Space-earth integrated load balancing method based on improved particle swarm and hippopotamus algorithm

Abstract

The present invention discloses a space-ground integrated load balancing method based on improved particle swarm and hippopotamus algorithm. According to the space-ground integrated intelligent network structure, an objective optimization model and constraint conditions are established. Taking particles to represent the improved solutions of the space-ground integrated intelligent network structure, the set of particles forms a population, and an adaptive meta-heuristic mapping is introduced. The objective optimization model is solved, a selective reset mechanism is introduced, and the hippopotamus optimization algorithm is adopted to simulate the defense behavior and escape behavior of particles, and a better particle is selected as the optimal space-ground integrated intelligent network structure output. This method has good adaptability, robustness and scalability, improves efficiency, can respond to changes in network status, and provides an efficient traffic management solution for the link load balancing problem in the space-ground integrated intelligent network.

Description

Technical Field

[0001] The present invention relates to the field of information engineering, and particularly to a space-ground integrated load balancing method based on improved particle swarm and hippopotamus algorithm. Background Art

[0002] As a comprehensive communication network system, the space-ground integrated network aims to achieve seamless coverage and information transmission globally by integrating space-based satellite network and ground network resources, providing users with the ability of ubiquitous access and on-demand services, while ensuring the security and reliability of the network. However, with the explosive growth of the number of Internet users and data traffic, the space-ground integrated network is facing unprecedented challenges, such as network congestion and data security problems, which seriously threaten the efficient operation of the network and the normal circulation of data. To solve these problems and ensure the reasonable allocation and optimization of network resources, network load balancing algorithms have emerged. The load balancing algorithm is a key technology to improve network performance, reduce network load and enhance network service quality. In the current research of load balancing algorithms, heuristic algorithms have received extensive attention internationally due to their effectiveness in solving complex optimization problems.

[0003] For heuristic algorithms, especially genetic algorithms and ant colony algorithms, researchers have done a lot of work on these basic algorithms, developing various variants and improved algorithms to improve the convergence speed and solution quality of the algorithms. The prior art integrates heuristic algorithms into a memetic algorithm and uses it as an initialization method and mutation operator; some solutions improve and optimize the existing PSO load balancing algorithm by dynamically adjusting the parameters or mechanisms of the PSO algorithm, such as adaptively adjusting the inertia mechanism, improving the compression mechanism, dynamic neighborhood scheduling strategy, double-population cross learning, combined with divide-and-conquer algorithm and traffic scheduling.

[0004] The above-mentioned heuristic algorithms are suitable for solving complex optimization problems and are widely applied. However, when solving problems, heuristic algorithms are prone to falling into local optimal solutions and premature convergence. A local optimal solution is the optimal solution in a certain area of the search space, but not necessarily the global optimal solution. Heuristic algorithms rely on local information and are easily attracted to fall into local optima. For example, in genetic algorithms, poor quality of the initial population or conservative selection operations may lead to the loss of excellent genes, causing the algorithm to converge prematurely. Premature convergence means that the algorithm stops searching before finding the global optimal solution, usually due to improper parameter settings or ineffective search strategies. For example, in particle swarm optimization, if the inertia weight is too large, the particles will move quickly and cannot search carefully, resulting in premature convergence. Insufficient population diversity will also cause the algorithm to lose the ability to explore new solution spaces, thus leading to premature convergence. Summary of the Invention

[0005] Objective of the present invention: To address the deficiencies of the existing particle swarm optimization algorithm, such as premature convergence of particles, and the lack of global exploration ability and easy entrapment in local optima in the standard hippopotamus optimization algorithm, the present invention provides a space-earth integrated load balancing method (PH-PSO) based on improved particle swarm and hippopotamus algorithms. In the space-earth integrated network environment, this method realizes an innovative hybrid optimization method by integrating the advantages of the improved particle swarm optimization and the hippopotamus optimization algorithm (HO). This hybrid optimization method not only improves the adaptability and robustness of the method, but also enhances its efficiency and scalability, enabling the PH-PSO method to intelligently respond to changes in network status and providing an efficient traffic management solution for the link load balancing problem in the space-earth integrated network.

[0006] To achieve the above functions, the present invention designs a space-earth integrated load balancing method based on improved particle swarm and hippopotamus algorithms. For the space-earth integrated network, the following steps S1 - S3 are executed to complete the optimization of the space-earth integrated network structure:

[0007] Step S1: According to the space-earth integrated network structure, establish a network topology. Based on the multi-commodity flow characteristics of network traffic, link bandwidth constraints, and the optimization requirements of traffic allocation, establish an objective optimization model, and define corresponding constraint conditions for the objective optimization model.

[0008] Step S2: Use particles to represent the improved solutions of the space-earth integrated network structure. According to the characteristics of the space-earth integrated network structure, encode the positions and velocities of the particles. With particles as individuals, the set of all particles forms a population. In the population initialization stage, introduce an adaptive meta-heuristic mapping to increase the diversity of the population.

[0009] Step S3: Solve the objective optimization model. Introduce a selective reset mechanism to reset the positions of the particles. Adopt the hippopotamus optimization algorithm to simulate the defense behavior and escape behavior of the particles. Compare the new positions and fitness values of the particles after the defense behavior and escape behavior, and select the better particles as the optimal space-earth integrated network structure output to complete the optimization of the space-earth integrated network structure.

[0010] Beneficial effects: Compared with the prior art, the advantages of the present invention include:

[0011] 1. The present invention adopts an adaptive meta - heuristic mapping, a selective reset mechanism, and a hippopotamus optimization algorithm. The adaptive meta - heuristic mapping generates a high - quality and diverse initial population by selecting or combining different initialization methods. The selective reset mechanism enhances global search, enabling the algorithm to jump out of local optima and explore a wider solution space. The hippopotamus optimization algorithm increases diversity and local - optimum sensitivity, refining the search process. Multiple mechanisms complement each other, jointly promoting the search and optimization effects. Global search quickly locates high - quality solution regions, and local search deeply explores solution details, achieving a balance between global and local, efficiently optimizing network traffic allocation and load - balancing strategies, significantly improving network performance, and ensuring business QoS requirements.

[0012] 2. The present invention makes the links in the space - ground integrated network tend to be balanced, avoiding link congestion. At the same time, this method can balance the traffic in the network, ensuring the utilization rate of the links. BRIEF DESCRIPTION OF THE DRAWINGS

[0013] Figure 1 is a flowchart of the space - ground integrated load - balancing method based on an improved particle swarm and hippopotamus algorithm according to an embodiment of the present invention;

[0014] Figure 2 is a schematic diagram of the particle reset mechanism according to an embodiment of the present invention;

[0015] Figure 3 is a comparison chart of fitness convergence according to an embodiment of the present invention;

[0016] Figure 4 is a comparison chart of jitter according to an embodiment of the present invention;

[0017] Figure 5 is a link load display chart of the space - ground integrated load - balancing method based on an improved particle swarm and hippopotamus algorithm according to an embodiment of the present invention;

[0018] Figure 6 is a link load display chart of the improved particle swarm optimization algorithm according to an embodiment of the present invention;

[0019] Figure 7 is a link load display chart of the particle reset optimization algorithm according to an embodiment of the present invention;

[0020] Figure 8 is a link load display chart of the basic particle swarm algorithm according to an embodiment of the present invention. DETAILED DESCRIPTION OF THE EMBODIMENTS

[0021] The present invention will be further described below with reference to the drawings. The following embodiments are only used to more clearly illustrate the technical solution of the present invention and should not be used to limit the protection scope of the present invention.

[0022] The integrated space-ground load balancing method based on improved particle swarm and hippopotamus algorithm provided by the embodiment of the present invention is directed to the integrated space-ground network. Referring to Figure 1 , the following steps S1-S3 are executed to complete the optimization of the integrated space-ground network structure:

[0023] Step S1: According to the integrated space-ground network structure, a network topology is established. Based on the multi-commodity flow characteristics of network traffic, link bandwidth constraints, and the optimization requirements of traffic allocation, a target optimization model is established, and corresponding constraint conditions are defined for the target optimization model;

[0024] The specific steps for establishing the target optimization model in Step S1 are as follows:

[0025] Step S1.1: The network topology is a directed graph , where V is the set of nodes, E is the set of edges. For the link in the network from node i to node j , it is represented by the edge , and is used to represent the bandwidth that the edge can provide, is used to represent the reliability weight of the edge ; assume that there are G in the network K data streams transmitted simultaneously. For the data stream k , is used to represent the bandwidth it requires; on the edge , the traffic of the data stream k is represented by , and the traffic of the entire data stream k is represented by the vector ; the transmission cost per unit traffic in the data stream k is represented by the vector ; the length of the path k of the data stream is ;

[0026] Step S1.2: Each path is composed of many links. A linear function is used to define the transmission cost as follows:

[0027] ;

[0028] Among them, , is a constant, is the length of, is Utilization rate;

[0029] In a given network The objective function of the algorithm for the flow distribution based on the path length and link utilization rate is expressed as the following formula:

[0030] ;

[0031] Under the framework of the algorithm, the position of each particle corresponds to a fitness value calculated by the objective function, which is used to measure the quality of the particle solution; when the algorithm is applied to a constrained optimization problem, the key lies in two main steps: First, the specific solution of the problem needs to be transformed into the encoded form of the particle; Second, a fitness function that can accurately reflect the particle performance needs to be defined; this fitness function will guide the evolution process of the particle in the search space in order to find the optimal solution; so the final objective function model is expressed as:

[0032] ;

[0033] In the formula, Represents the objective optimization model, , Are constants, Is the data stream k Of the path Of the length, Is Utilization rate;

[0034] Step S1.3: For the objective optimization model, define the link bandwidth constraint, traffic request bandwidth constraint, non-negativity constraint, and traffic conservation constraint, that is, under the satisfaction of these constraint conditions, make the transmission cost of the entire network reach the minimum, specifically as follows:

[0035] Link bandwidth constraint:

[0036] ;

[0037] In the formula, Represents the traffic of all links from node i To node j In the network;

[0038] The link bandwidth constraint ensures that the traffic distribution does not exceed the actual carrying capacity of the link, thus avoiding link congestion, optimizing the network resource utilization rate, and providing necessary constraint conditions for the optimization algorithm to achieve effective load balancing and network performance improvement.

[0039] Traffic request bandwidth constraint:

[0040] ;

[0041] The traffic request bandwidth constraint ensures that the traffic demands of each data flow are fully met in path allocation, providing a clear allocation goal for the optimization algorithm, thus supporting the implementation of load balancing and ensuring the efficient operation of the network.

[0042] Non-negativity constraint:

[0043] ;

[0044] The role of the non-negativity constraint is to ensure the rationality of traffic allocation. It stipulates that the traffic on each link must be greater than or equal to zero, thus avoiding unrealistic negative traffic situations. This constraint, combined with the link bandwidth constraint and the traffic request constraint, provides clear boundary conditions for the optimization algorithm, ensuring that the traffic allocation scheme solved by the algorithm is feasible in the actual network environment, while guaranteeing the stability of network operation and the implementation of load balancing.

[0045] Traffic conservation constraint:

[0046] ;

[0047] where and represent the source node and the destination node of data flow k respectively;

[0048] The traffic conservation constraint ensures that the traffic input and output of each node are balanced in the traffic allocation problem, that is, the sum of the traffic entering a node is equal to the sum of the traffic leaving the node. This is necessary in the actual network environment because data does not appear or disappear out of thin air during transmission and must follow the conservation principle. The traffic conservation constraint provides basic constraint conditions for the optimization algorithm, ensuring that the traffic allocation scheme solved by the algorithm is feasible in the actual network.

[0049] Step S2: Represent the improved scheme of the space-ground integrated network structure with particles. According to the characteristics of the space-ground integrated network structure, encode the position and velocity of the particles. Use the particles as individuals, and the set of all particles constitutes a population. In the population initialization stage, introduce an adaptive meta-heuristic mapping to increase the diversity of the population;

[0050] The specific steps of Step S2 are as follows:

[0051] Step S2.1: Each particle represents a potential improved scheme of the space-ground integrated network structure. Each particle not only explores the best position found in its own history (personal best, denoted as ), but also searches for the best position known among all particles (global best, denoted as ); During the iteration process, the particle adjusts its velocity and position according to these two extreme values to get closer to these extreme values; the update rules for defining the velocity and position of the particle are as follows:

[0052] ;

[0053] ;

[0054] Wherein, represents the number of iterations, and represent the velocity and position of the particle in the th iteration, and are learning factors, and the value range is between 0 and 2. The learning factors help the particle explore and utilize information in the search space, is the inertia weight, which is between 0.1 and 0.9. It controls the tendency of the particle to maintain the previous velocity, is a random number between 0 and 1. It introduces randomness to the update of the particle, increasing the diversity of the search, represents the individual best position of the particle, represents the global best position of the particle; the particle swarm continuously adjusts and optimizes the solution during the iteration process to seek the optimal solution to the problem;

[0055] Step S2.2: In the population initialization stage, an adaptive meta-heuristic mapping is introduced, including global optimization and local optimization; global optimization assigns initial traffic values to each data stream on all possible links, and local optimization performs dynamic traffic allocation according to the sharing degree of the links ;

[0056] Adaptive meta-heuristic mapping is a technique for dynamically adjusting the initialization strategy. It adaptively selects or combines different initialization methods according to the characteristics of the optimization problem (such as dimension, constraint conditions, etc.) to generate a high-quality and diverse initial population. This mapping method can further improve the global search ability, adaptability and robustness of the method of the present invention, reduce the risk of premature convergence, and at the same time reduce the computational cost. Adaptive meta-heuristic mapping is divided into two parts: global optimization (fixed dimension) and local optimization (dynamic dimension).

[0057] The goal of the global optimization described above is to assign initial traffic values to each data stream on all possible links to optimize the performance of the entire network. The specific method is as follows:

[0058] Since the dimension of the global optimization of the present invention is very large, is the total number of links, and the initialization strategy directly adopts a hybrid method of uniform distribution and normal distribution. For each data streamk and each link , the globally optimized value is initialized as:

[0059] ;

[0060] wherein, is uniform distribution sampling, is normal distribution sampling, means to limit the value of x within the range of [0, 1]:

[0061] ;

[0062] The globally optimized vector is composed of all :

[0063] ;

[0064] wherein, has a length of , is the total number of links, is the total number of data streams;

[0065] This method combines the randomness of the uniform distribution and the concentration of the normal distribution, and can generate initial solutions with better diversity and stability in the high-dimensional space.

[0066] The dimension of the local optimization described is dynamic and depends on the actual path length of each traffic ; for each data stream k , its path contains several links, represents the value of local optimization. For each data stream k and each link in its path , different sampling strategies are selected according to the path length :

[0067] When , uniform distribution sampling is adopted:

[0068]

[0069] ; represents a random number sampled from the uniform distribution to introduce randomness;

[0070] When , normal distribution sampling is adopted:

[0071] ;

[0072] represents a random number sampled from a normal distribution to introduce central tendency.

[0073] When , a mixed method of uniform distribution and normal distribution is adopted:

[0074] ;

[0075] By mixing the sampling results of these two distributions, an initial traffic allocation value with both randomness and a certain central tendency can be generated, thus providing a better starting point for the optimization process.

[0076] To consider the sharing degree of the link, it is assumed that the link is shared by multiple data streams, and its sharing degree, i.e., the number of usage times, is , then the adjusted traffic is:

[0077]

[0078] where represents the sharing degree of the link ;

[0079] Finally, the locally optimized vector is composed of all , and its length is the sum of the lengths of all paths:

[0080] ;

[0081] where the length is , is the total number of links, is the total number of data streams.

[0082] The dimension of local optimization and the sampling strategy both depend on the actual path length of the data stream . When the path length is short (e.g., ), uniform distribution sampling is used; when the path length is moderate (e.g., ), normal distribution sampling is used; when the path length is long (e.g., ), a mixed sampling of uniform distribution and normal distribution is used. This strategy can dynamically adjust the initialization method according to the complexity of the path, thus better adapting to the characteristics of different paths.

[0083] In the local optimization part, the traffic allocation will be based on the sharing degree of the link Make adjustments to ensure more reasonable traffic distribution and avoid performance issues caused by link overload.

[0084] Step S2.3: Construct particles as follows:

[0085] Each particle is composed of a global optimization part and a local optimization part:

[0086] ;

[0087] In the formula, represents the particle, represents the vector of global optimization, represents the vector of local optimization, represents the vector concatenation operation; the set of all particles constitutes the population:

[0088] ;

[0089] In the formula, represents the population, represents the particle in the population, represents the total number of all particles.

[0090] Step S3: Solve the objective optimization model, introduce a selective reset mechanism to reset the positions of the particles, adopt the hippopotamus optimization algorithm, simulate the defense behavior and escape behavior of the particles, compare the new positions and fitness values of the particles after the defense behavior and escape behavior, and select the better particles as the optimal space-ground integrated network structure output to complete the optimization of the space-ground integrated network structure.

[0091] The specific steps of Step S3 are as follows:

[0092] Step S3.1: Start iteration, introduce a selective reset mechanism, calculate the fitness value of each particle in the population to quickly locate the high-quality solution area, thereby improving the global search ability of the algorithm; when the convergence speed of the particle exceeds the set threshold, that is, when the particle quickly moves towards the local optimal solution, reset the position of the particle and place it at a random position in the search space;

[0093] In Step S3.1, for each particle in the population, if the difference between the current fitness of the particle and the individual best fitness is greater than 1% of the individual best fitness, then it is considered that the convergence speed of this particle is too fast, and the position of the particle is reset to a random position:

[0094] ;

[0095] In the formula, is the individual best fitness of the particle, is the current fitness of the particle;

[0096] If the current fitness of a particle is greater than the individual best fitness, update the individual best fitness. If the current fitness of a particle is greater than the global best fitness, update the global best fitness, and update the velocity and position of the particle, and output it as the optimal solution of the target optimization model.

[0097] The pseudocode for the above steps is shown in Table 1 below:

[0098] Table 1: Pseudocode of the particle reset mechanism

[0099]

[0100] The schematic diagram of the particle reset mechanism is as Figure 2 shown, Figure 2 In it, the red dots represent the initial positions of the particles, the blue dots represent the positions after reset, the green dots represent the global optimal solutions, and the orange dots represent the individual optimal solutions. It can be seen that when the particles move rapidly towards the local optimal solution, through the reset mechanism, the particles are reset to random positions in the search space, thus improving the global search ability of the algorithm.

[0101] Step S3.2: During the iteration process, for the particles with position reset, simulate the defense behavior and escape behavior of the hippopotamus optimization algorithm, calculate the new positions of the particles after the defense behavior and escape behavior, and according to the fitness values of the new positions of the particles after the defense behavior and escape behavior, so that the algorithm can deeply explore the details of the solution region, thereby improving the local search ability of the algorithm; select the better particle position as the final result;

[0102] Simulate the defense behavior and escape behavior of the hippopotamus optimization algorithm to enhance the local search ability. This simulation enables the particles to better avoid falling into the local optimal solution during the search process, while enhancing the detailed exploration of the solution space. When the particles approach the local optimal region, the defense behavior prompts them to disperse, while the escape behavior helps the particles break free from the bondage of the local optimum and continue to move towards a better solution.

[0103] The behavior of hippopotamuses defending against predators: In the natural environment, hippopotamuses will form a defensive formation to protect the cubs in the group from the threat of predators. In the optimization algorithm, this behavior is simulated as a repulsion mechanism between particles: when the particles are too close to each other, they will repel each other to avoid gathering near the local optimal solution. This mechanism is achieved by randomly selecting a "predator" particle and moving the current particle away from this "predator".

[0104] The behavior of hippopotamuses escaping from predators: When encountering predators, hippopotamuses will quickly escape to ensure their own safety. In the optimization algorithm, this behavior is simulated as the sensitivity of particles to the global optimal solution (gBest): if a particle finds that it is too close to gBest, it will be pushed in the direction away from gBest.

[0105] By simulating two key behaviors of the hippopotamus optimization algorithm, the diversity among particles and the sensitivity to local optima are increased, thereby improving the local search ability of the algorithm.

[0106] The specific steps of step S3.2 are as follows:

[0107] Step S3.2.1: For the particles whose positions are reset in step S3.1, if their random search probability is less than 0.5, that is, satisfying the following formula, simulate the defense behavior and escape behavior of the hippopotamus optimization algorithm for the particles:

[0108] ;

[0109] In the formula, represents the random search probability, which determines whether the current particle needs to perform local search with a probability of 50%. The role of this probability mechanism is to balance global search and local search: to avoid all particles entering local search simultaneously, thereby retaining a certain global search ability.

[0110] Execute steps S3.2.2 - S3.2.4 to simulate the behavior of a hippopotamus defending against predators:

[0111] Step S3.2.2: Generate the position of the predator as follows:

[0112] ;

[0113] Among them, is the lower bound of the global search space, is the upper bound of the global search space, ensuring that the initial position of the particle is within a reasonable range and avoiding the particle starting the search from an unreasonable position. is a uniformly distributed random number in the range of ; Its role is to generate a random position in the current search space as the position of the predator. By randomly generating the position of the predator, the scenario of a hippopotamus facing a predator is simulated, thereby introducing randomness and enhancing the diversity of the search.

[0114] Step S3.2.3: Calculate the distance between the current solution (hippopotamus) and the predator as follows:

[0115] ;

[0116] Among them, is the position of the current particle;

[0117] Step S3.2.4: Generate a defense strategy and calculate the new position of the particle after the defense behavior:

[0118] If the position of the predator has a lower fitness than the current solution, that is, it is better than the current solution, then adjust the position of the current solution according to the defense strategy:

[0119] ;

[0120] Otherwise, use a more conservative defense strategy:

[0121] ;

[0122] Among them, is the new position of the particle after the defense behavior, , that is, randomly generate a value from the interval [2, 4); , that is, randomly generate a value from the interval [1, 1.5); , that is, randomly generate a value from the interval [2, 3); is the angular parameter sampled from the uniform distribution, , that is, randomly generate a value from the interval [−2 π , 2 π ); is a random number of the uniform distribution, and the range is , is random number;

[0123] Execute steps S3.2.5 - S3.2.8 to simulate the behavior of the hippopotamus escaping from the predator:

[0124] Step S3.2.5: Define the local search range:

[0125] According to the current iteration number and the maximum iteration number T , dynamically adjust the upper and lower bounds of the local search:

[0126] ;

[0127] ;

[0128] Among them, is the upper bound of the local search, is the lower bound of the local search;

[0129] Step S3.2.6: Randomly select an escape strategy:

[0130] Randomly select an escape strategy , and the escape strategy is one of the following three:

[0131] ;

[0132] In the formula, is a uniformly distributed random number in the range [-1, 1);

[0133] ;

[0134] wherein, is a uniformly distributed random number in the range [0, 1);

[0135] ;

[0136] wherein, represents a random number of normal distribution;

[0137] Step S3.2.7: According to the selected escape strategy, calculate the new position of the particle after the escape behavior:

[0138] ;

[0139] wherein, is the new position of the particle after the escape behavior, is a random number within the range of; by multiplying by the random number, the size of the step length can be adjusted so that the step length of each escape behavior is slightly different, increasing the diversity of the search.

[0140] Step S3.2.8: Compare the fitness of the new positions of the particles after the defense behavior and the escape behavior, and select the better particle position as the final result:

[0141] ;

[0142] wherein, is the fitness of the new position of the particle after the escape behavior, is the fitness of the new position of the particle after the defense behavior;

[0143] Limit the generated new position within the range of the search space:

[0144] ;

[0145] wherein, represents a clipping operation.

[0146] This step is used to limit the particle position within the boundary (0 to 1) of the solution space, ensure that the particle is always within the range of valid solutions, prevent calculation errors or violations of problem constraints caused by exceeding the boundary, and thus ensure the stability and convergence of the algorithm.

[0147] Step S3.3: Determine whether the maximum number of iterations is satisfied. If so, output the space-air-ground integrated network structure corresponding to the individuals in the population after iteration as the optimal space-air-ground integrated network structure; otherwise, execute Step S3.1.

[0148] The comparison chart of the fitness convergence of the method of the present invention (PH-PSO) with the basic particle swarm optimization algorithm (PSO), the particle reset optimization algorithm (PR-PSO), and the improved particle swarm optimization algorithm (IPSO) is referred to Figure 3 , and the comparison chart of the jitter situation is referred to Figure 4 . It can be seen from the comparison that the method of the present invention has improved by 28.018%, 17.996%, and 11.111% respectively compared with the basic particle swarm optimization algorithm, the particle reset optimization algorithm, and the improved particle swarm optimization algorithm in terms of the jitter index;

[0149] The display diagrams of the link loads of the method of the present invention (PH-PSO), the basic particle swarm optimization algorithm (PSO), the particle reset optimization algorithm (PR-PSO), and the improved particle swarm optimization algorithm (IPSO) are respectively referred to Figure 5 , Figure 6 , Figure 7 , Figure 8 . Among them, different link connections represent different link utilization rates. The thin solid line represents that the utilization rate of the link is 0-0.3, the dotted line represents that the utilization rate of the link is 0.3-0.5, the dotted line represents that the utilization rate of the link is 0.5-0.7, and the dotted-dashed line represents that the utilization rate of the link is 0.7-1. The situation of link congestion (utilization rate greater than 0.3) generally appears in the other three algorithms. However, in the link load display diagram corresponding to the method of the present invention, it is basically a thin solid line, indicating that after adopting this method, the links tend to be balanced and there is no link congestion situation. At the same time, it also verifies that this method can balance the traffic in the network and well ensure the utilization rate of the links.

[0150] The embodiments of the present invention have been described in detail above with reference to the accompanying drawings. However, the present invention is not limited to the above embodiments. Within the scope of knowledge possessed by those of ordinary skill in the art, various changes can be made without departing from the gist of the present invention.

Claims

1. A space-ground integrated load balancing method based on improved particle swarm and Hippo algorithm, characterized in that: For the integrated space-ground intelligent network, the following steps S1 to S3 are performed to complete the optimization of the integrated space-ground intelligent network structure: Step S1: Establish a network topology based on the integrated network structure of the ground and the sky, establish a target optimization model based on the multi-commodity flow characteristics of network traffic, link bandwidth constraints, and optimization requirements for traffic distribution, and define corresponding constraints for the target optimization model; By building a directed graph ,in V is a collection of nodes, E is a set of edges. For nodes in the network i To Node j The link is edge To express, Represents edge Reliability weight; Assuming that in the network G There are K Data streams are transmitted simultaneously, data stream k The flow rate To represent the data flow k Path The length is , for Utilization rate, establish target optimization model : , , is a constant; Step S2: An improved scheme using particles to represent the integrated space-ground network structure. Aiming at the characteristics of the integrated space-ground network structure, the position and speed of the particles are encoded. The particles are taken as individuals, and the collection of all particles constitutes a population. Each particle is composed of a global optimization part and a local optimization part. The target optimization model is a fitness function that reflects the performance of the particles. In the population initialization stage, an adaptive meta-heuristic mapping is introduced, including global optimization and local optimization. The global optimization allocates initial flow values ​​for each data flow on all possible links, and the local optimization performs dynamic flow allocation according to the sharing degree of the links to increase the diversity of the population. Step S3: Solve the target optimization model, introduce a selective reset mechanism to reset the position of the particles, use the Hippo optimization algorithm to simulate the defense and escape behaviors of the particles, compare the new positions and fitness values ​​of the particles after the defense and escape behaviors, select better particles as the optimal ground-ground integrated network structure output, and complete the optimization of the ground-ground integrated network structure.

2. The space-ground integrated load balancing method based on improved particle swarm and Hippo algorithm according to claim 1 is characterized in that: The specific steps for establishing the target optimization model in step S1 are as follows: Step S1.1: The network topology is a directed graph ,in V is a collection of nodes, E is a set of edges, for nodes in the network i To Node j Link, with edge To express, and use Represents edge The bandwidth available, Represents edge Reliability weight; Assuming that in the network G There are K Data streams are transmitted simultaneously. k ,use Indicates the bandwidth required; On, data flow k The flow rate To represent the entire data flow k The flow rate is expressed as vector To express; Data Flow k The transmission cost per unit flow is represented by the vector To express the transmission cost Define as follows: ； in, , is a constant, for Length, for Utilization rate; Data Flow k Path The length is ; Step S1.2: In a given network In the above example, the target optimization model is established as follows: ； In the formula, represents the target optimization model, , is a constant, For data flow k Path Length, for Utilization rate; Step S1.3: For the target optimization model, define the link bandwidth constraint, traffic request bandwidth constraint, non-negativity constraint, and traffic conservation constraint as follows: Link bandwidth constraints: ； In the formula, Represents a slave node in the network i To Node j Traffic of all links; Traffic request bandwidth constraints: ； Non-negativity constraints: ； Conservation constraints for flow: ； in, and Respectively represent data flow k The source and destination nodes.

3. The space-ground integrated load balancing method based on improved particle swarm and hippo algorithm according to claim 1, characterized in that: The specific steps of step S2 are as follows: Step S2.1: Define the update rules for the particle's velocity and position as follows: ； ； In the formula, represents the number of iterations, , Represents the particle The speed and position of the round iteration, and is the learning factor, ranging from 0 to 2. is the inertia weight, between 0.1 and 0.9, is a random number between 0 and 1, represents the individual optimal position of the particle, represents the global optimal position of the particle; Step S2.2: In the population initialization phase, an adaptive meta-heuristic mapping is introduced, including global optimization and local optimization; global optimization allocates initial flow values ​​for each data flow on all possible links, and local optimization allocates initial flow values ​​based on the sharing degree of the links. Perform dynamic traffic allocation; Step S2.3: construct particles as follows: Each particle is composed of a global optimization part and a local optimization part: ； In the formula, represents particles, represents the globally optimized vector, represents the local optimization vector, represents a vector concatenation operation; the collection of all particles constitutes a population: ； In the formula, Represents the population, represents the particles in the population, Represents the total number of all particles.

4. The space-ground integrated load balancing method based on improved particle swarm and Hippo algorithm according to claim 3 is characterized in that: The specific method of the global optimization described in step S2.2 is as follows: For each data stream k and each link , is the total number of links, the globally optimized value Initialized to: ； in, For uniformly distributed sampling, is a normal distribution sample, Indicates that x The value of is restricted to the range [0,1]: ； Globally optimized vector By all constitute: ； in, Length is , is the total number of links, The total number of data flows.

5. The space-ground integrated load balancing method based on improved particle swarm and hippo algorithm according to claim 3 is characterized in that: The method of local optimization described in step S2.2 is as follows: Represents the local optimization value, for each data flow k and its path Each link in , according to the path length Choose a different sampling strategy: when When , uniform distribution sampling is used: ； Represents a uniform distribution The random number sampled from ; when When , normal distribution sampling is adopted: ； Represents a normal distribution The random number sampled from ; when When , a mixed method of uniform distribution and normal distribution is used: ； Assume Link If it is shared by multiple data flows, the adjusted traffic is: ； in, Indicates link the degree of sharing; The final locally optimized vector By all Its length is the sum of all path lengths: ； in, Length is , is the total number of links, The total number of data flows.

6. The space-ground integrated load balancing method based on improved particle swarm and Hippo algorithm according to claim 1 is characterized in that: The specific steps of step S3 are as follows: Step S3.1: Start iteration, introduce a selective reset mechanism, calculate the fitness value of each particle in the population, and reset the position of the particle when the convergence speed of the particle exceeds the set threshold; Step S3.2: During the iteration process, for particles whose positions are reset, the defense behavior and escape behavior of the Hippo optimization algorithm are simulated to calculate the new positions of the particles after the defense behavior and escape behavior. According to the fitness values ​​of the new positions of the particles after the defense behavior and escape behavior, a better particle position is selected as the final result; Step S3.3: Determine whether the maximum number of iterations is met. If so, the space-ground integrated network structure corresponding to the individuals in the iterated population is output as the optimal space-ground integrated network structure. Otherwise, execute step S3.

1.

7. The space-ground integrated load balancing method based on improved particle swarm and Hippo algorithm according to claim 6 is characterized in that: In step S3.1, for each particle in the population, if the particle satisfies the following formula, the position of the particle is reset to a random position: ； In the formula, is the individual optimal fitness of the particle, is the current fitness of the particle; If the current fitness of the particle is greater than the individual optimal fitness, the individual optimal fitness is updated; if the current fitness of the particle is greater than the global optimal fitness, the global optimal fitness is updated, and the particle's speed and position are updated as the optimal solution output of the target optimization model.

8. The space-ground integrated load balancing method based on improved particle swarm and Hippo algorithm according to claim 6, characterized in that: The specific steps of step S3.2 are as follows: Step S3.2.1: For the particle whose position is reset in step S3.1, if its random search probability is less than 0.5, the defense behavior and escape behavior of the Hippopotamus optimization algorithm are simulated for the particle; Perform steps S3.2.2-S3.2.4 to simulate the behavior of a hippopotamus defending itself against a predator: Step S3.2.2: Generate predator positions As follows: ； in, is the lower bound of the global search space, is the upper bound of the global search space, is a uniformly distributed random number in the range ; Step S3.2.3: Calculate the distance between the current solution and the predator As follows: ； in, is the current particle position; Step S3.2.4: Generate a defense strategy and calculate the new position of the particle after the defense behavior: If the predator's position has lower fitness than the current solution, the position of the current solution is adjusted according to the defense strategy: ； Otherwise, use a more conservative defense strategy: ； in, is the new position of the particle after the defensive action, ; ; ; is the sampling angle parameter in the uniform distribution, , is a uniformly distributed random number in the range , for A random number; Perform steps S3.2.5-S3.2.8 to simulate the behavior of a hippopotamus fleeing from a predator: Step S3.2.5: Define the local search range: According to the current number of iterations and the maximum number of iterations T , dynamically adjust the upper and lower bounds of local search: ； ； in, is the upper bound of the local search, is the lower bound of the local search; Step S3.2.6: Randomly select an escape strategy: Randomly select an escape strategy , the escape strategy is one of the following three: ； In the formula, is a uniformly distributed random number in the range of [-1,1); ； In the formula, is a uniformly distributed random number in the range [0,1); ； In the formula, represents normally distributed random numbers; Step S3.2.7: Calculate the new position of the particle after the escape behavior according to the selected escape strategy: ； in, is the new position of the particle after the escape action, for A random number in a range; Step S3.2.8: Compare the fitness of the new positions of the particles after the defense behavior and the escape behavior, and select the better particle position as the final result: ； In the formula, is the fitness of the new position of the particle after the escape behavior, is the fitness of the new position of the particle after the defensive behavior; Constrain the generated new positions to be within the bounds of the search space: ； in, Represents a crop operation.

Images