SD-MANET multi-controller deployment method based on improved piebald algorithm

Abstract

The invention discloses an SD-MANET multi-controller deployment method based on an improved piebald algorithm, and belongs to the technical field of mobile ad hoc networks. The method comprises the following steps: acquiring network topology information of the SD-MANET; constructing an undirected graph representing the network topology; defining a controller deployment problem as a multi-objective optimization problem; and solving the multi-objective optimization problem by adopting an improved emerald optimization algorithm to obtain an optimal controller deployment scheme, and deploying the controller and the management domain thereof in the SD-MANET environment according to the optimal controller deployment scheme. According to the method, three key indexes including the link failure rate, the time delay and the load balancing are optimized at the same time, the obvious effect is achieved in the aspect of reducing the link failure rate and the time delay, the link stability and the time delay performance are preferentially guaranteed, and therefore the overall reliability and efficiency of the SD-MANET network are improved.

Description

Technical Field

[0001] The present invention relates to a method for deploying an SD-MANET controller, specifically a method for deploying multiple SD-MANET controllers based on an improved pied kingfisher algorithm, belonging to the technical field of mobile ad hoc networks. Background Art

[0002] With the rapid development of information technology, especially the wide application of emerging technologies such as big data, cloud computing, and artificial intelligence, the form of modern warfare is undergoing profound changes. Against this background, the manned / unmanned cooperative combat system has emerged as the times require and has become a crucial innovative force in modern warfare. In this system, cooperative perception is a crucial link. It requires different combat units to be able to share battlefield information and status in real time and continuously, ensuring that every decision can be made based on global real-time data. However, in complex battlefield environments such as strong adversariality and variable climate, combat units are easily damaged, communication links are easily interrupted, and thus battlefield information cannot be provided in real time and continuously. Moreover, traditional mobile ad hoc networks (MANETs) are also difficult to cope with the above problems. Therefore, software-defined networking (SDN) is introduced into the MANET architecture to form a software-defined mobile ad hoc network (SD-MANET). By separating the control plane from the data plane, SDN provides centralized control and intelligent resource scheduling capabilities, enabling the network to dynamically adjust resource allocation from a global perspective and adapt to changes in the battlefield environment. However, in the SD-MANET network, the deployment location and number of controllers are important research issues because combat units in the battlefield environment are easily damaged, communication links are also easily damaged, and some nodes may receive a large amount of data, resulting in excessive load and failure, and thus data cannot be transmitted in real time and continuously, making it impossible for individual nodes to make correct judgments.

[0003] Existing research on SDN controller deployment generally has the problem of one-sided optimization objectives. Most of them focus on optimizing one or two of the single indicators of delay, load, or reliability, lacking a joint consideration and optimization mechanism for the internal conflicts and collaborative requirements of these three, and it is difficult to obtain a deployment plan with the optimal comprehensive performance. At the same time, the application scenarios of these studies are mainly limited to local area networks, aviation, and satellite networks. However, for the problem of controller deployment in the software-defined mobile ad hoc network (SD-MANET) environment, relevant specialized research is even scarcer, resulting in the difficulty of directly applying existing methods to effectively address the unique challenges of tactical scenarios. Currently, most studies on the controller deployment problem in the SD-MANET environment only target the single objective of delay and do not comprehensively consider other performance indicators. Summary of the Invention

[0004] Objective of the Invention: Aiming at the above problems, the objective of the present invention is to provide an SD-MANET multi-controller deployment method based on an improved kingfisher algorithm, with the goal of reducing the control link failure rate and network delay and ensuring controller load balancing. By improving the kingfisher optimization algorithm, a multi-controller deployment algorithm that can automatically determine the number and location of controllers is designed.

[0005] Technical Solution: An SD-MANET multi-controller deployment method based on an improved kingfisher algorithm of the present invention includes the following steps: Step 1, obtain the network topology information of SD-MANET, including the node set, node coordinates, and link set; Step 2, construct an undirected graph representing the network topology, calculate the physical distance of the direct link using the node coordinates, and apply the Dijkstra algorithm to calculate the shortest path distance between any two nodes in the network; Step 3, define the controller deployment problem as a multi-objective optimization problem to determine the optimal set of controller deployment positions and the mapping relationship between each node and the controller; Step 4, use the improved kingfisher optimization algorithm to solve the multi-objective optimization problem, obtain the optimal controller deployment plan, and deploy the controller and its management domain in the SD-MANET environment according to the optimal controller deployment plan.

[0006] Further, Step 3 includes: Calculate the propagation delay from the data plane node to the controller, and the formula is: , where V represents the set of data nodes and C represents the set of controllers , represents the i-th node, represents the j-th controller, represents the node and the controller the shortest path distance between them, Calculate the transmission delay from the data plane node to the controller, and the formula is: , where represents the transmission delay from the node to the controller ; Calculate the propagation delay between controllers, and the formula is: , where represents the controller and the controller The shortest distance between; Calculate the transmission delay between controllers. The formula is: , In the formula, represents the transmission delay from controller to controller ; Then the total communication delay of the controller domain is: .

[0007] Furthermore, step 3 also includes: According to the node failure probability and link failure probability in the control domain of each controller, calculate the control link reliability in the domain of controller , which is expressed as: , In the formula, represents the set of nodes managed by the j-th controller, represents the set of links e in the domain of controller j; Calculate the control plane link failure rate of the SD-MANET network, which is expressed as: , In the formula, m represents the number of controllers in the SD-MANET network.

[0008] Furthermore, step 3 also includes: Calculate the load of each controller in the SD-MANET network. The formula is: , In the formula, represents the number of flow requests of the node, represents whether there is a link between node and , represents the existence of a link, represents the non-existence of a link; Calculate the utilization rate of controller . The formula is: , In the formula, represents the maximum load capacity of controller ; Calculate the average utilization rate of all controllers in the SD-MANET network. The formula is: , Calculate the load balancing degree of the control controller, and the formula is: .

[0009] Furthermore, step 3 also includes: Construct a multi-objective optimization function F for controller deployment based on the total communication delay, control plane link failure rate, and load balancing rate within the controller domain, with the optimization objective of minimizing the function value; where the formula for the multi-objective optimization function is: , In the formula, respectively represent the total communication delay , the control plane link failure rate , and the weight coefficients of the load balancing degree of the controller; The constraint conditions for constructing the multi-objective optimization function are: , , , , , In the formula, V represents the set of data plane nodes , represents the th controller; The first constraint condition means that each node can only be managed by one controller, the second constraint condition means that a controller must manage at least one node, the third constraint condition means that the number of controllers in the SD-MANET network is m; the fourth constraint condition means that the controller cannot exceed its load capacity; the fifth constraint condition means that the weighted sum of the weight coefficients of the multi-objective optimization function is 1.

[0010] Furthermore, step 4 includes: Step 41, set the maximum number of iterations, use the specific deployment location of each controller as an individual in the population, and introduce the Chebyshev chaotic map to initialize the population, and the formula is: , In the formula, represents the position of the i-th individual in the j-th dimension, respectively represent the upper and lower boundaries of the search range of the j-th dimension of the search space, represents the value within (-1, 1) generated by the Chebyshev chaotic map, N represents the total number of individuals, represents the number of dimensions; Step 42: Introduce the Levy flight strategy in the exploration stage to update the population position. The update formula is as follows: , wherein, is the position of the individual at the current iteration s, is the position of another individual randomly selected from the population used to guide the movement and exploration of , is the state parameter, is the random direction parameter, is a random value within the normal distribution; is the final step size of the Levy flight, and the formula is: , wherein, and are both random numbers generated from the normal distribution, is the parameter of the Levy flight; Step 43: Introduce the Gaussian mutation strategy in the exploitation stage to update the population position. The update formula is as follows: , wherein, is the new position obtained under the influence of the probability , is the standard deviation, and the formula is , is a coefficient used to control the magnitude of the standard deviation, respectively represent the upper and lower boundaries of the search range; rand represents a random number in [0, 1], is the mutation probability, indicating that it perturbs the individual variables with a certain probability ; represents the new position obtained by the individual after introducing the Gaussian mutation; Step 44: In the symbiosis stage, update the population position at the current iteration. The update formula is as follows: , wherein, represents the positions of 2 individuals randomly selected from the population at the current iteration s, is the predation efficiency of the kingfisher, represents the hunting ability, represents a random number in [0, 1]; Step 45, merge the updated new population, calculate the fitness value of each individual in the new population, sort the fitness values from smallest to largest, and take the individual corresponding to the smallest fitness value as the optimal feasible solution under the current iteration number; determine whether it is greater than the iteration number, if so, output the optimal feasible solution, otherwise return to the exploration stage to solve and iterate again.

[0011] Beneficial effects: Compared with the prior art, the significant advantages of the present invention are as follows: (1) The present invention proposes an algorithm based on multi-objective optimization for the controller deployment problem in the SD-MANET network, aiming to simultaneously optimize three key indicators: link failure rate, delay, and load balancing. (2) The present invention effectively improves the convergence and search efficiency of the algorithm by improving the population initialization method of the spotted kingfisher algorithm and combining Levy flight and Gaussian strategy, and can automatically determine the optimal number and location of controller deployment. (3) The present invention has achieved remarkable results in reducing the link failure rate and delay, giving priority to ensuring link stability and delay performance, thereby improving the overall reliability and efficiency of the SD-MANET network. Especially in scenarios with extremely high requirements for delay and reliability, its advantages are more obvious. Description of the Drawings

[0012] Figure 1 is a flowchart of an SD-MANET multi-controller deployment method based on an improved spotted kingfisher algorithm; Figure 2 is the original topology diagram; Figure 3 is a flowchart of the improved spotted kingfisher optimization algorithm; Figure 4 is the topology diagram corresponding to the controller deployment scheme solved by the EPKO algorithm; Figure 5 is a comparison diagram of the fitness curves of EPKO, PKO, PSO, and SCSSA when the number and location of controller deployment are optimal; Figure 6 is a comparison diagram of the link failure rates of EPKO, PKO, PSO, and SCSSA when the number and location of controller deployment are optimal; Figure 7 is a comparison diagram of the total delays of EPKO, PKO, PSO, and SCSSA when the number and location of controller deployment are optimal; Figure 8 is a comparison diagram of the load balancing degrees of EPKO, PKO, PSO, and SCSSA when the number and location of controller deployment are optimal. Detailed Embodiments

[0013] To make the objectives, technical solutions, and advantages of this application more clear and understandable, the following further elaborates on this application in conjunction with the accompanying drawings and embodiments.

[0014] A method for deploying multiple controllers in SD-MANET based on an improved pied kingfisher algorithm aims to reduce the failure rate of control links, network latency, and ensure controller load balancing. By improving the pied kingfisher optimization algorithm, a multi-controller deployment algorithm that can automatically determine the number and location of controllers is proposed. This deployment method is mainly divided into three stages. In the first stage, the network topology of SD-MANET is obtained, and the information of nodes and links in the network is stored. In the second stage, based on the basic information of the network, an optimization objective of minimizing the failure rate, latency, and load of the control plane link is constructed. In the third stage, the improved pied kingfisher optimization algorithm is used for solving. Under the condition of low failure rate and latency of the control plane link, load balancing is ensured, and the optimal number and location of controller deployments are obtained.

[0015] Specifically, for a method for deploying multiple controllers in SD-MANET based on an improved pied kingfisher algorithm described in this embodiment, the flowchart is as Figure 1 shown, and this deployment method includes the following steps: Step 1, obtain the network topology information of SD-MANET, including the node set, node coordinates, and link set.

[0016] Software-defined mobile ad hoc network (SD-MANET) is a self-organizing network based on mobile nodes. This SD-MANET contains multiple nodes and communication links connecting the nodes. Obtain the real-time network topology information or preset network topology information of SD-MANET, including the node set V composed of nodes, the geographical or logical coordinates of each node, and the link set E composed of all communication links.

[0017] Step 2, construct an undirected graph representing the network topology, calculate the physical distance of direct links using node coordinates, and apply Dijkstra's algorithm to calculate the shortest path distance between any two nodes in the network .

[0018] In an example, as Figure 2 shown is the original topology diagram of 20 nodes within 2000m×2000m, that is, the undirected graph G=(V,E), and the serial numbers 1 to 20 represent nodes.

[0019] Step 3, define the controller deployment problem as a multi-objective optimization problem to determine the optimal set of controller deployment locations and the mapping relationship between each node and the controller.

[0020] The mapping relationship between each node and the controller refers to which controller should manage the remaining nodes after the controller position is determined, that is, to which controller the remaining nodes are assigned.

[0021] In this example, a comprehensive objective function containing the following core metrics is constructed , with minimizing the function value as the optimization objective: Total communication delay : Considering comprehensively the propagation delay and transmission delay between the data plane nodes and their managing controllers, as well as the synchronization propagation delay and transmission delay between the controllers. Among them, the propagation delay is calculated based on the shortest path distance and the signal propagation speed c, and the transmission delay is calculated based on the packet size and the link data transmission rate; Control plane link failure rate : Based on the predicted node failure probability and the link failure probability , calculate the reliability of each controller domain , and use the link failure rate of all control domains to quantify the overall link failure risk of the network; Controller load balance degree : Based on the flow request rate of each node and the maximum processing capacity of the controller, calculate the load and utilization rate of each controller, and use the standard deviation or variance of the controller utilization rate to measure the load balance degree, with the goal of minimizing this difference.

[0022] This objective function is the weighted sum of the above three metrics, that is . At the same time, it is also necessary to set constraint conditions: a) Each node must and can only be managed by one controller; b) Each deployed controller manages at least one node; c) The total number of controllers deployed in the network is fixed at m; d) The load of each controller shall not exceed its capacity; e) The weight coefficients of the objective function are non - negative and their sum is 1.

[0023] Specifically, step 3 includes: Calculate the propagation delay from the data plane nodes to the controller, and the formula is: , In the formula, V represents the set of data nodes C represents the set of controllers , represents the i - th node, represents the j - th controller, represents the node and the controller the shortest path distance between them, Calculate the transmission delay from the data plane node to the controller. The formula is: , In the formula, represents the transmission delay from node to the controller . The formula is: , In the formula, represents the packet size of the data plane node , represents the transmission rate between the data plane node and the controller . The formula is: , In the formula, B represents the bandwidth, represents the transmission power of the data plane node , represents the variance of Gaussian white noise; Calculate the propagation delay between controllers. The formula is: , In the formula, represents the shortest distance between controller and controller ; Calculate the transmission delay between controllers. The formula is: , In the formula, represents the transmission delay from controller to controller . The formula is: , In the formula, represents the packet size of controller , represents the transmission rate between controller and controller . The formula is: , In the formula, B represents the bandwidth, represents the transmission power of controller , represents the variance of Gaussian white noise; Then the total communication delay of the controller domain is: .

[0024] Specifically, step 3 further includes: According to the node failure probability within the control domain of each controller and the link failure probability calculate the reliability of the control link within the domain of the controller which is expressed as: , wherein represents the set of nodes managed by the j-th controller and represents the set of links e within the domain of controller j; Calculate the failure rate of the control plane links in the SD-MANET network, which is expressed as: , wherein m represents the number of controllers in the SD-MANET network.

[0025] Specifically, step 3 further includes: Calculate the load of each controller in the SD-MANET network, and the formula is: , wherein represents the number of flow requests of the node represents whether there is a link between the node and represents the existence of a link represents the non-existence of a link; Calculate the utilization rate of the controller , and the formula is: , wherein represents the maximum load capacity of the controller Calculate the average utilization rate of all controllers in the SD-MANET network, and the formula is: , Calculate the load balancing degree of the controller, and the formula is: .

[0026] Specifically, step 3 further includes: Construct a multi-objective optimization function F for controller deployment based on the total communication delay, control plane link failure rate, and load balancing rate within the controller domain, with the optimization goal of minimizing the function value; where the formula for the multi-objective optimization function is: , wherein respectively represent the total communication delay 、Failure rate of control plane links 、Load balancing degree of the controller ; weight coefficient The constraint conditions for constructing the multi-objective optimization function are as follows: , , , , , In the formula, V represents the set of data plane nodes , represents the th controller; The first constraint condition means that each node can only be managed by one controller. The second constraint condition means that a controller should manage at least one node. The third constraint condition means that the number of controllers in the SD-MANET network is m. The fourth constraint condition means that the controller cannot exceed its load capacity. The fifth constraint condition means that the weighted sum of the weight coefficients of the multi-objective optimization function is 1.

[0027] Step 4: Use the improved pied kingfisher optimization algorithm to solve the multi-objective optimization problem, obtain the optimal controller deployment plan, and deploy the controller and its management domain in the SD-MANET environment according to the optimal controller deployment plan.

[0028] In this example, the pied kingfisher optimization algorithm is improved. The key improvement points include: (1) Enhanced chaotic initialization: Use the Chebyshev chaotic map to replace the random initialization strategy of the standard algorithm to generate the initial population, aiming to improve the diversity and ergodicity of the initial solution and avoid premature convergence.

[0029] (2) Enhanced exploration with Levy flight: In the exploration stage of the algorithm, introduce the Levy flight mechanism to adjust the individual position update step size. Utilize the occasional large jump characteristic of Levy flight to enhance the global search ability of the algorithm and help jump out of the local optimal trap.

[0030] (3) Enhanced exploitation with Gaussian mutation: In the exploitation stage of the algorithm, introduce the Gaussian mutation strategy for the positions obtained through the original update rule. Apply a small perturbation conforming to the Gaussian distribution to the individual positions with a certain probability, aiming to improve the local search accuracy of the algorithm in the neighborhood of the optimal solution. The algorithm continuously updates the positions of individuals in the population (representing the controller deployment plan) by iteratively executing the exploration, exploitation, and symbiosis stages until the termination conditions are met, such as reaching the maximum number of iterations or the accuracy requirement of the solution, and output the best fitness, that is, the objective function value. The smallest individual is taken as the optimal solution.

[0031] Combine Figure 3 As shown, step 4 includes: Step 41, set the maximum number of iterations. Take the specific deployment location of each controller as an individual in the population, and introduce the Chebyshev chaotic map to initialize the population. The formula is: , In the formula, represents the position of the i-th individual in the j-th dimension, respectively represent the upper and lower boundaries of the search range of the -th dimension of the search space, represents the value within (-1, 1) generated by the Chebyshev chaotic map, N represents the total number of individuals, represents the number of dimensions; The population initialization formula of the Chebyshev chaotic map is as follows: , When the dimension is 1, provide the initial seed value for the Chebyshev chaotic map, and the value is between 0 and 1. When the dimension is greater than 1, use the value of the previous dimension to generate subsequent values through the Chebyshev map. In this example , ensure that no matter how close the initial value selections are, the iterated sequences are uncorrelated, that is, they are chaotic and ergodic within this range; Step 42, in the exploration stage, that is, when the random floating-point number Rand < 0.8, introduce the Levy flight strategy to update the population position. The large-step jumps of Levy flight can help the algorithm explore the search space more effectively, jump out of the local optimal trap, and discover potential optimal solutions in a wider area; the update formula is: , In the formula, is the position of the individual at the current iteration s, is the position of another individual randomly selected from the population to guide the movement and exploration of , is the state parameter, is the random direction parameter, is a random value within the normal distribution; is the final step size of Levy flight, and the formula is: , In the formula, and are both random numbers generated from the normal distribution, is the parameter of Levy flight. In the example, it can be set ; The strategies in the exploration stage of the pied kingfisher are divided into the perching strategy and the hovering strategy. The main difference lies in the state parameter value . When 0.5 < Rand < 0.8, it enters the perching strategy. When Rand ≤ 0.5, it enters the hovering strategy.

[0032] In the perching strategy the value formula is as follows: , wherein, represents the maximum number of iterations, is a constant with a set value of 8, is a random number; In the hovering strategy the value formula is as follows: , wherein, is the fitness of the -th and the -th pied kingfishers. The fitness function is the multi-objective optimization function F; Step 43, in the development stage, when the random floating-point number Rand ≥ 0.8, a Gaussian mutation strategy is introduced to update the population position. The update formula is: , wherein, is the new position obtained under the influence of the probability , is the standard deviation, and the formula is , is a coefficient used to control the magnitude of the standard deviation, respectively represent the upper and lower boundaries of the search range; rand represents a random number in [0, 1], is the mutation probability, indicating that it perturbs the individual variables with a certain probability ; represents the new position obtained by the individual after introducing Gaussian mutation; Step 44, in the symbiotic stage, the population position at the current iteration is updated. The update formula is: , wherein, represents the positions of randomly selected 2 individuals from the population at the current iteration s, is the predation efficiency of the pied kingfisher, represents the hunting ability, Represents a random number in the range [0, 1]; , wherein, are the maximum and minimum values of the predation efficiency of the pied kingfisher; Step 45: Combine the updated new population, calculate the fitness value of each individual in the new population, sort the fitness values from smallest to largest, and take the individual corresponding to the smallest fitness value as the optimal feasible solution at the current iteration number; Determine whether it is greater than the iteration number. If so, output the optimal feasible solution; otherwise, return to the exploration stage to solve and iterate again.

[0033] To further illustrate the effectiveness and superiority of the SD-MANET multi-controller deployment method based on the improved pied kingfisher algorithm described in the present invention, it is illustrated by the following examples.

[0034] Table 1 shows the parameters and their corresponding values selected when calculating the transmission delay in the ad hoc mode according to the IEEE 802.11ag protocol.

[0035] Table 1 Parameter Selection

[0036] Figure 4 is the topology diagram corresponding to the controller deployment scheme solved by the EPKO algorithm. The controllers are located at node 3, node 12, and node 18 respectively. Among them, the controller located at node 3 manages nodes 1 and 17; the controller located at node 12 manages nodes 4, 6, 9, 13, and 19; the remaining nodes are managed by the controller at node 18.

[0037] Denote the method described in the present invention as EPKO, Figure 5 is a comparison chart of the fitness curves of the EPKO algorithm, the pied kingfisher algorithm (PKO), the particle swarm optimization algorithm (PSO), and the sparrow optimization algorithm (SCSSA) that fuses the sine-cosine strategy and Cauchy mutation when the optimal controller deployment quantity and location are obtained. It can be seen that the performance of the EPKO algorithm is better than that of the PKO algorithm, the PSO algorithm, and the SCSSA algorithm, and the solved controller deployment location is better.

[0038] Figure 6 is a comparison chart of the control plane link failure rates of the EPKO algorithm, the PKO algorithm, the PSO algorithm, and the SCSSA algorithm when the optimal controller deployment scheme is solved. The lower the failure rate, the better. It can be seen that the link failure rate effect of the controller deployment scheme solved by the EPKO algorithm is better.

[0039] Figure 7It is a comparison chart of the delay when the EPKO algorithm, PKO algorithm, PSO algorithm, and SCSSA algorithm solve the optimal controller deployment scheme. The lower the delay, the better. It can be seen that the delay effect of the controller deployment scheme solved by the EPKO algorithm is better.

[0040] Figure 8 It is a comparison chart of the load balancing degree when the EPKO algorithm, PKO algorithm, PSO algorithm, and SCSSA algorithm solve the optimal controller deployment scheme. Although there is a certain gap between the EPKO algorithm and other algorithms in terms of load balancing, this is the result of a trade-off in multi-objective optimization, which gives priority to ensuring link stability and delay performance, thus improving the overall reliability and efficiency of the SD-MANET network.

Claims

1. An SD-MANET multi-controller deployment method based on an improved pied kingfisher algorithm, characterized in that, It includes the following steps: Step 1: Obtain the network topology information of SD-MANET, including the node set, node coordinates, and link set; Step 2: Construct an undirected graph representing the network topology, calculate the physical distance of direct links using node coordinates, and apply the Dijkstra algorithm to calculate the shortest path distance between any two nodes in the network; Step 3: Define the controller deployment problem as a multi-objective optimization problem to determine the optimal set of controller deployment locations and the mapping relationship between each node and the controller; Step 4: Use an improved spotted kingfisher optimization algorithm to solve the multi-objective optimization problem, obtain the optimal controller deployment plan, and deploy the controller and its management domain in the SD-MANET environment according to the optimal controller deployment plan.

2. The SD-MANET multi-controller deployment method based on the improved kingfisher algorithm according to claim 1, characterized in that Step 3 includes: Calculate the propagation delay from the data plane node to the controller, and the formula is: ， Wherein, V represents the set of data nodes and C represents the set of controllers . represents the i-th node and represents the j-th controller represents the node and the controller the shortest path distance therebetween Calculate the transmission delay from the data plane node to the controller, and the formula is: ， In the formula, represents the node to the controller transmission delay; Calculate the propagation delay between controllers, and the formula is: ， In the formula, represents the controller and the controller the shortest distance between; Calculate the transmission delay between controllers, and the formula is: ， In the formula, represents the transmission delay from the controller to the controller ; Then the total communication delay of the controller domain is: 。 3. The SD-MANET multi-controller deployment method based on the improved kingfisher algorithm according to claim 2, wherein Step 3 also includes: According to the node failure probability within the control domain of each controller and the link failure probability , calculate the reliability of the control link within the control domain of the controller , which is expressed as: ， wherein, represents the nodes managed by the j-th controller set, represents the set of links e within the domain of controller j; Calculate the control plane link failure rate of the SD-MANET network, which is expressed as: ， In the formula, m represents the number of controllers in the SD-MANET network.

4. A method for deploying multiple controllers in SD-MANET based on an improved kingfisher algorithm according to claim 3, characterized in that, Step 3 also includes: Calculate the load of each controller in the SD-MANET network as follows using the formula: ， In the formula, represents the number of flow requests of a node, represents a node and whether there is a link between them, represents the existence of a link, represents the non-existence of a link; Calculation controller The usage rate is calculated using the formula: ， In the formula, represents the maximum load capacity of the controller ; Calculate the average utilization rate of all controllers in the SD-MANET network, and the formula is: ， Calculate the load balancing degree of the controller, and the formula is: 。 5. A method for deploying multiple controllers in SD-MANET based on an improved kingfisher algorithm according to claim 4, characterized in that Step 3 also includes: Construct a multi-objective optimization function F for controller deployment based on the total communication delay, control plane link failure rate, and load balancing rate within the controller domain, and take minimizing the function value as the optimization goal; where the formula of the multi-objective optimization function is: ， In the formula, respectively represent the total communication delay , the failure rate of the control plane link , and the load balancing degree of the controller ; the weight coefficients The constraint conditions for constructing the multi-objective optimization function are: ， ， ， ， ， where V represents the set of data plane nodes ; represents the th controller The first constraint condition means that each node can only be managed by one controller, the second constraint condition means that a controller should manage at least one node, the third constraint condition means that the number of controllers in the SD-MANET network is m; the fourth constraint condition means that the controller cannot exceed its load capacity; the fifth constraint condition means that the weighted sum of the weight coefficients of the multi-objective optimization function is 1.

6. The method for deploying multiple controllers in SD-MANET based on the improved pied kingfisher algorithm according to claim 5, wherein Step 4 includes: Step 41: Set the maximum number of iterations. Take the specific deployment location of each controller as an individual in the population, and introduce the Chebyshev chaotic mapping to initialize the population. The formula is: ， Wherein, represents the position of the i-th individual in the j-th dimension, respectively represent the upper and lower boundaries of the search range of the -th dimension of the search space, represents the value within (-1, 1) generated by the Chebyshev chaotic map, N represents the total number of individuals, represents the number of dimensions; Step 42: Introduce the Levy flight strategy to update the population position in the exploration stage. The update formula is: ， Wherein, is the position of an individual at the current iteration number s, is another individual randomly selected from the population whose position is used to guide the movement and exploration of, is a state parameter, is a random direction parameter, is a random value within the normal distribution; is the final step size of the Lévy flight, and the formula is: ， wherein, and are both random numbers generated from a normal distribution, is the parameter of Levy flight; Step 43: Introduce the Gaussian mutation strategy to update the population position in the development stage. The update formula is: ， Wherein, is the new position obtained under the influence of probability ; is the standard deviation, and the formula is ; is a coefficient used to control the magnitude of the standard deviation; respectively represent the upper and lower boundaries of the search range; rand represents a random number in [0, 1]; is the mutation probability, indicating that it perturbs the individual variables with a certain probability ; represents the new position obtained by the individual after introducing Gaussian mutation; Step 44: In the symbiosis stage, update the population position at the current iteration. The update formula is: ， In the formula, represents the positions of randomly selecting 2 individuals from the population at the current iteration number s, is the predation efficiency of the pied kingfisher, represents the hunting ability, represents a random number in [0, 1]; Step 45: Merge the updated new population, calculate the fitness value of each individual in the new population, sort the fitness values from small to large, and take the individual corresponding to the smallest fitness value as the optimal feasible solution at the current iteration; judge whether it is greater than the number of iterations. If so, output the optimal feasible solution, otherwise return to the exploration stage to solve and iterate again.

Images