JP2013247667A - Route specifying method and device and route specifying program in network - Google Patents

Abstract

PROBLEM TO BE SOLVED: To provide a route specifying method and device, and a route specifying program, by which a processing load can be reduced when specifying three or more different routes which do not share a link/node on a way of the route connecting two nodes.SOLUTION: The method includes: specifying k route-containing closed routes (C200+C400, C300) which pass an arbitrary first node (N300) and second node (N500) and do not intersect each other based on the closed route closed by a plurality of nodes and links in a network; and specifying different (k+1) routes between the first node and the second node from the route-containing closed route.

Description

  The present invention relates to a method and apparatus for specifying a route in a network in which a plurality of nodes are linked and a route specifying program.

  In a network in which a plurality of nodes are linked, a method of repeatedly using a Dijkstra method (see Non-Patent Document 1) is the most representative method for specifying a redundant route. For example, when considering a graph in which a cost value is given to each link and the cost of the route is given by the sum of the costs of the links through which the route passes, first, the route with the lowest cost as a route connecting two end point nodes ( The first route) is specified according to the Dijkstra method. Next, all links or nodes included in the specified route are excluded, and a route (second route) with the lowest cost is specified according to the Dijkstra method again as a route connecting the end point nodes. In this way, it is possible to specify two different routes (first route and second route) that do not share links or nodes between the end point nodes.

  However, in the method of repeatedly applying the Dijkstra method described above, a detour route may not be specified if the topology of the graph includes a trap structure. For example, consider a case where the route connecting the node N100 and the node N400 is specified in the graph shown in FIG. First, when a route connecting links L100-L900-L500-L400 in this order is specified as a minimum cost route (first route) by the Dijkstra method, then all links included in the first route ( (Broken link in the figure) is excluded. Then, it becomes impossible to specify a route connecting the node N100 and the node N400 any more. That is, it can be said that the topology of this graph includes a trap structure.

  In order to cope with such a problem, the Bhandari method is known (see Non-Patent Document 2). According to this method, a part of the topology of the graph is converted by extending the process of repeatedly using the Dijkstra method, and the trap link that is the basis of the trap structure is specified. Subsequently, in the graph excluding the identified trap link, two different routes that do not share the link are identified by repeatedly using the Dijkstra method again. Therefore, it is possible to calculate a plurality of different routes by repeatedly using the Bhandari method.

E. W. Dijkstra, “A note on two problems in connexion with graphs”, Numerische Mathematik, 1 (1959), pp.269-271 Ramesh Bhandari, “Survivable Networks: Algorithms for diverse routing”, Springer (1999)

However, in the method using the Bhandari method, the amount of calculation for specifying the trap link and eliminating the influence of the trap link is relatively large. Therefore, when the Bhandari method is used repeatedly, the processing load for avoiding the trap structure increases. . For example, when calculating k different routes, the Dijkstra method or the alternative shortest route calculation method needs to be repeated {(k ² + 3k−2) / 2} at most, and the amount of calculation increases as the value of k increases. It increases in a quadratic function.

  Therefore, an object of the present invention is to provide a path specifying method and apparatus and a path specifying program capable of reducing the processing load when specifying three or more different paths that do not share a link / node in the middle of a path connecting two nodes. It is to provide.

The route specifying method according to the present invention is a route specifying method in a network in which a plurality of nodes are linked, and is based on a route closed by a plurality of nodes and links in the network (hereinafter referred to as closed circuit). K path-containing cycles that pass through any first node and second node and do not cross each other (k is an integer of 2 or more), and the first node and the second node are identified from the k path-containing cycles. And (k + 1) different paths between and are identified.
The route specifying device according to the present invention is a route specifying device in a network in which a plurality of nodes are linked, and is based on a route closed by a plurality of nodes and links in the network (hereinafter referred to as closed circuit). A path specifying means for specifying k (k is an integer of 2 or more) path-containing cycles that do not cross each other through any first node and second node, and the first node from the k path-containing cycles Route specifying means for specifying different (k + 1) routes from the second node.
A program according to the present invention is a program that causes a computer to function as a path specifying device in a network in which a plurality of nodes are linked, and is a path closed by a plurality of nodes and links in the network (hereinafter referred to as a closed path). ) To specify k path-containing cycles that do not cross each other through any first node and second node (k is an integer greater than or equal to 2), and A function of specifying different (k + 1) paths between a node and the second node is realized by the computer.

  According to the present invention, it is possible to reduce the processing load when specifying three or more different routes that do not share a link / node in the middle of a route connecting two nodes.

FIG. 1 is a diagram showing a network topology for explaining a route specifying method according to the background art. FIG. 2 is a schematic block diagram showing a functional configuration of the path identifying device according to the first embodiment of the present invention. FIG. 3 is a schematic network diagram for explaining the route specifying method according to the first embodiment. FIG. 4 is a schematic diagram showing basic graph information of the network shown in FIG. FIG. 5 is a flowchart for generating a closed adjacency graph in the first embodiment. FIG. 6 is a schematic diagram conceptually showing the spanning tree in the first embodiment. FIG. 7 is a schematic diagram showing the closed circuit information of the closed circuit C100 in FIG. FIG. 8 is a schematic diagram showing the closed circuit information of the closed circuit C200 in FIG. FIG. 9 is a schematic diagram showing the closed circuit information of the closed circuit C300 in FIG. FIG. 10 is a schematic diagram showing the closed circuit information of the closed circuit C400 in FIG. FIG. 11A is a diagram showing a closed adjacency graph for explaining the route specifying method according to the first embodiment of the present invention, and FIG. 11B is a closed circuit corresponding to the closed adjacency graph of FIG. It is a schematic diagram which shows adjacent graph information. FIG. 12 is a schematic diagram showing a cycle search graph using the cycle adjacency graph shown in FIG. FIG. 13 is a flowchart showing a route specifying method according to the first embodiment. FIG. 14 is a schematic diagram showing a graph obtained by removing a search exclusion link from the cycle search graph shown in FIG. FIG. 15 is a schematic diagram showing cycle information representing a route-containing cycle according to the first embodiment. FIG. 16 is a diagram showing a closed adjacency graph for explaining a route specifying method according to the second embodiment of the present invention. FIG. 17 is a schematic diagram showing a cycle search graph using the cycle adjacency graph shown in FIG. FIG. 18 is a schematic diagram showing merged cycle information stored in the graph information storage unit of the route specifying device according to the third embodiment of the present invention. FIG. 19 is a flowchart showing a route specifying method according to the fourth embodiment of the present invention. FIG. 20 is a schematic block diagram showing a functional configuration of a path identifying device according to the fifth embodiment of the present invention.

  According to the embodiment of the present invention, in a network in which a plurality of nodes are connected by links, k (> = 2) path-containing cycles including any first node and second node as a common vertex are identified. Then, (k + 1) different routes that do not overlap are identified from the k route-containing cycles. The procedure for specifying the k or more route-containing cycles is performed by reducing the shortest route calculation k times in the cycle search graph created based on the topology of the network. In this way, it is possible to specify three or more different routes that do not share a link / node in the middle of the route connecting the first node and the second node with a small amount of calculation.

  The route specifying device according to the embodiment of the present invention can be configured as an information processing device, for example, a server device in a network, a network switch, a router, a personal computer, a mobile phone terminal, a PHS (Personal Handyphone System), a PDA (Personal Data). Assistance or Personal Digital Assistant), a smartphone, a car navigation terminal, a game terminal, or the like. That is, the path specifying device according to the present embodiment includes a central processing unit (CPU), a primary storage device (RAM), and a secondary storage device (hard disk drive (HDD)). It can be mounted on an information processing apparatus.

  Further, the route determination function according to the present embodiment may be realized by executing a program stored in the secondary storage device on the CPU, or may be realized by hardware such as a circuit. Further, the program may be stored in a computer-readable recording medium instead of the secondary storage device. For example, a portable medium such as a flexible disk, an optical disk, a magneto-optical disk, and a semiconductor memory can be used as the recording medium. Hereinafter, embodiments and examples of the present invention will be described in detail with reference to the drawings.

1. 1. First Embodiment 1.1) Configuration As shown in FIG. 2, a route specifying device 1 according to a first embodiment of the present invention controls the operation of the entire route specifying device, and executes various data processing and operations. 10, a graph information storage unit 11, a cycle specifying unit 12, a cycle search space specifying unit 13, a route-containing cycle specifying unit 14, a route specifying unit 15, and a route selecting unit 16.

a) Graph information storage unit 11
The graph information storage unit 11 stores basic graph information, cycle information, cycle adjacency graph information, cycle search graph information, and route-containing cycle information. Although details will be described later, the outline of the information stored in the graph information storage unit 11 is as follows.

  The basic graph information includes a basic node identifier for identifying each of a plurality of basic nodes constituting the basic graph, a basic link identifier for identifying each of the plurality of basic links, and any two of the plurality of basic nodes. Information for specifying a basic link connecting two basic nodes (that is, information indicating whether or not two basic nodes are connected by a certain basic link) (specifically, FIG. 3 and FIG. 4) reference). The basic graph information is expressed in a format such as an adjacency matrix or an adjacency list, but may be expressed in other formats that can be processed by the information processing apparatus.

  The cycle information is information representing a cycle (unit cycle) in the basic graph. The closed circuit information includes a closed circuit identifier for identifying the closed circuit and information for specifying the basic link constituting the closed circuit (specifically, refer to FIGS. 7 to 10).

  The cycle adjacency graph information includes a cycle node identifier for identifying each of a plurality of cycle nodes constituting the cycle adjacency graph, a cycle adjacent link identifier for identifying each of the plurality of cycle adjacent links, and a plurality of cycle nodes Information for specifying a closed link adjacent to any two closed nodes (that is, information indicating whether or not the two closed nodes are connected by the closed link). 11 (B)). The closed adjacency graph information is expressed in a format such as an adjacency matrix or an adjacency list, but may be expressed in other formats that can be processed by the information processing apparatus.

  Here, in the cycle adjacency graph, each of a plurality of cycles (unit cycles) in the basic graph is represented by a cycle node, and a basic graph element (basic node or basic link) in the basic graph is shared by two cycles. This is a graph that represents the presence of a closed link adjacent to the two closed circuits (specifically, refer to FIG. 11A).

   The cycle search graph information is information representing a cycle search graph. Here, the cycle search graph is a graph in which an inclusion node and an inclusion link are added to the cycle adjacency graph (specifically, see FIG. 12). An inclusion node is a node that represents a basic node through which the cycle represented by the cycle node passes in the basic graph (that is, included in the cycle). The inclusion link is a link that represents that the cycle represented by the cycle node passes through the inclusion node in the basic graph, and that connects the cycle node and the inclusion node. The cycle search graph information includes, in addition to the cycle adjacency graph information, an inclusion node identifier for identifying each of the plurality of inclusion nodes, and for each of the arbitrary combinations of inclusion nodes and cycle nodes, the inclusion node and Information indicating whether or not there is an inclusive link connecting the closed node. Therefore, the graph information storage unit 11 can hold the cycle adjacent graph information as a part of the cycle search graph information, or can hold it separately from the cycle search graph information.

  The route-containing cycle information is information representing the route-containing cycle in the basic graph. A route-containing cycle is a cycle that passes through two nodes (a first basic node and a second basic node) as end points of a route in the basic graph. The route-containing cycle information includes a route-containing cycle identifier for identifying the route-containing cycle and information for specifying a basic link that constitutes the route-containing cycle.

b) Closure specifying unit 12
The cycle specifying unit 12 includes a plurality of first cycles that pass through the first basic node and second cycles that pass through the second basic node based on the basic graph information stored in the graph information storage unit 11. Identify the cycle of. The cycle specifying unit 12 stores cycle information indicating the specified cycle in the graph information storage unit 11.

c) Cycle search space specifying unit 13
The cycle search space specifying unit 13 selects and selects a cycle that can be used to specify a cycle that passes through the first basic node and the second basic node among the plurality of cycles specified by the cycle specifying unit 12. The selection information for acquiring the closed cycle information or the selected closed cycle is stored in the graph information storage unit 11.

d) Route-containing cycle specifying unit 14
The route-containing cycle specifying unit 14 specifies a route-containing cycle by merging the cycles sharing at least one of the basic graph elements among the cycles selected by the cycle search space specifying unit 13. Here, when the two closed circuits share at least one basic link, the route-containing closed circuit specifying unit 14 includes all the basic links that constitute a part excluding the shared link from one of the two closed circuits, A closed circuit obtained by connecting all the basic links constituting the portion excluding the shared link from the other of the two closed circuits is used as a closed circuit obtained by merging the two closed circuits. Here, the shared link is a basic link shared by the two closed circuits.

  Specifically, the route-containing cycle specifying unit 14 generates a cycle adjacent graph based on the cycle information stored in the graph information storage unit 11 and stores the generated cycle adjacent graph information in the graph information storage unit 11. To do. In addition, the route-containing cycle specifying unit 14 specifies a route-containing cycle in the basic graph based on the generated cycle adjacency graph. At this time, the route-containing cycle specifying unit 14 specifies a route-containing cycle in the basic graph so as to minimize the sum of basic link costs set in advance for each of the basic links in the basic graph. In addition, only the portion corresponding to the cycle selected by the cycle search space specifying unit 13 in the cycle adjacent graph is used. The route-containing cycle specifying unit 14 stores route-containing cycle information representing the specified route-containing cycle in the graph information storage unit 11.

e) Route specifying unit 15
The route specifying unit 15 uses the route-containing cycle information specified by the route-containing cycle specifying unit 14 stored in the graph information storage unit 11 to connect the first basic node and the second basic node, In addition, two different routes that do not share the basic link are specified, and the route information is stored in the graph information storage unit 11.

f) Route selection unit 16
The route selection unit 16 connects the first basic node and the second basic node from the route information recorded in the graph information storage unit 11 and does not share the basic link, and is different by k + 1 (k ≧ 1) or more. Select a route.

1.2) Operation Next, the route specifying operation of the route specifying device 1 will be described using the basic graph shown in FIG. 3 as an example. However, here, for the sake of simplicity, a basic graph having a relatively small scale is illustrated, but the route specifying device 1 according to the present embodiment is also used for a basic graph having a larger scale or a basic graph having a smaller scale. Needless to say, it is applicable.

  The basic graph shown in FIG. 3 is composed of seven basic nodes N100 to N700 and nine basic links L100 to L900, and this basic graph is represented by the basic graph information (adjacency matrix) shown in FIG. In the adjacency matrix shown in FIG. 4, “N100” to “N700” are basic node identifiers, respectively, and the value “1” in the adjacency matrix indicates two basic nodes respectively associated with the row and column of the value. Indicates that they are connected by a basic link.

1.2.1) Generation of a cycle adjacency graph First, the control unit 10 of the route identification device 1 controls the cycle identification unit 12 according to the procedure shown in FIG. 5 to generate a cycle adjacency graph from the basic graph.

  In FIG. 5, the cycle identification unit 12 calculates a spanning tree T in the basic graph (step S101). Various methods such as a depth-first search method, a breadth-first search method, a Dijkstra (Dijkstra) method, or a prim method can be used for obtaining the spanning tree. Here, it is assumed that the spanning tree shown in FIG. 6 is acquired for the basic graph shown in FIG. That is, the spanning tree in this example is composed of basic nodes (N100 to N700) and basic links (L100, L400 to L700, and L1000) indicated by solid lines, as shown in FIG.

  Next, the closed loop specifying unit 12 generates a set E ′ of basic links not included in the acquired spanning tree T (step S102). Therefore, according to the example shown in FIG. 7, the basic link set E ′ in the initial state includes the broken basic links L200, L300, L600, L800, and L900 in FIG. 6 as members. Hereinafter, the cycle identifying unit 12 identifies the cycle (unit cycle) by adding the basic links read from the basic link set E ′ one by one to the spanning tree illustrated in FIG. 6 in steps S103 to S106. A closed adjacency graph is generated based on the specified closed cycle. Specifically, it is as follows.

  First, the cycle identification unit 12 determines whether or not the basic link set E ′ is an empty set (step S103). If the basic link set E ′ is not an empty set (step S103; No), the basic link set E ′ is determined from the basic link set E ′. One cycle is read, and a closed path formed by adding this basic link to the spanning tree T is specified, and the read basic link is deleted from the basic link set E ′ (step S104).

  Subsequently, the cycle identification unit 12 adds a cycle node representing the identified cycle to the cycle adjacency graph (step S105), the cycle represented by the cycle node added up to the previous time, and the cycle newly added this time A closed link connecting the closed route represented by the node is added to the closed route adjacency graph (step S106). The processes in steps S103 to S106 are repeated until the basic link set E ′ becomes an empty set. When the basic link set E ′ becomes an empty set (step S103; Yes), the closed-circuit adjacent graph generation process is terminated. .

  The cycle specifying unit 12 stores the cycle information and the cycle adjacent graph information obtained by the above processing in the graph information storage unit 11.

  The cycle adjacency graph generation processing shown in FIG. 5 will be described using the spanning tree shown in FIG. 6 as an example. First, the basic link L200 is read from the basic link set E ′ and added to the spanning tree shown in FIG. 6 to obtain a closed path C100 composed of paths N100-N200-N600-N100, which becomes a closed node C100. Similarly, by adding the basic link L300, the closed node C200 from the route N200-N300-N700-N200 is added, and by adding the basic link L800, the closed node C300 from the route N300-N400-N500-N700-N300 is changed to the basic. By adding the link L900, the closed node C400 is obtained from the route N200-N700-N500-N600-N200, respectively.

  The cycle C100 is represented by the cycle information (adjacency matrix) shown in FIG. 7, and the cycle identifier is “C100”. Similarly, the closed circuit C200 is represented by the closed circuit information (adjacency matrix) shown in FIG. 8 and the closed circuit identifier is “C200”, and the closed circuit C300 is represented by the closed circuit information (adjacent matrix) shown in FIG. 9 and the closed circuit identifier is “C300”. C400 is represented by the cycle information (adjacency matrix) shown in FIG. 10, and the cycle identifier is “C400”.

  As shown in FIG. 6, the closed circuit C100 and the closed circuit C400 share the basic link L400. The closed circuit C200 and the closed circuit C300 share the basic link L700, the closed circuit C200 and the closed circuit C400 share the basic link L500, and the closed circuit C300 and the closed circuit C400 share the basic link L1000. Note that when a plurality of linked basic links are shared between two closed circuits, a node located at a position other than both ends of a section shared by the two closed circuits is called a critical node.

  The cycle adjacency graph generated in this way can be expressed as shown in FIG. That is, the closed node C100 and the closed node C400 are connected by the adjacent link A100, the closed node C200 and the closed node C400 are connected by the adjacent link A200, the closed node C200 and the closed node C300 are connected by the adjacent link A300, and the closed node C300 is connected. Each of the closed nodes C400 is wrought by the adjacent link A400. In this case, the adjacent link A100 is associated with the basic link L400, the adjacent link A200 is associated with the basic link L500, the adjacent link A300 is associated with the basic link L700, and the adjacent link A400 is associated with the basic link L1000. Hereinafter, a state in which the adjacent link A is associated with the basic link L is simply referred to as “the adjacent link A includes (or includes) the basic link L”. An adjacent link including a plurality of connected basic links includes a critical node.

  The closed adjacency graph shown in FIG. 11A is represented by closed adjacency graph information (adjacency matrix) shown in FIG. In FIG. 11B, the value “1” in the adjacency matrix indicates that there is an adjacent link connecting between the closed nodes respectively associated with the row and column of the value.

  Note that the route identification device 1 may generate a closed adjacency graph by a method other than the method described above. However, in the method of generating the cycle adjacency graph, each basic link in the basic graph is included in at least one cycle node (first condition), and the cycle adjacency graph is continuous (second). It is preferable to satisfy both of the above conditions. In the method described above, each basic link in the basic graph is included in the spanning tree or is a basic link added to form a cycle, so the first condition is satisfied, and each cycle is one of the spanning trees. The second condition is also satisfied.

1.2.2) Generation of Cycle Search Graph Subsequently, the control unit 10 controls the cycle search space specifying unit 13 to add an inclusion node to the cycle adjacency graph, and an inclusion that connects the cycle node and the inclusion node. A cycle search graph is generated by adding a link to the cycle adjacency graph. FIG. 12 shows a cycle search graph when the basic graph shown in FIG. 3 is used. In FIG. 12, broken lines connecting the closed circuit C and the comprehensive node N represent the inclusion links E101 to E114.

1.2.3) Specifying Route Next, the control unit 10 controls the route containing cycle specifying unit 14, the route specifying unit 15, and the route selecting unit 16 to specify a route based on the cycle adjacency graph. Hereinafter, the route specifying operation according to the present embodiment will be described with reference to FIGS.

  As described above, the graph information storage unit 11 stores basic graph information, cycle information, and cycle adjacency graph information. As described below, the control unit 10 specifies a route based on these pieces of information. Control to do.

First, as basic graph information, a basic link cost preset for each basic link is set to w ₀ (x). The parameter x is information for specifying the basic link. The basic link cost w ₀ (x) may be information input by the user of the route identification device 1. The basic link cost w ₀ (x) has a scalar value, and is preferably set based on the length of the basic link or other attributes of the basic link. By the way, the cost W (p) of the path p (= {l ₀ , l ₁ ,..., L _i }) as a basic link sequence is expressed by the following equation (1).

Here, l _j is the j-th basic link, and j is an integer from 0 to i.

Further, the control unit 10 calculates the adjacent link cost w (x) for each of the adjacent links in the closed circuit adjacent graph based on the basic link cost w ₀ (x). When the parameter x is an adjacent link connecting the closed node c ₀ and the closed node c ₁ , the control is performed when the adjacent link includes neither the first basic node nor the second basic node as a critical node. The unit 10 calculates the adjacent link cost w (x) based on the following equation (2).

w (x) = W (c _o ) + W (c ₁ ) −2W (c _o ∩c ₁ ) (2)

On the other hand, when the parameter x is an adjacent link connecting the closed node c ₀ and the closed node c ₁ , when the adjacent link includes at least one of the first basic node or the second basic node as a critical node, The control unit 10 calculates the adjacent link cost w (x) based on the following equation (3).

w (x) = ∞ (3)

 When the parameter x is an inclusion link and the inclusion link is a link from the first basic node to the closed node c, the control unit 10 calculates the search link cost w (x) by the following equation (4) ).

w (x) = W (c) (4)

 When the parameter x is an inclusive link and the inclusive link is a link from the closed node c to the second basic node, the control unit 10 sets the search link cost w (x) to the following formula (5) ).

w (x) = 0 (5)

  In addition, when the parameter x is an inclusion link, when the inclusion link connects a basic node (inclusion node) other than the first basic node and the second basic node and a closed node, the route specifying device 1 calculates the search link cost w (x) based on the following equation (6).

w (x) = ∞ (6)

  Hereinafter, it is assumed that the first basic node and the second basic node are the basic nodes N300 and N500, respectively, and the number of routes to be obtained between the nodes N300 and N500 is k = 3, with reference to FIG. One route specifying operation will be described.

  In FIG. 13, first, the control unit 10 determines whether or not a value obtained by subtracting the already determined route number m from the route number k to be obtained is greater than 0 (step S201). Note that the initial value of the determined route number m is zero. If k−m> 0 (step S201; Yes), the control unit 10 performs the following process; otherwise (step S201; No), the route specifying process ends.

  If k−m> 0 (step S201; Yes), the cycle search space specifying unit 13 specifies a searchable region of the cycle search graph (step S202). Specifically, in the cycle search graph shown in FIG. 12, a set of cycles that can be used to construct a cycle that passes through the first basic node N300 and the second basic node N500 is given. In the initial state, the entire cycle search graph becomes a searchable region, but thereafter, as will be described later, the remainder of the initial searchable region excluding adjacent links between the closed nodes identified so far is the searchable region. It becomes.

  The availability of the above-mentioned closed circuit is determined by the basic node (communication device) and the basic link (communication path / communication line) in addition to the exclusion process of the closed circuit included in the path-containing circuit specified by the path-containing circuit specifying unit 14 described later. It may be determined in consideration of external information such as the operational status (failure, load, etc.) and management / operation policy for them. In this example, for simplicity, it is assumed that there are no basic nodes or basic links to be excluded in the initial state.

  Subsequently, the route-containing cycle specifying unit 14 specifies the minimum cost route in the searchable area in the specified cycle search graph (step S203). The minimum cost route is a route that gives a closed circuit having a minimum cost among the closed circuits passing through the first basic node N300 and the second basic node N500. As a method for specifying the minimum cost route, as already described, a shortest route calculation algorithm that allows a negative link cost such as a breadth-first search method or a modified Dijkstra method (Bandari method; Non-Patent Document 2) can be used. .

  The searchable area in the cycle search graph can be expressed by setting the cost of an adjacent link or an inclusion link connected to an unusable cycle node in the cycle search graph to infinity ∞. This is because the link with the cost ∞ is excluded from the route search candidates in the shortest route calculation algorithm.

  Alternatively, as illustrated in FIG. 14, it can be expressed by excluding an adjacent link or an inclusion link connected to an unusable cycle node in the cycle search graph and a link of cost ∞ from the graph information. For example, when the inclusion link E can pass only in one direction in FIG. 14, the cost in the reverse direction may be set to ∞. Furthermore, in FIG. 14, the nodes N100, N200, N400, N600, and N700 that do not have a link may be excluded from the graph information. Alternatively, the searchable area in the cycle search graph may be expressed by using the above expression method together.

  The control unit 10 determines whether or not the minimum cost route in the cycle search graph has been identified (step S204). When the minimum cost route is not specified (step S204; No), the control unit 10 ends the route specifying process. Here, it is assumed that the route N300-C200-C400-N500 is specified as the minimum cost route in the closed loop search graph. Since the minimum cost route has been specified (step S204; Yes), the control unit 10 controls the route specifying unit 15 to specify the route-containing cycle by merging the cycles through which the minimum cost route passes in the cycle search graph ( Step S205). That is, a basic link (shared link) shared by two closed circuits is removed from each of the two closed circuits C200 and C400, and a closed circuit in which all the remaining basic links are connected is merged with the two closed circuits. Used as (path containing cycle). Since the closed circuit information is generated as an adjacency matrix as shown in FIGS. 7 to 10, a closed circuit obtained by merging two closed circuits (path-containing circuit) calculates an exclusive OR of the closed circuit information (adjacent matrix). Can be obtained.

  Specifically, as shown in FIG. 6, by combining the closed circuit C200 and the closed circuit C400, the route-containing closed circuit N300-N200-N600-N500-N700-N300 can be specified. That is, by calculating the exclusive OR of the cycle information (adjacency matrix) of FIG. 8 showing the cycle C200 and the cycle information (adjacency matrix) of FIG. 9 showing the cycle C400, the path-containing cycle shown in FIG. 15 is represented. Cycle information can be obtained. However, the method of merging the closed circuits is not limited to this.

Subsequently, the route specifying unit 15 specifies the minimum cost route (first route) connecting the first basic node N300 and the second basic node N500 in the specified route-containing cycle (step S206). The route specifying unit 15 specifies the minimum cost route by using the breadth-first search method or the Dijkstra method based on the basic link cost w ₀ (x). Alternatively, the above-mentioned route-containing closed circuit may be divided into two routes by the first basic node N300 and the second basic node N500, and the minimum cost route may be specified by comparing the respective costs. Here, it is assumed that N300-N700-N500 is specified as the minimum cost route (first route).

  Similarly, the route specifying unit 15 determines a redundant route (second route) different from the first route that connects the first basic node N300 and the second basic node N500 in the specified route-containing cycle. Specify (step S207). Here, it is assumed that N300-N200-N600-N500 is specified as the redundant route (second route). As a method for specifying the second route, for example, the second route is obtained by removing a portion corresponding to the first route specified in step S206 from the circular list representing the route-containing cycle specified in step S205. Is identified. Note that the second route can be obtained by using the depth-first search method or the Dijkstra method in a state where a condition for disabling the intermediate basic node or the basic link in the first route specified in step S206 is provided. Can also be specified.

  Next, the control unit 10 controls the route selection unit 16 to update the calculation result with the identified first route and second route (step S208). Specifically, if the two routes specified this time do not overlap any of the routes obtained so far, the two routes obtained this time are added to the calculation result. If there is an overlap in a part of the route so far, the route is deleted from the calculation result, and the two routes specified this time are added to the calculation result.

  In the case of this example, since the two paths between the first basic node N300 and the second basic node N500 are calculated for the first time in the above step S206 and step S207, the calculation result is empty. Therefore, the route selection unit 16 records the first route and the second route obtained this time in the graph information storage unit 11 as they are. After updating the total number m of determined paths in this way (in this case, after updating to m = 2), the process returns to step S201.

  Specifically, since k-2> 0 (step S201; Yes), in step S202, the adjacent link A200 through which the route N200-C200-C400-N500 specified in the previous step S203 passes is excluded ( (See FIG. 14). As described above, this can be realized by setting the cost of the adjacent link A200 to ∞ or deleting it from the cycle search graph.

  Next, the control part 10 specifies the minimum cost path | route in the searchable area | region of a closed circuit search graph similarly to last time in step S203. This time, the route N300-C300-N500 is specified in FIG. Since the minimum cost route has been identified, step S205 is executed subsequent to step S204.

  In step S205, the route-containing cycle specifying unit 14 specifies a new route-containing cycle by combining the routes through which the minimum cost route N300-C300-N500 passes. However, in this example, since only one closed circuit C300 is passed, the actual combined processing of closed circuits is not performed. As a result, a route-containing cycle of N300-N400-N500-N700-N300 is identified.

  Next, the minimum cost route between the first basic node N300 and the second basic node N500 is specified as in the previous time for the route-containing cycle N300-N400-N500-N700-N300 specified in step S205. In this example, N300-N700-N500 is obtained as the third route.

  Subsequently, the redundant route between the first basic node N300 and the second basic node N500 is specified for the route-containing cycle N300-N400-N500-N700-N300 specified in step S205 as in the previous case. As a result, N300-N400-N500 is obtained as the fourth route.

  Then, the route calculation result is updated using the third route and the fourth route specified in step S206 and step S207. (Step S208). Of the third route and the fourth route specified this time, the fourth route is the same as the first route specified in the previous step S206, the basic link L700, and the basic link L1000. Therefore, the first route is discarded from the calculation result, and the third route and the fourth route are added to the calculation result. At this stage, a total of three routes (m = 3) of the second route N300-N200-N600-N500, the third route N300-N700-N500, and the fourth route N300-N400-N500 are calculated. ) Has been identified.

  Thus, when returning to step S201, since m = 3 at this stage, “No” is obtained in step s201, and the route specifying process is terminated. As described above, when the desired value k is given, steps S201 to S208 are repeated until k−m becomes 0 (step S201; No) or until the minimum cost path is not specified (step S204; No). It is possible to specify the number k of routes specified in (1). Note that step S201 for controlling repetition may be provided after step S208.

  In this way, the route specifying device 1 connects the first basic node N300 and the second basic node N500 by using the closed loop search graph, and specifies k different routes that do not share the basic link. Can do.

1.3) Effects As described above, according to the first embodiment of the present invention, it is possible to reliably specify three or more different paths that connect two basic nodes and do not share a basic link. Furthermore, it is possible to reduce the processing load when two basic nodes are connected and two different routes that do not share a basic link or a basic node in the middle of the route are specified.

  Furthermore, according to the first embodiment of the present invention, the route-containing cycle in the basic graph is specified so as to minimize the total sum of basic link costs set in advance for each of the basic links in the basic graph. By specifying route-containing cycles, two basic nodes are connected to minimize the sum of basic link costs, and the calculation load is increased for three or more different routes that do not share basic links or basic nodes along the route. It is possible to specify without making it.

2. Second Embodiment Next, a path identification device according to a second embodiment of the present invention will be described. According to the present embodiment, when the route-containing cycle specifying unit 14 merges two cycles and specifies the route-containing cycle, even when the two cycles share only one basic node, It differs from the first embodiment in that it is to be merged.

  For example, in FIG. 6, the closed circuit C100 and the closed circuit C200 have the node N200 as the only shared node. However, according to the present embodiment, the closed circuit C100 and the closed circuit C200 are combined and treated as one route-containing closed circuit. It can be used for route identification. A state where two closed circuits share a basic link (L500 in FIG. 6) like closed circuits C200 and C400 is also called “strongly adjacent state”, and two closed circuits like closed circuits C100 and C200 A state in which only a basic node (N200 in FIG. 6) is shared without sharing a link is also referred to as a “weakly adjacent state”. Hereinafter, the characteristic points of the second embodiment of the present invention will be mainly described.

2.1) Configuration Since the route specifying device according to the present embodiment has basically the same functional configuration as the route specifying device 1 according to the first embodiment shown in FIG. 2, description of the configuration is omitted, and FIG. The route identification function will be described using the same reference numbers.

2.2) Operation 2.2.1) Generation of a cycle adjacency graph The cycle identification unit 12 of the path identification device 1 identifies a cycle as in the first embodiment described above. Since the closed circuit C100 and the closed circuit C200 share only the basic node N200 without sharing the basic link, they are in a weakly adjacent state. As illustrated in FIG. 16, the control unit 10 adds a weakly adjacent link A500 (indicated by a dotted line), which is an adjacent link representing a weakly adjacent state, to the closed adjacency graph. Note that the only basic node shared by two cycles in a weakly adjacent state is considered a critical node. That is, any weakly adjacent link does not include a basic link but includes one critical node.

2.2.2) Generation of Cycle Search Graph Subsequently, the control unit 10 controls the cycle search space specifying unit 13 to add an inclusion node to the cycle adjacency graph, and an inclusion that connects the cycle node and the inclusion node. A cycle search graph is generated by adding a link to the cycle adjacency graph. FIG. 17 shows a cycle search graph when the basic graph shown in FIG. 3 is used. In FIG. 17, a broken line connecting the closed circuit C and the comprehensive node N represents the inclusion links E101 to E114.

2.2.3) Specifying Route Next, the control unit 10 controls the route containing cycle specifying unit 14, the route specifying unit 15, and the route selecting unit 16 to select a route based on the cycle adjacency graph shown in FIG. Identify. Hereinafter, the characteristic operation of the second embodiment will be mainly described with reference to the flowchart of FIG. 13 as in the first embodiment. In the following description, description of operations similar to those of the first embodiment is simplified or omitted.

In formulas (1) to (6) in the first embodiment described above, an adjacent link parameter x connecting the closed node c ₀ and closed nodes c _1, and closed and closed node representing the closed node c ₀ c _{When the} closed circuit represented by ₁ is in a weakly adjacent state, if the basic node included in the adjacent link (weakly adjacent link) is not the end point of the route to be obtained, the adjacent link cost w (x) is expressed by the following equation (7). Calculated.

w (x) = W (c ₀ ) + W (c ₁ ) (7)

  However, since the weakly adjacent link does not include the basic link, Expression (4) is equivalent to Expression (2). Therefore, the adjacent link cost w (x) shown by the equations (1) to (6) in the first embodiment can be used as it is in the second embodiment.

  Hereinafter, in order to explain the path specifying function according to the second embodiment, it is assumed that the first basic node is the node N300, the second basic node is the node N600, and the number of paths k to be calculated is k = 3.

  As shown in FIG. 17, assuming that the route N300-C200-C400-N600 is specified as the minimum cost route in the cycle search graph in the initial state, steps S201 to S208 in FIG. 13 are described as described in the first embodiment. Is executed to identify the first route N300-N200-N600 and the second route N300-N700-N500-N600 (see FIG. 6), and add them to the calculation result of the graph information storage unit 11.

  Subsequently, as illustrated in FIG. 17, assuming that the route N300-C300-C400-N600 is specified as the minimum cost route in the cycle search graph, the third step is performed by executing steps S201 to S207 in FIG. Route N300-N700-N200-N600 and the fourth route N300-N400-N500-N600 are specified (see FIG. 6). However, in step S208, the first route already specified and the third route specified this time are overlapped by the basic link L400, and the second route and the fourth route are the basic link L900. Duplicate. For this reason, the route selection unit 16 discards the first route and the second route from the calculation result, and adds the third route and the fourth route to the calculation result.

  Subsequently, as illustrated in FIG. 17, it is assumed that the route N300-C200-C100-N600 is specified as the minimum cost route in the closed route search graph. In this case, in step S205, as in the previous time, the route containing the minimum cost route is merged to specify the route-containing cycle, and the route-containing cycle N300-N200-N100-N600-N200-N700-N300 is specified.

  In the subsequent step S206, the route specifying unit 15 determines the minimum value between the first basic node N300 and the second basic node N600 from the route-containing cycles N300-N200-N100-N600-N200-N700-N300 as in the previous time. Identify the cost path. Here, it is assumed that the fifth route N300-N700-N200-N600 is specified. Similarly, in step S207, the route specifying unit 15 specifies the sixth route N300-N200-N100-N600 from the route-containing closed circuit N300-N200-N100-N600-N200-N700-N300.

  However, the fifth route is equivalent to the third route N300-N700-N200-N600 recorded in the graph information storage unit 11. Therefore, in step S208, the third route is deleted, and the fifth route and the sixth route are added to the calculation result of the graph information storage unit 11.

  At this point, three routes, that is, a fourth route N300-N400-N500-N600, a fifth route N300-N700-N200-N600, and a sixth route N300-N200-N100-N600 are identified as calculation results. Therefore, the route specifying operation ends in step S201.

2.3) Effects As described above, the route specifying device according to the first embodiment is obtained by merging the cycles that share only one basic node like the cycles C100 and C200 in the same manner as in the first embodiment. The same operation and effect as 1 can be obtained.

3. Third Embodiment In the route specifying device according to the third embodiment of the present invention, the closed circuit information acquired during the process of specifying the route is held, and the held closed circuit information is used in the process of specifying another route. To do. As a result, the route specifying process can be speeded up. Hereinafter, this difference will be mainly described.

  As shown in FIG. 18, the graph information storage unit 11 in the third embodiment stores merged cycle information in addition to the information shown in FIG. The merged cycle information is information in which, when two certain cycles are merged, information indicating the cycle generated by the merge is associated with the two pieces of cycle information.

  As shown in FIG. 18, the merged cycle information is a table including a cycle identifier for identifying each of the two cycles to be merged and a cycle identifier for identifying the cycle generated by the merge. is there. That is, the identifier that is a value in this table is an identifier of a cycle generated by merging the cycles identified by the cycle identifier respectively associated with the row and column of the value. The merged cycle information may be represented not in such a table format but in another format that can be processed by the information processing apparatus.

  The path identifying device 1 according to the present embodiment executes processes other than step S205 of FIG. 13 as in the first embodiment.

  When the control unit 10 proceeds to step S205, the control unit 10 first determines whether or not merged cycle information related to two cycles to be merged is stored in the graph information storage unit 11. When merged cycle information relating to two cycles to be merged is stored, the control unit 10 acquires information representing a cycle generated by merging from the merged cycle information.

  On the other hand, when the merged cycle information related to the two cycles to be merged is not stored, the control unit 10 generates a route-containing cycle by merging the two cycles as in the first embodiment. . Furthermore, in this case, the closing identifiers of the two closings to be merged are associated with the closing identifiers of the closings generated by the merging and stored in the above table format.

  For example, it is assumed that the closed circuit C500 is generated by merging the closed circuits C200 and C400. In this case, the merged cycle information is in the state of T1 before the two cycles are merged, but when the cycle C500 is generated by merging the cycles C200 and C400, the merged cycle information is in the state of T2. . By using the merged cycle information held in this way, the route specifying process can be speeded up.

4). Fourth Embodiment In the route specifying device according to the fourth embodiment of the present invention, a route is specified by using a closed loop search graph only when a route cannot be specified based on a basic graph. The other points are the same as those in the first to third embodiments described above, and thus the differences will be mainly described.

  In FIG. 19, the route specifying device 1 specifies a route based on the basic graph (that is, without using the closed loop search graph) (step S401). Here, the Dijkstra method is repeatedly used to specify two or more different paths that connect the basic nodes and do not share the basic nodes. That is, the first route is specified, and then the second route is specified based on the graph obtained by removing the basic node through which the first route passes from the basic graph. Similarly, the (i-1) th route is specified, and then the ith route is specified based on the graph obtained by removing the basic nodes through which the first to (i-1) th routes pass from the basic graph. However, 1 <i ≦ k. Note that the route may be specified by other methods.

  Here, the operation for removing the basic node from the basic graph is to remove the basic node from the data structure of the basic graph, and to perform a masking process for hiding the basic node to be removed from the data structure of the basic graph. Alternatively, a process that does not select the basic node by the Dijkstra method or another algorithm used for route calculation, such as setting the cost of the basic link connected to the basic node to ∞, may be used.

  In addition, as in the second embodiment or the third embodiment based on it, when a weakly adjacent link in a closed adjacency graph is allowed, instead of removing the basic node through which the identified route passes from the basic graph, Remove the basic links that the specified route passes from the basic graph.

  The first route specified by the process of step S401 is a route with the lowest cost in the basic graph. On the other hand, the first route and the second route identified in this way do not necessarily have the smallest sum of the cost of the first route and the cost of the second route.

  Next, in step S402, the control unit 10 determines whether a route is specified. If no route is specified in step S401, a closed adjacency graph is generated in step S403 as in the first embodiment, and two different routes are specified based on the generated closed adjacency graph. . Even when the route is specified, the route may be determined as “No” when the specified route does not satisfy a preset condition. For example, this condition is a condition that the cost of the second route is larger than a preset reference value. When the route is specified in step S401 (step S402; Yes), the execution of the process is terminated.

  As described above, according to the fourth embodiment of the present invention, it is possible to obtain the same effect as in the first embodiment, and the sum of the costs of the first to kth (k ≧ 2) routes is minimized. It is also possible to specify other routes.

5. Fifth Embodiment With reference to FIG. 20, a route identifying device according to a fifth embodiment of the present invention will be described. A route identification device 1000 according to the fifth embodiment includes a cycle identification unit 1001, a route-containing cycle identification unit 1002, a route identification unit 1003, a route selection unit 1004, and a cycle search space identification unit 1005.

  The cycle specifying unit 1001 specifies a plurality of cycles including a first cycle that passes through the first basic node and a second cycle that passes through the second basic node. The route-containing cycle specifying unit 1002 combines the first basic node and the second by combining the cycles sharing at least one of the basic graph elements that are the basic nodes or basic links of the plurality of specified cycles. A route-containing cycle that is a cycle passing through the basic node is specified. The route specifying unit 1003 connects the first basic node and the second basic node in the specified route-containing cycle, and specifies two different routes that do not share the basic node or the basic link. The route selection unit 1004 updates the calculation result holding the specified route based on the two different routes specified above. The cycle search space identification unit 1005 records the merge between cycles performed by the route-containing cycle identification unit 1002 as a process that cannot be applied from the next time.

  Further, the series of processing of the route-containing cycle specifying unit 1002 to the cycle search space specifying unit 1005 specifies the number of routes that connect the first basic node and the second basic node and do not share the basic node or basic link. It is configured to repeat until it is possible or until it is determined that it is impossible to specify that number. With this configuration, two or more different paths that connect two basic nodes and do not share a basic link can be reliably specified, and the processing load can be reduced.

6). Others In the first embodiment to the fourth embodiment described above, the route specifying device 1 continues the process of generating a closed loop search graph (first process) and the process of specifying a route (second process). Was configured to run. By the way, the route identification device 1 may be configured to hold the execution result of the first process and then execute the second process a plurality of times by using the held execution result. .

  Further, the program for generating the cycle adjacency graph or the cycle search graph and the program for specifying the route may constitute one program, or may be two programs independent of each other. Good. In addition, the path specifying device 1 physically allocates calculation resources necessary for executing a program such as a central processing unit, a primary storage device, and a secondary storage device for each of two independent programs. Or you may be comprised so that it may allocate logically separately.

  In each of the above embodiments, each function of the path identification device is realized by the CPU executing a program (software), but may be realized by hardware such as a circuit. In each of the above embodiments, the program is stored in the storage device, but may be stored in a computer-readable recording medium. For example, the recording medium is a portable medium such as a flexible disk, an optical disk, a magneto-optical disk, and a semiconductor memory.

  Although the present invention has been described with reference to the above embodiment, the present invention is not limited to the above-described embodiment. Various changes that can be understood by those skilled in the art can be made to the configuration and details of the present invention within the scope of the present invention. In addition, as another modified example of the above-described embodiment, any combination of the above-described embodiments and modified examples may be employed.

7). Additional Notes Part or all of the above-described embodiments may be described as the following additional notes, but are not limited thereto.
(Appendix 1)
A method for specifying a route in a network in which a plurality of nodes are linked,
Based on a route closed by a plurality of nodes and links in the network (hereinafter referred to as a closed route), k pieces that do not cross each other through any first node and second node (k is an integer of 2 or more) Identify the path-containing cycles of
Identifying different (k + 1) paths between the first node and the second node from the k path-containing cycles;
A route specifying method characterized by the above.
(Appendix 2)
The route specifying method according to claim 1, wherein the k route-containing cycles are specified based on a route cost between the first node and the second node.
(Appendix 3)
The route specifying method according to claim 1 or 2, wherein two routes are specified from each of the k route-containing cycles, and the (k + 1) routes are specified by deleting duplicate routes. .
(Appendix 4)
Each of the k path-containing cycles is specified by merging a plurality of unit cycles including a unit cycle passing through the first node and a unit cycle passing through the second node. The route specifying method according to any one of -3.
(Appendix 5)
A routing device in a network in which a plurality of nodes are linked,
Based on a route closed by a plurality of nodes and links in the network (hereinafter referred to as a closed route), k pieces that do not cross each other through any first node and second node (k is an integer of 2 or more) A cycle specifying means for specifying a path-containing cycle of
Route specifying means for specifying different (k + 1) routes between the first node and the second node from the k route-containing cycles;
A path specifying device comprising:
(Appendix 6)
The route specifying device according to appendix 5, wherein the cycle specifying means specifies the k route-containing cycles based on a route cost between the first node and the second node.
(Appendix 7)
The route specifying means specifies two routes from the k route-containing cycles, and specifies the (k + 1) routes by deleting duplicate routes. The route identification device described in 1.
(Appendix 8)
The closed circuit specifying means includes:
Appendix 5- wherein each of the k path-containing cycles is sequentially identified by merging a plurality of unit cycles including a unit cycle passing through the first node and a unit cycle passing through the second node. The route specifying device according to any one of 7.
(Appendix 9)
A program for causing a computer to function as a path identification device in a network in which a plurality of nodes are linked,
Based on a route closed by a plurality of nodes and links in the network (hereinafter referred to as a closed route), k pieces that do not cross each other through any first node and second node (k is an integer of 2 or more) The ability to identify the path-containing cycles of
A function of identifying different (k + 1) paths between the first node and the second node from the k path-containing cycles;
Is realized by the computer.
(Appendix 10)
A route specifying method for specifying a route connecting a first basic node and a second basic node in a basic graph composed of a basic node and a basic link connecting two basic nodes,
a) identifying a plurality of cycles including a first cycle through the first basic node and a second cycle through the second basic node;
b) The first basic node and the second basic are merged by merging the cycles sharing at least one of the basic graph elements that are basic nodes or basic links among the plurality of specified cycles. Identify a first path containing cycle that is a cycle through the node;
c) After specifying the (k-1) th path-containing circuit that is a circuit passing through the first basic node and the second basic node (where k is an integer of 2 or more), the specified plural A basic graph that is a basic node or a basic link in a set of closed loops obtained by removing a closed loop merged in a specific process of the (k-1) th path-containing loop from the first path-containing loop. Merging cycles that share at least one of the elements to identify a kth path-containing cycle that is a cycle through the first basic node and the second basic node;
d) From the k path-containing cycles identified by repeating the above c) (k−1) times while increasing k by 1 from k = 2, the first basic node and the second basic node And (k + 1) different routes that do not share a basic link or a basic node other than the first basic node and the second basic node.
A route specifying method characterized by the above.
(Appendix 11)
The route specifying method according to attachment 10, wherein
Indicating a set of cycles that can be used to identify a path connecting the first basic node and the second basic node; or a path connecting the first basic node and the second basic node A route specifying method comprising: selecting a closed cycle that can be used for specifying a route connecting the first basic node and the second basic node by indicating a set of closed cycles that cannot be specifically used.
(Appendix 12)
The route specifying method according to appendix 10 or 11,
Any one of the (2 * i-1) and (2 * i) paths connecting the first basic node and the second basic node obtained from the i-th (i> 1) path-containing cycle , Any one of the (2 * j-1) and (2 * j) paths connecting the first basic node and the second basic node obtained from the j (j> i) path-containing cycle Are sharing a basic node or basic link excluding the first basic node and the second basic node, the corresponding (2 * i-1) path or the (2 * i) The route specifying method is characterized by discarding the route or both of them and specifying the (k + 1) different route.
(Appendix 13)
The route specifying method according to any one of appendices 10-12,
When two cycles share at least one basic link, all the basic links that constitute a portion excluding the shared link that is the shared basic link from one of the two cycles; A route specifying method characterized in that a closed circuit obtained by connecting all basic links constituting a portion excluding the shared link from the other closed circuit is used as a closed circuit obtained by merging the two closed circuits.
(Appendix 14)
A route identification method according to attachment 13, comprising:
When two cycles share only a basic node without sharing a basic link, all the basic links constituting one of the two cycles and all the basic links constituting the other of the two cycles A path specifying method characterized by using a closed circuit connecting the two closed circuits as a closed circuit.
(Appendix 15)
The route specifying method according to appendix 13 or 14,
Each of the plurality of identified cycles is represented by a cycle node, and the basic graph element in the basic graph is shared by two cycles, by an adjacent link connecting the cycle nodes representing each of the two cycles Generate a cycle adjacency graph that represents
The generated cycle adjacency graph represents an inclusion node representing each of the basic nodes through which the cycle represented by the cycle node passes in the basic graph, and the cycle represented by the cycle node passes through the inclusion node in the basic graph and Generate a cycle search graph with an inclusion link connecting the cycle node and the inclusion node,
Identifying the path-containing cycle in the basic graph based on the generated cycle search graph;
A route specifying method characterized by the above.
(Appendix 16)
The route specifying method according to any one of appendices 10-15,
The route-containing cycle specifying means specifies the route-containing cycle in the basic graph so as to minimize the total sum of basic link costs set in advance for each of the basic links in the basic graph. The route identification method.
(Appendix 17)
A route specifying device for specifying a route connecting a first basic node and a second basic node in a basic graph composed of a basic node and a basic link connecting two basic nodes,
A cycle specifying means for specifying a plurality of cycles including a first cycle that passes through the first basic node and a second cycle that passes through the second basic node;
A cycle search space identifying means for selecting a set of cycles that can be used to identify a path connecting the first basic node and the second basic node among the plurality of identified cycles;
Of the cycles included in the identified cycle search space, the first basic node and the second are shared by merging the cycles sharing at least one of the basic graph elements that are basic nodes or basic links. A route-containing cycle specifying means for specifying a route-containing cycle that is a cycle passing through the basic node of
In the specified route-containing cycle, the first basic node and the second basic node are connected, and a basic link, or a basic node other than the first basic node and the second basic node Route identification means for identifying two different routes that do not share
From 2 * k routes obtained by the route specifying means from k (k ≧ 1) route-containing cycles passing through the first basic node and the second basic node obtained by the route-containing cycle specifying means, a route selection means for selecting (k + 1) or more different routes;
A path specifying device comprising:
(Appendix 18)
A program for realizing in a computer a route specifying function for specifying a route connecting a first basic node and a second basic node in a basic graph composed of a basic node and a basic link connecting two basic nodes Because
a) identifying a plurality of cycles including a first cycle through the first basic node and a second cycle through the second basic node;
b) The first basic node and the second basic are merged by merging the cycles sharing at least one of the basic graph elements that are basic nodes or basic links among the plurality of specified cycles. Identify a first path containing cycle that is a cycle through the node;
c) After specifying the (k-1) th path-containing circuit that is a circuit passing through the first basic node and the second basic node (where k is an integer of 2 or more), the specified plural A basic graph that is a basic node or a basic link in a set of closed loops obtained by removing a closed loop merged in a specific process of the (k-1) th path-containing loop from the first path-containing loop. Merging cycles that share at least one of the elements to identify a kth path-containing cycle that is a cycle through the first basic node and the second basic node;
d) From the k path-containing cycles identified by repeating the above c) (k−1) times while increasing k by 1 from k = 2, the first basic node and the second basic node And (k + 1) different routes that do not share a basic link or a basic node other than the first basic node and the second basic node.
A path specifying program for causing the computer to function as described above.

  The present invention can be applied to a server device having a function of specifying a route connecting a first node and a second node.

1 Route Identification Device 11 Graph Information Storage Unit
12 Closing part
13 cycle search space specifying unit 14 route-containing cycle specifying unit
15 Route identification part
16 Route Selection Unit 1000 Route Identification Device
1001 Cycle identification unit 1002 Path-containing cycle identification unit
1003 Route specifying unit 1004 Route selecting unit 1005 Cycle search space specifying unit

Claims (9)

A method for specifying a route in a network in which a plurality of nodes are linked,
Based on a route closed by a plurality of nodes and links in the network (hereinafter referred to as a closed route), k pieces that do not cross each other through any first node and second node (k is an integer of 2 or more) Identify the path-containing cycles of
Identifying different (k + 1) paths between the first node and the second node from the k path-containing cycles;
A route specifying method characterized by the above.   The route specifying method according to claim 1, wherein the k route-containing cycles are specified based on a route cost between the first node and the second node.   3. The route specification according to claim 1, wherein two routes are specified from each of the k route-containing cycles, and the (k + 1) routes are specified by deleting duplicate routes. Method.   Each of the k route-containing cycles is specified by merging a plurality of unit cycles including a unit cycle passing through the first node and a unit cycle passing through the second node. The route specifying method according to any one of 1-3. A routing device in a network in which a plurality of nodes are linked,
Based on a route closed by a plurality of nodes and links in the network (hereinafter referred to as a closed route), k pieces that do not cross each other through any first node and second node (k is an integer of 2 or more) A cycle specifying means for specifying a path-containing cycle of
Route specifying means for specifying different (k + 1) routes between the first node and the second node from the k route-containing cycles;
A path specifying device comprising:   6. The route specifying apparatus according to claim 5, wherein the closed route specifying unit specifies the k route-containing closed routes based on a route cost between the first node and the second node.   The route specifying means specifies two routes from the k route-containing cycles, and specifies the (k + 1) routes by deleting duplicate routes. 6. The route specifying device according to 6. The closed circuit specifying means includes:
6. The k path-containing cycles are sequentially identified by merging a plurality of unit cycles including a unit cycle passing through the first node and a unit cycle passing through the second node. The route identifying device according to any one of -7. A program for causing a computer to function as a path identification device in a network in which a plurality of nodes are linked,
Based on a route closed by a plurality of nodes and links in the network (hereinafter referred to as a closed route), k pieces that do not cross each other through any first node and second node (k is an integer of 2 or more) The ability to identify the path-containing cycles of
A function of identifying different (k + 1) paths between the first node and the second node from the k path-containing cycles;
Is realized by the computer.