JPH09312648A - Stand-by facility designing method for switching alternate path of communication network - Google Patents

Abstract

PROBLEM TO BE SOLVED: To design a stand-by network for switching alternative paths at low cost by developing a target network by a 2nd phase according to guidelines that are provided by a 1st phase based on linear programming. SOLUTION: A sample network consisting of, for example, 7 cross-connect nodes and 13 links is used to show topology, link costs, and operation capacity allocations along the links. This network has 21 bidirectional terminal-to- terminal operation buses installed and combinations of different sources and destination nodes are connected at the lowest cost. Capacity units are so allocated to operation buses that, for example, 3 is for (1, 2), 1 for (1, 3), 2 for (1, 4), and 3 for (1, 5). Here, the distribution of the network with which minimum protection can be obtained is calculated according to the guidelines 1, 2, and 3. Then stand-by final constitutions SNFCs are enumerated and the constitution with top priority is selected.

Description

Detailed Description of the Invention

[0001]

BACKGROUND OF THE INVENTION 1. Field of the Invention The present invention relates to a backup network for a working network of a communication network using a guidance recovery protocol for selecting a network route that bypasses a fault, particularly a mesh network equipped with a cross-connect device. Regarding the design of.

[0002]

2. Description of the Related Art Various systematizations of design problems related to self-healing networks have been proposed and are realized by either heuristics or linear programming. Those methods do not seem to be suitable for large networks. Because the solution of the heuristic is far from the optimal solution,
On the other hand, linear programming requires a huge amount of calculation time.

Due to the design of recoverable networks,
Several reserve capacity allocation methods have been proposed as suitable for either link recovery or path recovery.

Many of these conventional methods are based on links.
It was developed in order to realize a reliable recovery method. Sakauchi et al. Developed an iterative linear programming method that minimizes the total reserve capacity ("A Self-Healing Ne.
works with an Economical
Spare Channel Assignment
(Self-healing network using economical backup channel allocation) "Proc. See GLOVECOM'90). A modification of this method has been proposed by Herzberg et al. ("An Optimal Spare-Capaci", taking into account the trade-off relationship between total reserve capacity and hop limit).
ty Assignment Model for S
urvival Networks with Hop
Limits (optimum reserve capacity allocation model for recoverable networks using hop limit) "Pro
c. GLOBECOM'94) Grover et al.
We first found a feasible solution, and then devised a way to reduce total reserve capacity while maintaining estimated recovery levels (see "Near Optimal Spare Capa").
city Planning in a Mesh R
Estorable Network (refer to Proc. GLOVECOM'91, a method for designing a near-optimal reserve capacity in a mesh type recoverable network). These methods achieve economical spare capacity allocation, but as the size of the network increases, the complexity increases and the calculation time rapidly increases.

The problem of design becomes even more complicated when one attempts to assume end-to-end replacement of faulty equipment in order to design a backup equipment for a path-based recovery method. This is mainly because, while benefiting from the degree of freedom of the cross-connect network, the sub-network of the entire network must be considered and optimized.
This is mainly because a large number of nodes are usually involved in the recovery, and the backup equipment (backup path) includes a large number of hops.

In order to support path recovery of cross-connect networks based on Asynchronous Transfer Mode (ATM) Virtual Path (VP), Kawamura et al. Have proposed a heuristic method for calculating a so-called backup VP (see Self-Healing ATM N
etwork Technologies Utilize
ng Virtual Paths (self-healing ATM network technology using virtual paths) "JSAC '
94). Another method has been proposed by Slominski et al. (“Planning Surva”).
bilityin ATM Networks (ATM
Design of resilience in network) "Proc. NE
See TWORKS '94). However, this method is mainly suitable for networks with equal cost of links and carrying a single service rather than multiple services.

[0007]

In view of the above, it is an object of the present invention to design a backup network that meets the resiliency requirements at the lowest cost possible.

A further object of the invention is to minimize network design execution time.

[0009]

SUMMARY OF THE INVENTION In order to achieve these objectives, the present invention contemplates the design of backup equipment in two phases. The first phase, which is based on linear programming, provides some guidelines to facilitate the design of the second phase. The second phase uses a heuristic that evolves the target network step by step.

That is, the present invention combines a new heuristic and a linear programming method to reduce the total cost of a backup network and a relatively short calculation time of a tool [tool] for a multi-service carrier network. It is possible to provide a network design tool capable of calculating the prepared backup equipment, especially the end-to-end path. The output of the tool includes the path configuration and capacity placement in the spare network. This tool is mainly used by M. M. Slominski
Guided recovery protocol ("Guide
d restoration of ATM cross
s-connect Networks (guided recovery of ATM cross-connect network) Proc. IEEE
It is realized in a recoverable network using ICC'94).

The design method according to the invention is to replace the entire faulty facility with a spare facility on an end-to-end basis.
The main purpose of the method is to pre-design the reserve capacity assigned to different reserves activated in different network failure patterns.

Calculating the minimum redundancy for recovering a meshed cross-connect network requires testing a huge number of candidate configurations. Linear programming derives optimal solutions, but its complexity and long computation time limit the number of practical applications. Here, we propose a design tool that keeps the number of network configuration candidates relatively small and does not need iterative design of the target network.

The design consists of two phases.
In the first phase, for each working path, an initial path that minimizes the protection cost is calculated, a set of backup network initial configurations is created from these initial paths, and the backup network final configuration (that is, the target network) is created. ) For any of the above, calculate the partial capacity allocation. In the second phase, the final path is designed to minimize the total cost of the target network. This is done by the gradual evolution of the target network. At each stage of development, first the preliminary network initial configuration is selected, then its components, the so-called initial paths, are ordered and rearranged one by one to meet the objectives characterized for the target network.

[0014]

DESCRIPTION OF THE PREFERRED EMBODIMENTS Next, the present invention will be described with reference to the drawings.

FIG. 1 shows a backup equipment designing method for switching a bypass route of a communication network according to claim 1 of the present invention, and a backup system for switching a bypass route of a communication network according to claim 2 of the present invention. 2 shows a facility design method, and FIG. 3 shows a preliminary facility design method for switching a bypass route of a communication network according to claim 3 of the present invention. FIG. 4 shows a method of designing a backup facility for switching a bypass route.

Next, a design method according to an embodiment of the present invention will be described with reference to FIG.

Definitions and Notations A ^(m) = {a _j ^(m) ; j = 1, ..., m _j }, m =
1, ..., M: A backup network initial configuration (SNIC) including an initial path activated in the mth failure pattern. a _j
^(M) must satisfy the following conditions.

(I) It has neither a common link with the protected service path nor a common transit node.

(Ii) its cost is less than the cost of any alternative satisfying condition (i). X ^(m) = {x _i ^(m) ; i = 1, ..., I}, m = 1,
..., M: capacitance arrangement in the m-th SNIC. x _i
^(M) is the total bandwidth of the initial paths sharing the i-th link.

B ^(m) = {b _j ^(m) ; j = 1, ..., m
_j }, m = 1, ..., M: backup network final configuration (SNF) consisting of final paths activated in the m-th failure pattern
C).

Y ^(m) = {y _i ^(m) ; i = 1, ...,
I}, m = 1, ..., M: Capacity allocation in the m-th SNFC. y _i ^(m) is the total bandwidth of the last path sharing the i th link.

Y (0) = {y _i (0); i = 1, ...,
I}: Partial capacity allocation in SNFC. y _i (0)
Indicates the minimum expected spare capacity for the i-th link.

Y (k) = {y _i (k); i = 1, ...,
I}, k = 1, ..., K ≦ M: Capacity allocation condition determined in the kth development stage of SNFC. y _i (k) is i
It shows the required capacity for the second link.

Y '(k) = {y' _i (k); i = 1,
..., I}: Supplemental capacity allocation conditions used in the current development stage of SNFC. y ′ _i (k) is max {y
_i (0), y _i (k)}.

Y = {y _i ; i = 1, ..., I}: Final capacity placement in PNB. y _i = y _i (K) for any of i = 1, ..., I.

(C ₁ , ..., C _i , ..., C _I ): A set of link cost parameters. c _i is the _i of all networks
It represents the cost of the second link.

Design Method Referring to FIG. 5, the design process consists of two phases. The first phase provides three guidelines for designing the target network. The second phase goes through K successive development stages of the target network. The value of K does not exceed the number M of fault patterns, and in most cases K <M.

Providing Guidelines Guideline 1 For each protection (service) path, an alternative is calculated that satisfies a number of requirements:

(I) The start node and end node are the same as those of the protection path, (ii) No common link with the protection path, and no common transit node, (iii) Condition (i)
Minimum cost of all alternatives that satisfy both (ii) and (ii). These are locally predetermined. Any of the calculated alternatives is called the initial pass.

Guideline 2 For each failure pattern, a subset of initial paths is created that collects only those that are activated in the pattern.
This is called a spare network initial configuration (SNIC) and is represented by the following equation.

A ^(m) = {a _j ^(m) ; j = 1, ..., m
_j }, m = 1, ..., M Here, a _j ^(m) is an initial path to be activated in the m-th failure pattern. The capacity arrangement state for each SNIC is expressed by the following equation.

X ^(m) = {x _i ^(m) ; i = 1, ...,
I}, m = 1, ..., M where x ^{i (m)} is the total bandwidth of the paths belonging to A ^(m) sharing the i th link (in the m th failure pattern).

Guideline 3 This guideline relates to optimal capacity placement in the target network under conditions that take into account only one-hop (incomplete) final path connections each departing from the source node.

This is called partial capacity allocation in SNFC (final network final configuration), and Y (0) = {y _i (0); i =
The result of this design represented by 1, ..., I} is
It provides information about the minimum reserve capacity expected to be allocated along each link of the entire network.

SNIC Trial The purpose of this design is to k-th (k =
1, ..., K ≦ M) One S to be used in the development stage
It is to select the NIC. To achieve the purpose,
Multiple SNICs are ranked and the highest priority is selected. The procedure is as follows (see FIG. 2 corresponding to claim 2). Search for SNICs containing only one initial path (others are final paths or none) and if this condition is met for only one SNIC, then initiate the initial path relocation procedure. (Described below). Otherwise, continue this work, including if there is no SNIC that meets this condition. These SNIC
A ^(u) that minimizes the function represented by the following Expression 9 is selected from among the above.

[0036]

[Equation 9] Here, when x _i ^(u) > y _i (k−1), d _i = x
_i ^(u) -y _i (k-1), otherwise d _i =
It becomes 0.

Alternatively, A ^(u) that minimizes the function represented by the following formula 10 is selected.

[0038]

(Equation 10) Here, when x _i ^(u) > y ′ _i (k−1), d ′ _i = x
_i ^(u) -y ' _i (k-1), otherwise d'
_i = 0.

Alternatively, A ^(u) that minimizes the function represented by the following equation 11 is selected.

[0040]

[Equation 11] Here, c _i represents the cost of the i-th link.

Alternatively, A ^(u) that minimizes the function represented by the following equation 12 is selected.

[0042]

(Equation 12) The output data of this procedure is the identification of the current highest priority SNIC.

Tries the path of the selected SNIC One S
Once the NIC is selected, the next goal is to try all the initial paths that belong to the SNIC and select the one with the highest priority. After relocating this path, this procedure continues until all initial paths have been relocated. The procedure for trying the SNIC path, that is, A ^(u) , is shown below.

1) Set the capacity allocation state for all final paths (already rearranged ⁾ belonging to A ^(u) .

Q = (q ₁ , ..., Q _i , ..., q _I ) (5) where q _i represents the total bandwidth of the final path that included the i-th link in the route selection.

2) Equation 13 below

[0047]

(Equation 13) For each initial path represented by, the condition (5) on the bandwidth required by the initial path is updated and the function represented by Equation 14 below is evaluated.

[0048]

[Equation 14]Where c (•) is a_j ^(U)Is the implementation cost of p
(•) is a_j ^(U)Is the number of SNICs included.
q_i> Y_iWhen (k-1) d_i= Q_i-Y_i (K-
1) otherwise d_i= 0.

The implementation cost of a _j ^(u) is equal to the initial path a _j
^It is given by the total cost of the links included in the route ^(u) .

Next, the initial path is selected so that the function (6) takes the minimum value. If this function takes the minimum value for at least two initial paths, then
Select the function with the smallest value.

[0051]

(Equation 15) Here, if q _i > y ′ _i (k−1), d ′ _i = q
_i −y ′ _i (k−1), otherwise d _i = 0.

The initial path selected by a _{v (u)} is indicated.

3) The initial path a _v ^(u) is ^replaced with the final path b _v
^The following initial path relocation procedure is used to relocate to ^(u) .

If A _v ^(u) satisfies the following condition for any i (i = 1, ..., I), the initial path a _v ^{(u) is changed} to the final path b _v ^(u). , And proceed to step 4).

Q _i −y _i (k−1) ≦ 0 (8) If condition (8) is satisfied for at least two alternatives, choose the one with lower cost (each cost is The maximum value, ie the estimated value, cannot be exceeded). Then use this alternative as b _v ^(u ) and proceed to step 4).

If a _{v (u)} satisfies the following condition for any i (i = 1, ..., I), then a _v
^{Use (u)} as final path b _v ^(u) , step 4)
Proceed to.

Q _i −y ′ _i (k−1) ≦ 0 (9) If the condition (9) is satisfied by two or more alternatives, the one with the least cost is selected. Then use this alternative as b _v ^(u ) and proceed to step 4).

If the considered initial path cannot be treated as a multiple path, the relocation procedure for it is continued. Otherwise, prioritize the paths according to their bandwidth and use each as a new initial path. Then, for any of these paths, the relocation procedure is performed from the beginning, starting with the one with the maximum bandwidth.

Consider a _v ^(u) and its substitutes as candidates for b _v ^(u) , and select a candidate from which the function represented by the following formula 16 can be minimized.

[0060]

(Equation 16) Here, c (b _v ^(u) ) is the cost of implementing the final pass, and when q _i ≧ y (k−1), d _i = q _i −y _i
(K-1), otherwise d _i = 0.

Then, this path candidate is used as the final path b _v
^Accept as ^(u) and proceed to step 4). The rule given by (10) is called the rearrangement first criterion.

Consider a _v ^(u) and its substitutes as candidates for b _v ^(u) , and select a candidate that minimizes the function represented by the following formula (17).

[0063]

[Equation 17] Here, if q _i ≧ y ′ (k−1), d ′ _i = q _i −
y ′ _i (k−1), otherwise d _i = 0. Then, the candidate path is accepted as the final path b _v ^(u) . The rule given by (11) is called the rearrangement second criterion.

4) Update the capacitance placement state 5) based on the results obtained for b _v ^(u) , and step 2) to step 4 until all initial paths of this SNIC have been converted to final paths. ).

5) The capacity arrangement state is updated based on the current design result expressed by the following equation.

Y _i (k) = max {y _i (k−1), q _i }} (12) y ′ _i (k) = max {y ′ _i (k−1), q _i }} (13 ) Then, for any of the other SNICs (those that remain for relocation), check the state given in (1).

The parameter d for all i = 1, ..., I for any of the remaining sub-networks.
_{If i} is equal to 0: All initial paths of such SNIC are accepted as final paths, capacity allocation state Y (k) is regarded as the final capacity allocation state of the target network, and the design procedure is stopped.

Otherwise, carry out the procedure of claim 2 to continue designing the target network.

Concrete Example Data Input Main design steps will be described using a sample network consisting of 7 cross-connect nodes and 13 links. FIG. 6 shows the topology, the link cost, and the operating capacity allocation along the link.

This network has 21 bidirectional end-to-end working paths, each connecting a different source and destination node combination at minimum cost. The capacity units assigned to each operation path are as follows.

3 for (1, 2), 1 for (1, 3), 2 for (1, 4), 3 for (1, 5),
5 for (1,6), 1, for (1,7), (2,
3), 4 for (2, 4), 3 for (2, 5), 1 for (2, 6), 1 for (2, 7), (3, 2 for 4), 2 for (3, 5),
2 for (3, 6), 5 for (3, 7), (4,
5), 3 for (4, 6), 4 for (4, 7), 4 for (5, 6), 4 for (5, 7), (6, 5 for 7).

Providing Guideline 1 It was assumed that for any of the 21 protected paths, the initial path (and also the final path) has no common links to the working path and no common transit nodes. Next, the initial path was calculated as the one with the minimum cost.

Providing Guideline 2 After assuming all single link failure patterns, 13 SNIs for each one pattern using the initial path.
C was created. m-th (m = 1, ..., 13) SNIC
For, the route selection is A ^(m) = {a _j ^(m) ; j =
, ..., m _j } and the capacitive placement is X ^(m)
= {X _i ^(m) ; i = 1, ..., I}.
Where x _i ^(m) is A sharing the i th link
It is the sum of the capacity units of the initial path belonging to ^(m) .
The result is shown in the matrix of FIG. 7 containing the elements corresponding to x _i ^(m) .

Guideline 3 provides a reserve capacity unit and distribution over the network to obtain the minimum protection cost, assuming that all of the final paths provided are limited to that first link. To be done. In order to explain this design step, the failure of L1 that affects two working paths will be taken as an example. That is,
(1,2) uses L1 and (1,5) uses L1 and L2
Is used (see FIG. 6). Therefore, node A
Can distribute 3 capacity units to L2, L3 and L4, and node B can distribute 6 capacity units to L5 and L6. Here, the capacity unit allocated along the i-th link is denoted by y _i (0). These states are represented by the following equations.

Y (0) ₂ + y (0) ₃ + y (0) ₄ ≧ 3 (14) y (0) ₅ + y (0) ₆ ≧ 6 (15) All source nodes in all failure patterns are considered. After that, we solved the system of inequalities with the solutions that should minimize the following functions.

F (y (0) ₁ , ..., y (0) _i , ..., y (0) _I ) = c ₁ y (0) ₁ + ... + c _i y (0) _i + (16) where c ₁ , ..., C _i , ..., C _I are the link costs given in FIG. The solution obtained for the sample network is the vector Y (0) = in FIG.
It is shown as {y _i (0); i = 1, ..., I}.

Trial SNIC To illustrate the procedure of trying SNIC and selecting the highest priority, the fourth evolution stage (k) to the target network.
= 4) is taken as an example. In the three previous stages, A ⁽¹³⁾ , A ⁽²⁾ , and A ⁽⁵⁾ are B
⁽¹³⁾ , B ⁽²⁾ , and B ⁽⁵⁾ are rearranged, and at this time, as shown in FIG. 8, Y ⁽¹³⁾ ,
It is characterized by Y ⁽²⁾ and Y ⁽⁵⁾ . Here, the capacitance arrangements X ⁽¹⁾ , X ⁽⁷⁾ , and X shown in FIG.
Note that ⁽¹²⁾ is different than that shown in FIG. This is SNIC A
^{This is because (13)} , A ⁽²⁾ , and A ⁽⁵⁾ are updated based on the result of the previous stage, and at this time, each of the SNICs includes both the initial path and the final path.

According to this procedure, the parameter (1) is first calculated for each SNIC. Since the minimum value -7 is obtained for both the ^{X (3)} and ^{X (12),} attempts to distinguish between ^{X (3)} and ^{X (12)} to calculate the parameters (2). This attempt was successful and parameter (2)
Has a value of -7 for X ^{(3) and} a value of -2 for X ⁽¹²⁾ . Therefore, A ⁽³⁾ and X
^The SNIC given by ⁽³⁾ was ranked first.

Trial Paths of Selected SNIC The next task is to try all paths of A ⁽³⁾ and select the highest priority one. The SNIC includes a path _a ¹ to share the link L2 and L10 ^(3), has two initial paths pass _a ² to share the link L2 and L12 ^(3), 9 about them, _{x 1} ⁽³⁾ and x
₂ ⁽³⁾ is shown. For these initial paths,
Parameter (6) was calculated and the results are as shown in Equation 18 below and Equation 19 below.

[0080]

(Equation 18)

[0081]

[Equation 19] Here, the denominator is the product of the cost of the shared link, the required capacity, and the number of SNICs that contain the initial path.

That is, a ₁ ⁽³⁾ is ranked first. Since the condition (8) is satisfied for this, b
₁ ⁽³⁾ = a ₁ ⁽³⁾ .

Next, consider a ₂ ⁽³⁾ . Regarding this, both the conditions (8) and (9) are not satisfied, and thus the rearrangement of the initial path is necessary. The final path is assumed to be predesignable as a path whose components have been able to allocate at least one capacity unit. In this case, after applying functions (10) and (11), the final structure of b ₂ ⁽³⁾ consists of the following paths (see FIGS. 9 and 6). That is, 3 assigned
L4-L9-L8 having one capacity unit and L2-L1 each having one allocated capacity unit
2 and L2-L10-L7-L8. As a result, Y
^{(4) = Y (3)} and ^Y'(4) = a ^Y'(3).
This is the purpose of this development stage to the target network,
This means that the protection costs determined in the previous stage have been met.

This step is accomplished by updating all of the remaining SNICs for consideration based on current results. More specifically, the currently calculated final path also becomes the final path in some other SNIC, and the capacity placement of such SNIC must be updated. Then all SNICs are retried and the highest priority is selected and relocated in the next evolution to the target network.

[0085]

As described above, according to the present invention, it is possible to design a backup network that satisfies the recovery requirement at the lowest possible cost.

Further, according to the present invention, the network design time can be minimized.

[Brief description of drawings]

FIG. 1 is a diagram showing a method of designing a backup facility for switching a bypass route of a communication network according to claim 1 of the present invention.

FIG. 2 is a diagram showing a method of designing a backup facility for switching a bypass route of a communication network according to claim 2 of the present invention.

FIG. 3 is a diagram showing a method of designing a backup facility for switching a bypass route of a communication network according to claim 3 of the present invention.

FIG. 4 is a diagram showing a method for designing a backup facility for switching a bypass route of a communication network according to claim 4 of the present invention.

FIG. 5 is a diagram showing a design method according to an embodiment of the present invention.

FIG. 6 is a diagram used for explaining the design method of FIG.

FIG. 7 is another diagram used to describe the design method of FIG.

FIG. 8 is yet another diagram used to describe the design method of FIG.

9 is another diagram used to describe the design method of FIG.

Claims (4)

[Claims] 1. A method for designing a backup equipment for switching a detour path of a communication network, in particular, a broadband communication network using a cross-connect node technology, wherein the method comprises an entire network topology and a link cost (c).
₁ , ..., C _i , ..., C _I ) and a set of initial specifications,
Specification of protection path and set of initial configuration of protection network {A ^(m) ;
m = 1, ..., M} and the capacitance arrangement X in each of them
^(M) = {x _i ^(m) ; i = 1, ..., I} and spare capacity allocation Y (0) = {y _i (0 ); I =
1, ..., I} through a continuous rearrangement of the preliminary network initial configuration such that the overall protection cost is minimized,
Final configuration of backup network {B ⁽¹⁾ , B ⁽²⁾ , ..., B ^(K) }
(K ≦ M), and the backup equipment design method comprises: a backup network initial configuration trial step of trying a number of backup network initial configurations; and a number of trials of a number of initial paths of the backup network initial configuration. Initial path trial step, and an initial path rearrangement step of rearranging the initial path of the backup network initial configuration to the final path of the backup network final configuration of bypass path switching of the communication network. For designing spare equipment for 2. The standby equipment design method for switching a bypass route of a communication network according to claim 1, wherein the standby network initial configuration trial step comprises only one non-relocation initial path from the standby network initial configuration. , And if only one such preliminary network initial configuration is shown, then automatically making that configuration top-ranked, and such A preliminary network initial configuration is not indicated, trying them according to the trial procedure of the preliminary network initial configuration to identify the highest ranked one, used in the kth evolution stage to the target network. procedure of trial the preliminary network initial ^{_{configuration, X (u) = (x}} 1 (u), ..., x I (u)) and capacity requirements, Y (k-1) = (y (k-1) ₁ , ..., y (k-
1) _I ) is the capacity calculated in the previous development stage to the target network, x _i ^(u) > y _i (k−1)
, D _i = x _i −y _i (k−1), otherwise d
_{When i} = 0, for any of the u-th preliminary network initial configurations, the following equation 1 The step of evaluating the function represented by, and setting the function having the minimum value as the initial configuration of the backup network with the highest priority, y ′ (k−1) for any of i = 1, ... ) _I =
max {y (k-1) _i , y (0) _i }, x _i ^(u) >
When y ′ _i (k−1), d ′ _i = x _i ^(u) −y ′ _i
(K-1), or else if d _i = 0, then for each of the preliminary network initial configurations, A step of evaluating the function represented by the above, and setting the function having the minimum value as the initial configuration of the backup network with the highest priority, c _i is the link cost of the i-th time, and x _i ^(u) > y _i (k−
In the case of 1), if d _i = x _i ^(u) −y _i (k−1), and if d _i = 0 otherwise, for any of the preliminary network initial configurations, the following equation 3 Evaluating the function represented by the above, and setting the function having the minimum value as the initial configuration of the backup network with the highest priority, c _i is the i-th link cost, x _i ^(u) > y ′ _i (k
−1), d ′ _i = x _i ^(u) −y ′ _i (k−
1) otherwise, if d _i = 0, then for any of the preliminary network initial configurations, the following Equation 4 And a step of evaluating a function represented by the above, and setting the function having the minimum value as an initial configuration of a backup network of the highest priority, a method for designing a backup equipment for switching a bypass route of a communication network. 3. The backup equipment design method for switching a bypass route of a communication network according to claim 2, wherein the initial path trial step includes a backup network initial configuration selected according to a trial procedure of the backup network initial configuration. This is a step of trying an initial path, wherein the initial path trial step calculates capacity allocation states for all the relocated paths of the selected backup network initial configuration A (u), and then the other Determining, for each initial path, Q (k) = (q ₁ , ..., q _i , ..., qI), which is the updated capacity placement state based on the capacity requirements for said initial path; , Y (k−1) determines the total capacity state for all the protection network final configurations, c (·) is the implementation cost of a _j ^(u) , and p (·) is a _j ^(u). the preliminary network initial structure ^that) contains The number _of, q _i> when y i of _{(k-1), d i} = q
_i −y _i (k−1), and if d _i = 0 otherwise, for any of the initial paths a (u) _j ,
Equation 5 below Evaluating the function represented by, and making the function having the minimum value the initial path having the highest priority, Y '(k-1) = max {Y (k-1), Y (0)} ,
When q _i > y ′ _i (k−1), d _i = q _i −y ′ _i (k
−1), otherwise assuming d _i = 0, for each of the initial paths a (u) _j , And a step of evaluating a function represented by the expression (1) and setting the function having the minimum value as an initial path having the highest priority, a method for designing a backup facility for switching a detour path of a communication network. 4. The method for designing a backup facility for switching a bypass route of a communication network according to claim 3, wherein the initial path rearrangement step rearranges an initial path selected according to the initial path trial step, The step of accepting as the final path, and the step of rearranging the initial path includes:
_{i −} y _i (k−1)) becomes zero, the step of accepting the initial path and the parameter (q _i −y _i (for i = 1, ..., I, k-1)) accepting the least-cost alternative among those that are zero, and for any of i = 1, ..., I, the parameter (q
_i− y ′ (k−1)) becomes zero, the step of accepting the initial path and the parameter (q _i −y ′ _i) for any of i = 1, ..., I. (K-1)) accepting the lowest cost alternatives out of zero, and considering the initial paths as a number of initial paths, prioritizing according to their bandwidth, and each as a new initial path. Using either the initial path relocation procedure starting with the one with the maximum bandwidth, or a capacity reduction request satisfying at least one of the conditions of the four accepting steps. And the steps to consider and let ca be the cost of implementing the alternative, And the cost of implementing the alternative, where ca is the cost of implementing the alternative, then the following equation 8 And a step of accepting an alternative such that the second criterion obtained according to (1) is minimized.