, the peer group 20
If there is only one route as in the case between the peer group 204 and the peer group 204, and if it is determined that the connection between the peer group 203 and the peer group 204 cannot be set on the link in this case, Has a peer group "204"
There will be no alternative route to reach.

In such a case, the DTL stack (S
setup message) to the source node (A in this case) in the peer group before the ATM node “109” that is the source node in the peer group “203”.
TM node "101"), and again,
The routing / signaling process from the ATM node “101” to the peer group “204” is executed (process S782 in FIG. 3).

In the routing / signaling process at this time, the link or the ATM node for which it is determined that the connection cannot be set before is deleted from the topology information used there, or a very large weight is given. In other words, a method of executing a process of selecting a new alternative route by preventing the route from being selected again by the route search process can be considered.

DTL stack (Setup message) subjected to crankback processing by the above method
The ATM node “109” or the ATM node “101” that has received the above, executes the new route selection processing as described above, and confirms that it is the DTL stack (Setup message) after executing the crankback processing. Cr
ankback Information Element
The message (including the created DTL stack) specified using nt is transferred onto the selected route in the same manner as in the above-described method.

The connection setting availability determination algorithm in the above-described processes S710 and S720 executed after the crankback process is based on the fact that the transmitted DTL stack (Setup message) is created after the crankback process. , "Algorithm-1" in FIG.

[Process S703 of FIG. 3] Process S72 of FIG.
0, the ATM node "1" which is a link within and to the peer group "203".
When it is determined that the requested connection can be set up on the link between “12” and “114”, the DTL stack created by the TM node “109” that is the source node in the peer group “203” (Setu
p message) is transferred to the ATM node "114".

In the ATM node "114", a routing / signaling process is newly performed in the peer group "204". Since the ATM node “114” has the peer group to which the own node belongs as the transfer destination peer group to which the call requests, the actual transfer destination ATM node to which the call request requests is included in the peer group to which the self node belongs. It can be recognized that it exists.

In this embodiment, since the transfer destination ATM node is set to "116", the routing / connection in the peer group "204" for connecting the ATM node "114" to the ATM node "116" is performed here.
A signaling process is performed (process S703 in FIG. 3).

If the route [114-116] is selected by the routing process by the ATM node "114", the ATM node "114"
D in which the route information in the peer group “203” described in the TL stack (Setup message) is rewritten to the route information in the newly created peer group “204”.
A TL stack (Setup message) is created, and as described above, the DTL stack (SetupP message) is sequentially transferred along the newly selected path.

At this time, the ATM node existing on the newly selected route compares the QOS information of the requested call with the resource status of its own node and the adjacent link as in the previous case, and receives the request. If the QOS information can be satisfied, it is determined that the connection can be set, and the DTL stack (Setup message) is transferred to the next ATM node on the selected route.

[Process S730 of FIG. 3] ATM node “1”
At each ATM node that has received the DTL stack (Setup message) sent from the 14 ", it is determined whether or not the requested bandwidth can be set on the link between itself and the nodes (see FIG. 3). Process S73
0). At this time, the algorithm specifically executed in each ATM node is the algorithm shown in FIG. 4, as in the case described above.

Therefore, in each ATM node, if the transmitted DTL stack (Setup message) is created as a result of the first routing processing in the source node, the "Algorithm-2" in FIG.
Execute

The DTL stack (Se
If the “tup message” is created as a result of the routing processing after the crankback processing, the request is made as the largest bandwidth among the bandwidths requested by the call generated in the network. The bandwidth is compared to determine whether the connection can be set ("Algorithm-1" in FIG. 4).

If it is determined that connection setting within the peer group “204” is possible, the final transfer destination A
The DTL stack (SetUp message) is transferred to the TM node "116", and a response message (Connect message) to the call connection request is sent back to the terminal that issued the call connection request, thereby completing the call connection process. .

[Crankback Processing S773] If it is determined in the processing S730 of FIG. 3 that the connection cannot be set in the peer group “204”, the crankback processing is executed again (FIG. 3). Processing S7
73). In the crankback processing executed in this case, the ATM node "10" which is the source node
The DTL stack (Setup message) is not always returned to “1”, and the existence of an alternative route in the peer group “204” may first be confirmed (the processing S7 in FIG. 3).
50).

In this case, the peer group “2” in FIG.
04, it is determined that it is not possible to set up a connection on the link between the ATM nodes "114" and "116", but instead [114-113-115]
−116]. Therefore, at this time, the DTL stack (Setup message)
It returns to the node "114", and executes the routing / signaling process in the peer group "203" again (process S783 in FIG. 3).

However, for example, once the peer group “2”
After performing the crankback process within the "04", the processes S703 and S730 of FIG. 3 are executed again. If the result indicates that the reconnection cannot be set any longer, the ATM node "114""To ATM node" 11
An alternative route to 6 "will no longer exist.

At this time, the DTL stack (Setup message) is stored in the AT
It will be returned to the M node “109”.

At this time, the route selection process (the process S702 in FIG. 3) can be executed immediately, but not only by such a method but also by once the ATM node "109"
It is also possible to confirm whether there is an alternative route to the TM node “116” and then proceed to the route selection process.

FIG. 3 shows the latter method.
At this time, since it is already known that a connection cannot be set on the path from the ATM node "114" to the ATM node "116", the ATM node "109" is re-entered.
Does not find a route in which a new connection can be set.

Therefore, in this case, the DTL stack (Setup message) is transmitted to the source node AT
The path is returned to the M node 101, and the route search process from the ATM node "101" to the peer group "204" is executed again (process S784 in FIG. 3).

After the route search processing, the connection setting is attempted again while determining whether or not the connection on the selected route can be set. In the determination as to whether or not the connection can be set, which is executed after such a process, the transmitted DTL stack (Setup message) is the one after the crankback process, and thus “Algorithm-1” in FIG. 4 is used. Will be.

As described above, the bandwidth information used to determine whether the connection can be set in the ATM node according to the present invention is obtained by creating the DTL stack (Setup message) requesting the determination as to whether the connection can be set up before the crankback process. , Or created later, the effect of the crankback process can be provided irrespective of the bandwidth information required by a call occurring in the network.

In particular, when determining whether or not to establish a connection activated by a DTL stack (Setup message) created after the crankback processing, the effective bandwidth within the ATM node or link on which the connection is to be established, By comparing the value of the maximum bandwidth in the bandwidth required by the calls generated in the network with the judgment of the connection setting, all the calls generated in the network are affected by the crankback processing. Can be enjoyed equally.

[Internal Configuration of ATM Node] FIG. 7 is a block diagram showing an example of the internal functional configuration of the ATM node shown in this embodiment. In the example shown in FIG. 7, the ATM interface processing units 1104 and 1106 for executing the actual ATM cell transfer / exchange processing, and the ATM switch 110
There are five.

A PNNI routing / signaling process execution unit 1103 for executing a series of routing processes, a DTL stack (Setup message) transfer / reception process, a connection setting availability determination process, and the like, and a PNNI routing / signaling process execution unit 1103 therein. There is a crankback processing execution unit 1102 that stores a DTL stack (Setup message) for the crankback processing, determines an ATM node to be returned when the crankback processing occurs, and executes processing for confirming the existence of an alternative route. .

Further, the ATM node shown in this embodiment has a PNNI routing / signaling process execution unit 1
The DTL stack (Setup message) received by 103 determines whether or not it is after the crankback process, and determines / selects an algorithm to be used for determining whether connection can be set based on the result. There is a processing unit 1101.

As an example of the case where the actual signaling processing is executed by these processing units, the processing S70 in FIG.
2 to S720 and S740
The specific processing procedure of each function in the ATM node “109” will be described below.

In the ATM node "109" having the configuration as shown in FIG. 7, the ATM node "104"
A DTL stack (Setup message) is received via the interface processing unit 1106. And this D
The ATM node “109” that has received the TL stack
Transfers the received DTL stack (SetUp message) to the PNNI routing / signaling processing execution unit 1103.

The DTL stack (Setup message) received from the PNNI routing / signaling processing execution unit 1103 is stored in the algorithm selection processing unit 1.
Then, it is determined whether or not the received DTL stack (Setup message) has been subjected to crankback processing.

As a result, the algorithm selection processing section 110
1 indicates that if the DTL stack (Setup message) has not been subjected to the crankback process, A
It is instructed to use “algorithm-2” in FIG. 4 to determine whether or not a connection can be set up in the TM node “109”, and the DTL stack (Setup message) was created after the crankback processing was performed. In this case, an instruction is given to use “algorithm-1” in step S802 in FIG. 4 in determining whether a connection can be set in the ATM node “109”.

At this time, the algorithm selection processing unit 110
1 in the determination of the type of the received DTL stack (Setup message), the Crankback in the received DTL stack (Setup message)
Description of k InformationElement and C
This is possible by referring to the "use field".

As a result, using the selected algorithm, first, the requested bandwidth is changed to the ATM node “10”.
9 "is set or not. As a result, if it is determined that the setting is possible, the peer group" 2 "is set.
03 ”is searched for a connection setting route.

At this time, the own node belongs according to the virtual topology information of the network stored in the ATM node “109” and the entire route information and destination information described in the received DTL stack (Setup message). The route search process is executed in the peer group “203”.

As shown in FIG. 3, here [109-1]
10-112-204], the ATM node "109" and the ATM node "1"
Whether or not a connection can be set up on the link between 10 ”is evaluated by the previously selected algorithm. If it is determined that the connection can be set up, the DTL is newly added. The stack (Setup message) is rewritten, and the rewritten DTL stack (Setup message) is transferred to the next ATM node on the selected route. Is done.

An ATM node on the route selected by the ATM node "109", for example, the ATM node "11"
"0" checks whether or not the received DTL stack (Setup message) has been subjected to crankback processing as in the previous case, and selects a connection setting determination algorithm to be used.

If it is determined that the connection cannot be set, the DTL stack (Se
tup message) to the crankback processing execution unit 110
2 to the DTL stack (S
(the setup message).

At this time, the DTL stack (Setu)
p message) and the DTL stack (Setup
Message), and the reason for determining that the connection cannot be set is added as new information to the DTL stack (Setup message) to perform crankback processing (DTL stack (Setup message)). (Return processing).

As described above, the AT implementing the PNNI protocol
In a network composed of M nodes, whether the signaling process is a process after crankback,
By changing the algorithm for determining whether or not the connection can be set depending on whether the process is performed before crankback, it is possible to prevent the effect of the crankback process from being unfair depending on the type of connection.
A TM node can be provided. Conventionally, there has been unfairness due to the fact that the effect of the crankback processing in the PNNI protocol differs depending on the type of connection. In particular, this unfairness becomes remarkable in a multi-connection environment where calls having different requested bands exist. Therefore, in the present invention, at the time of the signaling process in the ATM node, the algorithm for determining whether or not the connection can be set is changed depending on whether or not the signaling process is performed after the crankback, so that the unfairness is corrected. did.

Note that the present invention is not limited to the above-described example, and can be implemented with various modifications.

[0161]

As described above, the AT according to the present invention is used.
According to the M node, in a call setup process in an ATM network constituted by ATM nodes connected by a PNNI interface, even in a multi-connection environment, the effect of crankback, which is one of the PNNI processes, can be reduced for all types of connections. Can be enjoyed equally.

[Brief description of the drawings]

FIG. 1 is a diagram for explaining the present invention, showing an example of a network topology using an ATM node of the present invention.

FIG. 2 is a diagram for explaining the present invention, showing an example of a network topology recognized by the ATM node of the present invention;

FIG. 3 is a diagram for explaining the present invention, and is a diagram showing an example of a flowchart showing a procedure of a signaling process executed in the ATM node of the present invention.

FIG. 4 is a diagram for explaining the present invention, showing a detailed example of a procedure of a signaling process executed in the ATM node of the present invention, in particular, a process of deciding whether connection setting is possible or not.

FIG. 5 is a diagram for explaining the present invention, and is a diagram illustrating a crankback information element (Crack);
back information Element)
FIG.

FIG. 6 is a diagram for explaining the present invention,
The figure which shows an example of a field (Cause Field).

FIG. 7 is a diagram for explaining the present invention, and is a block diagram showing an example of the internal configuration of the ATM node of the present invention.

FIG. 8 is a diagram showing the concept of a network topology recognition method in PNNI.

FIG. 9 is a diagram showing an example of topology information used in a conventional routing method.

FIG. 10 is a diagram for explaining the concept of a topology recognition method in the PNNI protocol.

FIG. 11 is a view for explaining an example of a signaling procedure using a DTL stack in the PNNI protocol.

FIG. 12 is a diagram for explaining an example of a signaling procedure using a DTL stack in the PNNI protocol.

FIG. 13 is a view for explaining an example of a crankback processing procedure in the PNNI protocol.

[Explanation of symbols]

101 to 116 ATM switch nodes (ATM nodes) 201 to 204 Peer groups 1101 Algorithm selection processing unit 1102 Crankback processing execution unit 1103 PNNI routing / signaling processing execution units 1104 and 1106 ATM interface processing unit 1105 ATM Switch 1110 ... ATM node.

Claims (11)

[Claims] The present invention relates to a network in which a resource on a communication path is selected and reserved in advance to start communication, and when a resource on a communication path selected first cannot be reserved, a process of selecting an alternative path is executed. In the connected node device, a plurality of determination means having different determination criteria are provided, and each time the communication path is selected, the determination means for designating whether or not a resource on the selected communication path can be reserved is provided. First signaling processing means for determining using the first communication means, an alternative path selection means for executing an alternative communication path selection process when the signaling processing means determines that the resource reservation is impossible, and communication selected by the alternative path selection means. A second signaling processing means for determining whether or not a resource can be reserved on a route; and a first signaling processing means used by the first signaling processing means. And a realism changing means for changing the designation according to whether or not the resource reservation of the communication route is determined by the selection. 2. An ATM node used in an ATM network for starting communication by selecting and reserving resources on a communication path in advance, operating according to a PNNI protocol, and selecting a path first selected by a source node when setting a connection. Signaling processing means for deciding whether or not a connection can be set up, or whether or not a connection can be set up on a selected alternative route; and performing crankback processing in the PNNI when the signaling processing means determines that connection cannot be set. An ATM node comprising: a crankback processing unit that performs the connection setting determination process; and a connection setting availability determination algorithm changing unit that changes the connection setting determination method performed by the signaling processing unit in accordance with the presence or absence of crankback execution. De. 3. An algorithm change area determining means for executing a change by said algorithm changing means for a call requesting a required bandwidth among calls generated in a network. The communication node according to claim 1. 4. An algorithm change area determining means for making a change by said connection setting availability determination algorithm changing means for a call requesting a required bandwidth among calls generated in the network. The ATM node according to claim 2, wherein 5. The method according to claim 1, further comprising the step of determining the algorithm change area, wherein the signaling processing means sets the connection of the communication path selected first when a bandwidth equal to or more than the required bandwidth of the generated call exists in the link. 2. The method according to claim 1, wherein the setting of the connection of the alternative route permits setting when a band equal to or more than a predetermined value determined by the algorithm change area determining unit exists in the link.
A communication node according to. 6. The method according to claim 1, further comprising the step of determining the algorithm change area, wherein the signaling processing means sets the connection of the communication path selected first when a bandwidth equal to or more than the required bandwidth of the generated call exists in the link. 3. The ATM node according to claim 2, wherein the setting of the connection of the alternative route is permitted when a band equal to or more than a predetermined value determined by the algorithm change area determining means exists in the link. 7. Bandwidth information collecting / storing means for collecting / storing requested bandwidth information requested by a call originating in the node or the entire network, and a request stored in the bandwidth information collecting / storing means. An algorithm changing area selecting means for selecting a required band for executing the algorithm changing means or the connection setting availability determining algorithm changing means among the bands, wherein the signaling processing means sets the connection of the communication path selected first. When a bandwidth equal to or more than the required bandwidth of the generated call exists in the link, the connection setting of the alternative route is set when a bandwidth equal to or greater than the predetermined value determined by the algorithm change area selecting means exists in the link. 2. The method according to claim 1, wherein the permission is granted.
A communication node according to. 8. Bandwidth information collecting / storing means for collecting / storing requested bandwidth information requested by a call originating in the node or the entire network, and request bandwidth information stored in said bandwidth information collecting / storing means. home,
An algorithm change area selecting means for selecting a required band to be executed by the algorithm changing means or the connection setting availability determination algorithm changing means, wherein the signaling processing means has set a connection setting of the communication path selected first. When a band equal to or more than the required band of the call exists in the link, the connection setting of the alternative route permits setting when a band equal to or more than the predetermined value determined by the algorithm change area selecting means exists in the link. The ATM node according to claim 2, wherein: 9. The communication apparatus according to claim 1, further comprising: a bandwidth information collecting / storing means, wherein the signaling processing means sets a connection of the communication path selected first, wherein a bandwidth equal to or more than a required bandwidth of the generated call exists in the link. At this time, the connection setting of the alternative route permits setting when a bandwidth equal to or more than the maximum required bandwidth is present in the link among the required bandwidths stored in the bandwidth information collection / storage means. The communication node according to claim 1, wherein 10. The apparatus according to claim 1, further comprising: said band information collecting / storing means.
The signaling processing means sets the connection setting of the communication path selected first when the bandwidth equal to or more than the requested bandwidth of the generated call exists in the link, and sets the connection setting of the alternative path to the bandwidth information collection / storage means. 3. The ATM node according to claim 2, wherein the setting is permitted when a bandwidth equal to or more than the maximum required bandwidth is present in the link among the stored required bandwidths. 11. A network in which communication is started by selecting and reserving resources on a communication path in advance, and when a resource on a communication path selected first cannot be reserved, a process for selecting an alternative path is performed. In the path selection connection method, a plurality of judgment algorithms having different judgment criteria are prepared, and each time the communication path is selected, the judgment algorithm for designating whether or not a resource on the selected communication path can be reserved is determined. If the resource reservation is not possible in the determination, a process of selecting an alternative communication path is performed, and it is determined whether or not a resource can be reserved on the selected communication path. The network according to claim 1, wherein the determination algorithm changes the designation according to whether or not the resource selection of the communication path is determined by the first selection. Route selection connection method.