WO2018181031A1 - Optical path design device and optical path design method - Google Patents

Abstract

The present invention addresses the problem of a sharp increase in the time required for optical path optimization due to design parameter increases and dynamic changes in a wavelength-multiplexed optical communication network. An optical path design device of the present invention comprises: a design result acquisition means which acquires an optical path design result in which a parameter factor that is a factor each of a plurality of design parameters for an optical path could take and a characteristic value indicating a characteristic of an optical path designed using the parameter factor are associated with each other; a correlation calculation means which, with respect to one design parameter including a modified-parameter factor obtained by adding an additional factor to the parameter factor which is not included in the optical path design result, calculates a correlation between modified-parameter factors by using the modified-parameter factor as an index; and a re-design parameter determination means which, by estimating from the optical path design result and the correlation a re-design characteristic value which is a characteristic value with respect to the additional factor, determines a re-design parameter factor which is a parameter factor of another design parameter providing the re-design characteristic value.

Description

Optical path design apparatus and optical path design method

The present invention relates to an optical path design apparatus and an optical path design method, and more particularly to an optical path design apparatus and an optical path design method used in an optical communication network using a wavelength division multiplexing system.

Due to the explosive spread of mobile terminals and the rapid expansion of high-definition video distribution services, expansion of communication capacity in the core network is required. This demand for capacity expansion tends to continue. In order to continuously increase the communication capacity at a limited cost, it is effective to increase network utilization efficiency by efficiently operating network resources.

Wavelength division multiplexing technology is used for the core network. And the communication service which connects between points is accommodated in an optical path. When the optical path capacity is sufficiently large, a plurality of communication services can be multiplexed and accommodated in one optical path. Here, in order to generate an optical path, that is, to perform optical communication between desired points, it is necessary to assign a route and a frequency resource (spectrum) (Routing and Spectrum Assignment: RSA) (for example, patents) Reference 1). That is, it is necessary to find a solution to the so-called RSA problem.

In the RSA problem, the solution may not be found if network resources are insufficient. This means that the generation of the optical path has failed, and the desired service cannot be provided between desired points.

In order to solve the RSA problem and check whether it is possible to generate an optical path, it is necessary to accurately grasp the resources of the wavelength division multiplexing network for which the optical path is to be generated and use it as an input parameter. The resources of a wavelength division multiplexing optical network include, for example, the number of nodes, the distance between nodes, the number of optical fibers connecting the nodes, the number of optical transceivers (transponders) deployed in the nodes, the wavelength band occupied by the optical path, and optical transmission. There are a wide variety of things such as various physical limitations in If these input parameters change, the calculation conditions will change, so the calculation must be repeated.

A wavelength division multiplexing network is generally composed of a large number of nodes and links, and a path connecting desired points, that is, an RSA problem solution is often not unique. The Dijkstra method is generally used for finding a solution, that is, searching for an optical path. In the Dijkstra method, even if there are a plurality of solutions, the solution having the shortest path length is selected, so that the solution can be obtained uniquely. Linear programming and heuristic methods are often used to find solutions. In linear programming, a completely optimal solution is required, but the problem is that it takes a very long time to obtain a solution. Depending on the calculation conditions, a solution cannot be obtained in a realistic time, and a solution may not be obtained substantially. On the other hand, in contrast to linear programming, the heuristic method takes a relatively short time to obtain a solution, but the obtained solution is not necessarily a completely optimal solution. Even when the heuristic method is used, for example, when the number of nodes is large or the number of optical paths to be handled is large, the time required to obtain a solution becomes long. In general, as network resources such as the number of nodes increase, the calculation time increases exponentially.

JP 2016-005063 A

In order to shorten the optical path design time in an optical communication network (wavelength division multiplexing optical communication network) using wavelength division multiplexing, it is not necessary to calculate all possible combinations of design parameters, but to narrow down the possible values. It is effective to do. In this case, a person skilled in optical path design can narrow down with a certain degree of accuracy from his experience and knowledge. However, the design parameters are diverse and are expected to continue to increase further in the future. For this reason, it is becoming increasingly difficult for even an expert to narrow down the possible values of design parameters.

Also, in the conventional wavelength division multiplexing optical communication network, the design parameter hardly changed with time. This is because the wavelength multiplexing optical communication network is a fixed network. On the other hand, with the advancement of digital coherent technology in recent years, it is predicted that the wavelength division multiplexing optical communication network will progress to a network that dynamically changes with time. In this case, even if the network is a fixed network, the optimization is very complicated because each parameter changes over time in optical path design that requires various design parameters. Along with this, the time required for optimization increases dramatically.

As described above, the wavelength division multiplexing optical communication network has a problem that the time required for the optical path optimization increases dramatically due to an increase in design parameters and a dynamic change.

The object of the present invention is to solve the above-described problem that in the wavelength division multiplexing optical communication network, the time required for optical path optimization increases dramatically due to an increase in design parameters and a dynamic change. An object of the present invention is to provide an optical path design apparatus and an optical path design method.

The optical path design apparatus of the present invention relates to an optical path in which a parameter element, which is an element that can be taken by each of a plurality of design parameters of an optical path, and a characteristic value indicating a characteristic of the optical path designed using the parameter element are related to each other. By indexing the changed parameter element with respect to one design parameter having the changed parameter element in which an additional element not included in the optical path design result is added to the design result acquisition means for acquiring the design result, Correlation calculation means for calculating the correlation between the parameter elements after change, and the redesign characteristic value is given by estimating the redesign characteristic value that is the characteristic value for the additional element from the optical path design result and correlation Redesign parameter determining means for determining a redesign parameter element which is a parameter element of another design parameter.

The optical path design method of the present invention relates to an optical path in which a parameter element, which is an element that can be taken by each of a plurality of design parameters of an optical path, and a characteristic value indicating the characteristic of the optical path designed using the parameter element are related to each other. By obtaining the design result and indexing the changed parameter element for one design parameter having the changed parameter element in which an additional element not included in the optical path design result is added to the parameter element, the parameter element after the change is indexed. Between the optical path design result and the correlation, and the redesign characteristic value, which is the characteristic value for the additional element, is estimated from the correlation result. Determine design parameter elements.

According to the optical path design apparatus and the optical path design method of the present invention, it is possible to reduce the time required for optical path optimization in a wavelength division multiplexing optical communication network.

It is a block diagram which shows the structure of the optical path design apparatus which concerns on the 1st Embodiment of this invention. It is a block diagram which shows the structure of the wavelength division multiplexing optical communication network system which the optical path design apparatus concerning the 1st Embodiment of this invention makes object. It is a figure which shows the data stored in the database with which the wavelength division multiplexing optical communication network system which the optical path design apparatus concerning the 1st Embodiment of this invention makes object is equipped. It is a figure for demonstrating the procedure for obtaining the optical path design result which the optical path design apparatus concerning the 1st Embodiment of this invention uses. It is a figure for demonstrating the procedure for obtaining the optical path design result which the optical path design apparatus concerning the 1st Embodiment of this invention uses. It is a figure for demonstrating the procedure for obtaining the optical path design result which the optical path design apparatus concerning the 1st Embodiment of this invention uses. It is a figure for demonstrating operation | movement of the optical path design apparatus which concerns on the 1st Embodiment of this invention. It is a figure for demonstrating operation | movement of the optical path design apparatus which concerns on the 1st Embodiment of this invention, Comprising: It is a figure which shows the relationship between the statistical value showing the characteristic of a design parameter and a design result. It is a figure for demonstrating operation | movement of the optical path design apparatus which concerns on the 1st Embodiment of this invention, Comprising: It is a figure which shows the relationship between the statistical value showing the characteristic of a design parameter and a design result. It is a figure for demonstrating operation | movement of the optical path design apparatus which concerns on the 1st Embodiment of this invention, Comprising: It is a figure which shows 4 * 4 mesh topology as an example of a parameter element. It is a figure for demonstrating operation | movement of the optical path design apparatus which concerns on the 1st Embodiment of this invention, Comprising: It is a figure which shows 16-node ring topology as an example of a parameter element. It is a figure for demonstrating operation | movement of the optical path design apparatus which concerns on the 1st Embodiment of this invention, Comprising: It is a figure which shows 16 node star topology as an example of a parameter element. It is a figure for demonstrating operation | movement of the optical path design apparatus which concerns on the 1st Embodiment of this invention, Comprising: It is a figure which shows the relationship between a parameter element and its magnitude relationship. It is a figure for demonstrating operation | movement of the optical path design apparatus which concerns on the 1st Embodiment of this invention, Comprising: When a parameter element is added, it is a figure which shows the relationship between a parameter element and its magnitude relationship. It is a figure for demonstrating the guard band allocation system which the optical path design apparatus concerning the 2nd Embodiment of this invention uses as a design parameter. It is a figure for demonstrating operation | movement of the optical path design apparatus which concerns on the 2nd Embodiment of this invention, Comprising: It is a figure which shows the calculation result of the blocking rate of the optical path in a 4x4 mesh network. It is a figure which shows the blocking rate of the optical path computed by the optical path design apparatus which concerns on the 2nd Embodiment of this invention, Comprising: The case where B = b2 and A = a1 is shown. It is a figure which shows the blocking rate of the optical path computed by the optical path design apparatus which concerns on the 2nd Embodiment of this invention, Comprising: The case where B = b3 and A = a2 is shown. It is a figure which shows the blocking rate of the optical path computed by the optical path design apparatus which concerns on the 2nd Embodiment of this invention, Comprising: The case where it is set as B = b2, A = a2, B = b3, and A = a1 is also united Show.

Hereinafter, embodiments of the present invention will be described with reference to the drawings.

[First Embodiment]
FIG. 1 is a block diagram showing a configuration of an optical path design apparatus 100 according to the first embodiment of the present invention.

The optical path design apparatus 100 includes a design result acquisition unit (design result acquisition unit) 110, a correlation calculation unit (correlation calculation unit) 120, and a redesign parameter determination unit (redesign parameter determination unit) 130.

The design result acquisition unit 110 associates a parameter element, which is an element that can be taken by each of a plurality of design parameters of an optical path, with a characteristic value indicating a characteristic of the optical path designed using the parameter element. Get the result. The correlation calculation unit 120 uses the changed parameter element as an index for one design parameter having the changed parameter element in which an additional element not included in the optical path design result is added to the parameter element. The correlation between is calculated. Then, the redesign parameter determination unit 130 estimates the redesign characteristic value, which is the characteristic value for the additional element, from the optical path design result and the correlation, thereby using the parameter elements of other design parameters that give the redesign characteristic value. A certain redesign parameter element is determined.

Thus, in the optical path design device 100 according to the present embodiment, the redesign characteristic value that is the characteristic value for the additional element is estimated from the optical path design result and the correlation. A redesign parameter element that is a parameter element of another design parameter that gives the estimated redesign characteristic value is determined. With this configuration, it is not necessary to recalculate the characteristic values for all parameter elements that can be taken by other design parameters in order to optimize the optical path for the additional elements. As a result, according to the optical path design device 100 of the present embodiment, the time required for optical path optimization can be shortened in the wavelength division multiplexing optical communication network.

Next, the optical path design method according to this embodiment will be described.

In the optical path design method according to the present embodiment, first, a parameter element that can be taken by each of a plurality of design parameters of the optical path, and a characteristic value indicating a characteristic of the optical path designed using the parameter element are obtained. Obtain the related optical path design results. Then, for one design parameter having a changed parameter element in which an additional element not included in the optical path design result is added to the parameter element, the post-change parameter element is indexed, thereby correlating the changed parameter elements. Is calculated. Finally, a redesign parameter element that is a parameter element of another design parameter that gives a redesign characteristic value by estimating a redesign characteristic value that is a characteristic value for the additional element from the optical path design result and correlation described above To decide.

This configuration eliminates the need to recalculate the characteristic values for all parameter elements that can be taken by other design parameters in order to optimize the optical path for the additional elements. As a result, according to the optical path design method of the present embodiment, it is possible to reduce the time required for optical path optimization in the wavelength division multiplexing optical communication network.

Next, the optical path design result described above will be described in more detail. FIG. 2 shows a configuration of a wavelength division multiplexing optical communication network system targeted by the optical path design apparatus 100 according to the present embodiment. Here, a case where the statistic Y of the optical path design result changes depending on the optical path design parameters A and B in the wavelength division multiplexing optical communication network 1000 will be described.

In the wavelength division multiplexing optical communication network 1000, the optical nodes 1001 to 1005 are connected by optical fibers. In the wavelength division multiplexing optical communication network 1000, the characteristic value Y indicating the characteristic of the optical path changes according to the parameters A and B. Here, the number of elements (values) that can be taken by the parameter A is, for example, m, specifically ai (i = 1, 2,..., M). The same applies to the parameter B, and there are n possible elements (values), specifically bj (j = 1, 2,..., N).

The wavelength division multiplexing optical communication network 1000 is managed by a network management system (NMS) 1100. The NMS 1100 is connected to the optical nodes 1001 to 1005, and is configured to give various instructions to the optical nodes 1001 to 1005 and collect information on the optical nodes 1001 to 1005.

Connected to the NMS 1100 is a PCE (Path Computation Element) 1200, which is a functional block for optical path design. The PCE 1200 is further connected to the analysis function unit 1300.

The PCE 1200 performs optical path design in the wavelength division multiplexing optical communication network 1000 using the design parameters ai and bj received from the NMS 1100. The analysis function unit 1300 analyzes the characteristics of the wavelength division multiplexing optical communication network 1000 in which the optical path is designed by the PCE 1200, and digitizes the analysis result. Here, the characteristic value of the wavelength multiplexing optical communication network 1000 calculated by the analysis function unit 1300 based on the result of the design performed by the PCE 1200 by inputting the design parameters A = ai and B = bj from the NMS 1100 is represented as yij.

Since the functions of the PCE 1200 and the analysis function unit 1300 can be regarded as functions, they are represented by F. That is, a series of operations for calculating the statistical value yij representing the characteristics of the wavelength multiplexing optical communication network 1000 from the design parameters A = ai and B = bj described above can be expressed as yij = F (ai, bj). In this way, the characteristic value yij varies depending on the possible parameter elements ai and bj. Here, if a set of yij is represented by Y, it can be represented as Y = F (A, B).

The analysis function unit 1300 is further connected to a database (DB) 1400. The DB 1400 stores Y = F (A, B), which is a result calculated by the PCE 1200 and the analysis function unit 1300. Data stored in the DB 1400 can be referred to from the NMS 1100.

FIG. 3 shows an example of data stored in the DB 1400. FIG. 3 shows a case where m × n results obtained by calculating characteristic values y for m parameter elements A and n parameter elements B are stored. When the values of parameters A and B are changed, that is, the number of elements of parameter A and parameter B is increased, the number of elements of Y also increases. As described above, the number of data stored in the DB 1400 increases as the optical path design is performed with different parameter values for the parameter A and the parameter B.

Next, the function F will be described with reference to FIGS.

First, the case where the value of the design parameter B is fixed to b1 will be described with reference to FIG. When the value of the design parameter A is a1 and the value of the design parameter B is b1, the PCE 1200 receives information that A = a1 and B = b1 from the NMS 1100 (step S1001). Then, an optical path design in the wavelength division multiplexing optical communication network 1000 is performed using a preinstalled RSA (Routing and Spectrum Assignment) algorithm (step S1002), and a design result is acquired (step S1003).

Here, the calculation time required for optical path design using the RSA algorithm generally increases exponentially according to the number of optical paths to be handled and the scale of the network. Therefore, when an optical path design is performed under conditions close to an actual network, there is a problem that the calculation may not be completed in a realistic time as described above.

The same applies when the value of parameter A changes. The case where the value that can be taken by the parameter A changes to ai (i = 1, 2,... M) and the value of the parameter B is fixed to b1 will be described. When A = a1 and B = b1, the optical path design is performed through steps S1001 to S1003 as described above. Since this operation can be regarded as a function h, it is expressed as hi1 = h (ai, b1) (i = 1, 2,... M). hi1 (i = 1, 2,..., m) represents a design result when A = ai and B = b1. That is, the state of the wavelength division multiplexing optical communication network 1000 has a distribution of hi1 according to the parameter A.

The analysis function unit 1300 performs a statistical process on the distribution of hi1 (step S1200) to obtain a correlation between the parameter A and hi1. That is, by performing statistical processing on the distribution of hi1, the parameter B is fixed to b1, and the statistical value y1 when the parameter A changes in the range of a1 to am can be obtained (step S1300).

FIG. 5 shows an operation for obtaining an optical path design result when the parameter B is changed from b1 to b2. The value taken by the parameter B is changed from b1 to b2, but the operation is the same as that shown in FIG. However, the obtained statistical value is y2 reflecting that the parameter B has changed from b1 to b2.

FIG. 6 summarizes the procedure described with reference to FIGS. 4 and 5. In the figure, when the design parameter A of the wavelength multiplexing optical communication network changes within the range of a1 to am and the parameter B changes within the range of b1 to bn, the statistical value Y representing the characteristics of the design result of the wavelength multiplexing optical communication network is , Y1 to yn.

By the operation procedure described above, the statistic Y of the optical path design result is obtained by the optical path design parameters A = [a1, a2,... Am] and B = [b1, b2,. Change.

Next, the case where the number of possible values of the optical path design parameter is increased will be described.
Here, the number of values that can be taken by the design parameter A of the optical path is the same, and the number of values that the parameter B can take is increased by one, that is, a case where a parameter element b (n + 1) is newly added. I will explain to you.

In this case, it can be considered that bn shown in FIG. 6 is replaced with b (n + 1). For this reason, y (n + 1) = f (A, b (n + 1)) is added to y (n) for calculation. At this time, the only parameter element added for the parameter B is b (n + 1), but this affects all of the parameters A having m possible values. Therefore, the number of calculations that need to be additionally performed is m. After the additional calculation here, the results are added and accumulated in the DB 1400 (see FIG. 3), and the DB 1400 is updated. The NMS 1100 can obtain the statistical value y (n + 1) corresponding to b (n + 1) by referring to the DB 1400.

Here, according to the present embodiment, even when the parameter element b (n + 1) is added, it is not necessary to repeat the operation shown in FIG. Therefore, the number of additional calculations using the RSA algorithm that tends to require a long time for calculation can be reduced, and the time required to update the DB 1400 can be shortened.

Hereinafter, the operation of the optical path designing apparatus according to the present embodiment that can obtain the above-described effects will be described in detail.

In the optical path design apparatus 100 according to the present embodiment, the characteristic value Y = [y1, y2,... Yn] of the wavelength division multiplexing optical communication network 1000 is acquired, and B = [b1, b2,. A new function g is introduced. By introducing the function g, it is possible to reduce the number m of necessary calculations described above. Here, it is assumed that Z = [z1, z2,... Zn] having a relationship of Z = g (B) can be obtained for the function g. FIG. 7 shows the relationship between Z, B, and g.

Next, the operation of reducing the number of additional calculations using the RSA algorithm by introducing the function g will be described.

FIG. 8A is a graph of the statistical value Y that represents the characteristics of the design result of the wavelength division multiplexing optical communication network described with reference to FIG. 6 when there are three possible values of the design parameter B, b1, b2, and b3. It is a thing. FIG. 8A shows a case where the maximum value is adopted as the statistical value Y. That is, the function f described with reference to FIG. 6 is the design parameter B = b1, and the optical path design result H = [h11, h21,... Hn1] for the design parameter A = [a1, a2,. On the other hand, it has a function for obtaining the maximum value.

8A, the curve of B = b1 takes the maximum value Y = y1 when A = a1. Similarly, when B = b2, the maximum value Y = y2 is taken at A = a2, and when B = b3, the maximum value Y = y3 is taken at A = a3.

Here, a case where b4 is newly added to possible values (parameter elements) of the design parameter B will be described. In this case, in order to obtain the maximum value Y = y4 when B = b4, it is necessary to repeat the operation represented by the function f for the case where B = b4. Since the function f includes calculation using the RSA algorithm, the calculation time is generally long.

On the other hand, the optical path design apparatus 100 of the present embodiment uses a correlation between the statistical value Y indicating the characteristics of the optical path that has already been obtained and the elements of the design parameter B. As this correlation, the magnitude relationship between the elements of the design parameter B can be used. That is, the correlation calculation unit 120 included in the optical path design device 100 can be configured to calculate a magnitude relationship obtained by quantifying the parameter elements after change as the correlation. Here, the changed parameter element is a parameter element of the design parameter B in which b4 is newly added in the above-described example.

In this case, the function g described above is a function that determines the magnitude relationship between the elements b1 to b3 of the design parameter B. If the design parameter B can clearly determine the magnitude relationship, the function g can be easily determined. For example, in the case of b1 <b2 <b3, if zi (i = 1 to 3) satisfying z1 <z2 <z3 is defined, zi = 1 × (bi) is established. Therefore, the function g is a simple linear function.

However, the magnitude relationship between the design parameters of the optical path in the wavelength division multiplexing optical communication network is not always determined. For example, the network topology as shown in FIGS. 9A, 9B, and 9C is used as a design parameter. If the element b1 of the design parameter B is a 4 × 4 mesh topology (FIG. 9A), b2 is a 16-node ring topology (FIG. 9B), and b3 is a 16-node star topology (FIG. 9C), the magnitude relationship between b1 to b3 is It cannot be quantified simply. Even in such a case, the function g is introduced so that the magnitude relationship between b1 and b3 can be defined. Therefore, the function g needs to be selected according to the design parameter B. When the design parameter B is the topology described above, the magnitude relationship can be defined using, for example, graph theory.

Here, there is a function g where Z = g (B), Z = [z1, z2, z3], and B = [b1, b2, b3], and the size between Z and B is as shown in FIG. A case where there is a relationship will be described. At this time, it is assumed that the parameter element b4 is newly added. It is assumed that z4 = g (b4), and the magnitude relationship shown in FIG. At this time, it can be seen from FIG. 10B that the magnitude relationship of b1 <b4 <b2 <b3 is established.

On the other hand, it is known that the relationship shown in FIG. 8A exists between A = [a1, a2, a3], B = [b1, b2, b3], and Z = [z1, z2, z3]. . As described above, it is known that b1 <b4 <b2 by introducing the function g. From the magnitude relationship between b1, b4, and b2 and the relationship shown in FIG. 8A, it is predicted that the relationship y1 <y4 <y2 holds for the Y value y4 for B = b4. The reason will be described below.

It is known that when the parameter elements of the design parameter B are b1 and b2, the distribution of the statistical value Y for the design parameter A is given as shown in FIG. 8B. Here, for example, when the value of B = b1 changes by a minute value Δb, the Y distribution Y1 ′ = (A, b1 + Δb) for A is the Y distribution Y1 for A when B = b1 = (A , B1). At this time, it is known that there is a relationship of b1 <b4 <b2.
Accordingly, similarly for b4 and b2, if Y distribution Y4 = (A, b4), Y2 = (A, b2) for A is obtained, distribution Y4 may exist between distribution Y1 and distribution Y2. I can expect. That is, it can be predicted that the distribution Y4 exists in the shaded area shown in FIG. 8B.

Y4 is the maximum value of the characteristic value y for the design parameter A at B = b4.
Thus, it can be expected that y4 exists between the maximum value y1 of y for A at B = b1 and the maximum value y2 of y for A at B = b2. That is, y4 can be predicted to exist in the area indicated by the broken line in the shaded portion shown in FIG. 8B. Here, since the values of A giving y1 and y2 are A = a1 and a2, respectively, it can be expected that the value a4 of A giving y4 is in the range of a1 <a4 <a2.

When the additional element b4 is newly added, in the case described with reference to FIGS. 4 and 5, the optical path design using the RSA algorithm is performed for all the possible values of the design parameter A (FIG. 5). Range indicated by arrow (2) in 8B). That is, the operation represented by the function h (Ai, b4), (i = 1, 2,... M) is performed to obtain y4 = f (Ai, b4), (i = 1, 2,... M). Become. In this case, since the additional calculation using the RSA algorithm is performed m times, as described above, a long time is required for the calculation.

On the other hand, according to the optical path design apparatus 100 of the present embodiment, it is not necessary to perform optical path design for all values (parameter elements) that the design parameter A can take. This is because by introducing the function g that determines the magnitude relationship of the design parameter B, the element a4 of the design parameter A corresponding to the statistical value y4 is in the range of a1 <a4 <a2, as shown in FIG. 8B. (The range indicated by the arrow (1) in FIG. 8B). Therefore, the number of calculations of the optical path design required for obtaining y4 can be reduced from m times to one. Here, the calculation time required to narrow the range of values that can be taken by the design parameter A using the function g is negligibly short compared to the time required for the optical path design using the RSA algorithm. Therefore, when b4 is added, the time required to update the DB 1400 according to an instruction from the NMS 1100 can be shortened to 1 / m. As a result, the optical path can be optimized even when the design parameters of the network change with time.

As described above, according to the optical path design device 100 and the optical path design method of the present embodiment, it is possible to shorten the time required for optical path optimization in the wavelength division multiplexing optical communication network. As a result, the optical path can be optimized even when the design parameters of the network change with time.

[Second Embodiment]
Next, a second embodiment of the present invention will be described. The configuration of the optical path design apparatus according to this embodiment is the same as that according to the first embodiment (see FIG. 1). Hereinafter, the operation of the optical path design apparatus according to the present embodiment will be specifically described.

The wavelength multiplexing optical communication network 1000 (see FIG. 2) can adopt a 4 × 4 square lattice mesh network as a network topology. The topology showing the connection state between the optical node and the optical fiber in this case is as shown in FIG. 9A.

When a 4x4 mesh network is adopted, the guard band allocation method can be selected as the optical path design parameter A. That is, the design parameters targeted by the optical path design apparatus according to the present embodiment include at least the node order centrality and the guard band allocation method. Specifically, as shown in FIG. 11, when the optical signal band of the same optical modulation system is adjacent to the a1 system that always provides guard bands at both ends of the optical signal band, One of the a2 methods that omits the guard band can be selected. As shown in FIG. 11, when the two optical signal bands are not adjacent, there is no difference between the a1 method and the a2 method.

Here, a case will be described in which, in the 4 × 4 mesh network, a method that can accommodate more communication traffic in the optical path is selected from the a1 method and the a2 method. This corresponds to the case where the maximum communication traffic volume that can be accommodated in the optical path is adopted as the statistic Y of the optical path design result and the topology is adopted as the design parameter B. That is, the design result acquisition unit 110 (see FIG. 1) included in the optical path design device 100 can be configured to use the maximum value of the amount of communication traffic that can be accommodated in the optical network as the characteristic value.

As described above, when one design parameter B among a plurality of design parameters is a network topology, b1 among parameter elements that can be taken by the design parameter B is a 4 × 4 mesh network. In this case, the function F representing the operation of the optical path design described in the first embodiment can be accommodated when two types of guard band allocation methods (a1, a2) are used in the 4 × 4 mesh network (= b1). It is a function for designing an optical path with the largest amount of communication traffic.

If the communication traffic volume can be increased or decreased in units of 200 Gbps (gigabits per second), the optical signal band shown in FIG. 11 increases or decreases in units of 200 Gbps. In addition, each optical fiber connecting between optical nodes in a 4 × 4 mesh network (= b1) has a capacity capable of accommodating, for example, 200 optical signal bands corresponding to a communication traffic amount of 200 Gbps. The communication traffic request is assumed to be a full mesh traffic that is generated in an equal number for each node, and the generation order is randomly generated.

As shown in FIG. 11, in order to prevent deterioration of optical signal quality during optical fiber transmission, a guard band may be added to the optical signal band accommodating communication traffic. In this case, the smaller the guard band to be added, the larger the amount of communication traffic that can be accommodated in the 4 × 4 mesh network (= b1). For this reason, it is considered that the amount of communication traffic that can be accommodated is always greater in the a2 method with fewer guard bands to be added, but this is not necessarily the case. This is because in the a2 method, the combined bandwidth of the optical signal and the guard band is not uniform depending on whether or not the guard band is provided. Therefore, in the case of the a2 method, there is a possibility that the fragmentation of the wavelength band becomes larger than that in the a1 method. The fragmentation of the wavelength band causes a reduction in communication traffic accommodation efficiency. Therefore, it cannot be easily determined which of the a1 method and the a2 method can accommodate more communication traffic.

Also, the situation where the wavelength band is fragmented depends on the topology. Therefore, when one optical node is added or when one optical fiber is added, the result of which of the a1 method and the a2 method can accommodate more communication traffic may change.

5 and FIG. 6, it is necessary to always perform the optical path design for both the a1 method and the a2 method every time the topology that is the design parameter B changes. If it is assumed that an optical path design under a certain condition takes one day, it takes two days to obtain an optical path design result using the a1 method and the a2 method for a b1 = 4 × 4 mesh network. This is because it is necessary to obtain respective values of the statistical value Y when the a1 method and the a2 method are applied, that is, y1 = f (A, b1), A = [a1, a2]. That is, if the statistical value Y is not obtained for A = [a1, a2], it is determined whether the a1 method or the a2 method is a guard band allocation method that can accommodate more communication traffic in a 4 × 4 mesh network. I can't.

In the example described above, a case where the parameter element b2 is newly added as a possible value of B, that is, a case where the optical path design is optimized also for the wavelength division multiplexing optical communication networks of different topologies will be described. In this case, since it is necessary to newly calculate the statistical value y2 for the parameter element b2, two more days are required to obtain the optical path design result.

Next, the operation of the optical path design apparatus according to the present embodiment will be described.

As described above, in the present embodiment, the statistical value Y is defined as the maximum value of the amount of communication traffic that can be accommodated in a wavelength-multiplexed optical communication network of a certain topology. According to this definition, Y indicates how small the amount of traffic that could not be accommodated when the communication traffic request was increased. That is, Y indicates how small the blocking rate of the optical path accommodating communication traffic is in the optical path design using the RSA algorithm.

FIG. 12 shows the result of calculating the blocking rate of the optical path in the 4 × 4 mesh network (b1) when a certain amount of communication traffic request is generated and the a1 method and the a2 method are applied. From FIG. 12, it can be seen that in the 4 × 4 mesh network (B = b1), the blocking rate of the optical path is smaller when the a1 method is applied.

Next, a case where a topology type (parameter element) as the design parameter B is added will be described. According to the optical path design apparatus of the present embodiment, the statistical value y1 for the parameter element b1 is obtained, and the function value g for determining the magnitude relationship of the parameter element b is determined, thereby obtaining the statistical value y2 for the parameter element b2. The calculation of can be omitted.

Below, an optical network EON16 (= b2, FIG. 9B) consisting of 16 nodes and an optical network NSFNET16 (= b3, FIG. 9C) consisting of 16 nodes are newly added as possible elements of the design parameter B (topology). An example will be described. A case where a function for calculating the degree centrality is employed as the function g will be described. That is, a case will be described in which the correlation calculation unit included in the optical path design apparatus of the present embodiment is configured to index the changed parameter elements (b1 to b3) by digitizing the characteristics of the network topology.

When calculating the degree centrality of each element of the design parameter B, ie, zi = g (bi), where zi is (i = 1, 2, 3), Z = [0.076, 0.086, 0.19]. It becomes. Here, the degree of centrality, that is, the calculation time required to obtain Z using the function g is shorter than the calculation time of the function f including the RSA algorithm, and can be ignored.

When the design parameter B is topology, the magnitude relationship between the three parameter elements b1 = 4 × 4 mesh, b2 = EON16, and b3 = NSFNET16 cannot be determined as it is. However, when the degree centrality of b1 to b3 is obtained by introducing the function g, a magnitude relationship of z1 = g (b1) <z2 = g (b2) << z3 = g (b3) is obtained. That is, the values of z1 and z2 are close, and the value of z3 is about twice as large as z1 and z2. From this, as explained using FIG. 10B, it is predicted that b1 and b2 are close to each other, and b3 is far from b1 and b2.

Here, it is obtained as an optical path design result that the value of A giving y1 is B1 when B = b1. The difference between the values of b1 and b2 is small, and the difference between b1 and b3 is extremely large. At this time, the possible value of the design parameter A is either a1 or a2. As described above, the value of A corresponding to B = b1 is a1. Therefore, the value of A corresponding to B = b2 is a1 close to the value of A corresponding to B = b1, and the value of A corresponding to B = b3 is a3 far from the value of A corresponding to B = b1. Can be predicted. That is, it can be predicted that the value of A giving y2 at B = b2 is a1, and the value of A giving y3 at B = b3 is a2.

13 shows the blocking rate of the optical path when B = b2 and A = a1, and FIG. 14 shows the blocking rate of the optical path when B = b3 and A = a2. FIG. 15 also shows the calculation result of the blocking rate of the optical path when B = b2, A = a2, B = b3, and A = a1. FIG. 15 shows that the prediction according to the present embodiment described above was correct.

According to the optical path design apparatus of this embodiment, for example, the calculation of the function f necessary for obtaining the result shown in FIG. 13 is only for A = a1. Since the operation time of the function g is negligibly short compared to the operation time of the function f, the blocking rate of the optical path is calculated in comparison with the above-described example in which A = a2 needs to be calculated. The required time can be reduced to ½.

As described above, according to the optical path design apparatus and the optical path design method of the present embodiment, it is possible to shorten the time required for optical path optimization in the wavelength division multiplexing optical communication network.

The elements of the design parameters A and B used by the optical path design apparatus described in this embodiment correspond to the case where m = 2 and n = 3 in the first embodiment. The values of m and n are not limited to these, and it is apparent that the effect of the present embodiment can be obtained even when other values are used.

In the present embodiment, a case has been described in which a function for calculating the degree centrality, which is a feature of the topology using graph theory, is adopted as the function g for determining the magnitude relationship of the parameter elements. However, the present invention is not limited to this, and a function that calculates features of other topologies may be adopted as the function g. Here, the topological features using graph theory include degree centrality, intermediary centrality, proximity centrality and their central tendency, Bonachic centrality, eigenvector centrality, Euclidean distance, diameter, density, average distance, Transitivity, φ coefficient, correlation matrix, distance matrix are included.

In the above-described embodiment, the case where the function g where Z = g (B) is employed to calculate the correlation (magnitude relationship) between the parameter elements has been described. At this time, for example, when the difference between the values of b1 and b4 or b4 and b2 shown in FIG. However, the accuracy of prediction can be improved by introducing a plurality of types of functions g. That is, a configuration in which a plurality of functions gi with Zi = gi (B), (i = 1, 2,... P) is introduced can be adopted. When the results of narrowing down the range of the a4 value by the function gi differ, it can be determined by, for example, majority vote. The prediction accuracy increases as the number of p increases. As the plurality of functions gi, for example, two or more types of functions that respectively calculate two or more of the topological features using the graph theory described above can be employed.

In this case, the configuration of the optical path design apparatus 100 shown in FIG. 1 will be described. The correlation calculation unit 120 included in the optical path design device 100 may be configured to calculate a plurality of correlations between the changed parameter elements by performing a plurality of types of indexing on the changed parameter elements. At this time, the redesign parameter determination unit 130 can be configured to determine the redesign parameter element by estimating the redesign characteristic value from the optical path design result and a plurality of correlations.

Although the present invention has been described with reference to the embodiments, the present invention is not limited to the above embodiments. 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.
This application claims the priority on the basis of Japanese application Japanese Patent Application No. 2017-070398 for which it applied on March 31, 2017, and takes in those the indications of all here.

DESCRIPTION OF SYMBOLS 100 Optical path design apparatus 110 Design result acquisition part 120 Correlation calculation part 130 Redesign parameter determination part 1000 Wavelength division multiplexing optical communication network 1001-1005 Optical node 1100 Network management system (NMS)
1200 PCE
1300 Analysis Function Unit 1400 Database (DB)

Claims (10)

Obtaining a design result that obtains an optical path design result that associates a parameter element that can be taken by each of a plurality of design parameters of the optical path and a characteristic value that indicates the characteristic of the optical path designed using the parameter element Means,
For one design parameter having a changed parameter element in which an additional element not included in the optical path design result is added to the parameter element, by indexing the changed parameter element, between the changed parameter elements Correlation calculating means for calculating the correlation;
A redesign parameter element that is a parameter element of another design parameter that gives the redesign characteristic value by estimating the redesign characteristic value that is the characteristic value for the additional element from the optical path design result and the correlation Redesign parameter determining means for determining, and an optical path design apparatus. In the optical path design device according to claim 1,
The optical path design apparatus, wherein the design result acquisition means uses a maximum value of the amount of communication traffic that can be accommodated in an optical network as the characteristic value. In the optical path design device according to claim 1 or 2,
The correlation calculation means calculates a plurality of correlations between the changed parameter elements by performing a plurality of types of indexing on the changed parameter elements,
The redesign parameter determining means determines the redesign parameter element by estimating the redesign characteristic value from the optical path design result and the plurality of correlations. In the optical path design device according to any one of claims 1 to 3,
The optical path design device, wherein the correlation calculation means calculates a magnitude relationship obtained by quantifying the changed parameter element as the correlation. In the optical path design device according to any one of claims 1 to 4,
The one design parameter is a network topology,
The optical path design apparatus, wherein the correlation calculation means indexes the changed parameter element by quantifying the characteristics of the network topology. In the optical path design device according to any one of claims 1 to 5,
The optical path design apparatus, wherein the design parameters include at least a node degree centrality and a guard band allocation method. Obtaining an optical path design result associating a parameter element that is an element that each of a plurality of design parameters of the optical path can take with a characteristic value indicating a characteristic of the optical path designed using the parameter element;
For one design parameter having a changed parameter element in which an additional element not included in the optical path design result is added to the parameter element, by indexing the changed parameter element, between the changed parameter elements Calculate the correlation,
A redesign parameter element that is a parameter element of another design parameter that gives the redesign characteristic value by estimating the redesign characteristic value that is the characteristic value for the additional element from the optical path design result and the correlation Determine the optical path design method. In the optical path design method according to claim 7,
Calculating the correlation includes calculating a plurality of correlations between the changed parameter elements by performing a plurality of types of indexing on the changed parameter elements;
Determining the redesign parameter element includes determining the redesign parameter element by estimating the redesign characteristic value from the optical path design result and the plurality of correlations. . In the optical path design method according to claim 7 or 8,
The calculating the correlation includes calculating a magnitude relationship obtained by quantifying the changed parameter element. In the optical path design method according to any one of claims 7 to 9,
The characteristic value is a maximum value of the amount of communication traffic that can be accommodated in the optical network,
The one design parameter is a network topology,
Calculating the correlation includes calculating a correlation between the changed parameter elements by digitizing features of the network topology and indexing the changed parameter elements. Path design method.

Images