FXC network expansion device, expansion method, and expansion program

Abstract

To automate wiring work in a communication building while suppressing initial investment costs.SOLUTION: An adding device 1 for an FXC network including intermediate FXCs of an intermediate layer and input / output FXCs of an input / output layer includes: a determination unit 11 that determines a final configuration of the FXC network; and a design unit 12 that designs an initial configuration of the FXC network. The design unit 12 causes the number of intermediate FXCs in the initial configuration to be smaller than the number of intermediate FXCs in the final configuration, and causes the number of initial fibers accommodated by the input / output FXCs in the initial configuration to be smaller than the number of final fibers by using the number of final fibers accommodated by each input / output FXC in the final configuration and the number of the intermediate FXCs in the final configuration.SELECTED DRAWING: Figure 20

Description

[Technical field]

[0001] The present invention relates to an FXC network expansion device, an expansion method, and an expansion program. [Background technology]

[0002] Although the installation of optical fiber is a tedious task, it is an essential foundation for fast and robust network services. The operational costs of these tasks in telecommunications buildings are high because they usually require the creation of work instructions, the dispatch of two workers to perform the connection at both ends of the fiber, the testing of the fiber continuity and power levels, and a supervisor to oversee these tasks. Furthermore, human operators can make mistakes in installing the fibers, which can lead to poor service quality and / or additional work to correct these mistakes. If the installation could be performed remotely and automatically without human intervention, the costs of these tasks could be reduced.

[0003] Fiber Cross Connect (FXC), which allows wiring work to be performed automatically and remotely, has been researched and developed (Patent Document 1, Non-Patent Documents 1 and 2). [Prior art documents] [Patent documents]

[0004] [Patent Document 1] Patent No. 6737747 [Non-patent literature]

[0005] [Non-Patent Document 1] T. Mano, T. Inoue, K. Mizutani, and O. Akashi, “Increasing capacity of the clos structure for optical switching networks,” in2019 IEEE Global Communications Conference (GLOBECOM), 2019, pp. 1-6. [Non-Patent Document 2] [2] C. Clos, “A study of non-blocking switching networks,” Bell System Technical Journal, vol. 32, no. 2, pp. 406-424, 1953. Summary of the Invention [Problem to be solved by the invention]

[0006] By constructing an FXC network and accommodating optical wiring within a telecommunications building, wiring work can be performed automatically and remotely. In other words, by connecting a large number of fibers to the FXC in advance and remotely changing the connections within the FXC, the connections of the fibers connected to the FXC can be changed without on-site manual work.

[0007] Since one FXC cannot accommodate all the wiring in a telecommunications building, it is necessary to configure an FXC network using multiple FXCs to automate the wiring work. In this case, in order to avoid blockages, not only FXCs that accommodate fibers but also FXCs that connect FXCs are required. Blockages are a state in which there are fiber pairs that cannot be connected no matter how the internal connection status of the FXCs in the FXC network is changed. When an FXC network is blocked, there are fiber pairs that cannot be connected by remote control, so manual work is required, and automation of wiring work cannot be realized.

[0008] The number of fibers (number of wiring) in a telecommunications building is small at first and increases over time, so it is desirable for an FXC network to have an expandable configuration in which the number of FXCs is small at the beginning and then the number of fibers that can be accommodated increases as the number of FXCs is increased. In this way, an FXC network is said to be expandable because the number of fibers that can be accommodated can be increased by adding FXCs. The act of adding FXCs to increase the number of wiring that can be accommodated is also called expansion.

[0009] The present invention has been made in consideration of the above circumstances, and an object of the present invention is to provide a technology that can automate wiring work within a communication building while reducing initial investment costs in an expandable FXC network. [Means for solving the problem]

[0010] In order to achieve the above-mentioned objective, one embodiment of the present invention is an FXC network expansion device, the FXC network comprising intermediate FXCs in an intermediate layer and input / output FXCs in an input / output layer, a determination unit for determining a final configuration of the FXC network, and a design unit for designing an initial configuration of the FXC network, wherein the design unit makes the number of intermediate FXCs in the initial configuration smaller than the number of intermediate FXCs in the final configuration, and makes the number of initial fibers accommodated by the input / output FXCs of the initial configuration smaller than the number of final fibers using the number of final fibers accommodated by each input / output FXC in the final configuration and the number of the intermediate FXCs in the final configuration.

[0011] One aspect of the present invention is a method for expanding an FXC network performed by an expansion device, the FXC network comprising intermediate FXCs in an intermediate layer and input / output FXCs in an input / output layer, and comprising a determination step for determining a final configuration of the FXC network and a design step for designing an initial configuration of the FXC network, the design step making the number of intermediate FXCs in the initial configuration smaller than the number of intermediate FXCs in the final configuration, and making the number of initial fibers accommodated by the input / output FXCs of the initial configuration smaller than the number of final fibers using the number of final fibers accommodated by each input / output FXC in the final configuration and the number of the intermediate FXCs in the final configuration.

[0012] One aspect of the present invention is an expansion program that causes a computer to function as the expansion device. Effect of the Invention

[0013] According to the present invention, it is possible to provide a technology that can automate wiring work within a communication building while suppressing initial investment costs in this expandable FXC network. [Brief description of the drawings]

[0014] [Figure 1A] This is a configuration diagram of FXC (N=4). [Figure 1B] FIG. 13 is a diagram showing an example of internal connections of FXC (N=3). [Diagram 2] Figure 2 shows an example of an FXC network configuration for the improved Clos configuration. [Diagram 3] Figure 3 shows an example of connecting two fibers in an improved Clos configuration. [Figure 4] FIG. 4 shows an example of a conventional Clos-based FXC network configuration. [Diagram 5] Figure 5 shows an example of connecting two fibers in a conventional Clos configuration. [Figure 6] Figure 6 shows an example of an expansion using the recursive method applied to the improved Clos configuration. [Figure 7] Figure 7 shows an example of an expansion using the recursive method applied to a conventional Clos configuration. [Figure 8] Figure 8 shows an example of the initial configuration when the intermediate placement method is applied to the improved Clos configuration. [Figure 9] Figure 9 shows an example of an expansion in which the intermediate placement method is applied to the improved Clos configuration. [Figure 10] FIG. 10 shows an example of the initial configuration when the intermediate placement method is applied to the conventional Clos configuration. [Figure 11] Figure 11 shows an example of an expansion using the intermediate placement method on a conventional Clos configuration. [Figure 12] FIG. 12 shows the final configuration of a specific example of this embodiment. [Figure 13] FIG. 13 shows the initial configuration of a specific example of this embodiment. [Figure 14] FIG. 14 shows the configuration of a specific example of this embodiment after one expansion. [Figure 15] FIG. 15 shows a configuration of a specific example of this embodiment after two expansions. [Figure 16] FIG. 16 shows the configuration of a specific example of this embodiment after three expansions. [Figure 17] FIG. 17 shows the configuration of a specific example of this embodiment after four expansions. [Figure 18] FIG. 18 shows the configuration of a specific example of this embodiment after five expansions. [Figure 19] FIG. 19 shows the configuration of a specific example of this embodiment after six expansions. [Figure 20] FIG. 20 is a diagram showing the configuration of an extension device of this embodiment. [Figure 21] FIG. 21 is an explanatory diagram showing the process of constructing an FXC network of an initial configuration. [Figure 22] FIG. 22 is an explanatory diagram showing the process of adding an FXC. [Diagram 23] FIG. 23 is a sequence diagram showing the processing at the time of initial construction and expansion. [Figure 24] FIG. 24 is a graph showing the number of FXCs used and the number of fibers that can be accommodated in the extension method and intermediate placement method of this embodiment. [Diagram 25] FIG. 25 shows an expanded version of the cable gland shown in FIG. 24 that can accommodate up to 15,000 fibers. [Figure 26] 2 is an example of a hardware configuration. DETAILED DESCRIPTION OF THE PREFERRED EMBODIMENTS

[0015] 1. FXC Configuration 1A is a diagram showing a schematic configuration of an FXC (optical fiber cross connect). The FXC has N receiving ports 101 on the front side and N transmitting ports 102 on the back side. The receiving ports 101 are optical fiber ports through which optical signals flow in the input direction as seen from the FXC. The transmitting ports 102 are optical fiber ports through which optical signals flow in the opposite direction as seen from the FXC, through which optical signals flow in the output direction.

[0016] In this embodiment, the number of receiving ports and the number of transmitting ports of each FXC are equal, and furthermore, the number of receiving ports of all FXCs is equal. The number of receiving ports of an FXC is represented by N, which is also called the FXC size. Figure 1A shows an example of the internal connection state of an FXC with an FXC size of 4.

[0017] FIG. 1B shows all possible internal connection states for an FXC with an FXC size of 3.

[0018] 2.FXC network configuration In this embodiment, as the configuration of an FXC network having multiple FXCs, the FXC network configuration shown in Fig. 2 and the FXC network configuration shown in Fig. 3 are assumed. The network configuration shown in Fig. 2 is called an improved Clos configuration, and the network configuration shown in Fig. 3 is called a conventional Clos configuration.

[0019] 2.1 Improved Clos configuration The improved Clos configuration shown in Figure 2 can accommodate the same number of wiring with fewer FXCs than the conventional Clos configuration when using a switch such as FXC that has the same number of receiving ports and transmitting ports (see non-patent document 1).

[0020] The improved Clos configuration comprises a set of input / output layer FXCs that accommodate fibers within a telecommunications building (fibers laid between buildings, fibers connected to network equipment, etc.), and a set of intermediate layer FXCs that interconnect these FXCs.

[0021] Specifically, the FXC network with the improved Clos configuration is equipped with k FXCs (input / output FXCs) that accommodate the fibers in the communication building and handle input / output, and l FXCs (intermediate FXCs) that interconnect and relay the input / output FXCs. The set of input / output FXCs is called the input / output layer, and the set of intermediate FXCs is called the intermediate layer.

[0022] Each input / output FXC has n receive ports and n transmit ports that accommodate the fibers in the telecommunications building. i The fiber connected to the j-th receiving port ofi j , the fiber connected to the jth transmitting port is r i j Each input / output FXC has m receiving ports and m sending ports connected to the intermediate FXCs, one for each intermediate FXC.

[0023] Similarly, each intermediate FXC has m receiving ports and m transmitting ports connected to input / output FXCs, one for each input / output FXC.

[0024] In this FXC network, each input / output FXC can accommodate n pairs of fibers, so the entire network can accommodate nk pairs of fibers. The probability of blocking can be reduced by increasing the number of intermediate FXCs l, or the number of wires m between the input / output FXCs and the intermediate FXCs. In addition, by making the number of intermediate FXCs l, or the number of wires m between the input / output FXCs and the intermediate FXCs sufficiently large, the probability of blocking can be reduced to 0, that is, the network becomes non-blocking (see Non-Patent Document 1).

[0025] Figure 3 shows an example of connecting two fibers accommodated in the improved Clos configuration shown in Figure 2 with an FXC network. In Figure 3, fiber t 1 1 and fiber 1 k Fiber t is connected. 1 1 The optical signal is input / output FXC P 1 The intermediate FXC P k+1 FXC P k and connected to the fiber r 1 k is entered into

[0026] 2.2 Conventional Clos configuration The conventional Clos configuration shown in Figure 4 comprises a set of FXCs in the input layer where light is input from fibers within a telecommunications building, a set of FXCs in the output layer that output light to the fibers, and a set of FXCs in an intermediate layer that interconnect these FXCs.

[0027] Specifically, the conventional Clos configuration has k FXCs (input FXCs) to which light from the fiber inside the communication building is input, k FXCs (output FXCs) that output light to the fiber, and l FXCs (intermediate FXCs) that interconnect them. The set of input FXCs is called the input layer, the set of output FXCs the output layer, and the set of intermediate FXCs the intermediate layer.

[0028] Each input FXC has n receive ports that accommodate the fibers in the telecommunication building, and m transmit ports that connect to each intermediate FXC. i The fiber connected to the j-th receiving port of i j Let us assume that.

[0029] Similarly, each output FXC has n transmit ports that accommodate the fibers in the telecommunication building, and m receive ports that connect to each intermediate FXC. i The fiber connected to the j-th transmitting port of i j Therefore, each intermediate FXC has m receiving ports connected to each input FXC, and m sending ports connected to each output FXC.

[0030] In this FXC network, each input FXC and output FXC can accommodate n fibers, so the entire network can accommodate nk pairs of fibers. The probability of blocking can be reduced by increasing the number of intermediate FXCs l, or the number of wirings m between the input FXCs and the output FXCs and the intermediate FXCs. In addition, by making the number of intermediate FXCs l or the number of wirings m sufficiently large, the probability of blocking can be reduced to 0, that is, the network is non-blocking (see Non-Patent Document 1).

[0031] Figure 5 shows an example of connecting two fibers that are conventionally accommodated in a Clos configuration with an FXC network. 1 1 and fiber 1 k Fiber t is connected. 1 1 The optical signal is input to the FXC P 1The intermediate FXC P k+1 and then output FXC P 2k+1 and connected to the fiber r 1 k is entered into

[0032] 3. Existing expansion method Next, we explain the existing methods of expanding FXC: the recursive method, the intermediate layer placement method, and the relocation method.

[0033] 3.1 Recursion The recursive method expands the FXC network by adding a hidden layer in addition to the hidden layers already placed. First, we explain its application to the improved Clos configuration, and then to the conventional Clos configuration.

[0034] Figure 6 shows an example of applying the recursive method to the improved Clos configuration. After the intermediate layer, another set of FXCs in the second intermediate layer is added to interconnect the intermediate FXCs. This method can be repeated recursively. The recursive method has the problem that the optical signal is attenuated as the number of FXCs used for fiber connection increases. For example, in Figure 3, the optical signal passes through three FXCs, but in Figure 6, it passes through a maximum of five FXCs.

[0035] Figure 7 shows an example of applying the recursive method to the conventional Clos configuration. An FXC set of two other hidden layers (second hidden layer, third hidden layer) is added after the hidden layer. As with the improved Clos configuration, the recursive method has the problem that the number of FXCs used for fiber connection increases, resulting in attenuation of the optical signal. For example, in Figure 5, the optical signal passes through three FXCs, but in Figure 7, it passes through a maximum of five FXCs.

[0036] 3.2 Intermediate layer placement method In the intermediate layer placement method, all intermediate FXCs are installed from the beginning, and input / output FXCs (or input FXCs and output FXCs) are added when expanding. Since no additional layers are added, the number of FXCs used for fiber connection does not change even after expansion, but since all intermediate FXCs are placed at the initial state, the number of FXCs required at the initial state is large.

[0037] Figure 8 shows an example of the initial state when the hidden layer placement method is applied to the improved Clos configuration. 2 ~P l+1 and one FXC P in the input / output layer. 1 Chisels are placed.

[0038] Figure 9 shows the layout of the system in Figure 8 with one input / output FXC P by using the intermediate layer placement method. 2 This is an example of adding n FXCs. As a result, the number of fiber pairs that the FXC network can accommodate increases by n. From this point on, input and output FXCs are added one by one. In the intermediate layer placement method, l+1 FXCs are required in the initial state as shown in Figure 8, which is a large number of FXCs required in the initial state.

[0039] Figure 10 shows an example of the initial state when the hidden layer placement method is applied to the conventional Clos configuration. 2 ~P l+1 FXC P is placed in the input layer. 1 , 1 FXC P in the output layer l+2 Place the following.

[0040] Figure 11 shows the hidden layer placement method for the input FXC P in Figure 10. 2 and one output FXC P l+4 This is an example of adding n FXCs. As a result, the number of fiber pairs that the FXC network can accommodate increases by n. After this, one input FXC and one output FXC are added. In the middle layer placement method, l+2 FXCs are required in the initial state as shown in Figure 10, which is a large number of FXCs required in the initial state.

[0041] 3.3 Relocation method The rearrangement method is an expansion method in which a small number of input / output FXCs (or input FXCs and output FXCs) and intermediate FXCs are placed in the initial state, and then input / output FXCs (or input FXCs and output FXCs) and intermediate FXCs are added with each expansion.

[0042] Unlike the intermediate placement method, almost all ports of the intermediate FXC are connected to input / output FXCs during the expansion. Therefore, when adding an input / output FXC (or an input FXC and an output FXC), it is necessary to rearrange unused wiring of the intermediate FXC that has already been placed and rearrange it between the FXC to be newly added. Also, like the intermediate placement method, almost all ports of the input / output FXC (or an input FXC and an output FXC) are connected to the intermediate FXC during the expansion. Therefore, when adding an intermediate FXC, it is necessary to rearrange unused wiring of the intermediate FXC that has already been placed and rearrange it between the intermediate FXC to be newly added.

[0043] In this rearrangement method, all the wiring of the intermediate FXCs is connected during the expansion process, so the number of fibers that can be accommodated in the FXC network increases, but each time an expansion is performed, rearrangement is required between the already installed input / output FXCs (or input FXCs and output FXCs) and the intermediate FXCs.

[0044] 4. Expansion design issues Next, we will explain the configuration of an FXC network that can add FXCs without rearranging the existing wiring, using the intermediate layer arrangement. The recursive method is a deep hierarchical configuration, and the rearrangement method requires rearrangement of the existing wiring between FXCs for each addition, so we will not explain it here.

[0045] When introducing fiber into a telecommunications building (laying fiber connecting buildings or installing new fiber connecting to network equipment), it is accommodated in the FXC network in advance (connecting one of the endpoints in Figure 2). After that, when it becomes necessary to connect the accommodated fibers together, they are connected by changing the internal connection status of the FXC in the FXC network (Figure 3).

[0046] Immediately after a telecommunications building is constructed, the number of inter-building fibers installed in the telecommunications building and the fibers connected to the network equipment are small, but over time these fibers will increase. Therefore, in order to reduce initial investment, the FXC network also needs to start with a small number of FXCs, and increase the number of fibers that can be accommodated by adding FXCs to the FXC network. Also, in order to reduce the labor costs during expansion, FXCs will be added without rearranging (re-laying) the wiring that has already been laid.

[0047] As mentioned above, the number of receiving ports and the number of sending ports of an FXC are equal, and all FXCs have the same number of receiving ports. The number of receiving ports of an FXC is represented as N, which is also called the FXC size (Figure 1).

[0048] The FXC network of this embodiment is premised on an improved Clos configuration or a conventional Clos configuration. The improved Clos configuration can be uniquely specified by four parameters k, l, m, and n. k is the number of input / output FXCs, and l is the number of intermediate FXCs. m is the number of receive ports (or transmit ports) that an input / output FXC has and that connect to each intermediate FXC. Similarly, the intermediate FXC has m receive ports and m transmit ports that connect to the input / output FXCs for each input / output FXC. n is the number of receive ports (or transmit ports) that each input / output FXC has to accommodate the endpoint fibers.

[0049] Similarly, the conventional Clos configuration can be uniquely specified by the four parameters k, l, m, and n. Therefore, the improved Clos configuration or the conventional Clos configuration is expressed by these four parameters (k, l, m, n). Let c be the symbol that represents the network configuration, and let c = (k, l, m, n).

[0050] 4.1 Expansion design issues in improved Clos configuration First, we explain the expansion design problem in the improved Clos configuration. We assume that the number of fibers that a communication building can accommodate or the number of FXCs that can be installed is known due to the physical constraints of the communication building (spatial constraints and power capacity constraints). Then, we assume that the final network configuration that can accommodate the final number of fibers or can be configured with the number of FXCs that can be installed is c∞ =(k ∞ ,l ∞ ,m ∞ ,n ∞ ) is given.

[0051] A suitable initial state c 0 =(k 0 ,l 0 ,m 0 ,n 0 Starting from c, we add FXCs to the FXC network without rearranging the existing wiring, and increase the number of wiring that the network can accommodate. Let the configuration after the i-th addition be c i =(k i ,l i ,m i ,n i ) The final network configuration c ∞ =(k ∞ ,l ∞ ,m ∞ ,n ∞ In other words, if the expansion is completed in the jth expansion, then c j =c ∞ It must be as follows.

[0052] It is also desirable that the configuration during expansion is less likely to become blocked than the final configuration. This is because the number of wires m connected to the middle layer of the input / output FXC is i l i and the number of fiber pairs to be accommodated, n i Ratio of m i l i / n i is the ratio of the final configuration m ∞ l ∞ / n ∞ To exceed m i l i / n i ≧m ∞ l ∞ / n ∞ This can be achieved by:

[0053] The expansion design problem is the final network configuration c ∞ =(k ∞ ,l ∞ ,m ∞ ,n∞ ) is given, and the network configuration sequence c 0 ,c 1 ,...,c j =c ∞ This method of generating a network configuration sequence is called the extension method. However, in the extension, that is, the network configuration after the i-th extension, c i Network configuration after the i+1th expansion from c i+1 When making changes to the above, it is not possible to rearrange (re-lay) wiring that has already been installed.

[0054] This is before and after the expansion. i ≦k i+1 ,l i ≦l i+1 ,m i ≦m i+1 ,n i ≦n i+1 This is equivalent to the fact that the number of fibers accommodated in the FXC network, n i ·k i For the number of FXC units used, k i +l i In particular, from the viewpoint of reducing initial capital investment, the smaller the number of FXCs used is within the range of the number of accommodated fibers, the better the extension method. In this embodiment, an extension method that achieves this is used.

[0055] 4.2 Expansion design issues in conventional Clos configuration Next, we will explain the expansion design problem in the conventional Clos configuration. As explained below, when the number of FXC units used is 2k i +l i Except for the above, this is the same as the expansion design for the improved Clos configuration.

[0056] As with the improved Clos configuration, the expansion design problem for the conventional Clos configuration is ∞ =(k ∞ ,l ∞ ,m ∞ ,n ∞ ) is given, and the network configuration sequence c 0 ,c 1 ,...,c j =c ∞However, the problem is to find the network configuration c i Network configuration after the i+1th expansion from c i+1 When changing the wiring, it is not possible to rearrange (re-lay) the wiring that has already been installed. i ≦k i+1 ,l i ≦l i+1 ,m i ≦m i+1 ,n i ≦n i+1 This is equivalent to the following being true:

[0057] Number of fibers accommodated in the FXC network n i ·k i 2k FXC units used i +l i In particular, from the viewpoint of reducing initial capital investment, the better the extension method is, the smaller the number of FXCs used is within the range of the number of accommodated fibers. In this embodiment, an extension method that achieves this is used.

[0058] 5. Expansion of the FXC network in this embodiment The following describes the addition of FXC in this embodiment.

[0059] 5.1 Expansion method of this embodiment In this embodiment, expansion is performed in accordance with the following explanation in both the improved Clos configuration and the conventional Clos configuration.

[0060] In this expansion method, the final network configuration c ∞ =(k ∞ ,l ∞ ,m ∞ ,n ∞ ), the initial network configuration c 0 Let be the following number 1. k, l, m, and n are as described above.

[0061]

number

[0062] The network configuration after the (i+1)th expansion is determined as follows.

[0063] (a) k i =k ∞ ∧l i =l ∞ In this case, no more FXC can be added, so no expansion is required.

[0064] (b) k i =k ∞ ∧l i <l ∞ In this case, an intermediate FXC will be added.

[0065] (c) k i <k ∞ ∧l i =l ∞ In this case, add input / output FXC (or input FXC and output FXC).

[0066] (d) Other than the above i ≧n i If so, add an intermediate FXC (d1); if not, add an input / output FXC (or an input FXC and an output FXC) (d2).

[0067] Here, the integer d is defined as the following number 2.

[0068]

number

[0069] In the above (b) or (d1), when an intermediate FXC is added, c i+1 =(k i ,l i+1 ,m i ,n i +d).

[0070] In the above (c) or (d2), when adding an input / output FXC (or an input FXC and an output FXC), i+1 =(k i+1 ,l i ,m i ,ni )

[0071] Currently being expanded i ≦k i+1 ,l i ≦l i+1 ,m i ≦m i+1 ,n i ≦n i+1 Therefore, there is no need to rearrange (re-lay) the wiring that has already been installed.

[0072] In this method, the number of fibers connected to the input and output FXCs (or the input and output FXCs) in the initial state is n i By reducing the number of intermediate FXCs required, the number of intermediate FXCs required can be reduced. In addition, this method compares the increase in the number of fibers that can be accommodated when an FXC is added to the input / output layer (or the input and output layers) or to an intermediate layer, and selects the layer in which to add the FXC so that the number of fibers that can be accommodated is increased.

[0073] 5.2 Specific examples of this expansion method A specific example of expansion using this expansion method will be explained using the improved Clos configuration. ∞ =(k ∞ ,l ∞ ,m ∞ ,n ∞ )=(72,37,14,476).

[0074] Figure 12 shows the final configuration of this expansion method. At this time, the initial configuration, the configuration after one expansion, the configuration after two expansions, the configuration after three expansions, the configuration after four expansions, the configuration after five expansions, and the configuration after six expansions, c 0 , c 1 ,c 2 ,c 3 ,c 4 ,c 5 ,c 6 are as follows. The integer d calculated by the above formula 2 is 13.

[0075] c0=(1,1,14,12) c1=(1,2,14,25) c2=(2,2,14,25) c3=(2,3,14,38) c4=(3,3,14,38) c5=(3,4,14,51) c6=(4,4,14,51). Figures 13, 14, 15, 16, 17, 18 and 19 show 0 ,c 1 ,c 2 ,c 3 ,c 4 ,c 5 ,c 6 The network configurations of the two are shown in Fig. 1. In each figure, the added FXC is drawn with a dotted line. 1 and intermediate FXC P 2 Only the wiring is shown, and other wiring is omitted.

[0076] 5.3 Configuration of the expansion device 20 is a configuration diagram showing the configuration of an extension device 1 of an FXC network in this embodiment. The extension device 1 adds FXC using the extension method of 5.1. The extension device 1 shown in the figure has a final network determination unit 11, a design unit 12, an extension judgment unit 13, a network construction unit 14, and a network status management unit 15. The extension device 1 can be used for both the improved Clos configuration and the conventional Clos configuration.

[0077] The final network determination unit 11 determines the final configuration c of the FXC network by using the physical constraints of the communication building (such as spatial constraints and power supply capacity constraints) and demand forecasts. ∞ For example, the final network determination unit 11 may calculate the maximum number of FXCs that can be installed using spatial constraints of the installation location, constraints on power supply capacity, etc., and calculate the configuration that maximizes the number of fiber pairs that can be accommodated from the calculated maximum number of FXCs as the final configuration. Alternatively, the final network determination unit 11 may calculate the number of fiber pairs that ultimately needs to be accommodated from a demand forecast, and calculate the configuration that minimizes the number of FXCs used among configurations that achieve that number of fiber pairs (accommodation number) as the final configuration. The method of Patent Document 1 may also be used to determine the final configuration.

[0078] The design unit 12 is a design unit for designing the initial configuration of the FXC network. 0 In this embodiment, the design unit 12 designs the number of intermediate FXCs in the initial configuration by multiplying the number of intermediate FXCs in the final configuration by l ∞ The number of initial fibers accommodated by the input / output FXC in the initial configuration is made smaller than the number of final fibers accommodated by each input / output FXC in the final configuration, n ∞ The design unit 12 may calculate the number of the initial fibers using the number of the final fibers and the number of the intermediate FXCs in the final configuration.

[0079] Specifically, the design unit 12 designs the FXC network of the final configuration as follows: ∞ =(k ∞ ,l ∞ ,m ∞ ,n ∞ ), the initial configuration of the FXC network is changed to the initial network configuration c 0 Here, k is the number of input / output FXCs, and l is the number of intermediate FXCs. m is the number of internal wiring between each input / output FXC and each intermediate FXC. Specifically, m is the number of receiving ports (transmitting ports) that each input / output FXC has and connects to each intermediate FXC. m is also the number of receiving ports (transmitting ports) that each intermediate FXC has and connects to each input / output FXC. n is the number of fibers that each input / output FXC accommodates. Specifically, n is the number of receiving ports (transmitting ports) that each input / output FXC has and accommodates the fibers at its endpoints.

[0080] In addition, the design unit 12 designates the current network configuration c i The design unit 12 calculates a network configuration ci+1 with an FXC added from the above. The design unit 12 may select a layer in which to place the added FXC based on the number of fibers that each input / output FXC can accommodate when an FXC is added to an input / output layer and the number of fibers that each input / output FXC can accommodate when an FXC is added to an intermediate layer. That is, the design unit 12 compares the increased number of fibers that can be accommodated when an FXC is added to an input / output layer or an intermediate layer, and selects a layer in which to add an FXC so that the number of fibers that can be accommodated is the largest.

[0081] Specifically, the design unit 12, as described above, i ≧n i In the case of , add an intermediate FXC to the middle layer and dk i <n i In this case, an input / output FXC of the input / output layer (or an input FXC of the input layer and an output FXC of the output layer) is added. The integer d is calculated by the above-mentioned formula 2.

[0082] The expansion determination unit 13 periodically monitors the state of the FXC network by referring to the network state management unit 15, and determines whether expansion is necessary. For example, the expansion determination unit 13 determines that expansion is necessary when the vacancy of the receiving port or transmitting port that accommodates the fiber at the end point of the input / output FXC (or the input FXC in the input layer and the output FXC in the output layer) becomes smaller than a predetermined threshold.

[0083] The network construction unit 14 constructs the network configuration ci set by the design unit 12. The network status management unit 15 stores and manages the status of the FXC network, such as the current network configuration, the internal connection status of each FXC, and connected fiber pairs.

[0084] 5.4 Handling additional devices FIG. 21 is an explanatory diagram showing an outline of a process in which the extension device 1 constructs an initial configuration of an FXC network. ∞ The design unit 12 determines the final configuration c and sends the final configuration to the design unit 12 (S11). ∞ Using the initial configuration c 0 Calculate the initial configuration c 0 The network construction unit 14 transmits the input initial configuration c 0 Then, the FXC network of the initial configuration is constructed according to the above, and the construction result is stored in the network status management unit 15 (S13).

[0085] FIG. 22 is an explanatory diagram showing an overview of the process in which the extension device 1 adds an FXC. The extension determination unit 13 periodically acquires information from the network status management unit 15 and determines whether or not an extension is necessary (S21). If the extension determination unit 13 determines that an extension is unnecessary, it does nothing. If it determines that an extension is necessary, the extension determination unit 13 requests the design unit 12 to add an FXC network (S22). Upon receiving the request, the design unit 12 receives the current network configuration c from the network status management unit 15. i The design unit 12 acquires the final configuration c calculated by the final network determination unit 11 when constructing the initial configuration (S23, S24). ∞ (S25).

[0086] The design section 12 is a current network configuration c i and the final configuration c ∞ Configuration after expansion using c i+1 Calculate the configuration after expansion c i+1 The network construction unit 14 sends the input configuration c i+1 The FXC network is expanded according to the above and the expansion result is stored in the network status management unit 15 (S27).

[0087] FIG. 23 is a sequence diagram showing the processing at the time of initial construction and expansion.

[0088] The final network determination unit 11 determines the final configuration c ∞ The design unit 12 determines the final configuration c and sends the final configuration to the design unit 12 (S31). ∞ Using the initial configuration c 0 Calculate (S32) and use the calculated initial configuration c 0 The network construction unit 14 transmits the input initial configuration c 0 Then, the FXC network of the initial configuration is constructed according to the above, and the construction result is stored in the network status management unit 15 (S34).

[0089] The expansion device 1 repeats the expansion process until the FXC network reaches the final configuration (S40). The expansion determination unit 13 periodically monitors the state of the FXC network in the network status management unit 15 (S41) and determines whether or not an FXC expansion is necessary (S42). Every time a fiber in a communication building is accommodated in an FXC, the network status management unit 15 stores the connection / disconnection state of the accommodated fiber pair (S43).

[0090] The expansion determination unit 13 periodically monitors the status of the FXC network in the network status management unit 15 (S44), and if it determines that an FXC expansion is necessary (S45), it requests the design unit 12 to expand the FXC network (S46).

[0091] Upon receiving the request, the design unit 12 receives the current network configuration c from the network status management unit 15. i The design unit 12 acquires the final configuration c calculated by the final network determination unit 11 (S47, S48). ∞ Get the current network configuration i and the final configuration c ∞ Configuration after expansion using c i+1 Calculate (S49) and the configuration after expansion c i+1 The network construction unit 14 sends the input configuration c i+1 The FXC network is expanded according to the above, and the expansion result is stored in the network status management unit 15 (S51).

[0092] 5.5 Comparison of the expansion method of this embodiment with the intermediate placement method Here, we use the improved Clos configuration to compare the number of FXCs used in the expansion method of this embodiment and the intermediate placement method. ∞ =(k ∞ ,l ∞ ,m ∞ ,n ∞ The number of FXCs in use and the number of fibers that can be accommodated in each network configuration being expanded when 14, 147, 148, 149, 150, 151, 152, 153, 154, 155, 156, 157, 158, 159, 160, 161, 162, 163, 164, 165, 1

[0093] In the hidden layer placement method, the following expansion is performed in the improved Clos configuration. Note that the conventional Clos configuration is the same as the improved Clos configuration.

[0094] In the hidden layer placement method, the initial state c 0 , c 0 =(1,l ∞ ,m ∞ ,n ∞ The network configuration after the i-th expansion is defined as c i =(i+1,l ∞ ,m ∞ ,n ∞ ) Let j be an integer k ∞ -1, and the expansion after the jth expansion is c j =c ∞ It will be. i +1=k i+1 ,l i =l i+1 ,m i =m i+1 ,n i =n i+1 Therefore, there is no need to rearrange (re-lay) the wiring that has already been installed.

[0095] FIG. 24 shows the network configuration column c generated by the expansion method and the intermediate placement method. 0 ,c 1 ,...,c j Number of FXC units used: k i +l i and the number of fibers that can be accommodated, n i k i It can be seen that when the number of fibers that can be accommodated is 10,000 or less, this expansion method uses fewer FXCs. When the number of fibers that can be accommodated is 10,000 or more, the two methods use roughly the same number of FXCs.

[0096] Figure 25 shows an expansion of the number of fibers that can be accommodated in Figure 24 to a range of 15,000 or less. It can be seen that when the number of fibers that can be accommodated is 4,000, the number of FXCs used with this expansion method is about 10 less than that of the intermediate placement method. Also, the smaller the number of available fibers, the fewer the number of FXCs used with this expansion method compared to the intermediate placement method.

[0097] 6. Effects of this embodiment In the FXC network expansion device 1 of this embodiment described above, the FXC network comprises intermediate FXCs in the intermediate layer and input / output FXCs in the input / output layer, a final network determination unit 11 that determines the final configuration of the FXC network, and a design unit 12 that designs the initial configuration of the FXC network, and the design unit 12 makes the number of intermediate FXCs in the initial configuration smaller than the number of intermediate FXCs in the final configuration, and makes the number of initial fibers accommodated by the input / output FXCs in the initial configuration smaller than the number of final fibers using the number of final fibers accommodated by each input / output FXC in the final configuration and the number of intermediate FXCs in the final configuration. The input / output layer may include an input layer and an output layer.

[0098] In this manner, in this embodiment, while based on the intermediate placement method, the number of fibers connected to the input and output FXCs is reduced in the initial configuration to reduce the number of intermediate FXCs required. Therefore, according to the expansion method of this embodiment, it is possible to build an FXC network that can be expanded over time while keeping initial investment low, and to automate wiring work within a telecommunications building. Furthermore, compared to the intermediate placement method, when the number of fibers to be accommodated is small, the number of FXCs used can be reduced, so that capital investment can be reduced. That is, in an expandable FXC network using multiple FXCs, it is possible to automate wiring work within a telecommunications building while keeping initial investment costs and costs for expanding FXCs low.

[0099] In this embodiment, the layer in which the FXC to be added is placed is selected based on the number of fibers that each input / output FXC can accommodate when the FXC is added to the input / output layer and the number of fibers that each input / output FXC can accommodate when the FXC is added to the middle layer. By selecting the layer in which the FXC is added so that the number of fibers that can be accommodated is large, the FXC can be added efficiently.

[0100] In this embodiment, the FXC network is maintained in a two-layer structure (input / output layer, middle layer) or three-layer structure (input layer, middle layer, output layer), so attenuation of optical signals due to multi-layering can be avoided. Also, in this embodiment, there is no need to rearrange (re-lay) the wiring that has already been laid when expanding, so the work required for expansion can be reduced.

[0101] 7.Hardware Configuration The above-described expansion device 1 can use, for example, a general-purpose computer system as shown in Fig. 26. The computer system shown in the figure includes a CPU (Central Processing Unit, processor) 901, a memory 902, a storage 903 (HDD: Hard Disk Drive, SSD: Solid State Drive), a communication device 904, an input device 905, and an output device 906. The memory 902 and the storage 903 are storage devices. In this computer system, the CPU 901 executes a predetermined program loaded on the memory 902, thereby realizing each function of the expansion device 1.

[0102] The extension device 1 may be implemented in one computer or in multiple computers, and may be a virtual machine implemented in a computer.

[0103] The program for the expansion device 1 can be stored on a computer-readable recording medium such as a HDD, SSD, USB (Universal Serial Bus) memory, CD (Compact Disc), or DVD (Digital Versatile Disc), or can be distributed via a network.

[0104] The present invention is not limited to the above-described embodiment and modifications, and various modifications are possible within the scope of the present invention. [Explanation of symbols]

[0105] 1: Expansion device 11: Final network decision section (decision section) 12:Design Department 13: Expansion judgment unit 14:Net construction department 15: Network status management unit

Claims

1. An expansion device for an FXC network including an intermediate FXC in an intermediate layer and an input / output FXC in an input / output layer, A determination unit for determining a final configuration of the FXC network; A design unit for calculating an initial configuration of the FXC network using the final configuration; an expansion determination unit that monitors a state of the FXC network constructed based on the calculations of the design unit, and requests the design unit to expand the FXC network when it is determined that an FXC expansion is necessary; The design unit includes: The number of intermediate FXCs in the initial configuration is made smaller than the number of intermediate FXCs in the final configuration; The number of initial fibers accommodated by the input / output FXC in the initial configuration is Using the number of final fibers accommodated by the input / output FXC and the number of intermediate FXCs in the final configuration, less than the number of said final fibers; The design unit includes: When an expansion of the FXC network is requested, the network configuration after expansion is calculated using the initial configuration or the current configuration and the final configuration. Expansion device.

2. The design unit selects a layer in which to place the FXC to be added based on the number of fibers that each input / output FXC can accommodate when the FXC is added to the input / output layer and the number of fibers that each input / output FXC can accommodate when the FXC is added to the intermediate layer. The expansion device according to claim 1 .

3. The FXC network of the final configuration is ∞ =(k ∞ ,l ∞ ,m ∞ ,n ∞ ), The design department will design the initial FXC network. [0010] year, k is the number of input / output FXCs, l is the number of intermediate FXCs, m is the number of internal wiring between each input / output FXC and each intermediate FXC, and n is the number of fibers contained in each input / output FXC. The expansion device according to claim 1 or 2.

4. The input / output layer includes an input layer and an output layer. The expansion device according to any one of claims 1 to 3.

5. A method for expanding an FXC network by an expansion device, comprising the steps of: The FXC network includes an intermediate FXC in an intermediate layer and an input / output FXC in an input / output layer; A determination step of determining a final configuration of the FXC network; a design step of calculating an initial configuration of the FXC network using the final configuration; an expansion decision step of monitoring a state of the FXC network constructed by the calculation in the design step, and requesting the expansion of the FXC network when it is determined that an FXC expansion is necessary; a calculation step of calculating a network configuration after the expansion using the initial configuration or the current configuration and the final configuration when the expansion of the FXC network is requested; The design step includes: The number of intermediate FXCs in the initial configuration is made smaller than the number of intermediate FXCs in the final configuration; The number of initial fibers accommodated by the input / output FXC in the initial configuration is made smaller than the number of final fibers using the number of final fibers accommodated by each input / output FXC in the final configuration and the number of the intermediate FXCs in the final configuration. How to add.

6. An expansion program that causes a computer to function as the expansion device according to any one of claims 1 to 4.

Images