JP5506538B2 - Route calculation method, apparatus and program - Google Patents

Description

  The present invention relates to a route calculation method, apparatus, and program, and more particularly, to a route calculation method, device, and program for searching for a route in a network.

  As a method for calculating a route, there are a Dijkstra method, a Bellman-Ford method, a method disclosed in Patent Document 1, and the like.

  The Dijkstra method, the Bellman-Ford method, and the method disclosed in Patent Document 1 say that the route candidate obtained by the route search from the start point to the end point matches the route candidate obtained by the route search from the end point to the start point. The route is searched on the premise.

  In addition, by using an undirected graph in an adjacency matrix or a connection matrix, data required for matrix representation can be reduced as compared with a directed graph, and a memory size required for calculation is reduced.

  For example, the difference between a directed graph and an undirected graph is as follows. When a network topology composed of 1 to N nodes is represented by an adjacency matrix, the adjacency matrix is an N × N matrix. At this time, for example, when the route search is advanced from the node “i” to the adjacent node “j”, the values of the row “i” and the column “j” of the adjacency matrix on the database are referred to. At this time, for example, if the value is zero, the connection is not established, and if the value is “1”, the connection is established. In the case of an undirected graph, the matrix components of row “i” and column “j” and the matrix components of row “j” and column “i” are the same. Therefore, even if the size of the adjacency matrix is N × N, the size of the adjacency matrix that should actually be stored in the database is equivalent to N × N ÷ 2, that is, half the size of the original adjacency matrix You only have to remember the minutes. On the other hand, in the case of a directed graph, the matrix components of row “i” and column “j” are different from the matrix components of row “j” and column “i”. Therefore, when the adjacency matrix is stored on the memory of the computer that executes the path calculation, the memory size necessary for storing the directed graph is required only for storing the N × N adjacency matrix itself. That is, assuming a certain network topology, the memory size of a database for storing a directed graph according to the topology is twice the memory size of a database for storing a similar undirected graph. Therefore, on a computer, an undirected graph can represent a larger network than a directed graph, and a route calculation method using an undirected graph is suitable for route search of a large-scale network.

  With network transmission technology, Ethernet (registered trademark), SONET / SDH (Synchronous Optical Network / Synchronous Digital Hierarchy), OTN (Optical Transport Network), etc. have the same path in both directions, so the network can be used with undirected graphs. Can be expressed. Therefore, undirected graphs are generally used for network route search.

JP 2008-301225A.

  The route calculation method based on the above premise cannot accurately search for a route from the start point to the end point when a communication device such as Reconfigurable Optical Add-Drop Multiplexer (ROADM) as shown in FIG. 7 exists in the network. . This is because, in ROADM, as shown in FIGS. 18 and 19, the route candidate obtained by the route search from the start point to the end point does not match the route candidate obtained by the route search from the end point to the start point. The concept of the ROADM route will be described with reference to FIGS. In FIG. 18, a signal input from the transponder in the interface unit to the ROADM switch unit is output in the right direction. However, it is not output in the left direction. On the other hand, in FIG. 19, the signal input from the ROADM switch unit to the same transponder as in FIG. 18 is input from the left to the ROADM switch unit and then input to the transponder. However, there is no input from the right to the ROADM switch section. That is, in the ROADM communication device, the route candidate obtained by the route search from the start point to the end point is different from the route candidate obtained by the route search from the end point to the start point. That is, in the network having ROADM as shown in FIG. 20, the feasible paths are different as shown in FIG. 21 depending on the setting of the start point and the end point.

  On the other hand, the adjacency matrix simply represents only a physical connection, and the above-described direction constraint cannot be represented only by the adjacency matrix.

  In order for the Dijkstra method, the Bellman-Ford method, and the method disclosed in Patent Document 1 to accurately obtain the route from the start point to the end point even in a network having ROADM, the direction restriction of the route search is managed to control the direction. It is necessary to search for a route that fits.

  The present invention has been made in view of the above points, and provides a route calculation method, apparatus, and program capable of searching a route in a network with directionality restrictions while suppressing the amount of memory for calculating an adjacency matrix. Objective.

In order to solve the above-mentioned problem, the present invention (Claim 1) searches for an adjacent point from a point “A” on the path from the start point to the end point, and finds a point “B”, and then When searching for a point adjacent to the point “B”, in the network where the adjacent point of the point “B” is limited by the point “A”, zero or more points from a certain point (start point) to a certain point (end point) A route calculation method for searching for a route ,
A route database storing the connection relationship between the base branch and the evaluation branch;
An adjacency matrix database representing the network by an adjacency matrix or an equivalent undirected graph matrix, and storing the matrix;
A direction constraint database that records pairs of input points and output points that are one or more output destination candidates for each point of identification information as a condition of route direction constraints,
A route calculation means;
In a device having
The route calculation means
A base branch search step for obtaining a point “A” (base branch) connected to the searched branch on the path from the start point to the end point with reference to the path database;
An adjacent point search step of searching for a point adjacent to the point “A” (base point branch) by referring to the adjacency matrix database and finding a point “B” (evaluation branch);
With reference to the direction constraint database, when the input point to the point "A" was searched-branches, determines whether include point "B" to the output point from the point "A", contained If you performed and direction constraints checking step of storing the point "a" (base branch) and the point "B" the connection relationship (evaluation branches) in the path database, the.

The present invention (Claim 2) searches for a point adjacent to a point “A” on the route from the start point to the end point, finds a point “B”, and then adjoins the point “B”. When searching for a point, a route calculation device that searches for zero or more routes from a certain point (start point) to a certain point (end point) in a network in which adjacent points of the point “B” are limited by the point “A” there is,
A route database storing the connection relationship between the base branch and the evaluation branch;
An adjacency matrix database representing the network by an adjacency matrix or an equivalent undirected graph matrix, and storing the matrix;
A direction constraint database that records pairs of input points and output points that are one or more output destination candidates for each point of identification information as a condition of route direction constraints,
A route calculation means;
Have
The route calculation means
Base point branch searching means for obtaining a point “A” (base point branch) connected to the searched branch on the route from the start point to the end point with reference to the route database;
An adjacent point search means for searching for a point adjacent to the point “A” (base point branch) by referring to the adjacency matrix database and finding a point “B” (evaluation branch);
With reference to the direction constraint database, when the input point to the point "A" was searched-branches, determines whether include point "B" to the output point from the point "A", contained If the direction constraint checking means for storing the point "a" (base branch) and the point "B" the connection relationship (evaluation branches) in the path database,
Have

  The present invention (Claim 3) is a program for causing a computer to function as each means constituting the route calculation apparatus according to Claim 2.

  Thereby, at a certain point “B”, the direction input to the point “B” and the output direction from the point “B” restricted based on the input direction can be managed.

  As described above, the present invention is based on an undirected graph by searching based on an adjacency matrix corresponding to an undirected graph and adding a direction constraint check in a route search in a network with a partially directional constraint. Therefore, it is possible to search for a route in a directional network while suppressing the memory capacity for calculation of the adjacency matrix.

It is a block diagram of the route calculation apparatus in the 1st Embodiment of this invention. It is an example of a network topology and its adjacency matrix. It is an example of the path | route information memorize | stored in the image example of the path | route database in the 1st Embodiment of this invention. It is a data structural example of the direction restriction database in the 1st Embodiment of this invention. It is a data example of the direction restriction database in the 1st Embodiment of this invention. It is a flowchart of the route calculation process of the route calculation apparatus in the 1st Embodiment of this invention. It is a block diagram of the route search apparatus (NRM) described in FIG. It is an extended example of the route search apparatus (NRM) in the 2nd Embodiment of this invention. 6 is an example of a network configuration list described in FIG. It is an example of an expansion of the network configuration list in the second embodiment of the present invention. FIG. 9 is an example of an available resource list described in FIG. 8 of Patent Document 1. FIG. It is an example of an expansion of the available resource list | wrist in the 2nd Embodiment of this invention. It is an example of the adjacency matrix described in FIG. It is an extended example of the adjacency matrix in the 2nd Embodiment of this invention. It is the route search method described in FIG. It is an extended example of the route search method in the 2nd Embodiment of this invention. This is an example of an existing device with direction restrictions and its mounted switch. It is an explanation (add (at the time of signal insertion)) of the direction restriction of the existing switch. This is an explanation of the direction restriction of an existing switch (drop (during signal branching)). It is an example of topology. It is an example of the path | route which can be implement | achieved on topology.

  Embodiments of the present invention will be described below with reference to the drawings.

[First Embodiment]
The present invention adds a function for managing the directionality of route search to the Dijkstra method, the Bellman-Ford method, and the method disclosed in Patent Document 1, and searches for a route that meets the direction constraints.

  FIG. 1 shows the configuration of a route calculation apparatus according to the first embodiment of the present invention.

  The route calculation device 10 includes a route calculation unit 11, an adjacency matrix database 12, a route database 13, and a direction constraint database 14.

  The adjacency matrix database 12 stores the adjacency matrix shown in FIG.

  The route database 13 stores the connection relationship between the base branch and the evaluation branch. FIG. 3 shows an example of the route database 13 and an example of routes stored on the route database 13. In this example, the start point is “Point 1”. The route is described for each row, and branches obtained from the starting point are listed. The branch at the right end of each row is basically an unsearched branch and is a candidate for a future base branch. However, it may be a searched branch.

  As shown in FIG. 4, the direction constraint database 14 includes a pair of input points and output points for each point identification information, and one or more output destination candidates are stored in the output point item. Specifically, data as shown in FIG. 5 is stored. The way of viewing the direction constraint database 14 will be described by taking the point of the identification information “4” as an example. When “point 5” is input to “point 4”, “point 4” can be connected to “point 3” or point 6. In addition, when “point 6” is input to “point 4”, “point 4” can be connected to “point 3”.

  The route calculation unit 11 refers to the adjacency matrix database 12, expresses the route database 13 network by an adjacency matrix, a connection matrix, or a matrix equivalent thereto, and a point (end point) from a point (start point) based on the matrix. And a route calculation algorithm for searching for zero or more routes up to.

  FIG. 6 is a flowchart of route calculation processing of the route calculation apparatus according to the first embodiment of the present invention.

  Step 101) First, the route calculation unit 11 refers to the route database 13 and selects one of the branches that have been connected so far (searched branches) and have not been searched yet. This is defined as “base branch”. At the beginning, the branch searched so far is only the start point, and the search is started with the start point as the base branch. At this time, the root branch is managed in the path database 13 as a searched branch.

  However, if there is no unsearched branch as a base branch, the search is unsuccessful. That is, there is no path connecting from the start point to the end point.

  Step 102) Next, if there is a base branch in Step 101, the base branch is stored in a memory (not shown), and the path calculation unit 11 refers to the adjacency matrix database 12 and has not searched for the base branch. Search for branches. If there is no unsearched branch viewed from the base branch, the process returns to step 101 to search for the base branch.

  Step 103) On the other hand, when an unsearched branch is found (determination of an evaluation branch), the branch is stored as an evaluation branch in a memory (not shown). The route calculation unit 11 refers to the direction constraint database 14 in FIG. 4 as the direction constraint check process, and evaluates the relationship between the base branch and the evaluation branch in the memory (not shown). If the base branch and the evaluation branch cannot be connected from the viewpoint of the direction constraint, the path calculation unit 11 proceeds to step 102 and refers to the adjacency matrix database 12 again to search for an unsearched branch viewed from the base branch.

  Step 104) If there is no problem of direction restriction from the base branch to the evaluation branch in the above Step 103, the path calculation unit 11 evaluates whether or not the evaluation branch is the end point next. If the evaluation branch is not the end point, the process proceeds to step 105. If the evaluation branch is the end point, the process proceeds to step 106.

  Step 105) The route calculation unit 11 stores the connection relation between the root branch and the evaluation branch in the route database 13.

  Step 106) When the evaluation branch is the end point in Step 104, the route calculation unit 11 stores the connection relation between the base branch and the evaluation branch in the route database 13 and obtains the route from the start point to the end point at that time.

  As described above, a route from the start point to the end point can be obtained in a network with direction restrictions.

  For example, when searching for a route with the start point “1” and the end point “6”, there are no routes other than the route of 1 → 5 → 4 → 6 from the viewpoint of the direction constraint and the adjacency matrix.

  On the other hand, when a route is calculated based only on the adjacency matrix without taking the direction constraint into consideration, there are the following four route candidates from the example of the topology of the adjacency matrix DB12 in FIG.

・ Example of route candidates 1 → 5 → 4 → 6
1 → 2 → 5 → 4 → 6
1 → 2 → 3 → 4 → 6
1 → 5 → 2 → 3 → 4 → 6
As described above, when the direction constraint is not taken into consideration, there is a possibility that a route that should not exist originally is calculated.

[Second Embodiment]
The present embodiment is a route calculation method with a direction constraint check based on Patent Document 1 and the first embodiment.

  FIG. 7 shows the configuration of the route search apparatus (NRM) described in FIG. 3 of Patent Document 1, while FIG. 8 shows an example of an extension of the route search apparatus (NRM) according to the second embodiment of the present invention. Indicates. The NRM shown in FIG. 8 is obtained by adding a direction constraint check unit 210 having a function corresponding to step 103 of FIG. 6 to the configuration of FIG.

  FIG. 9 shows an example of the network configuration list of FIG. On the other hand, FIG. 10 is an example of expansion of the network configuration list in the second embodiment of the present invention. The network configuration list in FIG. 9 includes the portion of the direction constraint database 13 in FIG. 4 of the first embodiment. It is added.

  FIG. 11 shows an example of the available resource list of FIG. On the other hand, FIG. 12 is an example of the available resource list in the second embodiment of the present invention, in which a portion of the direction constraint database 13 based on FIG. 10 is added to the available resource list of FIG.

  FIG. 13 shows an example of the adjacency matrix of FIG. On the other hand, FIG. 14 is an extension example of the adjacency matrix in the second embodiment of the present invention, in which the part of the direction constraint database 13 based on FIG. 12 is added to the adjacency matrix of FIG.

  FIG. 15 shows an example of the route search method of FIG. On the other hand, FIG. 16 is an extended example of the route search method according to the second embodiment of the present invention, and the direction constraint check step (step 300) of step 103 of FIG. 6 is added to the flow of the route search method of FIG. It is a thing.

  By performing the above addition, a direction constraint check step can be added to the route search method of Patent Document 1.

  The operations of the components shown in FIG. 1 and FIG. 8 in the first embodiment are constructed as a program, installed in a computer used as a route calculation device and a route search device, and executed, or via a network Can be distributed.

  In addition, the constructed program can be stored in a portable storage medium such as a hard disk, a flexible disk, or a CD-ROM, and can be installed or distributed in a computer.

  The present invention is not limited to the above-described embodiment, and various modifications and applications can be made within the scope of the claims.

DESCRIPTION OF SYMBOLS 10 Route calculation apparatus 11 Path calculation part 12 Adjacency matrix database 13 Path | route database 14 Direction restrictions database 100 Path | route search apparatus (NRM)
110 Core unit 120 Route search unit 125 Communication control unit 130 Connection matrix creation unit 140 Route check unit 150 Cost check unit 160 Limitation check unit 170 Data management DB
175 communication path control DB
180 Calculation path DB
190 Confirmed route DB
210 Direction constraint check unit 220 Direction constraint DB

Claims (3)

When searching for a point “B” by searching for a point adjacent to the point “A” on the path from the start point to the end point, and subsequently searching for a point adjacent to the point “B”, the point “B” A route calculation method for searching for zero or more routes from a certain point (starting point) to a certain point (ending point) in a network in which adjacent points of A are limited by the point “A” ,
A route database storing the connection relationship between the base branch and the evaluation branch;
An adjacency matrix database representing the network by an adjacency matrix or an equivalent undirected graph matrix, and storing the matrix;
A direction constraint database that records pairs of input points and output points that are one or more output destination candidates for each point of identification information as a condition of route direction constraints,
A route calculation means;
In a device having
The route calculation means
A base branch search step for obtaining a point “A” (base branch) connected to a searched branch on the path from the start point to the end point with reference to the path database;
An adjacent point search step of searching for a point adjacent to the point “A” (base point branch) by referring to the adjacency matrix database and finding a point “B” (evaluation branch);
Whether or not the point “B” is included in the output point from the point “A” when the input point to the point “A” is the searched branch with reference to the direction constraint database. If it is determined and included, the direction constraint check step of storing the connection relationship between the point “A” (base point branch) and the point “B” (evaluation branch) in the route database;
A route calculation method characterized by: When searching for a point “B” by searching for a point adjacent to the point “A” on the path from the start point to the end point, and subsequently searching for a point adjacent to the point “B”, the point “B” In a network in which adjacent points of A are limited by a point “A”, a route calculation device that searches for zero or more routes from a point (start point) to a point (end point) ,
A route database storing the connection relationship between the base branch and the evaluation branch;
An adjacency matrix database representing the network by an adjacency matrix or an equivalent undirected graph matrix, and storing the matrix;
A direction constraint database that records pairs of input points and output points that are one or more output destination candidates for each point of identification information as a condition of route direction constraints,
A route calculation means;
Have
The route calculation means includes
Base point branch searching means for obtaining a point “A” (base point branch) connected to the searched branch on the route from the start point to the end point with reference to the route database;
An adjacent point searching means for searching for a point adjacent to the point “A” (base branch) and finding a point “B” (evaluation branch) with reference to the adjacency matrix database;
Whether or not the point “B” is included in the output point from the point “A” when the input point to the point “A” is the searched branch with reference to the direction constraint database. If it is determined and included, the direction constraint check means for storing the connection relation between the point “A” (base branch) and the point “B” (evaluation branch) in the route database;
A route calculation apparatus comprising: The program for functioning a computer as each means which comprises the path | route calculation apparatus of Claim 2.

Images