CN101599778A - Designated lane guard method in a kind of WDM optical network plan - Google Patents
Designated lane guard method in a kind of WDM optical network plan Download PDFInfo
- Publication number
- CN101599778A CN101599778A CNA2009100791280A CN200910079128A CN101599778A CN 101599778 A CN101599778 A CN 101599778A CN A2009100791280 A CNA2009100791280 A CN A2009100791280A CN 200910079128 A CN200910079128 A CN 200910079128A CN 101599778 A CN101599778 A CN 101599778A
- Authority
- CN
- China
- Prior art keywords
- route
- link
- taboo table
- ant
- current
- Prior art date
- Legal status (The legal status is an assumption and is not a legal conclusion. Google has not performed a legal analysis and makes no representation as to the accuracy of the status listed.)
- Pending
Links
Images
Landscapes
- Data Exchanges In Wide-Area Networks (AREA)
Abstract
The embodiment of the invention discloses the designated lane guard method in a kind of WDM optical network plan.Key step comprises: the node in the collection network, link and business information, determine the cycle-index of single vocational work-protection route to calculating; Initiate two parallel ant processes simultaneously by the source node of single business, each ant process is selected route by the mode that distributed hop-by-hop calculates transition probability, but and by to separately independently the taboo table screen the link of every jumping transition with the attended operation of sharing the taboo table; Upgrade the pheromone concentration of globally optimal solution and each bar link of the whole network by evaluation work route and the total jumping figure of protection route; By initial hit is optimal working and protection route assignment wavelength; And by each business in the professional chained list of circular treatment.
Description
Technical field
The present invention relates to the communications field, relate in particular to and be used for the special-purpose path protection of WDM optical network plan.
Background technology
Along with the develop rapidly of optical transport technology, single channel transmission rate and single fiber multipling channel number constantly increase in wavelength division multiplexing (WDM) network, thereby cause the rapid expansion of network capacity.Meanwhile, in case break down in the network and bring for whole network performance to seriously influence.Therefore, the survivability of assurance optical-fiber network has very important significance.
The basic survivability mechanism that optical-fiber network adopts has: based on the protection mechanism and the real-time dynamic Restoration Mechanism of seeking available resources of private resource.The advantage of protection mechanism is because the route of protection path is reserved with the resource that needs, and the failure recovery time is very short, can be controlled at Millisecond, therefore is widely used in the planning of the professional route of optical transfer network.The research of present most of protection mechanisms aspect mainly contains based on the protection of passage with based on two kinds of schemes of protection of link.Wherein, path protection can be divided into two kinds of strategies again: designated lane protection and sharing channel protection.The designated lane protection distributes the protection passage of a special use for every service aisle, and the wavelength resource of reservation exclusively enjoys for it.The designated lane protection has the highest reliability, can also deal with the situation that the multistage link failure takes place on a service aisle.The standby wavelength that the sharing channel protection makes alternate channel reserve on each bar link can be shared between many alternate channels, as long as these alternate channels can not enabled simultaneously.Because the high reliability of designated lane protection in most of actual optical network plan work, mainly adopts this kind mode at the protection of work route at present.
In the WDM optical-fiber network, optical fiber link will pass through a plurality of pipelines and path in process of deployment.Have many pipelines in the same section path, comprise many optical fiber links in the same pipeline again.Therefore, when stretch footpath or pipeline break down, all optical fiber links by this path or pipeline will lose efficacy simultaneously.In order to improve optical-fiber network survivability mechanism, IETF working group has proposed shared risk link group (Shared Risk Link Group, notion SRLG) in the draft text of calendar year 2001.In brief, those optical fiber links in identical pipeline/path have constituted a SRLG.SRLG is meant one group of link of shared physical resource (promptly having shared failure risk).In the route planning problem of optical-fiber network, the designated lane protection needs consideration work route and the problem of protecting " SRLG separates " between the route.The route of promptly working can not be passed through same SRLG with the protection route.Like this, can reduce the possibility that work route and protection route lost efficacy simultaneously significantly, thereby improve the fault-resistant ability of whole network.
During designated lane protection scheme in considering optical network plan, also need to solve SRLG trap problem.The trap problem of SRLG can be divided into the true trap of SRLG and SRLG self-trapping.The former be meant specific source, destination node between, do not have the path of a pair of " SRLG separation " really.And the latter exists some defective to cause by used computational methods, and the imperfection unreasonable, that newly calculate to request work route of Resource Allocation in Networks can cause not have to be separated.The method of the self-trapping problem of present existing solution SRLG mainly contains integral linear programming method and some heuristics, as simulated annealing, and weighting algorithm, genetic algorithm etc.Because integral linear programming method oversize and intractable consuming time has the large scale network of a large amount of SRLG, use at present often is a heuristic.Ant group algorithm is a kind of in the heuristic.It is at first to propose a kind of simulative optimization searching algorithm that grown up by the true ant group of occurring in nature behavior in nineteen nineties by people such as Italian scholar M.Dorigo.(Ant Colony Optimization, ACO), ant can be found an optimal path from the ant nest to the food source to a kind of ant group optimization of ant group system (ACA:Ant Colony Algorithm) in the process of looking for food.Ant is in when motion, stayed the pheromone (Pheromore) of some on the path of process, and it can feel the existence and the power thereof of this material, tends to move towards the more direction of this amount of material.Thus, the behavior of a large amount of ant group colonies shows as a kind of positive feedback phenomenon: the ant of process is many more on certain paths, just there is more ant to select this path later on, quickens the search procedure of optimal solution, avoid the too early locally optimal solution that is absorbed in by Distributed Calculation by positive feedback.Yet traditional ant group algorithm mainly is the calculation process that adopts serial, and promptly Route Selection of ant and transition probability computational process are carried out one by one.In the designated lane protection scheme in optical network plan,, and ignore of the influence of protection route, be easy to be absorbed in the self-trapping problem of SRLG the work routing if only consider definite influence that selection produced of work route to the protection route.In addition, didactic routing search method major part is started with by the route solution space of source to destination from enlarging, and selects two routes that meet the SRLG separation condition and be used separately as work and protection from solution space.And in the solution space calculation process, seldom consider the restricted problem of SRLG, and cause a large amount of route calculation results invalid, reduced the efficient of route search.
Consider afore-mentioned, exist to overcome needs not enough in the correlation technique.
Summary of the invention
The object of the present invention is to provide a kind of based on ant group algorithm and consider the designated lane guard method of SRLG restriction.Be mainly used in the WDM optical network plan and protect route, and be respectively these work and protection route assignment wavelength for a plurality of plans of operation work route and designated lane thereof separately.This method can be used for having in the network of shared risk link group restriction.
Designated lane guard method in the WDM optical network plan provided by the invention is characterized in that comprising following steps:
Node in the collection network and link information, and extract shared risk link group information in the link information simultaneously, generate the topological chained list that contains shared risk link group information;
Extract source, the destination node information of current business in the professional chained list, determine the cycle-index of single vocational work-protection route, enter the calculation process that single vocational work-the protection route is right calculating;
In the right single calculation of single vocational work-protection route, the work route is calculated with the calculating of protection route and is walked abreast.Implementation method is to initiate two parallel ant processes by the source node of business, and the destination all is professional destination node.Two ant processes have separately independently taboo table and the taboo table of sharing.The available link of screening when the taboo table is used for the routing of ant process.The routing mode of single ant process is that the pheromone concentration according on each bar link of hop-by-hop calculates current transition probability and transfers to next node, till arriving destination node promptly professional destination node or the death of this ant process.Before the transition each time of single ant process constantly with transition after constantly, each ant process all needs respectively the taboo table to be finished full maintenance work.Constantly, the ant process needs the taboo table of self and shares the taboo table and safeguard before each transition.Rule is as follows: the link of no wavelength available in the current available link set is added shared taboo table; With another end points except that current point in the current available link set is this ant process has added self through the link of point taboo table; And after each transition constantly, the ant process needs to carry out again attended operation to sharing the taboo table.Rule is as follows: add with selected transition link and with its all links in same shared risk link group and share the taboo table;
After single work-protection route was finished calculating, system finished the renewal work of optimal solution and upgrades the pheromones of network link according to the quality of separating.When loop termination, the work of optimal solution correspondence-protection route will be to being assigned with wavelength.The strategy of Wavelength Assignment is the initial hit strategy.Current business is calculated the handling process then of finishing and is transferred to next professional in the professional chained list.After all Business Processing finish, each vocational work of system refresh network topology and report output and protection route, wavelength program results.
Designated lane guard method in the WDM optical network plan provided by the invention it is advantageous that the planing method that has adopted string and combination.Promptly in to the processing of professional chained list, adopt serial mode and optimal solution that the cyclic search vocational work-the protection route is right; And in the right single calculation of the work-protection route of single business, employing be the mode that two ant processes of walking abreast are searched for route jointly.Designated lane guard method in the WDM optical network plan provided by the invention is by adopting the mode of serial cyclic search optimal solution; only preserve the route result of calculation of global optimum; other effective Search Results is saved in the network topology with the form of pheromones; solved the memory space problem of feasible solution; and in route is calculated, formed effective pheromones positive feedback, thereby optimize follow-up solution procedure.Designated lane guard method in the WDM optical network plan provided by the invention has taken into full account the influence between vocational work routing and the protection routing, and by the computational methods of distributed parallel hop-by-hop it is coordinated control.Simultaneously in the processing of taboo table, add the SRLG restrictive condition, thereby solved the self-trapping problem of SRLG, and in route search, avoided a large amount of useless generations of separating, improved the efficient of route search.
Other features and advantages of the present invention will be set forth in the following description, and, partly from specification, become apparent, perhaps understand by implementing the present invention, purpose of the present invention and other advantage can be by specification, the claims in abbreviation, and the specifically noted structure realizes and obtains in the accompanying drawing.
Description of drawings
In conjunction with the accompanying drawing of having described various embodiment of the present invention, according to following detailed description, will be easier to understand above-mentioned and other advantage of the present invention to each invention of the present invention, wherein:
Fig. 1 shows the general flow chart of designated lane guard method provided by the invention;
Fig. 2 shows the calculation flow chart that single vocational work-the protection route is right in the designated lane guard method provided by the invention;
Embodiment
By with reference to following detailed description for exemplary non-limiting examples and figures, advantages and features of the invention and realize that method of the present invention is easier to understand.Yet the present invention can implement with multiple different form, and the embodiment that should not be construed as limited in this explaination.In addition, thus it will be completely and completely that these embodiment the disclosure are provided.And complete conveyed to those skilled in the art with design of the present invention, the present invention will only be defined by claims.In specification, identical label is indicated the parts of enjoying all the time.
Fig. 1 shows the general flow chart of designated lane guard method provided by the invention.In step 101, system is according to node and link information in the planning input data collection network, and generate corresponding chained list according to these information, and extract the shared risk link group information in the input data simultaneously, generate the topological chained list that contains shared risk link group information.At last, system is integrated into the form of chained list equally with the business datum of input, and the business that will be positioned at the chained list first place is as current business to be allocated.
In step 102, the handling process of current business begins.In step 103, the work of current business, the globally optimal solution G_Global of protection route are initialized as sky.Here, the work route of current business is kept among the G_Global with the form of protection route with chained list, and it all is empty that initialization will be specified these two chained lists.In addition, this step is specified the last digit call again M of single vocational work-protection route to calculating simultaneously.M has determined the ant group scale in the route calculating.2M ant process will be initiated and participation work and protection route calculation.At last, the plain concentration of the biological information in the network on each bar link is initialized to non-0 value, uses A
0Expression.A
0Value can get the link sum in the network according to the scale of network.
In step 104, M time recursive call begins.The work of being finished is that cyclic variable i is changed to 1.In step 105, single professional calculation process is activated, and then in step 106, waits for and receive the current G_Current of separating of the work of returning, protection route.Identical with G_Global, the work among the G_Current is preserved with the form of chained list equally with the protection route.
In step 107, judge whether current the separating of returning of route is available separating.Current availability of separating is by judging whether G_Current is empty decision.If current separating not for empty, flow process enters step 108; Otherwise flow process enters step 110.
In step 108, judge whether the current G_Current of separating is better than globally optimal solution G_Global.If current separating is better than current globally optimal solution or globally optimal solution for empty, then enter step 109; Otherwise flow process enters step 110.The method of judging is that more separately work route adds total jumping figure of protecting route.Jumping figure is few more, Xie Yueyou.Think that current globally optimal solution is better than current separating if jumping figure is the same.
In step 109, cover current globally optimal solution G_Global with the current G_Current of separating, enter step 110.Link in the step 110 pair network carries out the adjustment of pheromones.Regulation rule as shown in the formula:
A
l(i)＝A
l(i-1)+ΔA
l
Wherein, l represents certain bar link, A
l(i) represent link l adjusted pheromone concentration in the i time circulation, A
l(i-1) represent link l in the i time circulation, to adjust preceding pheromone concentration (A
l(0)=A
0), Δ A
lBe the pheromones increment on the link l in this adjustment.Δ A
lCalculate by following formula:
Wherein, function H (x) refers to and gets total jumping figure that route is separated x, and the route of promptly working is closed with the jumping figure of protecting route.Therefore, H (G Current) represents the total length of working among the current G_Current of separating and protecting the route chained list.In this step, flow process is according to the pheromone concentration on each bar link of above Policy Updates the whole network.
In step 111, flow process judges whether to reach the loop termination condition, and promptly whether cyclic variable i has equaled last digit call again M.If flow process enters step 112; Otherwise flow process enters step 113.In step 112, cyclic variable i gets back to step 105 then and initiates new round circulation from increasing.In step 113, judge whether whether current globally optimal solution G_Global is empty, promptly do not have available separating.If then flow process enters step 116; Otherwise flow process enters step 114.
In step 114, from the globally optimal solution G_Global that above flow process calculates, extract the work and protection routing iinformation of current business.Adopt the work that is respectively of initial hit (First_Fit) strategy to distribute wavelength piecemeal simultaneously with the protection route.In the WDM network planning, what the Wavelength Assignment mode adopted mostly is to prolong the mode that route is distributed piecemeal.Wherein, in the Wavelength Assignment process of single link, employing be newly-built principle after the first dilatation.Promptly preferentially take idle ONU, dispose new ONU again after ONU is fully loaded.Finish this step back flow process and enter step 115.
In step 115, business is marked as and is allocated successfully, and simultaneously current working practices and the respective wavelength of protecting the shared link of route is labeled as seizure condition.
In step 116, the current business in the professional chained list is labeled as the distribution failure, prepare report output.
In step 117, judge whether current managing business is positioned at the end of professional chained list, if then enter step 119; Otherwise flow process enters step 118.
In step 118, behind the pre-treatment position, moving in the professional chained list, flow process enters step 102, the next professional processing procedure of beginning.
In step 119, the business in the professional chained list all disposes, and the statistical service assignment information is exported the result with report form.
More than explanation is the main-process stream of the designated lane guard method that provides of this patent.Mainly comprise early stage to planning information; initialization procedure as network topology, SRLG information and Link State; return the statistics combing of separating and distribute wavelength to single vocational work-protection route right cycle calculations and to each route based on professional chained list mid-term, and the later stage is to the Information Statistics and the report output process of planning network.
Fig. 2 shows the calculation flow chart that single vocational work-the protection route is right in the designated lane guard method provided by the invention.In step 201, the current G_Current of separating is initialized as sky, wait available work to be filled and protection route right.
In step 202, to initiate two ant process Ant_A and Ant_B from the current source node place that manages business, and their current point is changed to current source node of managing business, point of destination is all current destination node of managing business.In step 203, the taboo table Tabu_B of taboo table Tabu_A, the Ant_B of initialization Ant_A and the taboo table Tabu_T that two ant processes are shared are for empty.The link information that must get rid of when the information that the taboo table of ant process is safeguarded is current ant process routing.The taboo table can adopt the structure of the elongated array of pointers of one dimension to realize, its each array element correspondence needs to get rid of the absolute position of link in the total network links chained list.In addition, this step is initialized as sky with the route chained list Route_A of Ant_A and the route chained list Route_B of Ant_B.The information that route chained list minute book ant process has been passed through link.
So far, following flow process is divided into two parallel routing flow processs, respectively corresponding ant process Ant_A and Ant_B.Two routing flow processs trigger simultaneously, separately the calculating of independent hop-by-hop from current source node of managing business the route to destination node.Step 204 to step 211 is the routing flow process of Ant_A, and the routing flow process of Bing Hang Ant_B comprises step 212 to step 219 with it.
In step 204, Ant_A judges whether current point is destination node, promptly current destination node of managing business.If not, continue step 205; If flow process changes step 220 over to.
In step 205, upgrade the current optional link set Link_A of Ant_A.At first will gather Link_A and be changed to sky, all of its neighbor link circuit resource with current some place adds among the Link_A then.In step 206, safeguard taboo table Tabu_A and the shared taboo table Tabu_T of Ant_A, rule is as follows: the link of no wavelength available among the Link_A is added shared taboo table Tabu_T; With another end points except that current point among the Link_A is that Ant_A adds taboo table Tabu_A through the link of point.
In step 207, safeguard current optional link set Link_A.If promptly the link among the Link_A is recorded in current taboo table Tabu_A or shared taboo table Tabu_T, then it is left out from Link_A.In step 208, judge whether also there is available link among the Link_A.If exist, flow process enters step 209; If there is no, then flow process enters step 211.
In step 209, Ant_A is calculated as follows the transition probability of each bar link,
And next jumps the link of this process to utilize the wheel disc method to select Ant_A.The similar roulette of the realization of wheel disc method, and promptly at first definite interval [0,100) be total interval.Distribute separately by stages according to the size of each bar link transition probability then.As, always having three links, the transition probability of each bar link is respectively: 1/4,1/4 and 1/2.Then distribute the by stages of these three links to be respectively: [0,25), [25,50) with [50,100).Get an interval [0,100) in random number x, x drops on and determines then in the by stages of which bar link that this link jumps the link of this process for next.Determine that next jumps the route chained list Route_A that this link is added behind link Ant_A, and this link another end points except that current point is changed to current point.In step 210, safeguard once more and share taboo table Tabu_T.Rule is as follows: the selected link of previous step is added Tabu_T; To add Tabu_T at all links of a shared risk link group with the selected link of previous step.After finishing maintenance work, flow process is got back to step 204, continues the routing circulation.
In step 211, Ant_A is labeled as death with the ant process, empties the route chained list Route_A of Ant_A, continues step 220.
And utilize the wheel disc method to select next to jump the link of process equally.Determine this link to be added behind the selected link route chained list Route_B of Ant_B, and this link another end points except that current point is changed to current point.In step 218, safeguard once more and share taboo table Tabu_T.Rule is as follows: the selected link of previous step is added Tabu_T; To add Tabu_T at all links of a shared risk link group with the selected link of previous step.After finishing maintenance work, flow process is got back to step 212, continues next jumping routing circulation of Ant_B.
In step 220, wait for the routing iinformation that reception Ant_A and Ant_B return, i.e. Route_A and Route_B.If all return, flow process enters step 221.
In step 221, system constitutes the current G_Current of separating with the routing iinformation of collecting and returns.Rule is as follows: when Route_A and Route_B are not sky; G_Current is not empty; two route chained list information Route_A of the current G_Current of separating record and Route_B and that chained list length is shorter are decided to be the work route, and that grow is decided to be the protection route.If two the chained list equal in length then selects one to be work at random, another is protection.So far, the calculation process that single vocational work-protection route is right finishes.
Though described embodiments of the present invention in conjunction with the accompanying drawings, those skilled in the art can make various distortion or modification within the scope of the appended claims in this area.
Claims (6)
1, special-purpose channel protection method in a kind of WDM optical network plan is characterized in that may further comprise the steps:
In the right single calculation of single vocational work-protection route, the work route of parallel computation business and protection route;
Described vocational work-protection the route that to return after the right single calculation of described single vocational work-protection route pair is compared with current globally optimal solution, and upgrades the described current globally optimal solution and the pheromone concentration of each bar link of the whole network.
2, method according to claim 1 is characterized in that the work route of described parallel computation business and the step of protecting route mainly may further comprise the steps:
Initiate two parallel ant processes simultaneously by the source node of described single business, and the destination all is the destination node of described single business;
Each described ant process is selected route by the mode that distributed hop-by-hop calculates transition probability, but and by to separately independently the taboo table screen the link of every jumping transition with the attended operation of sharing the taboo table.
3, method according to claim 1 is characterized in that described comparative approach adds total jumping figure of protecting route for the route of relatively working, and separating that jumping figure is few is excellent, thinks that if jumping figure is the same current globally optimal solution is excellent.
4, method according to claim 2, it is characterized in that described to separately independently the taboo table mainly may further comprise the steps with the step of sharing the attended operation of taboo table:
Before the transition of each described ant process constantly, to self the taboo table and share the taboo table and safeguard;
After the transition of each described ant process, constantly, carry out attended operation to sharing the taboo table.
5, method according to claim 4, it is characterized in that described before the transition of each described ant process constantly to self the taboo table and share the step that the taboo table safeguards and mainly may further comprise the steps:
The link of no wavelength available in the current available link set is added described shared taboo table;
With another end points except that current point in the set of current available link is that described ant process adds described self taboo table through the link of point;
6, method according to claim 4, it is characterized in that described after the transition of each described ant process constantly to self the taboo table and share the step that the taboo table safeguards and mainly may further comprise the steps:
Selected transition link is added described shared taboo table;
To add described shared taboo table at all links of same shared risk link group with described selected transition link.
Priority Applications (1)
| Application Number | Priority Date | Filing Date | Title |
|---|---|---|---|
| CNA2009100791280A CN101599778A (en) | 2009-03-04 | 2009-03-04 | Designated lane guard method in a kind of WDM optical network plan |
Applications Claiming Priority (1)
| Application Number | Priority Date | Filing Date | Title |
|---|---|---|---|
| CNA2009100791280A CN101599778A (en) | 2009-03-04 | 2009-03-04 | Designated lane guard method in a kind of WDM optical network plan |
Publications (1)
| Publication Number | Publication Date |
|---|---|
| CN101599778A true CN101599778A (en) | 2009-12-09 |
Family
ID=41421055
Family Applications (1)
| Application Number | Title | Priority Date | Filing Date |
|---|---|---|---|
| CNA2009100791280A Pending CN101599778A (en) | 2009-03-04 | 2009-03-04 | Designated lane guard method in a kind of WDM optical network plan |
Country Status (1)
| Country | Link |
|---|---|
| CN (1) | CN101599778A (en) |
Cited By (2)
| Publication number | Priority date | Publication date | Assignee | Title |
|---|---|---|---|---|
| WO2010149093A1 (en) * | 2009-12-15 | 2010-12-29 | 中兴通讯股份有限公司 | Wavelength routing method and system based on wavelength division multiplexing optical network |
| CN102196325A (en) * | 2011-05-24 | 2011-09-21 | 南京邮电大学 | Ant-colony-system-based dynamic routing and wavelength assignment method for optical network |
-
2009
- 2009-03-04 CN CNA2009100791280A patent/CN101599778A/en active Pending
Cited By (4)
| Publication number | Priority date | Publication date | Assignee | Title |
|---|---|---|---|---|
| WO2010149093A1 (en) * | 2009-12-15 | 2010-12-29 | 中兴通讯股份有限公司 | Wavelength routing method and system based on wavelength division multiplexing optical network |
| CN101711002B (en) * | 2009-12-15 | 2013-03-27 | 中兴通讯股份有限公司 | Wavelength routing method and system based on wavelength division multiplexing optical network |
| CN102196325A (en) * | 2011-05-24 | 2011-09-21 | 南京邮电大学 | Ant-colony-system-based dynamic routing and wavelength assignment method for optical network |
| CN102196325B (en) * | 2011-05-24 | 2013-12-04 | 南京邮电大学 | Ant-colony-system-based dynamic routing and wavelength assignment method for optical network |
Similar Documents
| Publication | Publication Date | Title |
|---|---|---|
| CN101052235B (en) | Business combing method and device for ASON special protection | |
| JP4837765B2 (en) | Resource management and recursive route calculation method and apparatus necessary for multi-tier resource transfer network route calculation | |
| CN105357124B (en) | A kind of MapReduce bandwidth optimization methods | |
| CN101883293A (en) | Method and device for realizing K optimal path algorithm under condition of multiple sides among neighboring nodes | |
| CN101808254B (en) | Static routing and wavelength allocation method based on layered graph | |
| CN108737011B (en) | The Wavelength allocation method of reduction crosstalk based on ant group algorithm | |
| Yuan et al. | A RMSA algorithm for elastic optical network with a tradeoff between consumed resources and distance to boundary | |
| Xiao et al. | Service-oriented DU-CU placement using reinforcement learning in 5G/B5G converged wireless-optical networks | |
| Ngo et al. | An ant-based approach for dynamic RWA in optical WDM networks | |
| CN102196325B (en) | Ant-colony-system-based dynamic routing and wavelength assignment method for optical network | |
| Yu et al. | A deep learning based RSA strategy for elastic optical networks | |
| CN102546380B (en) | Modified tree-based multicast routing scheme | |
| CN112104554A (en) | Comprehensive data network service differentiation protection method based on link failure | |
| Barpanda et al. | Genetic Algorithm techniques to solve Routing and Wavelength Assignment problem in Wavelength Division Multiplexing all-optical networks | |
| CN101599778A (en) | Designated lane guard method in a kind of WDM optical network plan | |
| CN105007223A (en) | Optical network dynamic multicast routing wavelength allocation method based on optical hierarchical structure | |
| CN102025615B (en) | Method and device for planning paths of small-granularity services in optical communication network | |
| CN101909223A (en) | Resource-based WDM optical network path protective P-cycle optimized configuration method | |
| Guo et al. | Multicast multi-granular grooming based on integrated auxiliary grooming graph in optical networks | |
| CN114726434B (en) | Millisecond-level rapid path-finding method suitable for large-scale optical network | |
| Ngo et al. | Adaptive routing and wavelength assignment using ant-based algorithm | |
| Kumar et al. | Traffic and fragmentation aware algorithm for routing and spectrum assignment in Elastic Optical Network (EON) | |
| CN201947293U (en) | Network, system and processor using multicast route based on tree | |
| CN110138444A (en) | A kind of shared guard method of the multi-area optical network dynamic multicast based on fuzzy game | |
| CN103490810A (en) | Wavelength conversion range limited 2-hop-segment-based dynamic optical multicast P-ring segment protection method |
Legal Events
| Date | Code | Title | Description |
|---|---|---|---|
| C06 | Publication | ||
| PB01 | Publication | ||
| C02 | Deemed withdrawal of patent application after publication (patent law 2001) | ||
| WD01 | Invention patent application deemed withdrawn after publication |
Application publication date: 20091209 |