CN102625198B - Intelligent light resource configuration method and system - Google Patents

Abstract

The invention discloses an intelligent light resource configuration method and an intelligent light resource configuration system. The method comprises the following steps of: dividing an entire network to be searched into un-overlapped sub-networks; searching for a preset number of shortest paths between inter-network nodes in each sub-network respectively; and performing entire-network searching according to the preset number of shortest paths between the inter-network nodes in each sub-network and the inter-network paths to obtain a second preset number of entire-network shortest paths as alternative intelligent light resource configuration schemes. According to the scheme, the entire network is divided into the sub-networks for path searching, so that computational efficiency can be improved; and the second preset number of finally obtained entire-network shortest paths serve as the alternative intelligent light resource configuration schemes which can be selected by a resource management expert to determine a final scheme, so that the final intelligent light source configuration scheme is accurate, and an optimal intelligent light source configuration scheme can be obtained.

Description

Light resources intelligent configuration method and system

Technical field

The present invention relates to communication technical field, particularly a kind of light resources intelligent configuration method, a kind of light resources intelligent configuration system.

Background technology

In general, the light resources of telecommunications refers to OLT (optical line terminal, optical fiber cable termination) the essential communication resource of composition optical communication network such as equipment, ONU (Optical Network Unit, optical node) equipment, optical fiber, optical cable, light joint.Along with the increase of user's access demand, rate requirement is also more and more higher, and therefore, development optical communication could meet the demand of growing access rate.

In the development of optical communication, the configuration of light resources intelligence is a routine work of common carrier, the configuration of opening, all related in the process such as engineering construction transformation light path resource in customer service with reconfigure, it is unpractical depending merely on the path that manually goes to find between Origin And Destination, even if to adopt office direction light path after engineering association as unit, data volume is also very huge.

In the time that light resources is configured, conventionally require to accomplish to reduce as much as possible resources costs, the response time of simultaneity factor processing is short, raising O&M efficiency also.Correspondingly, for the configuration of light resources, be also to adopt multiple tolerance conventionally, for example: device resource cost, circuit construction cost, delay, stability etc.When light resources being configured in existing scheme, general way is to be single metric parameter by variant metric parameter according to different weight conversions, thereby realize accordingly the configuration of light resources, and in order to ensure effective utilization of resource, also can regard the shortest path of searching from origin-to-destination as, but in practical operation, because the data volume of telecommunication resources is very large, and need to carry out in real time correspondingly, therefore need between resource utilization and response speed, do and weigh.

In telecommunication technology solution, the resource data amount of telecommunication administration is huge, traditional resource management adopts differentiated control, therefore, the resource searching scheme of existing inquiry shortest path, first to search the path of standing at Access Layer from user node to office, then from the path between convergence-level, the core layer office of searching station.Although sectioning search has improved the response of support system, also increase manually-operated complexity, and the scheme that this sectioning search method neither the whole network resource utilization optimum.In addition, because the development of optical-fiber network has increased plurality of access modes, for example: Fiber-To-The-Building, Fiber to the home, many, the outdoor nodes of outdoor equipment node may be connected to station multiple innings, adopt the method at the traditional office of being integrated into station to be connected to the light path that other innings are stood by automatic search, need to manually specify.Thereby for optical-fiber network, traditional multi-zone supervision mode is also not suitable for actual application.

Summary of the invention

For above-mentioned problems of the prior art, the object of the present invention is to provide a kind of light resources intelligent configuration method, a kind of light resources intelligent configuration system, it can improve the efficiency of light resources intelligence configuration, and can search out as far as possible accurately optimum light resources intelligence allocation plan.

For achieving the above object, the present invention by the following technical solutions:

A kind of light resources intelligent configuration method, comprises step:

The whole network to be searched is divided into nonoverlapping subnet;

Search for respectively the first preset number bar shortest path between all gateway nodes in each subnet;

Carry out the whole network search according to path between the first preset number bar shortest path between the gateway node in each subnet, subnet and obtain gateway node shortest path in second each subnet of preset number bar as light resources intelligence allocation plan to be selected.

A kind of light resources intelligent configuration system, comprising:

Cutting unit, for being divided into nonoverlapping subnet by the whole network to be searched;

Subnet search unit, for searching for respectively the first preset number bar shortest path between all gateway nodes in each subnet;

The whole network route searching unit, obtains gateway node shortest path in second each subnet of preset number bar as light resources intelligence allocation plan to be selected for carry out the whole network search according to path between the first preset number bar shortest path between the gateway node in each subnet, subnet.

According to the invention described above scheme, it is that the whole network to be searched is divided into multiple subnets, then on the basis of these multiple subnets, calculate respectively the first preset number bar shortest path between all gateway nodes in each subnet, and the first preset number bar shortest path corresponding according to this each subnet, between subnet, path is carried out the whole network search and is obtained gateway node shortest path in second each subnet of preset number bar as light resources intelligence allocation plan to be selected, after being divided into multiple subnets, can parallel computation for the path between the gateway node in each subnet, can utilize multinuclear, the mode such as multiprocessor or multiple servers is carried out distributed parallel processing, raise the efficiency, and after network is cut apart, nodes can reduce by 2 to 3 orders of magnitude conventionally, and then can further improve computational efficiency.Moreover, what finally carry out that the whole network search obtains is that gateway node shortest path in second each subnet of preset number bar is as light resources intelligence allocation plan to be selected, can be come to select to determine final scheme from these light resources intelligence allocation plans to be selected by resource management expert, thereby can make final light resources intelligence allocation plan more accurate, obtain optimum light resources intelligence allocation plan.

Brief description of the drawings

Fig. 1 is the schematic flow sheet of light resources intelligent configuration method embodiment of the present invention;

Fig. 2 is the schematic flow sheet that the whole network to be searched is divided into nonoverlapping subnet embodiment;

Fig. 3 is the schematic flow sheet that the whole network is carried out to the embodiment of multistage roughening processing;

Fig. 4 is that during multistage roughening is processed, afterbody roughening the whole network figure after treatment carries out the schematic flow sheet that collection of illustrative plates is cut apart;

Fig. 5 is the schematic flow sheet of the embodiment of cluster segmentation;

Fig. 6 is the structural representation of light resources intelligent configuration system embodiment of the present invention.

Embodiment

Below in conjunction with preferred embodiment wherein, the present invention program is described in detail.

The schematic flow sheet of light resources intelligent configuration method embodiment of the present invention has been shown in Fig. 1.As shown in Figure 1, the light resources intelligent configuration method in this embodiment of the present invention comprises step:

Step S101: the whole network to be searched is divided into nonoverlapping subnet, enters step S102;

Step S102: search for respectively the first preset number bar shortest path between the gateway node in each subnet, enter step S103;

Step S103: carry out the whole network search according to path between the first preset number bar shortest path between the gateway node in each subnet, net and obtain second preset number bar the whole network shortest path as light resources intelligence allocation plan to be selected.

According to the present invention program as above, it is that the whole network to be searched is divided into multiple subnets, then on the basis of these multiple subnets, calculate respectively the first preset number bar shortest path between all gateway nodes in each subnet, and the first preset number bar shortest path corresponding according to this each subnet, between net, the whole network search acquisition second preset number bar the whole network shortest path is carried out as light resources intelligence allocation plan to be selected in path, after being divided into multiple subnets, can parallel computation for the path between the gateway node in each subnet, can utilize multinuclear, the mode such as multiprocessor or multiple servers is carried out distributed parallel processing, raise the efficiency, and after network is cut apart, nodes can reduce by 2 to 3 orders of magnitude conventionally, and then can further improve computational efficiency.Moreover, what finally carry out that the whole network search obtains is that second preset number bar the whole network shortest path is as light resources intelligence allocation plan to be selected, can be come to select to determine final scheme from these light resources intelligence allocation plans to be selected by resource management expert, thereby can make final light resources intelligence allocation plan more accurate, obtain optimum light resources intelligence allocation plan.

Below in conjunction with one of them specific embodiment, the light resources intelligence allocation plan of the invention described above is described in detail.

The present invention program, in the time implementing, for the whole network to be searched, is first divided into nonoverlapping subnet.For ease of understanding, below first figure is cut apart and explained.

Suppose to have the figure G (V, E) of Weight, the wherein fixed-point set of V presentation graphs, E represents the set on limit, (i, j) represents to exist one from v _ito v _jlimit, so, it is exactly to find a kind of segmentation strategy that figure is cut apart, to meet V=V ₁∪ V ₂∪ ... ∪ V _p, with E _cutrepresent the limit of removing when cutting apart, so, figure is cut apart and will be found one and cut apart and make sum (E _cut) value minimum, make simultaneously | V _i| equate, finding optimum algorithm is NP-hard as far as possible.

The most classical algorithm of cutting apart without the figure of coordinate is Kernighan/Lin algorithm, and it is by the node of continuous exchange subgraph, and the effect that collection of illustrative plates is cut apart is more excellent.If figure N is divided into A, B, N=A ∪ B, and | A|=|B|, the limit E of calculating A and B _cut, find so subset X, Y belongs to A, B, and | A|=|B|, so X, Y are exchanged, after exchange, be A ', B ', the limit E ' between A ' and B ' _cut, E ' _cut<E _cut.In practical application, require the computational efficiency of evaluation function fast as far as possible, and in fact evaluation function calculating is very consuming time.

It is an important content that belongs to atlas analysis that figure is cut apart, in atlas analysis, will scheme G Laplacian Matrix and represent, and L (G)=D-A, wherein A is the adjacency matrix of cum rights, that is, and A _ij=w (i, j), if w (i, j) >0, D is diagonal matrix, D _ij=0, if vectorial X represents that a kind of two cut apart and are divided into S, S ', X component forms by 0,1, and 1 represents v _ibelong to S set, 0 represents v _ibelong to S set '; $X\&CenterDot;L\left(G\right)\&CenterDot;X^t=\underset{(i,j)\&Element;E}{\mathrm{\&Sigma;}}w(i,j){(x_i-x_j)}^2,$ Can find out if v _iwith v _jsame set corresponding item is 0; Theoretical proof λ ₂(representing the second little characteristic value of Laplacian Matrix) is 0, and presentation graphs is not communicated with; The value of X is relaxed to real number by (0,1), and works as by solving λ ₂characteristic vector, what can obtain being similar to cuts apart, and it is higher to cut apart mass ratio.But solving of characteristic value is very slow for super large matrix, is not suitable for for the network topology of telecommunications.

Based on this, in the present invention program, in the time that the whole network to be searched is divided into nonoverlapping subnet, can adopt the collection of illustrative plates partitioning scheme based on collection of illustrative plates and cluster, the schematic flow sheet of embodiment when the whole network to be searched is divided into nonoverlapping subnet has been shown in Fig. 2.It should be noted that, in the explanation of following embodiment, average in order to ensure each set sizes, can increase constraints, suppose network to be divided into p subnet SubG, set constraints e is the parameter of controlling set sizes.

As shown in Figure 2, the concrete segmentation step that the whole network to be searched is divided into nonoverlapping subnet can comprise:

Step S201: the whole network to be searched is carried out to multistage roughening processing, enter step S202;

Step S202: afterbody roughening the whole network figure after treatment in described multistage roughening processing is carried out to collection of illustrative plates and cut apart, enter step S203;

Step S203: the figure after cutting apart according to roughenings at different levels the whole network figure after treatment and described collection of illustrative plates in described multistage roughening processing carries out cluster segmentation, obtains described nonoverlapping subnet.

In above-mentioned steps S201, in the time that the whole network to be searched carries out multistage roughening processing, concrete mode can be:

Figure G while processing for any one-level roughening ⁿ, choose figure G ⁿnode to (i, j), a ⁿfor figure G ⁿadjacency matrix;

Determine new node, this new node comprises node and the figure G that node is merged into (i, j) ⁿin be not chosen for the right node of node;

Whether the nodes that judges the figure being determined by described new node is less than the 3rd preset number threshold value, if so, finishes multistage roughening processing procedure, if not, carries out next stage roughening processing procedure for the figure being determined by described new node.

Based on above-mentioned multistage roughening mode, in Fig. 3, illustrate the whole network to be searched has been carried out to the schematic flow sheet in an embodiment of multistage roughening processing.As shown in Figure 3, the process of multistage roughening processing can be specifically:

Note G is the figure being made up of node, limit, is the corresponding figure of the whole network to be searched;

In the time of iteration for the first time, note n=0, i.e. G=G ⁰;

According to G ⁰structure adjacency matrix A ⁰, (i, j) ∈ E, all the other are 0;

Choose G ⁰node to (i, j), until can not choose again, selected node does not repeat (i, j), if now do not choose node pair, exits, and returns to the process of above-mentioned roughening for the first time, i.e. the above-mentioned G=G that makes ⁰, according to G ⁰structure adjacency matrix A ⁰etc. process;

Node is merged into new node to (i, j), simultaneously using all can not choose also as new node;

Adjust weight, suppose node to merge into i ', node is merged into j ' to (k, l) (i, j), have $A_{i^{\&prime;}{j}^{\&prime;}}^1=A_{\mathrm{ik}}^0+A_{\mathrm{il}}^0+A_{\mathrm{jk}}^0+A_{\mathrm{jl}}^0;$

Make n=n++, i.e. n=n+1, makes n=1;

Judgement corresponding figure G ¹nodes whether be less than preset number (being above-mentioned the 3rd preset number threshold value), if not, the figure G of gained ¹be the whole network figure after multistage roughening, otherwise continue to carry out following process;

Choose G ¹node to (i, j), until can not choose again, selected node,, exits if now do not choose node pair not repeating (i, j);

Node is merged into new node to (i, j), simultaneously using all can not choose also as new node;

Adjust weight, suppose node to merge into i ', node is merged into j ' to (k, l) (i, j), have $A_{i^{\&prime;}{j}^{\&prime;}}^2=A_{\mathrm{ik}}^1+A_{\mathrm{il}}^1+A_{\mathrm{jk}}^1+A_{\mathrm{jl}}^1;$

Make n=n++, i.e. n=n+1, makes n=2;

Judgement corresponding figure G ²nodes whether be less than preset number (being above-mentioned the 3rd preset number threshold value), if not, the figure G of gained ²be the whole network figure after multistage roughening, otherwise continue to carry out the G for figure ²above-mentioned choose node to, node is combined as processes such as new nodes;

For any one figure G ⁿ, choose figure G ⁿnode to (i, j), until can not choose again, selected node does not repeat (i, j), if now do not choose node pair, exits;

Node is merged into new node to (i, j), simultaneously using all can not choose also as new node;

Adjust weight, suppose node to merge into i ', node is merged into j ' to (k, l) (i, j), have $A_{i^{\&prime;}{j}^{\&prime;}}^{n+1}=A_{\mathrm{ik}}^n+A_{\mathrm{il}}^n+A_{\mathrm{jk}}^n+A_{\mathrm{jl}}^n;$

Make n=n++, i.e. n=n+1;

Continue said process, until obtain the figure G that nodes is less than preset number (being above-mentioned the 3rd preset number threshold value) ⁿ⁺¹, this is less than the figure G of preset number ⁿ⁺¹be the figure G having realized after multistage roughening ⁿ⁺¹.

The preset number here, can carry out synthetic setting according to aspects such as Practical Calculation demand, operation times, and in the present invention program's a specific embodiment, above-mentioned preset number can be made as 3000.

Based on the schematic flow sheet shown in Fig. 3, the examples of program code in a specific implementation process can be as described below:

Complete after above-mentioned roughening processing procedure, can enter afterbody roughening the whole network figure after treatment G in multistage roughening is processed in above-mentioned steps S102 ⁿ⁺¹carry out the process that collection of illustrative plates is cut apart, its concrete collection of illustrative plates cutting procedure can comprise:

Afterbody roughening the whole network figure after treatment G during multistage roughening is processed ⁿ⁺¹carry out collection of illustrative plates and cut apart, generate the S set ubG of p subnet;

In the secondary splitting process of any one-level, obtain figure tmpG and the homography tmpA of divided least number of times in S set ubG; Solve the second little characteristic value and the characteristic of correspondence vector X of tmpA; The node that X component is greater than or equal to 0 correspondence is put into the first set, and the node that is less than 0 correspondence is put into the second set, and limit is divided in the first set, corresponding two subgraphs of the second set, generates G _{sub '}, G _{sub "};

Make SubG=(SubG ∪ { G _{sub '}, G _{sub "})-{ tmpG};

Whether the progression that judges secondary splitting is less than or equal to the first default iterations, if so, returns to pair set SubG and carry out the secondary splitting process of next stage, if not, finishes secondary splitting process.

Based on aforesaid way, obtain above-mentioned multistage roughening process in afterbody roughening the whole network figure after treatment G ⁿ⁺¹after, for this figure G ⁿ⁺¹carry out collection of illustrative plates and cut apart, generate the S set ubG of p subnet, i.e. SubG={G _sub, then can carry out secondary splitting repeatedly for the S set ubG of this subnet, in Fig. 4, illustrate that antithetical phrase net collective SubG carries out the schematic flow sheet of secondary splitting repeatedly, as shown in Figure 4, its detailed process can be:

Obtain figure tmpG and the homography tmpA of divided least number of times in SubG;

Solve the second little characteristic value and the characteristic of correspondence vector X of tmpA;

The node that X component is greater than or equal to 0 correspondence is put into a set (can be designated as the first set), by another set (can be designated as the second set) of putting into that is less than 0, and limit is also divided in these two corresponding two subgraphs of set, generate two subgraph G _{sub '}, G _{sub "};

Make SubG=(SubG ∪ { G _{sub '}, G _{sub "})-{ tmpG}, after the figure tmpG of divided least number of times removes from subnet S set ubG in subnet S set ubG, adds subnet S set ubG by two new subgraphs that generate;

Make i=i++, i.e. i=i+1;

Judge whether i is less than or equal to default iterations p, if, return to the process such as figure tmpG and homography tmpA that continues to carry out divided least number of times in the above-mentioned SubG of obtaining, until reach default iterations p, what need to be careful is, the number of the subnet that the default iterations p is here divided into while cutting apart with above-mentioned collection of illustrative plates is identical, this be because, above-mentioned collection of illustrative plates is cut apart p the subnet obtaining, be set to p and can guarantee that this p subgraph all likely accepts secondary splitting as far as possible by presetting iterations, otherwise, the current SubG obtaining is the S set ubG of the subgraph obtaining after collection of illustrative plates is cut apart, and adopt this current SubG obtaining to enter the process of follow-up cluster segmentation.

Based on the schematic flow sheet shown in Fig. 4, the examples of program code in a specific implementation process can be as described below:

Because cluster segmentation is to process and collection of illustrative plates carries out after cutting apart at roughening, therefore, also can be referred to as is the stage that refinement is cut apart.In the time carrying out cluster segmentation, can adopt various can with clustering algorithm cut apart, for example k-mean algorithm, to utilize the calculated performance of this algorithm with the advantage of scale linear growth.

Cluster is that a group is had to manifold object, be divided into several classes according to certain similarity, object in each class is more close, and the difference between class is larger, and this characteristic is similar with the figure ration of division, if a node is many with the limit of some subnet associations, just can think that it follows the node of this subnet all more similar, can, at same subnet, utilize this characteristic, in the present invention program, can cut apart network with clustering algorithm.

In the time carrying out cluster segmentation, need to be applied to the whole network figure (the whole network figure of the roughening processing at different levels here of roughening processing at different levels in above-mentioned multistage roughening processing, do not comprise afterbody roughening process after the whole network figure of obtaining) and the figure of above-mentioned collection of illustrative plates after cutting apart, in the time of the whole network figure of application roughening processing at different levels, can apply all the whole network figure of roughening processing at different levels, in reducing amount of calculation, improving the object of computational efficiency, can only choose a part in these roughenings at different levels the whole network figure after treatment and carry out the process of cluster segmentation.In the situation that selected part carries out cluster segmentation, the process of cluster segmentation can be specifically:

In processing, described multistage roughening in the whole network figure of roughening processing at different levels, chooses a 4th preset number figure;

Calculate the cluster centre of the each figure after described collection of illustrative plates is cut apart;

In any one-level iterative process, to any one node in the set of node that in described a 4th preset number figure, any one figure is corresponding, judge that this node is nearest from which in each described cluster centre, and in the time that the size of this nearest cluster centre place cluster is being preset in magnitude range, this node is added to the cluster at this nearest cluster centre place;

Recalculate cluster centre, and calculating is when the error of previous stage cluster centre and upper level cluster centre, in the time that not complete to described a 4th preset number figure iteration and error are being preset in error range, enter next stage iterative process for the next figure in described a 4th preset number figure, otherwise construct described nonoverlapping subnet according to the each cluster obtaining when previous stage iterative process.

According to cluster segmentation mode as implied above, the schematic flow sheet that carries out the embodiment of cluster segmentation is shown in Fig. 5.In this embodiment, suppose in roughenings at different levels the whole network figure after treatment, to choose 10 figure from multistage roughening is processed, as shown in Figure 5, the process of cluster segmentation can be:

In above-mentioned multistage roughening is processed in the whole network figure of roughening processing at different levels, choose a 4th preset number figure, for the consideration to arithmetic speed and computational efficiency, the 4th preset number here can arrange according to actual needs, to meet the needs of refinement level and the demand of arithmetic speed simultaneously, therein in a concrete example, the 4th preset number here can be chosen to be 10, in the time selecting, can adopt various possible modes to select, for example can average and choose by step-length, consider and may not be divided exactly by 10, when actual choosing, can choose by the mode of approximate average step length, to consider the roughening result in each stage in multistage roughening process,

Calculate each figure in the SubG after above-mentioned collection of illustrative plates is cut apart cluster centre;

Then, for the 1st figure in 10 selected figure, for giving the 1st any one node in set of node corresponding to figure, judge that this node is nearest from which in each cluster centre, and the size that judges this cluster centre place cluster or set is whether in default magnitude range;

If, in default magnitude range, do not return this node re-started to judgement, to find the nearer cluster centre of another one except above-mentioned cluster centre;

If in default magnitude range, this node is added to cluster or the set at this nearest cluster centre place;

After the 1st each node corresponding to figure is finished, the each node in the 1st figure has all added after the cluster or set at certain cluster centre place, recalculate the cluster centre of this cluster or polymerization, and calculate the error of this cluster centre and cluster centre before;

If above-mentioned 10 are schemed not the complete and error of cluster iteration in default error range time, carry out said process, that is: for the 2nd selected figure

In any one-level iterative process, to any one node in set of node corresponding to certain figure from these 10 figure, judge that this node is nearest from which in each current cluster centre, and in the time that the size of this nearest cluster centre place cluster is being preset in magnitude range, this node is added to the cluster at this nearest cluster centre place;

Recalculate cluster centre, and calculating is when the error of previous stage cluster centre and upper level cluster centre, in the time that not complete to described the second preset number figure iteration and error are being preset in error range, enter next stage iterative process for the next figure that not yet carries out cluster iterative process in these 10 figure, otherwise construct described nonoverlapping subnet according to the each cluster obtaining when previous stage iterative process.

Based on the schematic flow sheet shown in Fig. 5, the examples of program code in a specific implementation process can be as described below:

According to construct subgraph SubG _i;

$V\left({\mathrm{SubG}}_i\right)=c_i^0,$ $E\left({\mathrm{SubG}}_i\right)=\left\{(v_i,v_j)\right|v_i\&Element;c_i^0,v_j\&Element;c_i^0,(v_i,v_j)\&Element;E\},i=\mathrm{1,2,3}...p;$

Can obtain by the subgraph SubG after cluster segmentation refinement by above-mentioned steps _i.

Cutting apart the subgraph SubG obtaining after cluster segmentation refinement by above-mentioned collection of illustrative plates _iafter, then search respectively the each subgraph SubG after these refinements _imany shortest paths between interior all gateway nodes.

Searching respectively each subgraph SubG _iin all gateway nodes between many shortest paths time, the present invention program adopts the thought of Dynamic Programming, is converted to by solving a problem problem that scale is less that solves, until finish, problem can directly be solved.By Dynamic Programming, problem can be resolved by the mode of iteration.

Suppose that we finally need to find out k bar (being equivalent to the second preset number bar) shortest path, so, in each subnet, we can search u*k bar (being equivalent to the first preset number bar) path, due to u too senior general increase memory space, therefore the concrete value of u can be set according to actual conditions, and in a specific embodiment, the value of u can be 1<u<1.3 therein.

The mathematical principle that finds out k bar shortest path by searching u*k bar path can be as described below: supposing that n is subnet nodes, is the shortest path P of i if require limit number _st(i), min (P _sr(i-1)+w (r, t)), r=1,2,3 ..., n.If P represents gateway node and arrive the distance of all nodes, A represents the adjacency matrix of cum rights, P ' st=min (P _sr+ A (r, t)), r=1,2,3 ..., n, by continuous P iteration, will restrain i.e. P '=P during to certain single-step iteration.

Based on above-mentioned mathematical principle, in the present invention program, can adopt various possible modes to find out the k*u bar shortest path of gateway node in each subnet.In an embodiment, can be generalized to the shortest algorithm of k therein, represent the path from s to r, m=1,2,3 ... k, while application, can follow following rule in the present invention program:

Only search the mulitpath between gateway node;

Only to newly adding paths and process,

There is not loop in judgement in algorithm;

If $P_{\mathrm{sr}}^m+A(r,t)<d,d=\max \left(P_{\mathrm{st}}^m\right),m=\mathrm{1,2,3},...,k,$ There is the new route from s to t;

When can not find new path all k shortest paths have been found in explanation, need not carry out iteration again, can premature termination.

Cutting apart the whole network to be searched is being divided into after subnet by collection of illustrative plates, due to intensive in subnet, net between nodes sparse, subnet relatively little, and the gateway node of subnet is than great, and limit, conventionally than comparatively dense, therefore can adopt based on Dynamic Programming iterative algorithm and find out many shortest paths.

In the time adopting Dynamic Programming iterative algorithm to find out many shortest paths, concrete step can comprise:

Determining after a paths, search and whether have a shorter path, if exist, upgrade after path status, judge whether to exist new route, if there is new route, continue to search whether have shorter path, until can not find shorter path.

Based on above-mentioned four, in a specific embodiment, the program implementation of searching many shortest paths between the gateway node in subnet can be as described below:

Above-mentionedly finally obtain be exactly SubG path, wherein, s, t is boundary point.

After the above-mentioned SubG path that obtains the gateway node in each subnet, then can carry out the whole network search according to path between this SubG path, net and obtain K bar shortest path as light resources intelligence allocation plan to be selected.

In the time carrying out the whole network route searching, be the path of the gateway node based in the above-mentioned each subnet having obtained, search for the whole network path of the gateway node between each subnet.When specific implementation, can carry out the whole network route searching based on dijkstra's algorithm.The principle of dijkstra's algorithm is: if P _stthe shortest path from s to t, so for other node r on path, P _sralso be the shortest path from s to r.

In the present invention program's a specific embodiment, can, on the basis of dijkstra's algorithm, the search simple extension in the whole network path be arrived to k shortest path, specifically can follow following rule:

First, the topology of the whole network search is made up of three parts: circuit between net; The path of the gateway node in each subnet of the gateway node of subnet, the above-mentioned calculating of system storage; The topology of the subnet at starting point, terminal place;

Next, if be k article of shortest path from s to t, r is certain node except starting point, terminating point on path, so also be k ' article of shortest path from s to r, k '≤k, traditional dijkstra's algorithm is only to retain a paths for the each node in figure, and in the scheme after the present invention expansion, need to retain k paths;

Its three, require the path of searching not occur loop;

Its four, for k shortest path, be traversed k time if point to the limit of destination node, can search for by premature termination.

Comprehensive foregoing, in a specific embodiment, its corresponding software is realized program can be as described below:

Obtain treat exactly routing path, wherein,

According to the mode in above-mentioned concrete example, can obtain between k bar (the second preset number bar) net path as resource allocation proposal to be selected, can be come to select to determine final scheme from these light resources intelligence allocation plans to be selected by resource management expert, thereby can make final light resources intelligence allocation plan more accurate, obtain optimum light resources intelligence allocation plan.

But, because communication network is very large, and need to constantly carry out engineering upgrading and transformation, new resource data can be entered into system at any time, causes the network topology of the whole network constantly to change, thereby has influence on the search to the whole network.Accordingly, in the present invention program, give relevant resource and change the variation of strategy with reply network topology.

The change of resource generally includes following a few class: the line out of service of the equipment failure of subnet, the increase of the circuit of subnet, subnet, the equipment increase of the whole network, equipment failure of the whole network etc.

To this, in the present invention program, it can be as described below that the resource of making changes processing policy:

When detecting that server idle running and subnet exist in the situation of line resource change, carry out the process of many shortest paths between all gateway nodes of again searching in subnet immediately;

In the time detecting that circuit in the busy and subnet of server changes quantity and reaches certain quantity, again search for many shortest paths between all gateway nodes in subnet;

In the time detecting that circuit that the whole network increases is a lot of and reach some, re-start that collection of illustrative plates is cut apart and the subnet of counterweight new search collection of illustrative plates after cutting apart in all gateway nodes between many shortest paths;

When detecting that the whole network has increased node, and starting point, terminal are also unallocated during to certain subnet, now can, on the basis of original division, by the mode of cluster, this node newly increasing be assigned to certain subnet;

If subnet equipment failure, and be gateway node, the path of respective stored is also set to lose efficacy;

When the line out of service of subnet, the path of respective stored is also set to lose efficacy.

According to light resources intelligent configuration method of the present invention as above, below describe with regard to one of them concrete application case.

In this concrete application case, to choose certain and economize inside the province the at present network of a maximum branch company, nearly 70,000 nodes of this network, have the circuit of 100,000 left and right.

First network is carried out to dividing processing, choose and be divided into p=128 subnet, subnet size is controlled and is e gets 1.5, is segmented in 1 second and completes, and cuts apart the subnet nodes more than 700 of rear maximum, minimum subnet nodes more than 400, and splitting speed is than very fast.By the actual test of this branch company's network is found, be no more than 1000 the cutting apart all in 15 seconds of full figure at the subnet number of cutting apart.

On the above-mentioned basis of cutting apart, increase a node newly, adopt k-mean to carry out cluster, because original division is relatively good, thereby can within 1-2 second, restrain.

Set resource distribution path k=200 to be selected, redundancy factor u=1.1, if only need to upgrade each subnet, the operation time of each subnet is in 2 seconds, in this real case, adopt the CPU multithreading of 4 cores to carry out, be evenly distributed workload, each core distributes 32 subnet parallel computations, because subnet size is relatively average, substantially complete, the time is within 20 seconds altogether simultaneously.

The sum of all-ones subnet gateway node is at 3600, add subnet initial, terminal node, node sum is in 5000,10 times are dwindled than original figure, inquiry k shortest path only needs the 2-3 time of second, confirm that the present invention program can improve the efficiency of light resources intelligence configuration, and can search out as far as possible accurately the conclusion of optimum light resources intelligence allocation plan.

The invention described above scheme, also release without any similar technology and system in telecommunications industry, belong to and propose first and realize, it has proposed a kind of distributed, parallel light resource searching scheme, by cutting apart based on collection of illustrative plates and the algorithm of cluster is divided into the whole network multiple subnets fast, every pair of gateway node path in can parallel computation subnet on subnet; When inquiring about, user searches the k bar path of the whole network as the routing path for the treatment of of resource distribution on the basis of subnet path; When network changes and system when busy, the present invention program also proposes to adopt interim partitioning algorithm, rapidly newly-increased resource node is joined to the subnet of having cut apart, raising response speed in existing cutting apart on basis.Possess compared with high scalability, can be simply by increasing CPU or Server Extension in the time that performance is not enough.

According to the light resources intelligent configuration method of the invention described above, the present invention also provides a kind of light resources intelligent configuration system, the structural representation of light resources intelligent configuration system embodiment of the present invention has been shown in Fig. 6, as shown in Figure 6, in this embodiment, light resources intelligent configuration system of the present invention comprises:

Cutting unit 601, for being divided into nonoverlapping subnet by the whole network to be searched;

Subnet search unit 602, for searching for respectively the first preset number bar shortest path between the gateway node in each subnet;

The whole network route searching unit 603, obtains second preset number bar the whole network shortest path as light resources intelligence allocation plan to be selected for carry out the whole network search according to path between the first preset number bar shortest path between the gateway node in each subnet, net.

Wherein, in a specific embodiment, above-mentioned cutting unit 601 specifically can comprise:

Roughening processing unit 6011, carries out multistage roughening processing for the whole network to be searched;

Collection of illustrative plates cutting unit 6012, cuts apart for described multistage roughening processing afterbody roughening the whole network figure after treatment is carried out to collection of illustrative plates;

Cluster segmentation unit 6013, carries out cluster segmentation for the figure processing according to described multistage roughening after roughenings at different levels the whole network figure after treatment and described collection of illustrative plates are cut apart, and obtains described nonoverlapping subnet.

In another one preferred embodiment, above-mentioned subnet search unit 602 can include more than two.Thereby, can utilize these multiple subnet search units 602 to carry out parallel computation for the path between the gateway node in variant subnet, realize distributed treatment, improve treatment effeciency.Wherein, the subnet search unit 602 here, can realize by modes such as multinuclear, multiprocessor or multiple servers, does not repeat them here.

In addition, as shown in Figure 6, in another one specific embodiment, can also include detecting unit 604, for detection of system mode;

Now, above-mentioned subnet search unit 602, also for detecting that at detecting unit 604 line resource of server idle running and subnet changes or detects in the busy and subnet of server that circuit changes quantity and reaches while setting amount threshold, re-executes the process of the first preset number bar shortest path between the above-mentioned gateway node of searching for respectively in each subnet;

Above-mentioned cutting unit 601, also sets when amount threshold for detecting that at detecting unit 604 number of, lines that the whole network increases reaches, and re-executes the described process that the whole network to be searched is divided into nonoverlapping subnet;

Above-mentioned cluster segmentation unit 6013, also when detecting that at detecting unit 604 the whole network has increased new node and starting point, terminal and is not yet divided into certain subnet, is assigned to subnet by cluster by the node that this newly increases.

Wherein, in the light resources intelligent configuration system of the invention described above, what the specific implementation of each unit can be with the light resources intelligent configuration method of the invention described above is identical, does not repeat them here.

The above embodiment has only expressed several execution mode of the present invention, and it describes comparatively concrete and detailed, but can not therefore be interpreted as the restriction to the scope of the claims of the present invention.It should be pointed out that for the person of ordinary skill of the art, without departing from the inventive concept of the premise, can also make some distortion and improvement, these all belong to protection scope of the present invention.Therefore, the protection range of patent of the present invention should be as the criterion with claims.

Claims (8)

1. a light resources intelligent configuration method, is characterized in that, comprises step:

The whole network to be searched is divided into nonoverlapping subnet;

Search for respectively the first preset number bar shortest path between all gateway nodes in each subnet;

Carry out the whole network search according to path between the first preset number bar shortest path between the gateway node in each subnet, subnet and obtain gateway node shortest path in second each subnet of preset number bar as light resources intelligence allocation plan to be selected;

Wherein, the step that the whole network to be searched is divided into nonoverlapping subnet comprises:

The whole network to be searched is carried out to multistage roughening processing;

Afterbody roughening the whole network figure after treatment in described multistage roughening processing is carried out to collection of illustrative plates to be cut apart;

Figure after cutting apart according to roughenings at different levels the whole network figure after treatment and described collection of illustrative plates in described multistage roughening processing carries out cluster segmentation, obtains described nonoverlapping subnet.

2. light resources intelligent configuration method according to claim 1, is characterized in that, the process of the whole network to be searched being carried out to multistage roughening processing comprises:

Figure G while processing for any one-level roughening ⁿ, choose figure G ⁿnode to (i, j), a ⁿfor figure G ⁿadjacency matrix;

Determine new node, this new node comprises node and the figure G that node is merged into (i, j) ⁿin be not chosen for the right node of node;

Whether the nodes that judges the figure being determined by described new node is less than the 3rd preset number threshold value, if so, finishes multistage roughening processing procedure, if not, carries out next stage roughening processing procedure for the figure being determined by described new node.

3. light resources intelligent configuration method according to claim 2, is characterized in that, during multistage roughening is processed, afterbody roughening the whole network figure after treatment carries out the step that collection of illustrative plates cuts apart and comprises:

Afterbody roughening the whole network figure after treatment G during multistage roughening is processed ⁿ⁺¹carry out collection of illustrative plates and cut apart, generate the S set ubG that comprises P subnet;

In the secondary splitting process of any one-level, obtain figure tmpG and the homography tmpA of divided least number of times in S set ubG; Solve the second little characteristic value and the characteristic of correspondence vector X of tmpA; The node that X component is greater than or equal to 0 correspondence is put into the first set, and the node that is less than 0 correspondence is put into the second set, and limit is divided in the first set or corresponding two subgraphs of the second set, generates G _{sub '}, G _{sub "};

Make SubG=(SubG ∪ { Gsub ＇, Gsub 〞 })-{ tmpG};

Whether the progression that judges secondary splitting is less than or equal to the first default iterations, if so, returns to pair set SubG and carry out the secondary splitting process of next stage, if not, finishes secondary splitting process.

4. light resources intelligent configuration method according to claim 3, it is characterized in that, in processing according to described multistage roughening, the whole network figure of roughening processing at different levels and the described collection of illustrative plates figure after cutting apart carries out cluster segmentation, the step that obtains described nonoverlapping subnet comprises:

In processing, described multistage roughening in the whole network figure of roughening processing at different levels, chooses a 4th preset number figure;

Calculate each figure after described collection of illustrative plates is cut apart cluster centre;

In any one-level iterative process, to any one node in the set of node that in described a 4th preset number figure, any one figure is corresponding, judge that this node is nearest from which in each described cluster centre, and in the time that the size of this nearest cluster centre place cluster is being preset the situation in magnitude range, this node is added to the cluster at this nearest cluster centre place;

Recalculate cluster centre, and calculating is when the error of previous stage cluster centre and upper level cluster centre, in the time that not complete to described a 4th preset number figure iteration and error are being preset in error range, enter next stage iterative process for the next figure in described a 4th preset number figure, otherwise construct described nonoverlapping subnet according to the each cluster obtaining when previous stage iterative process.

5. according to the light resources intelligent configuration method described in claim 1 to 4 any one, it is characterized in that, also comprise following any one or combination in any in every:

In the time that the line resource that server idle running and subnet detected changes, re-execute the process of the first preset number bar shortest path between described all gateway nodes of searching for respectively in each subnet;

Detecting in the busy and subnet of server that circuit changes quantity and reaches while setting amount threshold, re-execute the process of the first preset number bar shortest path between described all gateway nodes of searching for respectively in each subnet;

The number of, lines increasing at the whole network reaches while setting amount threshold, re-executes the described process that the whole network to be searched is divided into nonoverlapping subnet;

In the time that the whole network has increased new node and starting point and terminal and is not yet divided into certain subnet, by cluster, the node that this newly increases is assigned to subnet.

6. a light resources intelligent configuration system, is characterized in that, comprising:

Cutting unit, for being divided into nonoverlapping subnet by the whole network to be searched;

Subnet search unit, for searching for respectively the first preset number bar shortest path between all gateway nodes in each subnet;

The whole network route searching unit, obtains gateway node shortest path in second each subnet of preset number bar as light resources intelligence allocation plan to be selected for carry out the whole network search according to path between the first preset number bar shortest path between the gateway node in each subnet, subnet;

Wherein, cutting unit comprises:

Roughening processing unit, carries out multistage roughening processing for the whole network to be searched;

Collection of illustrative plates cutting unit, cuts apart for described multistage roughening processing afterbody roughening the whole network figure after treatment is carried out to collection of illustrative plates;

Cluster segmentation unit, carries out cluster segmentation for the figure processing according to described multistage roughening after roughenings at different levels the whole network figure after treatment and described collection of illustrative plates are cut apart, and obtains described nonoverlapping subnet.

7. light resources intelligent configuration system according to claim 6, is characterized in that, described subnet search unit comprises more than two.

8. according to the light resources intelligent configuration system described in claim 6 to 7 any one, it is characterized in that, also comprise: detecting unit, for detection of system mode;

Described subnet search unit, also for detecting that at described detecting unit the line resource of server idle running and subnet changes or detects in the busy and subnet of server that circuit changes quantity and reaches while setting amount threshold, re-executes the process of the first preset number bar shortest path between described all gateway nodes of searching for respectively in each subnet;

Described cutting unit, also sets when amount threshold for detecting that at described detecting unit number of, lines that the whole network increases reaches, and re-executes the described process that the whole network to be searched is divided into nonoverlapping subnet;

Described cluster segmentation unit, also when detecting that at described detecting unit the whole network has increased new node and starting point and terminal and is not yet divided into certain subnet, is assigned to subnet by cluster by the node that this newly increases.