Reliable optical path transmission method for shared backup path protection and spectrum fragment sensing service

Abstract

The invention relates to a shared backup path protection and spectrum fragment sensing service reliable optical path transmission method in an elastic optical network, and belongs to the technical field of optical fiber communication. According to the method disclosed by the invention, a service protection and spectrum fragment sensing resource allocation method is designed aiming at possible faults of an optical fiber link of an elastic optical network; in the service path selection stage, the fault probability and resources of the path are considered, and a path cost function is designed to determine a service transmission path; resource allocation is divided into two stages: in the resource partitioning stage, a dynamic spectrum partitioning method considering path resources and fault probability is adopted, and in the path resource allocation stage, a spectrum window allocation formula for sensing fragments in working resources and a spectrum window allocation formula for sensing fragments in backup resources and backup resource sharing are designed; when a link fault occurs, a method of activating a backup path or rerouting is used for recovering the service. The method can improve the reliability of service transmission and the utilization rate of spectrum resources.

Description

Technical Field

[0001] The present invention belongs to the technical field of optical fiber communication and reliable transmission of optical signals, and relates to a reliable optical path transmission method for shared backup path protection and spectrum fragmentation perception services. Background Art

[0002] In the digital age, with the emergence of bandwidth-intensive applications such as the Internet of Things, 5G mobile services, and high-definition video calls, Internet services are growing exponentially, making limited network resources even more scarce. Optical networks need to continuously improve their flexible and dynamic transmission capabilities to adapt to complex and changing networks. Improving network resource utilization has become urgent. Elastic Optical Networks (EONs) can dynamically allocate spectrum resources according to the bandwidth requirements of different services, thereby improving the flexibility and utilization of resource allocation, and are considered to be an effective solution to improve network resource utilization.

[0003] During the service transmission process of EONs, it is inevitable that EONs will be damaged by the physical layer. For example, failures caused by natural disasters and accidental cutting of network optical fibers by humans will cause transmission link failures. Any optical path failure may cause a large amount of data loss or service interruption. Therefore, the research on reliable service transmission is crucial. Even if the reliable transmission of services is guaranteed, the reasonable allocation of resources for service transmission must still be considered. Since service requests in EONs are dynamically established and terminated, and optical paths are also dynamically established and released, discontinuous resource fragments will appear in the link, which will directly affect the utilization of spectrum resources and network performance, and will also lead to waste of spectrum resources. Subsequent services often find it difficult to use these fragmented frequency slots, and the spectrum utilization rate in the network will deteriorate. Therefore, when allocating resources, it should be ensured that the allocated resources not only meet the service bandwidth requirements, but also avoid the generation of additional fragments, thereby effectively reducing the blocking probability of services.

[0004] After EONs encounter link failure, protection and recovery methods can be used to ensure that services continue to be transmitted. However, the recovery method requires re-finding paths and allocating resources for services, which takes too long. Therefore, the use of protection methods will greatly reduce the time consumption. In the protection method, according to whether the protection resources can be shared, it can be divided into dedicated path protection and shared backup path protection. Shared backup path protection allows backup resources to be shared between multiple protection optical paths. Compared with dedicated path protection, shared backup path protection can save more resources. The use of this method can also greatly save resource consumption. In the resource allocation stage, we can focus on the resource fragmentation problem and effectively improve the successful transmission rate of services. The fragmentation-aware method is used. The fragmentation-aware solution is to reduce the generation of spectrum fragmentation by preventing the fragmentation of incoming service requests during resource allocation. The focus is on the design of fragmentation metrics, which needs to consider various factors, such as considering fragments of various sizes and the number of fragments. In EONs based on shared backup path protection, in addition to the fragmentation metric of the working path, resource sharing and resource fragmentation on the backup path must also be considered. Summary of the invention

[0005] In view of this, an object of the present invention is to provide a shared backup path protection and spectrum fragmentation aware service reliable optical path transmission method for ensuring reliable transmission of services.

[0006] In order to achieve the above object, the present invention provides the following technical solutions:

[0007] In view of the problem that services in elastic optical networks may suffer data loss due to link failures and spectrum resource allocation of services may easily cause spectrum fragmentation, a method for reliable optical path transmission of services with shared backup path protection and spectrum fragmentation awareness is proposed. In this algorithm, based on the shared backup path protection method, the failure probability of the service path and the number of path resources are weighed, and the path cost function of the path combination is designed to determine the service transmission path. A dynamic spectrum partitioning method is adopted, so that the service working path and the backup path can adopt different resource allocation methods, and the service can make full use of the resources in the existing resource area to avoid the generation of excessive fragments. In terms of resource allocation, a spectrum window (SW) allocation formula that is aware of fragmentation in working resources and a spectrum window allocation formula that is aware of fragmentation in backup resources and sharing of backup resources are proposed, both of which aim to reduce the generation of spectrum fragmentation in the network. Finally, when a link failure occurs, a fault response method is proposed to restore the service as much as possible. The method specifically includes the following steps:

[0008] S1: Initialize network service requests, network topology, link failure probability, and divide link resources into three parts as the initial spectrum partitioning scheme, with protection frequency gap GB = 1 frequency gap, C f =12.5GHz, assuming α1, α2, ω1, ω2, K=3, Go to S2;

[0009] S2: Service r(s, d, B) arrives. According to the service source node s, destination node d and transmission rate B, the service working path set and backup path set are obtained. According to the service path failure probability, path hop count, path resources and path modulation format, the path cost function of the service working path and backup path combination is calculated and put into the set P in ascending order. cost , select the path combination corresponding to the minimum value as the working path and backup path of the service, calculate the number of frequency slots required for the service, and go to S3;

[0010] S3: Check whether there are available spectrum blocks that meet the service rate requirements on the selected path, create a sliding spectrum window on the selected path, if there are available spectrum blocks, calculate the path fragmentation rate of each spectrum window for the service working path and the backup path of the non-shared backup resources, select the spectrum window position corresponding to the minimum path fragmentation rate to allocate the service working path resources, calculate the spectrum adaptability function of each spectrum window for the backup path of the service shareable backup resources, select the spectrum window position corresponding to the maximum spectrum adaptability function to allocate the service backup path resources, if the service working path and the backup path resources can be successfully allocated, go to S5, otherwise, block the service and end the algorithm; if there are no available spectrum blocks, go to S4;

[0011] S4: If there is no idle spectrum block that meets the service requirements on the selected path, the dynamic spectrum partitioning method is executed according to the path resource occupancy of the service and the average path failure probability. After the execution, the remaining links in the network are combined as a new network spectrum partitioning scheme, and then the process goes to S3 for service resource allocation;

[0012] S5: When the working path and backup path resources of the service are successfully allocated and the service is successfully transmitted until completion, the algorithm ends; when a link failure occurs on the working path of the service, the backup path of the service is activated for transmission. If the service cannot be switched to the backup path, a rerouting fault response method is used to restore the service.

[0013] Furthermore, the method for selecting the service working path and the backup path in S2 specifically includes the following steps:

[0014] S201: Based on the elastic optical network topology, the source node s of the service, and the destination node d of the service, a K shortest path algorithm is executed to find K shortest working paths for the service, and the shortest working paths are stored in a candidate working path set P. w In which, K=3, go to S202;

[0015] S202: For each working path p wi, temporarily delete the working path in the network topology, obtain a new elastic optical network topology, execute the K shortest path algorithm on the topology again, obtain a backup path that does not intersect with the working path link, and store it in the candidate backup path set P b , finally obtaining K pairs of working path and backup path combinations, and then going to S203;

[0016] S203: Calculate the path cost function of the combination of the working path and the backup path of the service according to the path failure probability, the number of path hops, the path idle resources, and the path modulation level of the service Store in ascending order into set P cost , go to S204;

[0017] The calculation formula of the path cost function of the path combination is:

[0018]

[0019] In the above formula, Indicates the working path p selected by a certain service r wi With backup path p bi The path cost function of the path combination; hop pwi Represents the working path p wi The number of hops pbi Indicates the backup path p bi ; α1 is the path weight coefficient, α2 is the reliability weight coefficient, satisfying α1+α2=1; M pwi Represents the working path p wi Modulation level; M pbi Indicates the backup path p bi The modulation level; Represents the working path p wi The number of idle frequency slots; Indicates the backup path p bi The number of idle frequency slots; R(r) represents the reliability of service r;

[0020] The reliability calculation formula of service r is:

[0021] R(r)=1-P(p wi ,p bi )-P(p wi )

[0022] In the above formula, P(p wi ,p bi ) represents the probability of service transmission failure caused by simultaneous failure of the working path and the backup path; P(p wi ) represents the probability of service transmission failure when the working path fails and the backup path is normal;

[0023] The probability calculation formula for service transmission failure caused by simultaneous failure of the working path and the backup path is:

[0024]

[0025] In the above formula, p wi represents a working path of business r; p bi represents a backup path of service r; p a,b represents the failure probability of link (a, b) on the working path; p m,n represents the failure probability of link (m,n) on the backup path;

[0026] The probability calculation formula for service transmission failure when the working path fails and the backup path is normal is:

[0027]

[0028] In the above formula, N s Indicates the number of services competing with the current service for backup resources;

[0029] S204: Select set P cost The path combination corresponding to the minimum value is used as the working path and backup path of the service, and the process goes to S205;

[0030] S205: Calculate the number of frequency slots N required for the service working path wi and the number of frequency slots required for the backup path N bi .

[0031] The number of frequency slots required for the service is calculated as follows:

[0032]

[0033] In the above formula, B represents the bandwidth required by the service (Gbps); C f is the unit frequency slot bandwidth, which is 12.5 GHz; M is the modulation level; GB is the protection frequency slot, which is 1 frequency slot;

[0034] The relationship between modulation level and modulation format is:

[0035]

[0036] Furthermore, the S3 allocates service path resources, specifically including the following steps:

[0037] S301: If the working resource area of ​​the service working path has available spectrum blocks Not less than the number of frequency slots N required by the service working path wi ,Right now Create a sliding spectrum window on the selected path, calculate the path fragmentation rate of each spectrum window, select the spectrum window position corresponding to the minimum path fragmentation rate to allocate the working path resources of the service, and go to step S202; otherwise, the service working path resource allocation fails;

[0038] The path fragmentation rate formula is implemented in the following way:

[0039]

[0040] Among them, ψ SWi represents the path fragmentation rate of the i-th spectrum window from the left to the right in the path resource; Indicates the number of idle spectrum blocks in the path working resource area; represents the number of frequency slots of the ith idle spectrum block in the path working resource area; Indicates the number of frequency slots of the smallest idle spectrum block in the path working resource area;

[0041] S302: If the backup resource area of ​​the service backup path has available spectrum blocks Not less than the number of frequency slots N required by the service backup path bi ,Right now A sliding spectrum window is created on the selected path. For the backup path of the service that can share backup resources, the spectrum adaptability function of each spectrum window is calculated, and the spectrum window position corresponding to the maximum value of the spectrum adaptability function is selected to allocate the backup path resources of the service. For the backup path of the service that cannot share backup resources, the path fragmentation rate after each spectrum window is allocated is calculated, and the spectrum window position corresponding to the minimum value of the path fragmentation rate is selected to allocate the working path resources of the service, and then go to step S203; otherwise, the service backup path resource allocation fails;

[0042] The path spectrum adaptability function formula is implemented in the following way:

[0043]

[0044] in, SFS represents the spectrum adaptability function of the i-th spectrum window from right to left in the path resource; ol Indicates the total number of frequency slots occupied by services after sharing backup resources; Indicates the number of idle spectrum blocks in the path backup resource area; represents the number of frequency slots of the ith idle spectrum block in the path backup resource area;

[0045] S303: If the working path and backup path resources of the service are successfully allocated, the service is successfully transmitted, and the process goes to S5; otherwise, the service transmission fails and the service is blocked.

[0046] Furthermore, the dynamic spectrum partitioning method in S4 specifically includes the following steps:

[0047] S401: If the business r cannot find available resources, the resource area where the available resources cannot be found will be expanded, and the remaining resource areas will not be expanded, and the process will go to S402;

[0048] S402: Obtain the number N of working path links for service r w The number of backup path links N b , calculate the average link failure probability of the working path and the backup path of r, and go to S403;

[0049] The average failure probability formula of the working path link is implemented in the following way:

[0050]

[0051] in, Represents the working path p of business r wi The average failure probability of the link;

[0052] The average failure probability formula of the backup path link is implemented in the following way:

[0053]

[0054] in, Indicates the backup path p for service r bi The average failure probability of the link;

[0055] S403: jointly consider the link resource situation, resource occupancy and average link failure probability of the service, respectively calculate the working resource area measurement function and the backup resource area measurement function, and then go to S404;

[0056] The work resource area measurement function is implemented in the following ways:

[0057]

[0058] Among them, θ w is the working resource area measurement function; ω1 is the resource weight coefficient, ω2 is the failure probability weight coefficient, satisfying ω1+ω2=1; N w Indicates the number of working path links; represents the number of frequency slots occupied by the working resource area of ​​link (a, b) on working path r; and Respectively represent the total number of frequency slots in the working resource area and backup resource area of ​​the link (a, b) on the working path r;

[0059] The backup resource area measurement function is implemented in the following ways:

[0060]

[0061] Among them, θ b Measuring function for backup resource area; N b Indicates the number of backup path links; represents the number of frequency slots occupied by the backup resource area of ​​link (m,n) on the backup path r;

[0062] S404: When θ w ≥θ b When , the capacity of the working resource area of ​​each link of the working and backup paths of r increases, the working resource area expands to the right, and the resources of 1 / 3 of the public resource area are merged into the working resource area. The calculation result is rounded up and the process goes to S405;

[0063] The calculation method of the capacity expansion of the working resource area is specifically implemented in the following way:

[0064]

[0065] in, represents the total number of frequency slots in the working resource area of ​​the ith link on the path of service r in the new spectrum partitioning scheme; They respectively represent the total number of frequency slots in the working resource area and the public resource area of ​​the ith link on the path of service r in the original spectrum partitioning scheme;

[0066] S405: When θ w <θ b When r, the capacity of the backup resource area of ​​each link of the working and backup paths increases, the backup resource area expands to the left, and the resources occupying 1 / 3 of the public resource area are merged into the backup resource area. The calculation result is rounded up and the process goes to S406;

[0067] The calculation method for the capacity expansion of the backup resource area is specifically implemented in the following way:

[0068]

[0069] in, FS represents the total number of frequency slots in the backup resource area of ​​the ith link on the path of service r in the original spectrum partitioning scheme; li B 'respectively represent the total number of frequency slots in the backup resource area of ​​the ith link on the path of service r in the new spectrum partitioning scheme;

[0070] S406: The adjusted links are combined with the remaining unadjusted links in the network as a new network spectrum partitioning scheme, and the process goes to S3.

[0071] Furthermore, the fault handling method in S5 specifically includes the following steps:

[0072] S501: When the working path and backup path resources of the service are successfully allocated, the service can be transmitted normally until the end. If the service transmission is successful, the algorithm ends. Otherwise, go to S502;

[0073] S502: When a link failure occurs on the working path of the service, a backup path of the service is activated to ensure normal transmission of the service, and the process goes to S503;

[0074] S503: The service selection P of the backup path cannot be activated cost The second one The corresponding path combination performs rerouting. BRIEF DESCRIPTION OF THE DRAWINGS

[0075] In order to make the purpose, technical solution and beneficial effects of the present invention clearer, the present invention provides the following drawings for illustration:

[0076] Figure 1 It is a schematic diagram of resource allocation for business work paths;

[0077] Figure 2 A schematic diagram of resource allocation for business backup paths;

[0078] Figure 3 is a schematic diagram of a dynamic spectrum partitioning method;

[0079] Figure 4 Flowchart of a reliable optical path transmission method for shared backup path protection and spectrum fragmentation awareness services; DETAILED DESCRIPTION

[0080] The preferred embodiments of the present invention will be described in detail below with reference to the accompanying drawings.

[0081] The shared backup path protection and spectrum fragmentation-aware service reliable optical path transmission method provided by the present invention aims to reduce the resource fragmentation rate as much as possible, avoid the generation of resource fragmentation, and ensure the reliable transmission of services under the premise that link failures may exist in the elastic optical network. The present invention can also be implemented or applied through other different specific implementation methods, and the details in this specification can also be modified or changed in various ways based on different viewpoints and applications without departing from the spirit of the present invention. It should be noted that the illustrations provided in the following embodiments only illustrate the basic concept of the present invention in a schematic manner, and the following embodiments and the features in the embodiments can be combined with each other without conflict.

[0082] Among them, the figures are only used for illustrative explanations, and they only represent schematic diagrams rather than actual pictures, and should not be understood as limitations on the present invention. In order to better illustrate the embodiments of the present invention, some parts in the figures may be omitted, enlarged or reduced, and do not represent the size of the actual product. For those skilled in the art, it is understandable that some well-known structures in the figures and their descriptions may be omitted.

[0083] As attached Figure 1 The service working path resource allocation diagram shown in the figure adopts the method of creating spectrum windows to select the appropriate resource allocation position for the service. There is a path p1 with a total of 10 frequency slot resources. Service r1 is on it. The existing service r2 arrives and requires 2 frequency slots. If SW1 or SW3 is used to allocate resources for r2, therefore, If SW2 is used to allocate resources for r2, therefore, If SW4 is used to allocate resources for r2, therefore, Since 0.833<2.5<7, SW1 or SW3 is the most suitable location to allocate resources to r2, which can leave more continuous idle spectrum blocks for subsequent services.

[0084] As attached Figure 2 The service backup path resource allocation diagram is shown in the figure. There is a link l1 with a total of 10 frequency slot resources. There are already backup resources for three services r1, r2, and r3. The existing service r4 arrives. Its working path does not intersect with the working path of r3, and there is a common link l1. Therefore, they can share resources. Assuming that the number of backup frequency slots required by r4 is 3, r3 and r4 can share the backup resources of 2 frequency slots. Two spectrum windows (SW1 and SW2) can be found on this link to allocate resources to r4. When SW1 is used to allocate backup resources for r4, r3 and r4 occupy a total of 4 frequency slots, and SFS ol =4, therefore, When SW2 is used to allocate r4 backup resources, SFS ol =4, therefore, Since 0.625>0.25, the resource allocation position of SW1 is more suitable for resource sharing. After allocation, less resource fragmentation is caused, and the idle spectrum blocks are more continuous, so that more continuous idle spectrum blocks can be reserved for subsequent services.

[0085] As attached Figure 3 The spectrum partitioning method is shown in Figure 1. Each link has 30 frequency slots, and each resource area has 10 frequency slots. A new service arrives, and its working path pw For BCF, 3 frequency slots are required, and its backup path p b For BDEF, 4 frequency slots are required. The backup resource can be shared or not. Therefore, the backup resource area can allocate resources to it or not. However, the working resource area on the working path of the service cannot allocate resources to it. Therefore, the proposed spectrum partitioning method is needed. The failure probability of these links can be obtained from the figure, thus obtaining N w =2,N b =3; at the same time, through the frequency slot occupancy of the path, we can get Therefore, we can substitute into the formula to get Since 0.928>0.816, the working resource areas of the service working path and the backup path are expanded to occupy 1 / 3 of the total frequency slots of the public resource area. Therefore, after the proposed method is applied to the working resource areas of the service working path and the backup path, the total number of frequency slots of the working resource areas becomes This method solves the problem of new services being unable to find available resources as much as possible, and takes into account the impact of failures on services.

[0086] The following will be combined with the attached Figure 4 The resource allocation method for reliable transmission of services based on shared backup path protection and fragmentation awareness of the present invention is introduced in more detail. The specific overall process can be divided into the following steps:

[0087] Input: Initialize network service request, network topology, link resources are divided into three parts as the initial spectrum partitioning scheme,

[0088] Link failure probability, protection frequency gap GB = 1 frequency gap, C f =12.5GHz, assuming α1, α2, ω1, ω2, K=3,

[0089]

[0090] Output: business path selection results, business resource allocation results, business transmission results

[0091] S1: Based on the source node s and destination node d of the service, execute the K shortest path algorithm to find K shortest working paths for the service and store them in the candidate working path set P w In which, K=3, go to S2;

[0092] S2: For each working path p wi , temporarily delete the working path from the network topology, obtain a new elastic optical network topology, and then execute the K shortest path algorithm on the topology to obtain a backup path that does not intersect with the working path link, and store it in the candidate backup path set Pb , finally get K pairs of working path and backup path combinations, and go to S3;

[0093] S3: Calculate the path cost function of the service's working path and backup path combination based on the service's path failure probability, path hop count, path idle resources, and path modulation level Put them into set P in ascending order cost , go to S4;

[0094] S4: Calculate the number of frequency slots N required for the service working path wi and the number of frequency slots required for the backup path N bi , go to S5;

[0095] S5: Available spectrum blocks in the working resource area of ​​the service working path Not less than the number of frequency slots N required by the service working path wi ,Right now Create a sliding spectrum window on the selected path, calculate the path fragmentation rate of each spectrum window, select the spectrum window position corresponding to the minimum path fragmentation rate to allocate the working path resources of the service, and go to S6;

[0096] S6: Available spectrum blocks in the backup resource area of ​​the service backup path Not less than the number of frequency slots N required by the service backup path bi ,Right now A sliding spectrum window is created on the selected path. For the backup path of the service that can share backup resources, the spectrum adaptability function of each spectrum window is calculated, and the spectrum window position corresponding to the maximum value of the spectrum adaptability function is selected to allocate the backup path resources of the service. For the backup path of the service that cannot share backup resources, the path fragmentation rate after each spectrum window is allocated is calculated, and the spectrum window position corresponding to the minimum value of the path fragmentation rate is selected to allocate the working path resources of the service, and then go to S12;

[0097] S7: If the business r cannot find available resources, the resource area where the available resources cannot be found will be expanded, and the remaining resource areas will not be expanded, and the process will go to S8;

[0098] S8: Get the number of working path links N for service r w The number of backup path links N b , calculate the average link failure probability of the working path and backup path of r, and go to S9;

[0099] S9: jointly consider the link resource situation, resource occupancy and average link failure probability of the service, respectively calculate the working resource area measurement function and the backup resource area measurement function to execute the dynamic spectrum partitioning method, and then go to S10;

[0100] S10: When θw ≥θ b When θ is θ, the capacity of the working resource area of ​​each link of the working and backup paths of r increases, the working resource area expands to the right, and 1 / 3 of the resources in the public resource area are merged into the working resource area. w <θ b When r, the capacity of the backup resource area of ​​each link of the working and backup paths of r increases, the backup resource area expands to the left, and the resources occupying 1 / 3 of the public resource area are merged into the backup resource area. The calculation result is rounded up and the process goes to S11;

[0101] S11: The adjusted links are combined with the remaining unadjusted links in the network as a new network spectrum partitioning scheme, and then the process goes to S5;

[0102] S12: When the working path and backup path resources of the service are successfully allocated, the service does not encounter link failures and can be transmitted normally until the end, the service transmission is successful, the algorithm ends, and the result is returned; otherwise, go to S13;

[0103] S13: When a link failure occurs on the working path of the service, the backup path of the service is activated to ensure normal transmission of the service, and then the process goes to S14;

[0104] S14: Service selection P of the backup path cannot be activated cost The second one The corresponding path combination performs rerouting to restore the service.

[0105] Finally, it should be noted that the above preferred embodiments are only used to illustrate the technical solutions of the present invention rather than to limit it. Although the present invention has been described in detail through the above preferred embodiments, those skilled in the art should understand that various changes can be made in form and details without departing from the scope defined by the claims of the present invention.

Claims

1. A method for reliable optical path transmission of shared backup path protection and spectrum fragmentation aware services, characterized in that: In this method, a shared backup path protection method is used to protect the service; in the service path selection stage, the failure probability and resources of the path are considered, and a path cost function is designed to determine the service transmission path; After determining the transmission path, a dynamic spectrum partitioning method that considers path resources and failure probability is used to implement different resource allocation methods for different paths; in the resource allocation stage, a spectrum window (SW) allocation formula that perceives fragments in working resources and a spectrum window allocation formula that perceives fragments and backup resources sharing in backup resources are designed; when a link failure occurs, a backup path is activated or re-routed to restore the service. The method specifically includes the following steps: S1: According to the elastic optical network topology, initialize the failure probability of the link, divide all link resources into three spectrum resource areas as the initial spectrum partitioning scheme; calculate the path cost function of the combination of the working path and the backup path of the service described in claim 2 according to the path failure probability, path hop number, path resource, and path modulation format of the service, and put them into the set P in ascending order cost , select P cost The path combination corresponding to the minimum value is selected as the working path and backup path of the service, and the number of frequency slots required by the service is calculated; The calculation formula of the path cost function of the path combination is: In the above formula, Indicates the working path p selected by a certain service r wi With backup path p bi The path cost function of the path combination; Represents the working path p wi The number of hops; Indicates the backup path p bi ; α1 is the path weight coefficient, α2 is the reliability weight coefficient, satisfying α1+α2=1; Represents the working path p wi The modulation level; Indicates the backup path p bi The modulation level; Represents the working path p wi The number of idle frequency slots; Indicates the backup path p bi The number of idle frequency slots; R(r) represents the reliability of service r; The reliability calculation formula of service r is: R(r)=1-P(p wi ,p bi )-P(p wi ) In the above formula, P(p wi ,p bi ) represents the probability of service transmission failure caused by simultaneous failure of the working path and the backup path; P(p wi ) represents the probability of service transmission failure when the working path fails and the backup path is normal; The probability calculation formula for service transmission failure caused by simultaneous failure of the working path and the backup path is: In the above formula, p wi represents a working path of business r; p bi represents a backup path of service r; p a,b represents the failure probability of link (a, b) on the working path; p m,n represents the failure probability of link (m,n) on the backup path; The probability calculation formula for service transmission failure when the working path fails and the backup path is normal is: In the above formula, N s Indicates the number of services competing with the current service for backup resources; The calculation formula for the number of frequency slots required for the service path is: Where B represents the bandwidth required by the service (Gbps); C f is the unit frequency slot bandwidth, which is 12.5 GHz; M is the modulation level; GB is the protection frequency slot, which is 1 frequency slot; S2: using the method of claim 3 to allocate resources to the working path and backup path of the service, finding available spectrum blocks in the service path that meet the frequency gap size required by the service, creating a sliding spectrum window on the selected path, calculating the path fragmentation rate of each spectrum window for the service working path and the backup path of the non-shared backup resources, and selecting the spectrum window position corresponding to the minimum path fragmentation rate to allocate the service working path resources; for the backup path of the service that can share backup resources, calculating the spectrum adaptability function of each spectrum window, and selecting the spectrum window position corresponding to the maximum spectrum adaptability function to allocate the service backup path resources; S3: If there is no available idle spectrum block on the service path, the path resource occupancy and average path failure probability of the service are counted, and the dynamic spectrum partitioning method according to claim 4 is adopted to expand the capacity of the spectrum resource area; S4: If a link failure occurs in the working path of the service, the fault coping method of activating the backup path or rerouting as described in claim 5 is adopted to determine a new working path for the service to restore the service.

2. The method for reliable optical path transmission of shared backup path protection and spectrum fragmentation-aware services according to claim 1, characterized in that: The initial spectrum partitioning scheme, the selection of the service working path and the backup path in S1 specifically include the following steps: S101: Divide all link resources in the elastic optical network into three spectrum resource areas, the left 1 / 3 frequency slot resources of the link resources are the working resource area, the right 1 / 3 frequency slot resources of the link resources are the backup resource area, and the middle 1 / 3 frequency slot resources of the link resources are the public resource area. The frequency slots in the public resource area cannot be used at the beginning. The working resource area can only allocate resources for the working path of the service, and the backup resource area can only allocate resources for the backup path of the service, and then go to step S102; S102: After the service arrives, find K shortest working paths for the service according to the K shortest path algorithm, where K=3, and go to step S103; S103: for each working path, delete the working path from the elastic optical network topology to obtain a new elastic optical network topology, execute the K shortest path algorithm on the topology, thereby obtaining a backup path that does not intersect with the working path link, and finally obtain K pairs of working path and backup path combinations, and go to step S104; S104: Calculate the path cost functions based on the path combinations and put them into the set P in ascending order. cost , go to step S105; S105: Choose P cost The path combination corresponding to the minimum value is used as the working path and the backup path of the service, and the process goes to step S106; S106: Calculate the number of frequency slots required for the service and go to S2.

3. The method for reliable optical path transmission of shared backup path protection and spectrum fragmentation-aware services according to claim 1, characterized in that: The method S2 for allocating resources to the working path and the backup path of the service specifically includes the following steps: S201: If the working resource area of ​​the service working path has available spectrum blocks Not less than the number of frequency slots N required by the service working path wi ,Right now Create a sliding spectrum window on the selected path, calculate the path fragmentation rate of each spectrum window, select the spectrum window position corresponding to the minimum path fragmentation rate to allocate the working path resources of the service, and go to step S202; otherwise, the service working path resource allocation fails; The path fragmentation rate formula is calculated in the following way: in, represents the path fragmentation rate of the i-th spectrum window from the left to the right in the path resource; Indicates the number of idle spectrum blocks in the path working resource area; represents the number of frequency slots of the ith idle spectrum block in the path working resource area; Indicates the number of frequency slots of the smallest idle spectrum block in the path working resource area; S202: If the backup resource area of ​​the service backup path has available spectrum blocks Not less than the number of frequency slots N required by the service backup path bi ,Right now A sliding spectrum window is created on the selected path. For the backup path of the service that can share backup resources, the spectrum adaptability function of each spectrum window is calculated, and the spectrum window position corresponding to the maximum value of the spectrum adaptability function is selected to allocate the backup path resources of the service. For the backup path of the service that cannot share backup resources, the path fragmentation rate after each spectrum window is allocated is calculated, and the spectrum window position corresponding to the minimum value of the path fragmentation rate is selected to allocate the working path resources of the service, and then go to step S203; otherwise, the service backup path resource allocation fails; The path spectrum adaptability function formula is calculated in the following way: in, SFS represents the spectrum adaptability function of the i-th spectrum window from right to left in the path resource; ol Indicates the total number of frequency slots occupied by services after sharing backup resources; Indicates the number of idle spectrum blocks in the path backup resource area; represents the number of frequency slots of the ith idle spectrum block in the path backup resource area; S203: If the working path and backup path resources of the service are successfully allocated, go to S4; otherwise, block the service.

4. The method for reliable optical path transmission of shared backup path protection and spectrum fragmentation-aware services according to claim 1, characterized in that: The dynamic spectrum partitioning method of S3 specifically includes the following steps: S301: If the new service cannot find available resources, the resource area where no available resources can be found will be expanded, and the process goes to step S302; S302: Calculate the working resource area measurement function and the backup resource area measurement function respectively according to the link failure probability, the number of links, and the link resource status of the service path, and then go to step S303; The work resource area measurement function is calculated in the following way: In the above formula, θ w is the working resource area measurement function; ω1 is the resource weight coefficient, ω2 is the failure probability weight coefficient, satisfying ω1+ω2=1; N w Indicates the number of working path links; represents the number of frequency slots occupied by the working resource area of ​​link (a, b) on working path r; and Respectively represent the total number of frequency slots in the working resource area and backup resource area of ​​the link (a, b) on the working path r; Represents the working path p of business r wi The average failure probability of the link; Indicates the backup path p for service r bi The average failure probability of the link; The backup resource area measurement function is calculated in the following way: In the above formula, θ b Measuring function for backup resource area; N b Indicates the number of backup path links; represents the number of frequency slots occupied by the backup resource area of ​​link (m,n) on the backup path r; Among them, business r working path p wi The average failure probability of a link is calculated as follows: Among them, service r backup path p bi The average failure probability of a link is calculated as follows: S303: When θ w ≥θ b When r, the working resource area capacity of each link of the working path and the backup path increases, the working resource area expands to the right, and the resources of 1 / 3 of the public resource area are merged into the working resource area. The calculation result is rounded up, and the process goes to step S304; The calculation method of the capacity expansion of the working resource area is as follows: In the above formula, represents the total number of frequency slots in the working resource area of ​​the ith link on the path of service r in the new spectrum partitioning scheme; They respectively represent the total number of frequency slots in the working resource area and the public resource area of ​​the ith link on the path of service r in the original spectrum partitioning scheme; S304: When θ w <θ b When , the backup resource area of ​​r is extended to the left, and the resources occupying 1 / 3 of the public resource area are merged into the backup resource area, and the calculation result is rounded up, and the process goes to step S305; The calculation method for the capacity expansion of the backup resource area is as follows: In the above formula, represents the total number of frequency slots in the backup resource area of ​​the ith link on the path of service r in the new spectrum partitioning scheme; represents the total number of frequency slots in the backup resource area of ​​the ith link on the path of service r in the original spectrum partitioning scheme; S305: After the spectrum partitioning scheme is adjusted, the remaining unadjusted links in the network are combined as a new network spectrum partitioning scheme and the process proceeds to S2.

5. The method for reliable optical path transmission of shared backup path protection and spectrum fragmentation-aware services according to claim 1, characterized in that: The fault response method in S4 specifically includes the following steps: S401: When the working path and backup path resources of the service are successfully allocated, the service can be transmitted normally until the end. If the service transmission is successful, the algorithm ends. Otherwise, go to step S402; S402: When a link failure occurs on the working path of the service, a backup path of the service is activated to ensure normal transmission of the service, and the process goes to step S403; S403: The service selection P of the backup path cannot be activated cost The second one The corresponding path combination performs rerouting.

Images