JP5168189B2 - Optical network design support system and program - Google Patents

Description

  The present invention relates to an optical network design support system and program for supporting the design of a PDS (Passive Double Star) type optical network.

  In a PDS type optical network, that is, a so-called PON (Passive Optical Network) network, when a large number of subscribers are allocated to a plurality of accommodation stations, the conventional technique has been relied on the manual work of skilled technicians. In Non-Patent Documents 1 and 2, it is assumed that PON subscribers' homes are uniformly distributed in a circular area centered on the accommodation station. A minimizing computer approach is described.

Daisuke Udagawa, Takao Matsumoto “Study on Minimum Optical Fiber Length of Multi-Stage Optical Access Network” IEICE General Conference, B-8-29, 2003 Daisuke Udagawa, Takao Matsumoto "Study on cost minimization conditions for multistage branch access network" IEICE General Conference, B-8-11, 2004

  Manual operation by a skilled engineer cannot confirm whether an optimal solution in terms of cost or a preferable solution above a certain level has been obtained. Moreover, enormous effort and time are required for creation.

The methods described in Non-Patent Documents 1 and 2 cannot be applied to cases where subscribers' homes are not uniformly distributed, or where the shape of the area covered for each accommodation station is not circular. Moreover, such a situation is more realistic.

  Therefore, the present invention presents an optical network design support system and program that presents a preferable solution regarding the optical fiber cost, assuming that a large number of subscribers' homes and a plurality of accommodation stations are distributed non-uniformly. Objective.

  An optical network design support system according to the present invention includes a position information database for storing position information of a accommodating station and a subscriber, parameter storage means for storing a maximum drop cable length, an optical splitter branch number, and an initial number of clusters; An optical splitter position determining unit that divides the applicants into clusters equal to or greater than the initial number of clusters and determines an optical splitter position in each cluster, wherein the number of applicants in each cluster Optical splitter position determination that determines the optical splitter position in each cluster so that the distance from the optical splitter candidate position in each cluster to the subscriber who wants to join in the same cluster is equal to or less than the maximum drop cable length. And the optical splitter at each of the optical splitter positions determined by the optical splitter position determining means. The characterized by comprising an optical splitter / acquisition station connection means for connecting to either the acquisition station liter.

The optical network design support program according to the present invention uses a computer to refer to the location information of the accommodation station and the subscription applicant stored in the location information database for a plurality of toll stations and a plurality of subscription applicants. under the maximum drop cable length and the optical splitter branches of the condition number, the position of the optical splitter, and an optical network design support program to determine the acquisition station for acquisition of the optical splitter, to the computer, the plurality of subscribers This is an optical splitter position determination function that divides the applicants into clusters that are equal to or greater than the initial number of clusters, and determines the optical splitter position within each cluster. Yes, the distance from the optical splitter candidate position in each cluster to the desired subscriber in the same cluster is the maximum drop An optical splitter position determining function of determining the optical splitter position within each cluster to be a flop cable length below, the optical splitter of the optical splitters each position determined by one of ordinary optical splitter position determination function of the acquisition station characterized in that to realize an optical splitter / acquisition station connectivity to connect to either.

  According to the present invention, when a large number of subscribers 'homes and a plurality of accommodation stations are distributed non-uniformly, it is possible to support a design with reduced optical fiber costs based on the location information of the subscribers' residences and the accommodation stations. .

It is a schematic block diagram of one Example of this invention. It is a flowchart of the network design process of a present Example. Oh Ru in the description example shown the expropriation station distribution of subscribers who wish. It is a detailed flowchart of the clustering process of FIG. It is an example which divided | segmented the example shown in FIG. 3 into a cluster. This is an example in which one cluster is further divided into two in the division example shown in FIG. FIG. 7 is an explanatory diagram of a state in which a nearby subscriber is connected to each optical splitter in the cluster division example shown in FIG. 6. It is a flowchart of the process which connects an optical splitter and a receiving station with the flow shown in FIG. It is explanatory drawing of the state which connected the optical splitter and the collection station by the flow shown in FIG. 3 is another flowchart of the clustering process of FIG. 2.

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

  FIG. 1 shows a schematic block diagram of a design support system according to the present invention. An input device 12 such as a keyboard and a display device 14 are connected to the CPU 10 of the computer. The design support program of this embodiment operates on the CPU 10. The parameter storage device 16 stores various parameters input by the input device 12 necessary for the design support program on the CPU. The parameter storage device 16 includes, for example, a RAM constituting the computer or a hard disk device as a secondary storage device.

The location information database (DB) 18 holds location information on a two-dimensional plane of PON subscribers and accommodation stations located in the design target area of the PON network. The location information of the accommodation station is C _i (i = 1 to N), and the location information of the applicant is X _j (j = 1 to M). In this example, it is assumed that there are N accommodation stations and the number of subscribers is M. C _i (i = 1 to N) indicates a case where each accommodation station is indicated and a case where the position coordinates are indicated. Similarly, X _j (j = 1 to M) may indicate the applicant for the subscription itself or may indicate one coordinate thereof.

The maximum accommodated line number database (DB) 20 indicates the maximum accommodated line number Nf _i (i = 1 to N) of each accommodated station C _i (i = 1 to N). For example, the maximum number of accommodated lines of C ₁ is Nf ₁ .

  It is obvious that the database may be configured to have a collecting station database that holds the position information of the collecting station and the maximum number of lines to be collected, and a subscription applicant database that holds position information related to the subscriber.

  The CPU 10 includes the following functions 22 to 30 as functional elements of the design support program. FIG. 2 shows an operation flowchart of the main flow of the design support program executed by the CPU 10.

The optical splitter arrangement design function 22 reads the location information C _i (i = 1 to N) of the accommodation station and the location information X _j (j = 1 to M) of the subscribers from the location information DB 18 (S1). In this example, there are N accommodation stations, and there are M applicants. FIG. 3 shows an example of the distribution of applicants and accommodation stations. In the distribution example shown in FIG. 3, it is determined by examining which of the three receiving stations to which 16 subscribers should be connected. M = 16 and N = 3. Assume that each receiving station can accommodate one or more PON systems.

  The operator uses the input device 12 to input basic parameters of the optical network design, here, the maximum drop cable length Lmax, the number of optical splitter branches Ns, and the number of clusters (initial value) Nc to the CPU 10 (S2). Of course, the input of these parameters (S2) may be performed prior to step S1. The maximum drop cable length Lmax is the maximum allowable distance from the optical splitter to the subscriber's home. The optical splitter branching number Ns is the number of subscribers who can be accommodated in one optical splitter. The number of clusters Nc is the number of PON systems, and here, since one optical splitter is assumed for one PON system, it is also the number of optical splitters. The number of clusters Nc may be increased or decreased during the calculation depending on the number of accommodation stations or the number of subscribers. In the example shown in FIG. 3, since the number of toll stations is 3, the initial value of the cluster number Nc is set to 3.

  Next, using the clustering processing function 24, the subscribers are clustered with a cluster number equal to or greater than the initial value of the cluster number Nc, and an optical splitter is assigned to each cluster (S4). Specifically, according to the conditions and initial values set in steps S1 and S2 and stored in the parameter storage device 16 by the clustering processing function 24, the number of applicants in each cluster is Ns or less, and the drop cable length Are clustered under the condition that the maximum length Lmax does not exceed the maximum length Lmax, and the positions of the optical splitter candidates are determined (S4).

FIG. 4 shows a detailed flowchart of the clustering process (S4). In accordance with the target area and the number of clusters Nc delivered from the main routine, Nc centroids Y _k (k = _{1 to} Nc) that become optical splitter position candidates are randomly set in the target area (S11).

Through a predetermined clustering process, all applicants who are present in the target area are divided into Nc clusters (S12). This is because, in each cluster, the affiliation of the applicant to each cluster and the position of each centroid so that the sum of the Euclidean distances between the centroid _Yk and the applicants belonging to the cluster is minimized. It is a process to determine. As such a clustering processing method, there is a k-means method widely used in the field of data mining. Each centroid determined by this clustering process becomes a position candidate of the optical splitter.

FIG. 5 shows a result of dividing the subscribers into three clusters by the clustering process (S12) with respect to the example shown in FIG. Dashed lines indicate cluster boundaries. A black circle is a centroid Y _k (light splitter candidate).

For each centroid _Yk , it is checked whether or not the desired number of subscribers Ns that can be accommodated in the optical splitter and the maximum drop cable length Lmax are satisfied (S14, S15). For this purpose, first, a variable k for designating a cluster / centroid for condition determination is initialized with 1 (S13). Then, it is checked whether or not the subscribers who belong to the kth cluster are Ns or less (S14), and all the distances between the subscribers in the kth cluster and the centroid _Yk are less than the maximum drop cable length Lmax. Is checked (S15).

  When any of the conditions (S14, S15) is not satisfied, the kth cluster is divided into two clusters by the clustering processing function (S18). FIG. 6 is an example shown in FIG. 5 and shows the result of further dividing this cluster into two parts in the lower right cluster where the condition of step S14 is not satisfied. In this case, the number of clusters (or the number of centroids) is 4. As the software program, step S18 is performed by recursively executing an argument with the kth cluster as the target region and the number of clusters being 2 by passing it to the clustering routine or function shown in FIG. Is realized.

The applicants who belong to the kth cluster are Ns or less (S14), and the Euclidean distance between the applicants in the kth cluster and the centroid _Yk is all less than the maximum drop cable length Lmax. If there is (S15), k is incremented (S17), and it is confirmed whether or not the conditions of steps S14 and S15 are satisfied for the next cluster.

  If it is confirmed that the conditions of steps S14 and S15 are satisfied for all clusters, the process returns to FIG. 3 to determine the centroid as the optical splitter position and logically connect the applicants in each cluster to the optical splitters in the same cluster. (S5). At this point, the placement of the optical splitter and the design of the drop cable are complete. It is known that a clustering method such as the k-means method converges to a local optimum solution. Therefore, the design result of the obtained PON network is a quasi-optimal solution for the optical fiber line length. FIG. 7 shows a state in which the optical splitter is arranged at the optical splitter candidate position (centroid) in the example shown in FIG. 6, and the subscriber is connected to the optical splitter in the same cluster in step S5.

  The receiving station determination function 26 determines to which receiving station each optical splitter whose position has been determined in step S5 is connected (S6). A detailed flowchart of step S6 is shown in FIG.

  The expropriation station determination function 26 instructs the Voronoi area calculation function 28 to divide the design target range into Voronoi areas using the positions of the N accommodation stations as a generating point (S21). The Voronoi region represents a set of points whose distance from the mother point is shorter than the distance from other mother points. A region obtained by dividing the entire region into a plurality of Voronoi regions is called a Voronoi diagram. FIG. 9 is a result of dividing the arrangement of the receiving station and the optical splitter shown in FIG. 7 into Voronoi regions, and a broken line indicates a boundary line of the Voronoi region. Examples of the Voronoi diagram drawing method include a sequential addition method and a recursive bisection method.

The expropriation station determination function 26 executes the following processing for all of the obtained Voronoi regions. First, a variable k designating a Voronoi region to be processed is initialized with 1 (S22). The toll station determining function 26 obtains the maximum accommodated line number Nf _k of the _kth accommodated station from the maximum accommodated line number DB 20 (S23). It is determined whether or not the sum of the number of optical splitters existing in the kth Voronoi region and the number of lines already allocated to the _kth accommodation station is less than or equal to Nfk (S24). That is, it is examined whether or not a PON system equal to the number of optical splitters existing in the kth Voronoi region can still be accommodated in the kth receiving station.

When the sum of the number of optical splitters existing in the kth Voronoi region and the number of lines already allocated to the _kth accommodation station is Nfk or less (S24), each optical splitter existing in the kth Voronoi region is Connect to the k-th accommodation station (S25). If k is not equal to Nc (S26), k is incremented (S27), and S23 and subsequent steps are repeated. If k is equal to Nc (S26), the process shown in FIG. 8 is terminated.

If the sum of the number of optical splitters existing in the kth Voronoi region and the number of lines already allocated to the _kth accommodation station exceeds Nfk (S24), all the light existing in the kth Voronoi region no longer exists. The splitter cannot be connected to the kth accommodation station. Therefore, an adjacent Voronoi region j (where j ≠ k), and there is still an available station even if the number of all optical splitters in the Voronoi region j and the number of optical splitters in the already connected adjacent region are added. It is checked whether or not there is one having S (S28). That is, the possibility of connection to a receiving station in an adjacent Voronoi area is examined. If there is no such adjacent Voronoi region j (where j ≠ k) (S28), the toll station determination function 26 displays an error message indicating that the PON network cannot be designed on the display device 14 (S29).

When there is an adjacent Voronoi region j (where j ≠ k) having a free acquisition station (S28), the adjacent Voronoi region j (where j ≠ k) in the optical splitter in the kth Voronoi region. the closest optical splitter acquisition station _{C j)} of connecting to acquisition station _{C j} (S30). Then, step S24 and subsequent steps are repeated. As a result, the number of optical splitters to be considered in step S24 is reduced by one. By repeating step S30 as many times as necessary, the condition of step S24 is finally satisfied, and the process proceeds to the next Voronoi region as described above (S26, S27).

  In this way, the all-optical splitter arranged in step S5 can be connected to a receiving station as close as possible. This completes the design of the PON network as shown in FIG. The design result output function 30 outputs the design result to the display device 14. Of course, the design result may be printed out and stored in a hard disk (not shown).

  In the flow shown in FIG. 4, the cluster in which the condition of the number of subscribers desired or the condition of the drop cable length in the cluster is not satisfied is divided into two by recursive processing. May be re-executed. FIG. 10 shows a flowchart in which FIG. Instead of step S18 of FIG. 3, the cluster number Nc is incremented (S18a), and step S11 and subsequent steps are re-executed.

  In this method, the clustering is repartitioned and takes more time than the flow shown in FIG. 3, but the whole is reconsidered, so that a good distribution of all applicants can be realized as a whole.

  According to the present invention, for example, in the case where a large number of PON subscribers 'homes and a plurality of accommodation stations are non-uniformly distributed, the optical fiber cost is determined based only on the location information of the subscribers' residences and the accommodation stations. A PON network that gives a suboptimal solution can be designed. As a result, it is possible to reduce the PON network design burden of the communication carrier that develops the PON service.

  Although the invention has been described with reference to specific illustrative embodiments, various modifications and alterations may be made to the above-described embodiments without departing from the scope of the invention as defined in the claims. This is obvious to an engineer in the field to which the present invention belongs, and such changes and modifications are also included in the technical scope of the present invention.

10: CPU
12: input device 14: display device 16: parameter storage device 18: position information database (DB)
20: Maximum number of accommodated lines database (DB)
22: Optical splitter arrangement design function 24: Clustering processing function 26: Acquired station determination function 28: Voronoi area calculation function 30: Design result output function

Claims (8)

A location information database for storing location information of the containment station and applicants;
Parameter storage means for storing the maximum drop cable length, the number of branches of the optical splitter, and the initial number of clusters;
An optical splitter position determining unit that divides the applicants into clusters equal to or greater than the initial number of clusters and determines an optical splitter position in each cluster, wherein the number of applicants in each cluster Optical splitter position determination that determines the optical splitter position in each cluster so that the distance from the optical splitter candidate position in each cluster to the subscriber who wants to join in the same cluster is equal to or less than the maximum drop cable length. Means,
An optical network design support system comprising: an optical splitter / acquisition station connection means for connecting an optical splitter at each optical splitter position determined by the optical splitter position determination means to any of the acquisition stations. The optical splitter position determining means is
A dividing means for optimizing the centroid by randomly setting a number of centroids corresponding to the initial number of clusters and dividing all applicants into a number of clusters corresponding to the initial number of clusters. ,
For each cluster generated by the dividing means, the first condition in which the number of applicants in the cluster is equal to or less than the number of optical splitter branches, and the desire to join in the same cluster from the position of the centroid in each cluster Determining means for determining whether or not the second condition that the distance to the person is equal to or less than the maximum drop cable length is satisfied;
Recursion means for executing the dividing means and the determining means for a cluster that does not satisfy at least one of the first condition and the second condition;
2. The optical network design support according to claim 1, wherein the discriminating means comprises means for setting the position of the centroid of the cluster satisfying the first condition and the second condition as the optical splitter position. system. The optical splitter position determining means is
A dividing means for optimizing the centroid by randomly setting a number of centroids corresponding to the initial number of clusters and dividing all applicants into a number of clusters corresponding to the initial number of clusters. ,
For each cluster generated by the dividing means, the first condition in which the number of applicants in the cluster is equal to or less than the number of optical splitter branches, and the desire to join in the same cluster from the position of the centroid in each cluster Determining means for determining whether or not the second condition that the distance to the person is equal to or less than the maximum drop cable length is satisfied;
Means for re-executing the dividing means and the determining means when there is a cluster that does not satisfy at least one of the first condition and the second condition;
The optical network according to claim 1, further comprising means for setting the position of the centroid of each cluster as the optical splitter position when all the clusters satisfy the first condition and the second condition. Design support system. The optical splitter / acquisition station connection means is
Voronoi area dividing means for dividing the area including the applicant to the Voronoi area with the receiving station as a base point;
In each Voronoi region divided by the Voronoi region dividing means, it is connected when an optical splitter in the same Voronoi region can be connected to a receiving station in the Voronoi region, and the same Voronoi region is connected to a receiving station in the Voronoi region. 4. A connection means for checking whether connection is possible with a receiving station in an adjacent Voronoi area when there is no space for connecting an optical splitter in the optical splitter, according to claim 1. Optical network design support system. Using a computer, refer to the location information of the accommodating station and the subscriber who are stored in the location information database for a plurality of receiving stations and a plurality of subscribers, and determine the maximum drop cable length and the number of optical splitter branches. Under the conditions, an optical network design support program for determining the position of the optical splitter and the acquisition station to acquire the optical splitter, the computer ,
An optical splitter position determining function for dividing the plurality of subscribers into clusters equal to or greater than the initial number of clusters and determining an optical splitter position in each cluster, wherein the number of subscribers in each cluster The optical splitter position that determines the optical splitter position in each cluster so that the distance from the optical splitter candidate position in each cluster to the subscriber who wants to join in the same cluster is less than or equal to the maximum drop cable length. A decision function;
Optical network design supporting the light splitter has been the optical splitters each position determined by those optical splitter positioning function, characterized in that to realize an optical splitter / acquisition station connectivity to connect to either of the acquisition station program. The optical splitter position determination function
The number of centroids that correspond to those of the cluster number initial value in terms of set randomly, all subscribers who wish to divide the number of clusters corresponding to the number of the clusters initial value, divided optimize the centroid Function and
For each cluster generated by those said dividing function, the first condition subscriber seekers number in the cluster is less than the optical splitter number of branches, and subscription in the same cluster from a position of the centroid within each cluster a determination function to determine whether the second or satisfies the distance to the seeker is less than the maximum drop cable length,
A recursive function for executing the division function and the discrimination function to the cluster that do not meet at least one of those first and second conditions,
In those該判another feature, the optical network according to claim 5, characterized in that it comprises a position of the centroid of the first condition and the second condition is satisfied cluster function to the corresponding optical splitter position Design support program. The optical splitter position determination function
The number of centroids that correspond to those of the cluster number initial value after having allowed to set at random, all subscribers who wish to divide the number of clusters corresponding to the number of the clusters initial value, that optimize the centroid Split function and
For each cluster generated by those said dividing function, the first condition subscriber seekers number in the cluster is less than the optical splitter number of branches, and subscription in the same cluster from a position of the centroid within each cluster a determination function to determine whether the second or satisfies the distance to the seeker is less than the maximum drop cable length,
If the cluster is not satisfied at least one of those first and second conditions are present, the ability to re-execute the division function and the determination function,
If all clusters have the first condition and the second condition is satisfied, the light according to the position of the centroid of each cluster to claim 5, characterized in that a function and shall be the said optical splitter position Network design support program. The optical splitter / acquisition station connection function,
A Voronoi tessellation ability to divide those該収for station areas including the subscriber seeker as a base point in the Voronoi region,
In those the Voronoi tessellation each Voronoi regions divided by function, when the optical splitter of the Voronoi same Voronoi region to acquisition stations in the area can be connected is connected, the same expropriation station of the Voronoi region either when there is no vacant for connecting the optical splitter of the Voronoi region are of claims 5 to 7, characterized by comprising a connection function Inspect connectable against acquisition station of the adjacent Voronoi regions 1 The optical network design support program according to the item.

Images