A dual-homing ring network topology and path optimization system
Through the dual-home ring network topology and path optimization system, combined with the Suurballe algorithm and Gurobi solver with SRLG constraints, the fiber resource consumption in the telecommunications network is optimized, and the problems of high fiber resource consumption and computational complexity in the existing technology are solved, and the problem of high fiber resource consumption and computational complexity is realized. More efficient fiber resource management and network carrying capacity.
Patent Information
- Application Number
- CN202210949946.7
- Authority / Receiving Office
- CN · China
- Patent Type
- Patents(China)
- Current Assignee / Owner
- Filing Date
- 2022-08-09
- Publication Date
- 2024-11-19
- Estimated Expiration
- 2042-08-09
AI Technical Summary
The prior art is difficult to effectively optimize the ring network topology in telecommunications networks, resulting in high fiber resource consumption, and the existing algorithms have high computational complexity when processing large-scale networks, making it difficult to achieve real-time optimization.
Using dual-home ring network topology and path optimization system, combined with the Suurballe algorithm and Gurobi solver with SRLG constraints, optimize the mapping relationship between the logic layer and the physical layer, and realize the fiber optic resource through network monitoring, path optimization, topology optimization and traffic planning modules. Effective planning of minimum consumption and traffic paths.
It significantly reduces the consumption of fiber resources and improves the solution speed. It saves 17% of fiber resources in the unchanged node scenario. The network access time of nodes in the newly added node scenario is shortened to 5 seconds. The load-raising capacity is increased by 13% while ensuring the network. Reliability and Agg node loop formation rate.
Smart Images

Figure CN115664525B_ABST
Abstract
Description
Technical Field
[0001] The invention belongs to the field of network topology optimization, and in particular relates to a dual-home ring network topology and path optimization system. Background Art
[0002] In the past few decades, the low-cost design of telecommunication networks has attracted more and more attention. The growing communication demand requires constant re-planning of the communication structure. For an optical network consisting of a group of nodes connected by optical fibers, it includes a physical layer and a logical layer. The physical layer consists of network nodes and optical fibers connecting the nodes. The logical layer is a set of optical transmission channels called optical paths established between nodes through optical cross-connects (OXC, optical cross-connects). Optical paths are composed of bidirectional optical channels that connect two nodes in the network through the path of the physical layer. In order to adapt to the dynamically changing traffic in the optical network and reduce the cost and resource consumption, it is necessary to optimize the network topology and path, that is, to effectively plan the physical layer topology and traffic path according to the traffic situation on the logical layer, and establish an effective mapping relationship between the logical layer topology and the physical layer topology. A reasonable configuration scheme can make the traffic transmission rate faster, the latency lower, and the resource consumption less.
[0003] At present, the academic community defines network topology and path optimization problems as multi-commodity flow problems. The methods for solving multi-commodity flow problems can be divided into two categories: exact algorithms and heuristic algorithms. Exact algorithms can obtain the exact optimal solution of the model, but their disadvantages are that they cannot handle problems with more decision variables or more constraints within a limited computing time under existing computer technology. The solution scale is large, the solution time complexity is high, and the performance requirements of the algorithm are extremely high. At the same time, the network topology structure is difficult to display and difficult to model in a formula. Heuristic algorithms are proposed relative to optimization algorithms. They are algorithms based on intuition or experience. They can give a feasible solution for each instance of the combinatorial optimization problem to be solved at an acceptable cost (referring to computing time and space). Although they can handle problems with more decision variables, the optimal solution they obtain is an approximate optimal solution and is more likely to fall into a local optimal solution. The gap between the approximate optimal solution and the actual optimal solution cannot be measured and estimated. The main technology for completing network topology optimization in the industry is the expert system, which relies on the experience and technology of experts. However, as the network scale becomes larger and larger, the network planning, design and management mechanism becomes more and more complex. It is necessary to continuously update the expert system to design a reasonable network topology. Summary of the invention
[0004] In response to the above problems, the present invention provides a dual-belonging ring network topology and path optimization system for the ring network topology structure. The system will effectively plan the physical topology and traffic path according to the traffic conditions on the logical layer network topology, and achieve the goal of less fiber optic resource consumption by adjusting the effective mapping relationship between the logical layer topology and the physical layer topology.
[0005] To achieve the above object, the present invention adopts the following scheme to achieve:
[0006] A dual-homing ring network topology and path optimization system, comprising:
[0007] The network monitoring module is used to collect relevant inputs of the optimization algorithm and pass them to the path optimization module;
[0008] The path optimization module is used to receive the data transmitted by the network monitoring module and plan the primary and backup paths for all traffic according to the traffic demand by using the Suurballe algorithm combined with SRLG constraints, so that the physical layer topology link corresponding to the planned path consumes the least optical fiber resources;
[0009] Topology optimization module: It is used to receive data transmitted by the path optimization module, combine the topology constraints of the dual-homed ring network logical layer with the path constraints of the physical layer, and use the Gurobi solver to perform topology optimization to achieve the goal of minimizing fiber resource consumption;
[0010] Traffic planning module: used to simulate traffic transmission of the physical layer topology G obtained by the topology optimization module to verify whether the topology G strictly meets the traffic-related constraints;
[0011] The scenario adaptation module is used as an auxiliary module to help the topology obtained by the topology optimization module meet the traffic-related constraints. It is also responsible for incorporating the newly added nodes into the existing network topology at the lowest cost in the scenario of disrupting the loop and adding points.
[0012] A further improvement of the present invention is that the network monitoring module collects network status information, including flow demand, network topology, SRLG constraint information and network topology structure constraint information, and the network status information is transmitted to the path optimization module as input.
[0013] A further improvement of the present invention is that the specific operation of the path optimization module is as follows: first, the undirected graph is updated according to the SRLG constraint; then the priority of the traffic is calculated and sorted from high to low according to the priority; next, based on the Suurballe algorithm, two primary and backup paths with non-intersecting edges are planned for each traffic; after each path of a traffic is planned, the weight of the edge with traffic load is reduced to 0, and then the path planning for the next traffic continues until all traffic is successfully planned.
[0014] A further improvement of the present invention is that the method for updating the undirected graph according to the SRLG constraint is as follows:
[0015] Let G = {V, E} be a weighted directed graph, V is the set of all nodes, E is the set of all edges, and R is i = {uv 1 ,uv 2 ,…,uv k} is the i-th SRLG constraint, that is, edge uv 1 ,uv 2 ,…,uv k Share the same optical fiber;
[0016] First, for each SRLG constraint, a virtual node d is introduced i , then V'(G)=V(G)∪{d 1 ,d 2 ,…,d k};
[0017] Then, traverse all edges in the SRLG:
[0018] If R i Each edge uv in j appears only once, then R i The edge in is deleted in G, and a new edge ud is added i and {d i v 1 ,d i v 2 ,…,d i v k}, edge i The weight of edge d is 0. i v j The weight is the original uv j The weight of
[0019] If there are two sets of SRLG constraints that have an intersection, that is, uv∈R i ∩R j , d i For R i The virtual node introduced, d j For R j The virtual node introduced, then R i and R j Delete all the edges in and add a new edge ud i and j , weight is 0; add edge d i v and side d j v, weight is ∞, the purpose of setting the weight to ∞ is to avoid selecting the same edge; add edge d id j , the weight is the weight of the original edge uv; finally, a virtual node is added to R i and R j The edges of other nodes in .
[0020] A further improvement of the present invention is that the method for calculating the priority of traffic is as follows:
[0021] In the process of planning paths for traffic, the priority is defined as the average cost of the edges of the path, that is,
[0022]
[0023] Among them, cost(p) represents the total cost of the edges contained in path p, and len(p) represents the number of edges contained in path p;
[0024] The specific process of Suurballe algorithm is as follows:
[0025] Let G = {V, E} be a weighted directed graph, V is the set of all nodes, E is the set of all edges, ω ij represents the weight of edge (i,j) in G, v s is the source point, t sj Represents the source point v s The shortest distance to node j;
[0026] (1) Use Dijkstra algorithm to find the source point v s To the destination d The shortest path path1;
[0027] (2) Transform the network by changing the direction of the edges in the calculated path1 and the weights of other edges, with the goal of reducing the total length of the path. First, reverse the edges contained in path1, and then update the weights of all edges to ω′ ij :
[0028] ω′ ij =ω ij +t si +t sj
[0029] (3) Use Dijkstra algorithm to find the source point v in the updated network s To the destination d The shortest path path2;
[0030] (4) Delete the common edges in path1 and path2 and rearrange them to obtain two non-intersecting paths with the smallest total length.
[0031] A further improvement of the present invention is that the specific operation of the topology optimization module is as follows: the topology result obtained based on the path optimization module is used to simplify the constraints related to the dual-homing ring network topology originally established, and under the simplified topology constraints, the Gurobi solver is used to perform topology optimization and solve the problem with the goal of achieving the highest Agg node ring rate, and the optimized logical layer and physical layer topology results are output;
[0032] Among them, the simplified topology constraint model is as follows:
[0033] Symbolic variables:
[0034] G = {V, E}: undirected graph, where V = V Core ∪V DIS ∪V Agg is the set of all nodes, V Core , V DIS , V Agg They are the collection of Core nodes, DIS nodes and Agg nodes respectively, and E is the collection of edges, where Core nodes are core nodes, Agg nodes are Aggregation nodes, DIS nodes are Distribution nodes, the upper layer is Core nodes, and the lower layer is Agg nodes;
[0035] L: is a set of logical rings;
[0036] cost ij : The distance between node i and node j;
[0037] e il : binary variable,
[0038] g ijl : binary variable,
[0039] Decision variables:
[0040] x ij : binary variable,
[0041] y il : binary variable,
[0042] Auxiliary variables:
[0043] x l : binary variable,
[0044] Optimization goal:
[0045] 1. Agg has the highest ring formation rate
[0046]
[0047] 2. Consumes minimal physical layer resources
[0048]
[0049] Constraints:
[0050] 1. Constraints on the node belonging ring
[0051]
[0052]
[0053]
[0054]
[0055] 2. Constraints on Consumption of Physical Layer Resources
[0056]
[0057] A further improvement of the present invention is that, in order to determine whether the physical layer topology G obtained by the topology optimization module satisfies the traffic-related constraints, the traffic planning module performs traffic transmission simulation on the topology G, plans the primary and backup paths for traffic transmission based on the Dijkstra algorithm, and obtains the physical layer topology G' corresponding to the traffic-related constraints, compares G and G', and if the topologies are consistent, the process ends and the logical layer and physical layer topology results obtained by the topology optimization module are output; if the topologies are inconsistent, the missing links are marked and input into the scenario adaptation module to add the missing links.
[0058] A further improvement of the present invention is that the scene adaptation module adds missing links or newly added nodes to the physical layer links where their adjacent nodes are located, and assigns them to the relevant logical layer rings.
[0059] The present invention has at least the following beneficial technical effects:
[0060] The present invention provides a dual-home ring network topology and path optimization system, which plans two main backup paths for traffic that do not intersect and do not share optical fiber resources based on the Suurballe algorithm combined with SRLG (Shared Risk Link Group) constraints, so that the optical fiber resources consumed by the planned path are minimized, and topology optimization is performed based on the Gurobi solver, and finally effective planning of physical topology and traffic paths is achieved under the traffic conditions on the known logical layer network topology. Compared with the existing optimal algorithm, the algorithm of the present invention has been significantly improved in both the node unchanged scenario and the new node scenario and the load increased scenario. The node unchanged scenario refers to optimizing the existing topology to achieve the purpose of consuming less optical fiber resources while ensuring that the network carrying capacity and reliability are not reduced. In this scenario, compared with the optimal algorithm, the algorithm of the present invention saves 17% of optical fiber resources, while ensuring that the Agg node ring rate is 100%, all traffic is successfully planned as the main backup path that meets the SRLG constraints, the running time is short, and the solution speed is increased by three times. The new node scenario refers to the implementation of the minimum cost of adding new nodes to the network without affecting other nodes. The algorithm of the present invention can add new nodes to the ring network within 5 seconds. The load increase scenario refers to improving the network's carrying capacity as much as possible without affecting reliability. The algorithm of the present invention can achieve a 13% increase in the traffic load that the topology can carry. In summary, the algorithm proposed in the present invention can address the disadvantage of high time complexity of precise algorithm solutions, while better combining the advantages of high accuracy of precise algorithms and fast solution speed of heuristic algorithms, so that the dual-home ring network topology and path optimization problems are well solved. BRIEF DESCRIPTION OF THE DRAWINGS
[0061] Figure 1 A schematic diagram of the mapping relationship between the logical layer topology and the physical layer topology;
[0062] Figure 2 It is a schematic diagram of SRLG constraints;
[0063] Figure 3 It is a schematic diagram of the basic framework of the dual-homing ring network topology and path optimization system;
[0064] Figure 4 This is a schematic diagram of the algorithm flow architecture of the path optimization module;
[0065] Figure 5 It is a schematic diagram of the change of the number of dual-homing nodes and fiber resource consumption with parameters in the scenario where the nodes remain unchanged;
[0066] Figure 6 It is the topology diagram of the physical layer in the scenario where the nodes remain unchanged;
[0067] Figure 7 It is the topology diagram of the logical layer in the scenario where the nodes remain unchanged;
[0068] Figure 8 This is a comparison diagram between the present invention and the existing optimal algorithm in a scenario where nodes remain unchanged;
[0069] Fig. 9 This is a comparison chart of the node before and after it joins the network in the scenario of adding a new node;
[0070] Fig.10 Overflow load variation curve for load increase scenario;
[0071] Fig.11 Link load conditions before adding bandwidth constraints in the load increase scenario;
[0072] Fig.12 This figure shows the link load after adding bandwidth constraints in the load increase scenario. DETAILED DESCRIPTION
[0073] Exemplary embodiments of the present disclosure will be described in more detail below with reference to the accompanying drawings. Although exemplary embodiments of the present disclosure are shown in the accompanying drawings, it should be understood that the present disclosure can be implemented in various forms and should not be limited by the embodiments set forth herein. On the contrary, these embodiments are provided in order to enable a more thorough understanding of the present disclosure and to be able to fully convey the scope of the present disclosure to those skilled in the art. It should be noted that, in the absence of conflict, the embodiments of the present invention and the features described in the embodiments can be combined with each other. The present invention will be described in detail below with reference to the accompanying drawings and in combination with the embodiments.
[0074] The present invention focuses on the dual-homing ring network topology and path optimization issues, effectively plans the physical layer topology and traffic path according to the traffic situation on the logical layer network topology, and achieves less fiber resource consumption by adjusting the effective mapping relationship between the logical layer topology and the physical layer topology. Figure 1 , showing the mapping relationship from logical layer topology to physical layer topology. At the same time, in order to improve the reliability of traffic transmission, optical networks generally adopt a dual-homing ring topology. Its structural characteristics ensure that data transmission has a primary backup path, and the primary path cannot share any common resources with the links on the backup path. When the primary path fails, the backup path will be activated, which to a certain extent guarantees the reliability of transmission, improves the fault tolerance of computer networks, and protects the link from SRLG failures. Reference Figure 2 , showing two fiber links sharing risk resources.
[0075] At present, the academic community defines this problem as a multi-commodity flow problem. The exact algorithm can obtain the exact optimal solution of the model, but its solution scale is large, the solution time complexity is high, and the performance requirements of the algorithm are extremely high. In addition, the dual-home ring network topology is difficult to display and difficult to model formally. The heuristic algorithm can give a solution to the problem at an acceptable cost, but the optimal solution obtained is an approximate optimal solution and is prone to fall into a local optimal solution. The main technology in the industry is the expert system, which relies on the experience and technology of experts. However, as the network scale becomes larger and larger, the network planning, design and management mechanism becomes more and more complex. It is necessary to continuously update the expert system to design a reasonable network topology.
[0076] In summary, the present invention proposes a dual-homing ring network topology and path optimization system. The system has a fast solution speed and high accuracy, and can effectively plan the physical topology and traffic path according to the traffic situation on the logical layer network topology, so that less fiber resources are consumed.
[0077] refer to Figure 3 , showing the overall framework of the dual-homing ring network topology and path optimization system. The system consists of five modules: network monitoring module, path optimization module, topology optimization module, traffic planning module and scenario adaptation module.
[0078] The network monitoring module is mainly responsible for collecting network status information, including traffic demand, network topology, SRLG constraint information and network topology structure constraint information, which are passed as input to the path optimization module. After receiving the relevant input, the path optimization module will plan the primary and backup paths for all traffic according to the traffic demand using the Suurballe algorithm combined with SRLG constraints, so that the physical layer topology link corresponding to the planned path consumes the least fiber resources. After that, the topology optimization module will receive the physical layer topology results obtained from the path optimization module, and simplify the related constraints of the dual-home ring network topology originally established based on the topology results. Under the simplified topology constraints, the Gurobi solver is used to perform topology optimization and solve the problem with the highest ring rate of the Agg node (Aggregation) as the target, and output the optimized logical layer and physical layer topology results. The traffic planning module will perform traffic transmission simulation for the obtained physical layer topology G, mainly based on the Dijkstra algorithm to plan the primary and backup paths for traffic transmission, and obtain the physical layer topology G' corresponding to the strict satisfaction of traffic-related constraints. Compare G and G'. If the topology is consistent, the process ends and the logical layer and physical layer topology results obtained by the topology optimization module are output; if the topology is inconsistent, the missing links are marked and input to the scenario adaptation module for adding missing links. The scenario adaptation module will add the missing edges to the logical ring and the corresponding physical ring under the premise of satisfying the traffic and topology constraints. At the same time, the scenario adaptation module is also responsible for incorporating the newly added nodes into the existing network topology at the lowest cost in the scenario of breaking the ring and adding points.
[0079] With the coordinated cooperation of the above five modules, we can achieve joint optimization of the topology and path of the dual-home ring network.
[0080] refer to Figure 4 , showing the algorithm flow architecture of the path optimization module. The algorithm first updates the undirected graph according to the SRLG constraint; then calculates the priority of the traffic and sorts it from high to low; then plans two non-intersecting primary and backup paths for each traffic based on the Suurballe algorithm. After planning the path of each traffic, the weight of the edge with traffic load is reduced to 0, and then continue to plan the path for the next traffic until all traffic is successfully planned.
[0081] The present invention compares the results of the present invention with those of the existing optimal algorithm in a scenario where the nodes remain unchanged, and shows the results in a scenario where new nodes are added and a scenario where the load is increased.
[0082] In the scenario where nodes remain unchanged, we evaluate the physical topology obtained by the path optimization module from two perspectives: the number of dual-homing nodes and fiber resource consumption. The results are shown in Table 1. Figure 5. As k increases, the overall cost shows an upward trend, while the change in the number of dual-homing nodes does not have an obvious pattern. Among them, the physical topology when k=14 can achieve the lowest cost under the premise of ensuring that the dual-homing rate of Agg nodes is 100%. At the same time, the physical topology calculated by the path optimization module can ensure that all traffic can find the primary path and backup path under the premise of satisfying the SRLG constraints. The topology diagram of the physical layer and the logical layer is referenced Figure 6 and Figure 7 Under the condition that the network carrying capacity and reliability are not reduced, the comparison chart of the present invention and the existing optimal algorithm is shown in Figure 8 ,The algorithm of the present invention has the advantages of 17% fiber resource saving, 100% Agg looping rate, all traffic is successfully planned on the ,primary backup path that satisfies the SRLG constraints, short running time, and three times the ,solving speed.
[0083] In the scenario of adding a new node, the comparison chart of the new node (node 26) before and after joining the network is referenced Fig. 9 By connecting a node to its two adjacent nodes, new nodes can be added to the network at the minimum cost.
[0084] In the load increase scenario, the overflow load change curve reference Fig.10 The results show that the topology obtained by the path optimization algorithm has additional carrying capacity to a certain extent. Without affecting the reliability, the maximum load that the topology can carry is increased by 13%. The impact of bandwidth constraints on topology bandwidth utilization Fig.11 and Fig.12 , the topology without considering bandwidth constraints obviously has more high-load edges (i.e., bandwidth utilization is higher than 85%); and in the topology considering bandwidth constraints, some of the original high-load edges are alleviated. This result verifies the effectiveness of bandwidth constraints, that is, bandwidth constraints can make the bandwidth utilization of the topology more balanced.
[0085] Although the present invention has been described in detail above with general descriptions and specific embodiments, it is obvious to those skilled in the art that some modifications or improvements may be made thereto based on the present invention. Therefore, these modifications or improvements made without departing from the spirit of the present invention all fall within the scope of protection claimed by the present invention.
Claims
1. A dual-homing ring network topology and path optimization system, characterized in that: include: The network monitoring module is used to collect relevant inputs of the optimization algorithm and pass them to the path optimization module; The path optimization module is used to receive the data transmitted by the network monitoring module and plan the primary and backup paths for all traffic according to the traffic demand by using the Suurballe algorithm combined with SRLG constraints, so that the physical layer topology link corresponding to the planned path consumes the least optical fiber resources; Topology optimization module: It is used to receive data transmitted by the path optimization module, combine the topology constraints of the dual-homed ring network logical layer with the path constraints of the physical layer, and use the Gurobi solver to perform topology optimization to achieve the goal of minimizing fiber resource consumption; Traffic planning module: used to simulate traffic transmission of the physical layer topology G obtained by the topology optimization module to verify whether the topology G strictly meets the traffic-related constraints; The scenario adaptation module is used as an auxiliary module to help the topology obtained by the topology optimization module meet the traffic-related constraints. It is also responsible for incorporating the newly added nodes into the existing network topology at the minimum cost in the scenario of breaking the loop and adding nodes. The specific operation of the path optimization module is as follows: first, the undirected graph is updated according to the SRLG constraint; then the priority of the traffic is calculated and sorted from high to low; next, two non-intersecting primary and backup paths are planned for each traffic based on the Suurballe algorithm; after planning the path of each traffic, the weight of the edge with traffic load is reduced to 0, and then the path planning for the next traffic is continued until all traffic is successfully planned; The method to update an undirected graph according to the SRLG constraints is as follows: Let G = {V, E} be a weighted directed graph, V is the set of all nodes, E is the set of all edges, and R is i ={uv1, uv2, ..., uv k } is the i-th SRLG constraint, i.e., the edges uv1, uv2, ..., uv k Share the same optical fiber; First, for each SRLG constraint, a virtual node d is introduced i , then V'(G)=V(G)∪{d1,d2,…,d k }; Then, traverse all edges in the SRLG: If R i Each edge uv in j appears only once, then R i The edge in is deleted in G, and a new edge ud is added i and {d i v1,d i v2,…,d i v k }, edge i The weight of edge d is 0. i v j The weight is the original uv j The weight of If there are two sets of SRLG constraints that have an intersection, that is, uv∈R i ∩R j , d i For R i The virtual node introduced, d j For R j The virtual node introduced, then R i and R j Delete all the edges in and add a new edge ud i and j , weight is 0; add edge d i v and side d j v, weight is ∞, the purpose of setting the weight to ∞ is to avoid selecting the same edge; add edge d i d j , the weight is the weight of the original edge uv; finally, a virtual node is added to R i and R j The edges of other nodes in The traffic priority is calculated as follows: In the process of planning paths for traffic, the priority is defined as the average cost of the edges of the path, that is, Among them, cost(p) represents the total cost of the edges contained in path p, and len(p) represents the number of edges contained in path p; The specific process of Suurballe algorithm is as follows: Let G = {V, E} be a weighted directed graph, V is the set of all nodes, E is the set of all edges, ω ij represents the weight of edge (i, j) in G, v s is the source point, t sj Represents the source point v s The shortest distance to node j; (1) Use Dijkstra algorithm to find the source point v s To the destination d The shortest path path1; (2) Transform the network by changing the direction of the edges in the calculated path1 and the weights of other edges, with the goal of reducing the total length of the path. First, reverse the edges contained in path1, and then update the weights of all edges to ω′ ij : oh' ij =ω ij +t si +t sj (3) Use Dijkstra algorithm to find the source point v in the updated network s To the destination d The shortest path path2; (4) Delete the common edges in path1 and path2 and rearrange them to obtain two non-intersecting paths with the smallest total length.
2. A dual-homing ring network topology and path optimization system according to claim 1, characterized in that: The network monitoring module collects network status information, including traffic demand, network topology, SRLG constraint information and network topology constraint information. The network status information is passed as input to the path optimization module.
3. A dual-homing ring network topology and path optimization system according to claim 1, characterized in that: The specific operation of the topology optimization module is as follows: based on the topology results obtained by the path optimization module, the constraints related to the dual-homing ring network topology originally established are simplified. Under the simplified topology constraints, the Gurobi solver is used to perform topology optimization with the goal of maximizing the Agg node ring rate, and the optimized logical layer and physical layer topology results are output; Among them, the simplified topology constraint model is as follows: Symbolic variables: G = {V, E}: undirected graph, where V = V Core ∪V DIS ∪V Agg is the set of all nodes, V Core , V DIS , V Agg They are the collection of Core nodes, DIS nodes and Agg nodes respectively, and E is the collection of edges, where Core nodes are core nodes, Agg nodes are Aggregation nodes, DIS nodes are Distribution nodes, the upper layer is Core nodes, and the lower layer is Agg nodes; L: is a set of logical rings; cost ij : The distance between node i and node j; e il : binary variable, g ijl : binary variable, Decision variables: x ij : binary variable, y il : binary variable, Auxiliary variables: x l : binary variable, Optimization goal:
1. Agg has the highest ring formation rate 2. Consumes minimal physical layer resources Constraints:
1. Constraints on the node belonging ring 2. Constraints on Consumption of Physical Layer Resources 4. A dual-homing ring network topology and path optimization system according to claim 3, characterized in that: In order to determine whether the physical layer topology G obtained by the topology optimization module meets the traffic-related constraints, the traffic planning module performs traffic transmission simulation on the topology G, plans the primary and backup paths for traffic transmission based on the Dijkstra algorithm, and obtains the physical layer topology G' that meets the traffic-related constraints. Compare G and G'. If the topologies are consistent, the process ends and the logical layer and physical layer topology results obtained by the topology optimization module are output; If the topology is inconsistent, the missing links are marked and input into the scenario adaptation module to add the missing links.
5. A dual-homing ring network topology and path optimization system according to claim 4, characterized in that: The scenario adaptation module adds missing links or newly added nodes to the physical layer links where their adjacent nodes are located and assigns them to the relevant logical layer rings.
