CN100499534C - Implementing method for optimizing multiple services in optical synchronization data transmission network - Google Patents

Abstract

The method includes following steps: (A) dividing network into annular sub networks in advance; (B) setting up service need table, assigning routes and defining information of route according to principle of minimum total quantity of network resources utilized by services to be created in table, and load balancing; (C) based on information of route determined in step (B), assigning resources for services to be created. Determining whether resources in whole network meet targets: index of degree of adaptability for new added service, and minimum low order intercross resources; if yes, the meted is ended, otherwise returning back to (B).

Description

A kind of method that realizes multi-business optimization of photo synchronous digital transmission network

Technical field

The present invention relates to the network multi-service optimization problem, relate in particular to a kind of Optical synchronization digital (SDH) of realizing and transmit the method for net multi-business optimization.

Background technology

For the optical transport network of an operation, how fully, effectively utilize Internet resources, just how to carry out the optimization of business configuration to admit more business, all be a very crucial problem all the time.This optimization problem is the problem of a non-multinomial complete (NP-Complete), can be described by integral linear programming (ILP), as everyone knows, under the bigger situation of network size, the globally optimal solution of finding the solution such optimization problem almost is impossible.Transmit in the net specific to Optical synchronization digital (SDH), can be with its optimization problem qualitative description: under the situation of known physical network and traffic matrix, reasonable distribution resource how takies to reach that the Internet resources total amount is minimum, NLB, makes the surplus resources in the network can satisfy in the future more traffic growth and use four minimum optimization aim of low order interlace algorithm resource.But owing to transmit some complicated constraintss of existence in the net at SDH, such as the restriction of node crossing aspect, the link circuit resource asymmetry that special mapping structure causes etc. will cause problem more complicated.

Also there is not at present to transmit at SDH specially the research document of multi-business optimization problem in the net, immediate with it have only multi-business optimization way to solve the problem in dense wave division multipurpose (DWDM) optical-fiber network, i.e. static routing and Wavelength Assignment (RWA:Routing ﹠amp; Wavelength Assignment) way to solve the problem, in the method, owing to consider the complexity of optimization problem, be two subproblems usually with PROBLEM DECOMPOSITION, be respectively static routing assignment problem and assignment of wavelength, these two problems also are the NP difficulty separately.Wherein, the route assignment problem generally adopts heuritic approach, its thinking is to obtain initial solution by certain routing policy, again according to the resource occupation state of the whole network, partial service is carried out heavy-route calculate, reach desired value up to index, perhaps numerical convergence, perhaps calculate certain number of times and stop, provide more excellent separating then.And assignment of wavelength can be converted into a map colouring problem, and the existing ripe standard drawing colouring algorithm of this type of problem is found the solution, but considers efficiency, usually also according to actual conditions, adopts heuritic approach to obtain more excellent separating.

In general, the DWDM network belongs to general mesh network (Mesh net), this web frame is simple, inside has only interconnecting between the node, and do not have other syndetons such as ring, owing to be applied to heuritic approach on the DWDM network, only consider the simple connection between the node based on this Mesh net, and do not consider concrete network topology characteristic, so it is very big to optimize the uncertainty of effect.In addition, the resource allocation in the DWDM network is base unit with the wavelength, and the resource allocation that SDH transmits in the net is a base unit with the time slot then, and has specific mapping structure, is not suitable for SDH transmission net so be applied to the method for DWDM network.In addition, at optimization problem, also have methods such as genetic algorithm or neural net model establishing, the thinking of these methods is problem not to be split, and directly finds the solution.But, it is less that method such as genetic algorithm or neural net model establishing generally is only applicable to network size, the simple situation of constraints, increase along with network size, especially the interstitial content in network is greater than in 50, these methods need spend the long duration and find the solution, and can not satisfy the time efficiency requirement of operation network probably, and in most of the cases possibly can't obtain optimal solution.

Summary of the invention

For addressing the above problem, the inventor had once proposed the method that a kind of SDH of realization transmits route assignment in the net multi-business optimization in another patent application, the realization thought of this method is: transmit the characteristics that net is made of ring according to SDH, be divided into different subnets with transmitting net by ring, and the load balance adjustment of the whole network is decomposed into adjusts in each annular subnet and the whole adjustment of the whole network, thereby make the adjustment of the whole network simpler, efficiently, factor by further analyzing influence load balance again, can't carry out being in network edge and separating of balance, guarantee to take the balance of the minimum and offered load of offered load with the subnet that network has only a link to link to each other.

But, this method has only realized that preceding two optimization aim promptly take the minimum and NLB of Internet resources total amount in the SDH transmission net multi-business optimization problem, and for latter two optimization aim, that is: make surplus resources in the network can satisfy in the future more traffic growth and use that the low order interlace algorithm resource is minimum but can't to be realized at all.

In view of this, main purpose of the present invention provides a kind of Optical synchronization digital (SDH) of realizing and transmits the method for net multi-business optimization, make it when taking into account optimization efficiency and performance balance, assurance takies that the Internet resources total amount is minimum, NLB, make the surplus resources in the network can satisfy in the future more traffic growth and use the low order interlace algorithm resource minimum, and then reaches the best service optimization effect.

In order to achieve the above object, the invention provides the method that a kind of SDH of realization transmits the net multi-business optimization, its key is that this method may further comprise the steps:

A. in advance the Optical synchronization digital transmission network network is divided into annular subnet, and the business demand table is set according to current network resource;

B. calculate initial route to be created each in the demand schedule is professional,, determine routing iinformation by the initial route of principle optimization that takies the minimum and load balance of Internet resources total amount;

C. according to the determined routing iinformation of step B, by the Virtual Channel high-order is that unit gives the traffic assignments time interval resource, check that Resources allocation satisfies Added Business fitness index and business takies the low order interlace algorithm resource the most after a little while, the ending resource assigning process, otherwise the business of regular similar route is that unit gives the traffic assignments time interval resource by the Virtual Channel low order in each subnet;

D. it is minimum to judge whether resource in the whole network satisfies the Added Business fitness index and the low order interlace algorithm resource occupation of network, if satisfy index and take the low order interlace algorithm resource minimum, then finishes, otherwise returns step B.Wherein, the Added Business fitness index of setting network is the shared node number of professional route and the sum of products of professional shared low order time interval resource number in each subnet.

The initial route of the described calculating of above-mentioned steps B is: call shortest path first and calculate initial route;

The described optimization route of step B may further comprise the steps:

B1. respectively to link circuit resource between annular subnet and in the annular subnet take situation inspection, to exceed standard all business on the link of resources occupation rate, recomputate route;

B2. check the load balance situation of annular subnet, business in the annular subnet that does not satisfy the load balance index requirement is recomputated route, check the situation that takies of link circuit resource in the annular subnet then;

B3. the Optical synchronization digital transmission network network is divided into again core sub and edge subnet, whether the overall target of judging core sub in the whole network then satisfies the condition of convergence, as satisfying, then finish route assignment,, check then whether the route calculation number of times reaches thresholding as not satisfying, if reach thresholding, then finish route assignment, otherwise, carry out step B1.

The business of regular similar route described in the above-mentioned steps C is: with the pairing business of different low order time slots between any two nodes in the subnet, merge in the same high-order time slot by high-order time slot rank;

It after the described regular similar route of step C, in each subnet, is further to comprise before unit gives the traffic assignments time interval resource: pairing business of above high-order time slot in the coupling subnet by the Virtual Channel low order.

By such scheme as can be seen, key of the present invention is: according to the characteristics of SDH network, network is divided into annular subnet, on the basis that route assignment realizes, carry out relatively independent resource allocation in each subnet, the computational efficiency of carrying out resource allocation with raising; In each subnet, business demand is mapped to different solids then, mate again, merge, promptly pressing high-order or low order refinement distributes, not only solved the problem of low order interlace algorithm resource limit preferably, and reflected to have under the time slot continuity constraint condition well, how to measure the good and bad situation of resource allocation result, thereby finish optimization aim.

Therefore, the method for realization SDH transmission net multi-business optimization provided by the present invention has following characteristics and advantage:

1) the present invention is according to the characteristics of SDH network, network is divided into annular subnet,, has improved the computational efficiency of when business is larger, carrying out resource allocation by in each subnet, carrying out relatively independent resource allocation, and in a short period of time, can provide goodish approximate optimal solution.

2) the present invention is mapped to business demand on the different solids, in whole resource allocation process, by reasonable accumulation and combination to solid, and, solved the problem of low order interlace algorithm resource limit preferably to the analysis of professional route similitude, at last by introducing " Added Business fitness " index, reflected well and had under the time slot continuity constraint condition, how to measure the good and bad situation of resource allocation result, thereby finish optimization aim

3) the present invention is applied to the network of middle and small scale and it is tested after, the result that can in 30 seconds, be optimized, this explanation the present invention can be advantageously applied in the engineering.

Description of drawings

The overview flow chart that Fig. 1 realizes for the present invention;

Fig. 2 is the realization flow schematic diagram of route assignment;

Fig. 3 is the schematic diagram that core sub and edge subnet are divided;

Fig. 4 transmits the resource schematic diagram of BLSR in the net for SDH;

Fig. 5 is for realizing the flow chart of resource allocation;

The schematic diagram that Fig. 6 merges for the VC4 chest;

Fig. 7 is for using network topology structure figure of the present invention.

Embodiment

The present invention is on general heuritic approach basis, by SDH network topology form being done in-depth analysis, the optimization method targetedly that puts forward.Consider the complex nature of the problem, similar with general solution, it is two subproblems that this method is also disassembled problem, is respectively route assignment problem and resource allocation problem and finds the solution.Wherein, the route assignment stage is determined preceding two in four optimization aim, promptly guarantee to take the minimum and NLB of Internet resources total amount, the method of route assignment is identical in the realization SDH transmission net multi-business optimization that the solution in this stage and inventor propose in another patent application, and the resource allocation stage is determined in four optimization aim latter two, even the surplus resources in the network can satisfy in the future more traffic growth and use the low order interlace algorithm resource minimum.The low order interlace algorithm resource here refers to the professional shared low order time interval resource through same node point, is arranged in different high-order time interval resources.

General optical transport network is by a lot of network elements (NE) interconnected compositions, and network element (NE) can be regarded node as, and optical fiber can be seen and connect.Set up a business, the source node of given business, place (or being called purpose) node, bandwidth and other constraints are exactly determined the node of service needed process, the thing that route assignment that Here it is need be finished; Determine professional concrete which resource used in the node behind the routing node, the thing that resource allocation that Here it is need be finished.

Be example with optimization flow process of the present invention below, the implementation procedure of resource allocation of the present invention is described in conjunction with the accompanying drawings, as shown in Figure 1, the present invention may further comprise the steps at least:

Step 101:, therefore, at first will carry out the distribution of professional route on the basis of route assignment because service optimization method of the present invention is to finish.So, the detailed process of this route assignment is as shown in Figure 2:

A. at first SDH is transmitted net and be divided into different subnets by ring;

B. then according to the network resource information of input, calling service shortest path first to be created listed in the business demand table is calculated its initial route;

C. more respectively to link circuit resource between annular subnet and in the annular subnet take situation inspection, to exceed standard all business on the link of resources occupation rate, recomputate route;

D. check the load balance situation of annular subnet, business in the annular subnet that does not satisfy the load balance index requirement is recomputated route, and then check the situation that takies of link circuit resource in the annular subnet;

E. again network is divided into core sub and edge subnet, whether the overall target of judging core sub in the whole network then satisfies the condition of convergence, as satisfying, then enter step 102,, check then whether the route calculation number of times reaches thresholding as not satisfying, if reach thresholding, then enter step 102, otherwise, recomputate route.Wherein, so-called edge subnet is meant and is in network edge and with having only a link-attached subnet portion between other subnet, and core sub promptly refers to remove in the whole network the residue subnet portion of edge subnet.Give an example, as shown in Figure 3, for a network topology structure 300 in the practical application, R represents annular subnet, L represents link, by ring this net is divided into different subnet R1～R7 after, when further dividing edge subnet and core sub again, R4, R5, R6, R7, L5, L6 constitute the edge subnet, and R1, R2, R3, L1, L2, L3, L4 then constitute core sub.Described overall target obtains with the weight coefficient sum of products separately by calculating behind the routing optimality resource occupying total amount and overall load balance index.

Step 102: carry out resource allocation.According to routing iinformation is traffic assignments resource to be created, that is to say, by the service path of determining by route assignment from the source node to the destination node, concrete selection source node, destination node and each intermediate node provide the part resource in the resource to realize the distribution of resource, and wherein routing iinformation comprises the intermediate path of business required process from the process of source node arrival destination node.

Step 103: judge whether resource allocation is successful.It is minimum that successful condition is exactly that resource in the whole network satisfies the Added Business fitness index and the professional low order interlace algorithm resource that takies of network, if the resource allocation success then finishes optimizing process, recomputates route otherwise return step 101.

Further specifically introduce the implementation procedure of resource allocation below.

At first introduce the characteristics of SDH link circuit resource,, thereby caused the complexity of SDH resource allocation just because of these characteristics.

The link circuit resource of SDH is to be unit with the time slot, according to certain mapping structure tissue.In brief, resource can be divided into two main ranks: VC4 rank and VC12 rank.Article one, link may comprise a plurality of VC4 time interval resources according to its speed rank, and each VC4 passage comprises 63 VC12 time slots.The VC4 time slot generally can intersect arbitrarily at the node place, and promptly the VC4 time slot does not have continuity constraint by a node time.But this point is false on two-direction duplexing section rotating loop (BLSR), and BLSR is a kind of trail protection ring, by finishing the protection switch function for the collaborative work of byte definition Bit-Oriented Protocol, can be two fibres, can be four fibres also, and be operated in two-way.The node failure owing to need protection, so on BLSR, the VC4 time slot has continuity constraint, this point will specifically be discussed in the back.The intersection of VC12 time slot then has the restriction of two aspects, at first is the restriction of node low order interlace algorithm capacity, promptly is not that whole link circuit resources can carry out low order interlace algorithm; Be the timeslot number continuity constraint on the other hand.In the SDH of reality network, may have only part of nodes to have the time-division crossing, these nodes possess the time slot adjustment capability, no timeslot number continuity constraint, and the node of other no time-division crossing then has the timeslot number continuity constraint.

Because BLSR is the most frequently used protected mode, transmit in the net at SDH and to occupy very big ratio, institute is so that the surplus resources in the network can satisfy in the future more the optimization aim of traffic growth at BLSR.In BLSR, because the restriction of timeslot number continuity is arranged, the service efficiency of time interval resource is not only relevant with its physical capacity summation, and is also relevant with its distribution.Can regard whole time interval resources of BLSR as a resource pool, Fig. 4 shows the resource schematic diagram of BLSR, wherein, rectangle 401 expression node NE1 are the 1-20 VC12 time interval resource in the VC4-1 time slot to the link of NE2, rectangle 402 expression node NE2 are the 1-20 VC12 time interval resource in the VC4-2 time slot to the link of NE3, and rectangle 403 expression node NE1 cross NE2 the 1-20 VC12 time interval resource in the VC4-3 time slot to the link of NE3.Find out easily that from figure rectangle 401 equates with the total resources of rectangle 402 and rectangle 403, but the business that they can be supported is different.For instance, have a business to take from node NE1 through preceding 20 the VC12 time interval resources of NE2 to NE3 now, rectangle 403 obviously can be supported this business, but rectangle 401 or rectangle 402 just can not be supported, because its VC4 time slot difference on two sections links is subjected to continuity constraint, distributing can failure.From this example as can be seen, the Resource Block of continuous distribution in the horizontal direction, its width is big more, the business that can adapt to is just many more, therefore, in order to satisfy the defined optimization aim that makes surplus resources in the network can satisfy in the future more traffic growth, will under the fixed prerequisite of total resources, make the idling-resource piece continuous as far as possible.In order to illustrate how to weigh this target, introduce the notion and the computational methods of " Added Business fitness " index below.

The Added Business fitness D of a Resource Block, can be quantitative be expressed as:

$D=N\×\underset{i=1}{\overset{w}{\mathrm{\Σ}}}i$

Wherein N represents the number of VC12 in the Resource Block, and w represents the width of Resource Block.The Added Business fitness index of the whole network is exactly the summation of all Resource Block Added Business fitness.

Taking the minimum optimization aim of low order interlace algorithm resource has the low order interlace algorithm resource limit to be provided with at most of nodes of SDH network.Reach this target, business that need route is similar when the same node, is carried out regularly, for they distribute time slot in the same VC4, thereby avoids using the low order interlace algorithm resource.

SDH transmits the host and will be made of various rings, can the whole network be divided into different subnets by ring, chain, in case the professional conversion of interannular is professional in the ring of subnet for institute, then resource allocation problem just is converted into distribution time slot problem in each subnet.Before concrete discussion solution, introduce following several notion:

Any two internodal whole business were called a business unit during business unit: SDH transmitted and nets, and the number of VC12 is called the size of business unit in the business unit.

The Block piece: be used to put the rectangular vessel of or a part of business unit, it is according to the size of business unit and length customization.Block is placed in the VC4 chest, in the adjustment process to the Block piece, may split from height the Block piece.

VC4 chest: be a jumping or uncertain VC4 number long VC4 pipeline of a few jumping, be used for placing professional permanent high rectangular vessel.So-called notion of jumping, give an example: node NE1 is made as one to the distance between the node NE2 and jumps length, and then node NE1 is that double bounce is long to the distance between node NE3.Each business unit is assembled according to the VC4 rank, and the business unit of 63 VC12 of less than is handled according to the business unit that contains 63 VC12, puts among the VC4, claims other business of each VC4 level to be the VC4 chest.The purpose of introducing it is: in the relatively more idle network of business, adjust the professional logical business of just can joining according to the VC4 chest, but also can reserve the leeway of dilatation for many similar business.

Subnet container a: subnet is regarded as a container, be called the subnet container.The business configuration process can the image be described as a VC4 chest or the Block piece is stacked in the subnet container.The subnet container is layering, and every layer has a VC4 numbering, a VC4 resource of corresponding physical subnets.The number of plies of subnet container is defined as the height of subnet container, is in course of adjustment, and highly be arbitrarily.

Fig. 5 shows the flow chart of realizing resource allocation.Specify the implementation procedure of resource allocation below in conjunction with each step.

Step 501: in each subnet, be that unit distributes time interval resource to business unit by VC4.

Business unit in the subnet is assembled into VC4 chest one by one earlier, and these chests are equal to the solid of rectangle one by one.Optimization allocation to business just can be converted into these solids, i.e. the arrangement problems of VC4 chest reaches following purpose thereby just rationally pile up, make up these solids: the total height that (1) piles up solid is minimum; (2) the remaining space future that can admit, potential solid was maximum.The function of this step is exactly to finish the arrangement of VC4 chest according to above-mentioned optimization aim.If be allocated successfully, because each business unit is all monopolized a VC4, just do not need low order interlace algorithm, satisfy the optimization aim that the surplus resources that makes in the network can satisfy more business growth in the future naturally.

Step 502: judge the resource distributed whether satisfy whole network Added Business fitness index will whether minimum with business if taking the low order interlace algorithm resource, if resource satisfies index request and the low order interlace algorithm resource occupation is minimum, then direct ending resource assigning process, otherwise the business of execution in step 503 regular similar routes.

Step 503: the business of regular similar route.

Because when business unit is put the VC4 chest into, not necessarily be full of this chest, can stay some spaces.If Internet resources do not satisfy, just a plurality of business unit need be put into a chest.But, owing to will reduce taking of low order interlace algorithm, need similar business unit to be put in the same VC4 chest as far as possible, just carry out the regular of business.This step compares by the similitude to the different business route, finishes regular operation for the traffic assignments that has similar route through same network node the low order time interval resource in the same high-order time slot.

Step 504: in the regular business of the inner combination of subnet.

Regular by previous step makes that in a subnet the actual shape that takies resource of part of V C4 cabinet interior no longer is simple rectangle, and may be the zigzag that is formed by the overlapping back of a plurality of rectangles.The function of this step is exactly as far as possible some to be satisfied the VC4 chest of specified conditions to coupling, enables to merge into a chest, thereby reduces taking of VC4 time slot.Fig. 6 shows the schematic diagram that the VC4 chest merges.As can be seen from Figure, two VC4 chests 601,602 all are zigzag up and down, and just in time can merge into a VC4 chest.

Step 505: in each subnet, be that unit gives the traffic assignments time interval resource by VC12.

Because a Block piece comprises several VC12, therefore, can will be that unit is converted into solid for the traffic assignments time interval resource with VC12, i.e. the accumulation of Block piece and combination.Its concrete distribution thought is: make regular Block piece relative fixed, and Block piece that will be not regular is piled up arbitrarily on the vertical direction of subnet container.The optimization aim of this step is that unit give business unit Resources allocation by VC4 with step 501 in each subnet, also is feasible as far as possible: the total height that (1) piles up solid is minimum; (2) the remaining space future that can admit, potential solid was maximum, just the solid has here become the Block piece by the VC4 chest, though be rectangle equally, but the height of Block piece is inequality, and can in height decompose, be about to a Block and be split as two Block, this will cause the increase of method complexity.

In order to prove the validity of method provided by the present invention, utilize actual Operation Network, this method has been carried out the Computer Simulation test.The main optimization index of being added up during test is as follows:

Take total resources S

At first introduce the notion of resource units, (Resource Unit RU) just is meant a VC12 resource in the VC4 time slot of a multiplex section to so-called resource units.

$S=\underset{i\∈A}{\mathrm{\Σ}}\mathrm{RU}\left(i\right)$

Wherein, A is the business unit collection, the resource units that RU (i) expression business unit i takies, and its value is determined by following formula: VC12 sum * institute that RU (i)=business unit comprises is through link hop count.

Low order interlace algorithm resources occupation rate C

Number/node of the actual VC12 of taking of C=can provide the sum of VC12.

The load balancing situation

Add up each bar link residual resource situation, calculate the variance yields of link surplus resources in each subnet.

${\mathrm{\δ}}^2=1_{(n-1)}\underset{i=1}{\overset{n}{\mathrm{\Σ}}}{(a_i-\stackrel{\‾}{a})}^2$

A wherein _iSurplus resources in the representative ring on the i bar link, n represents the link sum of this ring.The mean value of surplus resources on all links in a representative ring.

The Added Business fitness

Index summation to each subnet.

Fig. 7 shows the topological structure of institute's test network.Can find out from figure, Bei'an, Qiqihar, grand celebration, Suihua and Harbin and the link between them 1,2,3,4,5 constitute subnet 701, Suihua, Yichun, Harbin, Hegang and Jiamusi and the link between them 6,7,8,9,10 constitute subnet 702, Mudanjiang, Jixi, Qitaihe, Jiamusi and Harbin and the link between them 11,12,13,14,15 constitute subnet 703, and wherein these three subnets are the BLSR net.

Table 1 has provided the business demand table between each department in the test network.From table, can know the business demand situation in the test network between the each department.Take 189 VC12 time interval resources such as the business demand between Harbin and the grand celebration.

	Harbin	Grand celebration	Qiqihar	The Bei'an	Suihua	Yichun	The Hegang	Jiamusi	Qitaihe	Jixi	Mudanjiang
												Harbin	? —	189	189	63	73	28	36	200	36	36	177
Grand celebration	—	—	48	20	10	12	15	36	21	21	21
												Qiqihar	—	—	—	60	20	20	21	42	20	20	23
The Bei'an	—	—	—	—	40	30	30	40	30	30	30
												Suihua	—	—	—	—	—	36	20	40	20	20	20
Yichun	—	—	—	—	—	—	20	40	20	20	20
												The Hegang	—	—	—	—	—	—	—	63	20	20	23
Jiamusi	? —	? —	? —	? —	—	? —	? —	? —	40	36	74
												Qitaihe	? —	? —	? —	? —	? —	? —	? —	? —	? —	36	36
Jixi	—	—	—	—	—	—	—	—	—	—	36

Mudanjiang

? —

? —

? —

? —

? —

? —

? —

? —

? —

? —

? —

Table 1

According to above-mentioned business demand table, the test result of artificial optimization's design is compared with the test result of utilizing the present invention to obtain, thereby can be clear that superiority of the present invention more.Table 2 has provided the actual result of test.

Table 2

Data rows comprises three parts in the table 2: first is the result of veteran engineer's design, configuration when also being the network actual motion, and the data of second portion are exactly the resulting result of method among utilization the present invention, and third part is the above two comparison.By the data in the contrast table, as can be seen, NLB and resource occupation total amount are basic identical; But Added Business fitness and low order interlace algorithm occupancy have greatly improved, and have improved 32.1% and 48.1% respectively.Therefore, after the data analysis in his-and-hers watches 2, can think that the present invention has obtained good optimization effect.

More than lifted preferred embodiment, the purpose, technical solutions and advantages of the present invention are further specified, institute it should be understood that it is not in order to restriction protection scope of the present invention.

Claims (4)

1, a kind of method that realizes multi-business optimization of photo synchronous digital transmission network is characterized in that this method may further comprise the steps:

A. in advance the Optical synchronization digital transmission network network is divided into annular subnet, and the business demand table is set according to current network resource;

B. calculate initial route to be created each in the demand schedule is professional,, determine routing iinformation by the initial route of principle optimization that takies the minimum and load balance of Internet resources total amount;

C. according to the determined routing iinformation of step B, by the Virtual Channel high-order is that unit gives the traffic assignments time interval resource, check that Resources allocation satisfies Added Business fitness index and business takies the low order interlace algorithm resource the most after a little while, the ending resource assigning process, otherwise the business of regular similar route is that unit gives the traffic assignments time interval resource by the Virtual Channel low order in each subnet;

D, to judge whether resource in the whole network satisfies the Added Business fitness index and the low order interlace algorithm resource occupation of network minimum, if satisfy index and take the low order interlace algorithm resource minimum, then finishes, otherwise return step B.

2, method according to claim 1 is characterized in that, the initial route of the described calculating of step B is: call shortest path first and calculate initial route;

The described optimization route of step B may further comprise the steps:

B1. respectively to link circuit resource between annular subnet and in the annular subnet take situation inspection, to exceed standard all business on the link of resources occupation rate, recomputate route;

B2. check the load balance situation of annular subnet, business in the annular subnet that does not satisfy the load balance index requirement is recomputated route, check the situation that takies of link circuit resource in the annular subnet then;

B3. the Optical synchronization digital transmission network network is divided into again core sub and edge subnet, whether the overall target of judging core sub in the whole network then satisfies the condition of convergence, as satisfying, then finish route assignment,, check then whether the route calculation number of times reaches thresholding as not satisfying, if reach thresholding, then finish route assignment, otherwise, carry out step B1.

3, method according to claim 1 is characterized in that the business of the described regular similar route of step C is: with the pairing business of different low order time slots between any two nodes in the subnet, merge in the same high-order time slot by high-order time slot rank;

It after the described regular similar route of step C, in each subnet, is further to comprise before unit gives the traffic assignments time interval resource: pairing business of above high-order time slot in the coupling subnet by the Virtual Channel low order.

4, according to claim 1 or 3 described methods, it is characterized in that: set described network Added Business fitness index and be the shared node number of professional route in each subnet and the sum of products of professional shared low order time interval resource number.

Images