Method for multi-layer protection and restoration and resource allocation in an optical network

Abstract

The application discloses a multi-layer protection recovery and resource allocation method of an optical network, and relates to the technical field of network communication.The application comprehensively considers multi-service characteristics and network resource dynamic scheduling, and proposes a hybrid method of multi-routing selection and network service slicing, so that protection based on a WDM optical network in a complex network environment is realized.Meanwhile, a work domain and a protection domain are arranged in a bandwidth allocation period, and a flexible resource allocation scheme is adopted, so that network resource utilization can be improved, and the work route can be protected.

Description

Technical Field

[0001] This invention relates to the field of network communication technology, specifically to a multi-layer protection, recovery, and resource allocation method for optical networks. Background Technology

[0002] Currently, wavelength division multiplexing (WDM) optical networks are widely used as the carriers of various services in my country's JS and civilian scenarios. In the evolution of optical networks from opaque networks using all-electric switching (where optoelectronic-to-optical (O / E / O) conversion is performed at each node) to transparent networks using all-optical switching (where intermediate nodes do not perform O / E / O conversion), a fundamental issue of network survivability has emerged. How to protect and restore optical networks under specific network requirements is a critical problem that urgently needs to be solved. Currently, various protection and fault recovery methods have been researched and applied to ensure reliable information transmission in the event of single or multiple component failures. However, protection and recovery methods based on the multi-service characteristics, dynamic wavelength, and scheduling characteristics of networks have not yet been fully studied. Summary of the Invention

[0003] The purpose of this invention is to address the shortcomings of existing technologies by proposing a multi-layer protection, recovery, and resource allocation method for optical networks. This method comprehensively considers the characteristics of multiple services and the dynamic scheduling of network resources. When a service's working route fails, it switches the service route to multiple alternative paths, thereby protecting the services affected by the failure. This approach aims to meet the needs of WDM-based optical networks in various scenarios involving high capacity, low latency, and ubiquitous network operation.

[0004] The technical solution to achieve the objective of this invention is:

[0005] A method for multi-layer protection recovery and resource allocation in an optical network includes the following steps:

[0006] 1) Multi-layer protection: Multi-layer protection includes multiple route alternatives, network service slicing, and setting working domains and protection domains within the bandwidth allocation cycle;

[0007] 1.1 Multiple alternative routes: There are multiple alternative routes between the source node and the destination node. If the working route fails, it can be switched to an alternative route for protection in a timely manner.

[0008] 1.1.1 In a complex multi-node mesh topology optical network, the KSP algorithm with K shortest paths is used to calculate n fixed alternative routes for each pair of physical nodes in the physical network, and the routing information is stored in the routing information table. Thus, for a given node, the routing information table stores the routing information of any other node in the multi-node mesh topology optical network, and for a specified source node and destination node, at most n fixed alternative routes will be pre-calculated.

[0009] 1.1.2: Set the route weight value L. This weight value is positively correlated with the bandwidth resources α in the candidate routes and negatively correlated with the hop count β of the routes. The control layer of the multi-node Mesh topology optical network obtains the remaining bandwidth resources in each candidate route in real time, which means that the route weight value is updated dynamically in real time.

[0010] 1.2 Network service slicing: Services with different Quality of Service (QoS) are transmitted by different routes. If one working route fails, the other two working routes will not be affected.

[0011] 1.2.1: Differentiated quality of service exists in multi-node mesh-type optical networks:

[0012] (1) EF services with accelerated forwarding: require sufficient bandwidth, low packet loss rate, low latency and jitter;

[0013] (2) Ensure forwarded AF services: require sufficient bandwidth, low packet loss rate, but are not sensitive to latency and jitter;

[0014] (3) Best-effort forwarding BE service: If there are no special requirements, do your best as long as bandwidth resources allow;

[0015] 1.2.2: Based on the routing weight values ​​obtained in step 1.1.2, three routes are selected to ensure differentiated service QoS transmission, that is, three services use three mutually isolated routes;

[0016] 1.3: Setting up working and protection domains within the bandwidth allocation cycle: The protection domain is used for switching when other links fail;

[0017] 1.3.1: Allocate bandwidth resources in T-cycles, and set up working domains and protection domains. The working domain accounts for 70% for normal service transmission, while the protection domain accounts for 30% for protection, in case other links fail and need to be switched to.

[0018] 1.3.2: The same wavelength scheduling domain can be occupied by multiple working routes at the same time; therefore, reasonable allocation of wavelengths and time slots can effectively guarantee service transmission and reduce transmission delay. Available resources that meet the service bandwidth are allocated with a period of T, where available resources include wavelengths and time slots.

[0019] 1.3.3: Time slots on different wavelengths will be filled sequentially according to time order. Only when an earlier time slot is filled will a later time slot be used.

[0020] 1.3.4: In conjunction with service QoS, prioritize EF services that require sufficient bandwidth, low packet loss rate, low latency and jitter, that is, prioritize EF services in the earliest time slot, then satisfy AF services that only require sufficient bandwidth and low packet loss rate, and finally use the remaining bandwidth resources for BF services.

[0021] 2) Route recovery:

[0022] 2.1: When the working route fails, the control layer of the multi-node Mesh topology optical network sends a ranging signal to confirm whether all nodes in the multi-node Mesh topology optical network are online. If all nodes are online, it means that the optical fiber link in the working route has failed.

[0023] 2.2: If there is no response after multiple inquiries about whether the node is online, it means that the malfunction is caused by the node going offline;

[0024] 2.3: The control layer of a multi-node mesh-type topology optical network selects a new working route from the candidate routes based on the route weight value and calculates the bandwidth resource allocation in the route;

[0025] 2.4: If the remaining bandwidth resources of a single alternative route are insufficient to meet the service requirements, multiple alternative routes will be upgraded to working routes in stages to utilize the bandwidth resources of multiple alternative routes to meet the service requirements.

[0026] 2.5: After completing steps 2.1 to 2.4, the control layer notifies all nodes in the route, enabling all nodes to prepare for the new route switch, thereby performing the route switch and realizing route protection.

[0027] This method takes into account the characteristics of multiple services and the dynamic scheduling of network resources. When a service working route fails, it switches the service route to multiple alternative paths, thereby protecting the services affected by the failure, so as to meet the multi-scenario application of WDM-based optical networks in terms of high capacity, low experimental cost and ubiquitous bearer. Attached Figure Description

[0028] Figure 1 This is a schematic diagram of a multi-layered network protection process as illustrated in the embodiment.

[0029] Figure 2 This is a schematic diagram of the network recovery process in an embodiment;

[0030] Figure 3 This is a schematic diagram of multiple alternative routes for an embodiment;

[0031] Figure 4 The network service slicing example is illustrated in the diagram.

[0032] Figure 5This is a schematic diagram of bandwidth resource allocation in an embodiment;

[0033] Figure 6 This is a schematic diagram of network recovery as an example. Detailed Implementation

[0034] The invention will now be described in further detail with reference to the accompanying drawings and specific embodiments, but this is not intended to limit the invention.

[0035] Example:

[0036] A multi-node mesh topology optical network based on WDM optical networks is divided into a control layer and a physical layer. The control layer is primarily responsible for routing planning and bandwidth resource scheduling, while the physical layer handles the transmission of service data. For example... Figure 3 As shown, the number of wavelengths varies among nodes in the physical layer. These differences result in different wavelengths on the fiber optic links between nodes. The wavelengths between nodes are shown below:

[0037] Node A - Node B: λ1, λ2, λ3;

[0038] Node A to Node C: λ1, λ2, λ3;

[0039] Node A to Node D: λ1, λ2, λ3;

[0040] Node B to Node C: λ1, λ2, λ3;

[0041] Node B to Node E: λ1, λ2, λ3;

[0042] Node C to Node D: λ1, λ2, λ3, λ4;

[0043] Node D - Node E: λ1, λ2;

[0044] Node E - Node F: λ1, λ2;

[0045] Node F - Node A: λ1, λ2.

[0046] Furthermore, three types of services with differentiated Quality of Service (QoS) exist in multi-node mesh topology optical networks:

[0047] (1) EF services with accelerated forwarding: require sufficient bandwidth, low packet loss rate, low latency and jitter;

[0048] (2) Ensure forwarded AF services: require sufficient bandwidth, low packet loss rate, but are not sensitive to latency and jitter;

[0049] (3) Best-effort forwarding BE service: If there are no special requirements, do your best as long as bandwidth resources allow.

[0050] A method for multi-layer protection and recovery and resource allocation in an optical network includes the following steps:

[0051] like Figure 1 As shown, in step 1: In a multi-node mesh topology optical network, each node reports a service request R(B,D) to the control layer with a period of T. That is, the R service request between the source node B and the destination node D. The control layer calculates 5 alternative routes from the source node B to the destination node D using the KSP algorithm. These are alternative route 1 (B, C, D), alternative route 2 (B, A, D), alternative route 3 (B, E, D), alternative route 4 (B, A, F, D), and alternative route 5 (B, A, C, D).

[0052] Step 1.1: The 5 routing information entries in Step 1 will be stored in the routing information table. If a node or fiber optic link in one of the routes fails, the working route can be switched in time.

[0053] Step 1.2: If node A fails and cannot forward data, a new working route can be selected from alternative routes 1 (B, C, D) and alternative route 3 (B, E, D).

[0054] Step 1.3: If the fault occurs in the fiber optic link between two nodes, such as the fiber optic link between node B and node C, a new working route can be selected from the four alternative routes: alternative route 2 (B, A, D), alternative route 3 (B, E, D), alternative route 4 (B, A, F, D), and alternative route 5 (B, A, C, D).

[0055] Step 2: After calculating the 5 candidate routes, set the route weight value L(α, β). This weight value is positively correlated with the bandwidth resource α in the route and negatively correlated with the number of routes β. Assuming that the bandwidth of each wavelength is the same, among the 5 routes, candidate route 1 has the most bandwidth resource and the fewest hops, followed by candidate route 2, then candidate route 3, and finally candidate route 4 and candidate route 5. Therefore, the initial route weight values ​​are L(α1, β1) > L(α2, β2) > L(α3, β3) > L(α4, β5) > L(α5, β5).

[0056] Step 3: The alternative route with the larger route weight value will be selected as the working route. Similarly, when a failure occurs, the new working route will give priority to the alternative route with the larger route weight value. In addition, the control layer periodically obtains the bandwidth resource usage and remaining status of each route in the multi-node Mesh topology optical network, and periodically updates the route weight value.

[0057] Step 4: Based on the route weight values ​​in Step 2, select three unrelated candidate routes from the candidate routes as the working routes for the three QoS services;

[0058] Step 4.1: As Figure 4 As shown, candidate route 1 has the largest route weight value L1, so it is selected as the working route for the EF service.

[0059] Step 4.2: Alternative route 3, which is not related to alternative route 1, is used as the working route for the AF service;

[0060] Step 4.3: Alternative route 5, which is unrelated to the two routes mentioned above, is used as the working route for the BE service;

[0061] Step 4.4: Three unrelated working routes. Except for the source node and destination node being the same, the other nodes in the routes are different from the fiber optic links. If any one of the three working routes fails, it will not affect the transmission of other services.

[0062] Step 5: The control layer obtains the service requests from each node in the multi-node mesh topology optical network and allocates bandwidth resources in a time period T to meet the service transmission needs of each node, such as... Figure 5 As shown, an allocation cycle includes a working domain and a protected domain, T op T is the start time of the period. end1 T is the end time of the working domain. end2 The period ends at λ1, λ2, λ3, and λ4, which represent four wavelengths. The working domain accounts for 70% and is used as the working route. The remaining 30% is used as the protection domain and is not allocated for the time being, in case other working routes fail and a temporary route switch is required.

[0063] Step 5.1: On the same fiber optic link, multiple virtual logical networks can occupy it simultaneously. Therefore, reasonable allocation of wavelengths and time slots can effectively guarantee service transmission and reduce transmission latency. Available resources that meet the service bandwidth are allocated in a period of T, where available resources include wavelengths and time slots.

[0064] Step 5.2: Time slots on different wavelengths will be filled in chronological order. Only when an earlier time slot is filled will a later time slot be used.

[0065] Step 5.3: In addition, prioritize EF services, then AF services, and finally use the remaining bandwidth resources for BF services, such as... Figure 5 As shown, the EF service is scheduled on time slots λ1, λ2, λ3, and λ4 that are close to the start time of the Top cycle. After all the time slots close to the Top have been allocated, the time slots allocated for subsequent times satisfy the AF service and the BF service.

[0066] Step 6: The recovery process after a failure occurs is as follows Figure 6 As shown, the route recovery steps are as follows: Figure 2As shown, the fiber optic link between node B and node C failed, causing the working route to malfunction.

[0067] Step 6.1: At this point, the control layer sends a ranging signal, indicating that the fault is the fiber optic link between node B and node C;

[0068] Step 6.2: Read the route weight value of the candidate route. At this time, the route weight value of candidate route 2 is larger, so it is selected as the new working route.

[0069] Step 6.3: After determining the route to be switched, calculate the bandwidth resource allocation in period T, mainly the resource allocation of the protection domain;

[0070] Step 6.4: If it is found at this time that the bandwidth resources are insufficient to meet the service requirements, alternative route 2 and alternative route 5 can be selected as the working routes at the same time;

[0071] Step 6.5: Notify all nodes in the route to prepare for the route switch. Once completed, normal data service transmission can be restored.

Claims

1. A method for multi-layer protection recovery and resource allocation in an optical network, characterized by comprising the following steps: 1) Multi-layer protection: Multi-layer protection includes multiple route alternatives, network service slicing, and setting working domains and protection domains within the bandwidth allocation cycle; 1.1 Multiple alternative routes: There are multiple alternative routes between the source node and the destination node. If the working route fails, it can be switched to an alternative route for protection in a timely manner. 1.1.1 In a complex multi-node mesh topology optical network, the KSP algorithm with K shortest paths is used to calculate n fixed alternative routes for each pair of physical nodes in the physical network, and the routing information is stored in the routing information table. Thus, for a given node, the routing information table stores the routing information of any other node in the multi-node mesh topology optical network, and for a specified source node and destination node, at most n fixed alternative routes will be pre-calculated. 1.1.2: Set the route weight value L. This weight value is positively correlated with the bandwidth resources α in the candidate routes and negatively correlated with the hop count β of the routes. The control layer of the optical network obtains the remaining bandwidth resources in each candidate route in real time, which means that the route weight value is updated dynamically in real time. 1.2 Network service slicing: Services with different Quality of Service (QoS) are transmitted by different routes. If one working route fails, the other two working routes will not be affected. 1.2.1: Differentiated quality of service exists in multi-node mesh-type optical networks: (1) EF services with accelerated forwarding: require sufficient bandwidth, low packet loss rate, low latency and jitter; (2) Ensure forwarded AF services: require sufficient bandwidth, low packet loss rate, but are not sensitive to latency and jitter; (3) Best-effort forwarding BE service: If there are no special requirements, do your best as long as bandwidth resources allow; 1.2.2: Based on the routing weight values ​​obtained in step 1.1.2, three routes are selected to ensure differentiated service QoS transmission, that is, three services use three mutually isolated routes; 1.3: Setting up working and protection domains within the bandwidth allocation cycle: The protection domain is used for switching when other links fail; 1.3.1: Allocate bandwidth resources in T-cycles, and set up working domains and protection domains. The working domain accounts for 70% for normal service transmission, while the protection domain accounts for 30% for protection, in case other links fail and need to be switched to. 1.3.2: The same wavelength scheduling domain can be occupied by multiple working routes simultaneously; available resources that meet the service bandwidth are allocated in a period of T, where available resources include wavelengths and time slots; 1.3.3: Time slots on different wavelengths will be filled sequentially according to time order. Only when an earlier time slot is filled will a later time slot be used. 1.3.4: In conjunction with service QoS, prioritize EF services that require sufficient bandwidth, low packet loss rate, low latency and jitter, that is, prioritize EF services in the earliest time slot, then satisfy AF services that only require sufficient bandwidth and low packet loss rate, and finally use the remaining bandwidth resources for BF services. 2) Route recovery: 2.1: When a working route fails, the control layer of the optical network sends a ranging signal to confirm whether all nodes in the network topology are online. If all nodes are online, it means that the fiber optic link in the working route has failed. 2.2: If there is no response after multiple inquiries about whether the node is online, it means that the malfunction is caused by the node going offline; 2.3: The control layer of the optical network selects a new working route from the candidate routes based on the route weight value, and calculates the bandwidth resource allocation in the route; 2.4: If the remaining bandwidth resources of a single alternative route are insufficient to meet the service requirements, multiple alternative routes will be upgraded to working routes in stages to utilize the bandwidth resources of multiple alternative routes to meet the service requirements. 2.5: After completing steps 2.1 to 2.4, the control layer notifies all nodes in the route, enabling all nodes to prepare for the new route switch, thereby performing the route switch and realizing route protection.

Images