Communication network scheduling method and system based on genetic algorithm

Through the communication network scheduling method based on genetic algorithm, the scheduling strategy of the communication network is optimized, the scheduling deviation caused by information collection errors is solved, the real-timeness of the system and resource utilization are improved, and a more efficient scheduling solution is realized.

CN119011674AInactive Publication Date: 2024-11-22GUANGZHOU POWER SUPPLY BUREAU GUANGDONG POWER GRID CO LTD

Patent Information

Application Number
CN202410872082.2
Authority / Receiving Office
CN · China
Patent Type
Applications(China)
Current Assignee / Owner
Filing Date
2024-07-01
Publication Date
2024-11-22
Estimated Expiration
Not applicable · inactive patent

AI Technical Summary

Technical Problem

When the information collection is inaccurate or errors exist, the existing communication network scheduling system causes the scheduling strategy to deviate from the optimal solution, reduces the system effectiveness, and requires a large amount of computing resources for real-time computing and analysis of capacity information of large-scale networks, resulting in the scheduling system. Reduced effectiveness.

Method used

Using a communication network scheduling method based on genetic algorithm, by collecting user's task request information, building multiple groups of individual chromosomes, calculating the fitness size and the probability of being selected, performing selection, crossing and mutating operations, iterative optimization, and finding the task request that completes Paths with the lowest energy consumption and network latency.

Benefits of technology

It improves the real-time and resource utilization of the system, provides a more efficient and reasonable scheduling solution, solves the problem of scheduling strategy deviation caused by inaccurate information collection, and improves the system operation speed through the parallel evolution of multiple groups.

✦ Generated by Eureka AI based on patent content.

Smart Images

  • Figure CN119011674A_ABST
    Figure CN119011674A_ABST
Patent Text Reader

Abstract

The invention discloses a communication network scheduling method and system based on a genetic algorithm, and relates to the field of communication networks. Comprising the following steps that task request information of all users at the current moment is collected, and the task request information of all the users comprises source nodes and destination nodes of the users; according to the collected source nodes and destination nodes of the users, calculating to obtain a set of all feasible paths for each user to complete the task; generating according to all feasible path sets, coding each path selected for each user into a gene, and forming individual chromosomes according to the obtained genes; the fitness size and the selected probability of each individual chromosome in the initial population are calculated, selection operation, crossover operation and mutation operation are carried out, two new individual chromosomes are obtained, and the two new individual chromosomes are placed in an iteration population; and finally, iterative operation is carried out, and an optimal solution is found from the initial population generated by the last iterative operation, so that the task request of the user is completed.
Need to check novelty before this filing date? Find Prior Art

Description

Technical Field

[0001] The invention belongs to the field of communication networks and relates to network scheduling technology, in particular to a communication network scheduling method and system based on genetic algorithm. Background Art

[0002] Communication is the exchange and transmission of information between people through some kind of media. A network is a data link that connects isolated workstations or hosts together with physical links. A communication network is a link that physically connects isolated devices to achieve information exchange between people, people and computers, and computers and computers, thereby achieving the purpose of resource sharing and communication.

[0003] However, when the communication network was first designed, the network energy consumption was not considered. The redundant design method was adopted, and the network equipment was simply added to meet the communication needs during peak hours. However, under normal circumstances, the load of the network equipment rarely reaches the limit of its own capacity, but it still consumes most of the energy, resulting in extremely low network energy utilization. With the addition of new network equipment, the scale of the Internet is further expanded, and the energy utilization rate of the communication network will continue to decline. Moreover, due to the rapid enhancement of technologies such as mobile Internet and data center networks, more network equipment will inevitably be added to the network to meet the communication needs of users, and the phenomenon of energy waste will become more serious. Therefore, reducing the energy consumption of communication network equipment and improving the energy utilization rate of communication networks has become a topic that needs to be solved as soon as possible, and it is also a very important factor for the sustainable development of society and the country.

[0004] In the prior art, publication number CN116016711A discloses a communication network scheduling system, method and system, which acquires communication information in the communication network through a communication information acquisition module, transmits the acquired communication information to an information classification module, and analyzes the acquired communication information; the information classification module classifies the analyzed communication information; the information monitoring module monitors the communication information and acquires the communication data; the data calculation module receives the communication data to acquire the communication network reference data; the communication network is scheduled by receiving the communication network reference data through the information scheduling module. This invention can acquire capacity information during the information transmission process when the communication network shares information, thereby forming an effective network scheduling method; however, the accuracy of the capacity information in this method directly affects the scheduling decision. If the capacity information is not collected accurately or there are errors, the effectiveness of the system is reduced, which may cause the scheduling strategy to deviate from the optimal solution and even cause network congestion. At the same time, for large-scale networks, real-time calculation and analysis of capacity information may require a large amount of computing resources, which may reduce the effectiveness of the scheduling system.

[0005] The above information disclosed in this Background section is only for enhancement of understanding of the background of the present disclosure and therefore it may contain information that does not constitute the prior art that is already known to one of ordinary skill in the art. Summary of the invention

[0006] The purpose of the present invention is to provide a communication network scheduling method and system based on genetic algorithm to solve the problems raised in the above background technology.

[0007] To achieve the above object, the present invention provides the following technical solutions:

[0008] A communication network scheduling method based on genetic algorithm, the specific steps include:

[0009] Collecting task request information of all users at the current moment, wherein the task request information of the user includes the source node and the destination node of the user, and calculating a set of all feasible paths for each user to complete the task based on the collected source node and the destination node of the user;

[0010] Construct multiple groups of individual chromosomes. Each individual chromosome is composed of a gene encoded by randomly selecting a path from the set of feasible paths for each user. Each individual chromosome has n gene positions. The gene corresponding to each gene position is a path randomly selected from the set of feasible paths for the corresponding user. Create a population based on the obtained individual chromosomes. The number of populations is not less than 1. Select m individual chromosomes as the initial population.

[0011] Calculate the fitness and selection probability of each individual chromosome in the initial population, perform a selection operation based on the calculated fitness and selection probability, select two individuals from the current population, perform a crossover operation on the two selected individuals, exchange the two selected individual chromosomes from a certain gene position and the chromosome fragment after the gene position, and obtain two crossover individual chromosomes;

[0012] Perform mutation operation on the two individual chromosomes after the crossover, mutate a gene on the two individual chromosomes after the crossover based on the adaptive mutation rate, obtain two new individual chromosomes, generate an iterative population, and place the two new individual chromosomes in the iterative population;

[0013] The initial population is subjected to selection, crossover and mutation operations in a cycle, and the two newly obtained individual chromosomes are placed in the iteration population each time until the number of individual chromosomes in the iteration population reaches m-1. The individual chromosome with the highest fitness in the initial population is copied to the iteration population, the initial population is cleared, and the m individual chromosomes in the iteration population are transferred to the initial population. At this time, the initial population contains a new generation of m individual chromosomes, and the individual chromosomes in the iteration population are cleared.

[0014] According to the initial population containing m individual chromosomes of the new generation, an iterative operation is performed, and the maximum number of iterations is set. After the maximum number of iterations is reached, the iterative operation is stopped, and the individual with the highest fitness is selected from the initial population generated by the last iterative operation as the optimal solution for communication network scheduling. The optimal solution is a path with the least total energy consumption and network delay to complete the task request of the corresponding user.

[0015] Furthermore, based on the collected source nodes and destination nodes of the users, the logic for calculating all feasible path sets for each user to complete the task is: use the depth-first search algorithm to find a feasible path set that meets the user's communication needs from the source node to the destination node for each user in the communication network, and mark T as the set of all feasible paths for all users, where T = {P1, P2, ..., P n}, where P u = {pu1, pu1, ..., pu k}, P u represents the set of all feasible paths of user u, n represents the number of users with task requests, and k represents the number of feasible paths owned by user u.

[0016] Furthermore, each path selected for each user is encoded as a gene, and an individual chromosome is formed according to the obtained gene. The individual chromosome is composed of n genes, where each feasible path for each user is a gene, i.e., pu1, pu2, ..., pu k are all optional values ​​of individual chromosome gene positions, and pu1, pu1, ..., pu k These k genes are alleles of each other, and the individual chromosome S i It consists of n genes, each of which is randomly selected from the feasible path set P of the corresponding user. i A path selected from , i = 1, 2, …, n.

[0017] Furthermore, the logic for creating a population based on the obtained individual chromosomes is: randomly select the feasible path set P of the corresponding user i A path selected in is used as a gene in one of the gene positions of the individual chromosome, and n genes are selected to fill in the gene positions of the corresponding individual chromosome to form a complete individual chromosome S i , a total of m individual chromosomes are formed, and the m individual chromosomes are selected to form an initial population Γ, where Γ={S1,S2,…,S m}.

[0018] Furthermore, the logic for calculating the fitness of each individual chromosome in the initial population and the probability of being selected is: the fitness of each individual chromosome in the initial population is calculated by the fitness function, and the formula based on the fitness function is:

[0019]

[0020] In the formula, f(S i ) represents the fitness of the i-th individual chromosome, delay(S i ) is the total network delay of the i-th individual chromosome, is the total energy consumption of the individual chromosome, a, b are the weight coefficients of the total network delay and the total energy consumption, a, b are both greater than 0 and a <b,W l is the bandwidth capacity of node l, w l is the load of node l;

[0021] The probability of being selected is calculated based on the formula:

[0022]

[0023] In the formula, f(S i ) represents the fitness of the i-th individual chromosome, pro(S i ) is the probability of the ith individual chromosome being selected. The roulette method is used to select individuals. The system randomly generates a random number in the interval [0, 1]. According to the generated probability of being selected, the random number is distributed in the selection interval of the individual chromosome to determine which individual chromosome to select in this round.

[0024] Furthermore, a crossover operation is performed on the two selected individuals. The logic of the crossover operation is: a single-point crossover method is used to process the two individual chromosomes selected in the selection operation process, which are represented as and in represents the value of the jth gene of chromosome i of individual i, randomly generates a random integer in the interval [1, n], which is the crossover point on the chromosome, and selects individual S i and individual S j The yth gene position of the chromosome is used as the crossover point, where 1≤y≤n, that is, and Then S i and S j The chromosomes from and The two intersection points are cut off and S i and S jThe chromosome segments composed of genes at and after the crossover point are exchanged to obtain two individual chromosomes S' after the crossover i and S′ j , respectively expressed as and

[0025] Furthermore, the logic for performing mutation operations on the chromosomes of the two individuals after the crossover is as follows: the mutation operation is performed on the two new individuals S′ generated by the crossover operation. i and S′ j To perform the operation, the system first randomly generates an integer in the interval [1, n] to represent the chromosome S′ i A gene position on m The selected gene is mutated into one of its alleles, and the mutated gene is used to replace the gene at the gene position, thereby obtaining a new individual S″ i , expressed as is the variant allele, where the mutation probability R m The formula is:

[0026]

[0027] In the formula, is the maximum mutation probability, f avg 、f best and f worst Respectively represent the average, maximum and minimum fitness of the current population.

[0028] Furthermore, the iterative operation is a step of cyclically performing selection, crossover and mutation operations on the initial population to generate an iterative population, and using the iterative population as a new initial population. When the set population size is greater than 1, a migration period G is set. According to the migration period G, a ring topology is used to migrate individuals in the population, that is, each population only exchanges individuals with adjacent populations. During each migration process, individuals in the top d% of fitness in each population are selected as migration candidate individuals, and the migration candidate individuals are replaced accordingly in adjacent populations to form a new complete population.

[0029] The present invention also provides a communication network scheduling system based on a genetic algorithm, wherein the communication network scheduling system based on a genetic algorithm is used to execute the above-mentioned communication network scheduling method based on a genetic algorithm, comprising:

[0030] A feasible path generation module is used to collect task request information of all users at the current moment, wherein the task request information of the user includes the source node and the destination node of the user, and calculate all feasible path sets for each user to complete the task based on the collected source node and the destination node of the user;

[0031] The population establishment module is used to construct multiple groups of individual chromosomes. Each individual chromosome is composed of a gene encoded by a path selected from the feasible path set of each user. Each individual chromosome has n gene positions. The gene corresponding to each gene position is a path randomly selected from the feasible path set of the corresponding user. A population is created based on the obtained individual chromosomes. The number of populations is not less than 1, and m individual chromosomes are selected as the initial population.

[0032] A selection crossover operation module is used to calculate the fitness size and selection probability of each individual chromosome in the initial population, and perform a selection operation based on the calculated fitness size and selection probability, select two individuals from the current population, perform a crossover operation on the two selected individuals, and exchange the two selected individual chromosomes from a certain gene position and the chromosome fragment after the gene position to obtain two crossed individual chromosomes;

[0033] A mutation operation module is used to perform a mutation operation on the two individual chromosomes after the crossover, mutate a gene on the two individual chromosomes after the crossover based on the mutation rate, obtain two new individual chromosomes, generate an iterative population, and place the two new individual chromosomes in the iterative population;

[0034] The population iteration module is used to perform selection, crossover and mutation operations on the initial population in a cycle, and place the two newly obtained individual chromosomes each time in the iteration population until the number of individual chromosomes in the iteration population reaches m-1, copy the individual chromosome with the highest fitness in the initial population to the iteration population, clear the initial population, and transfer the m individual chromosomes in the iteration population to the initial population. At this time, the initial population contains a new generation of m individual chromosomes, and the individual chromosomes in the iteration population are cleared;

[0035] The optimal solution acquisition module is used to perform an iterative operation based on an initial population containing m individual chromosomes of a new generation, set a maximum number of iterations, stop the iterative operation after reaching the maximum number of iterations, and select the individual with the highest fitness from the initial population generated by the last iterative operation as the optimal solution for communication network scheduling. The optimal solution is a path that completes the task request of the corresponding user with the least total energy consumption and network delay.

[0036] Compared with the prior art, the beneficial effects of the present invention are as follows: by collecting the task request information of all users at the current moment, the task request information of the users includes the source node and the destination node of the users, calculating all feasible path sets for each user to complete the task, constructing multiple groups of individual chromosomes, each individual chromosome is a gene composition encoded by selecting a path from the feasible path set of each user, selecting a number of individual chromosomes as the initial population, the number of the initial population can be multiple, calculating the fitness of the individual chromosomes in the initial population, performing selection, crossover and mutation operations on the population, updating the individual chromosome composition in the population, performing iterative operations on the population, ending the population iterative operation by setting the number of iterations, selecting the number of iterations The individual with the highest fitness in the population is taken as the optimal solution, and the optimal solution is the path with the least total energy consumption and network delay to complete the task requested by the corresponding user. The genetic algorithm is used to solve the optimal solution allocation problem of the communication network, and the path with the least total energy consumption and network delay to complete the user's task request is found. The problem of reduced system effectiveness due to inaccurate information collection or errors, which may cause the scheduling strategy to deviate from the optimal solution, is solved. At the same time, in the face of real-time calculations of a large number of computing resources, the use of multi-population parallel evolution improves the system operation speed, effectively improves the real-time performance of the system, and makes more full and comprehensive use of resources. It can provide a more efficient and reasonable scheduling plan for scheduling work to improve the task completion rate and resource utilization. BRIEF DESCRIPTION OF THE DRAWINGS

[0037] Figure 1 It is a schematic diagram of the overall method flow of the present invention;

[0038] Figure 2 It is a schematic diagram of the overall system structure of the present invention. DETAILED DESCRIPTION

[0039] In order to make the objectives, technical solutions and advantages of the present invention more clearly understood, the present invention is further described in detail below in conjunction with specific embodiments.

[0040] It should be noted that, unless otherwise defined, the technical terms or scientific terms used in the present invention should be understood by people with ordinary skills in the field to which the present invention belongs. The words "first", "second" and similar words used in the present invention do not indicate any order, quantity or importance, but are only used to distinguish different components. "Include" or "comprise" and similar words mean that the elements or objects appearing before the word include the elements or objects listed after the word and their equivalents, without excluding other elements or objects. "Connect" or "connected" and similar words are not limited to physical or mechanical connections, but may include electrical connections, whether direct or indirect. "Up", "down", "left", "right" and the like are only used to indicate relative positional relationships. When the absolute position of the described object changes, the relative positional relationship may also change accordingly.

[0041] Example:

[0042] See also Figure 1 , the present invention provides a technical solution:

[0043] A communication network scheduling method based on genetic algorithm, the specific steps include:

[0044] Step 1: Collect the task request information of all users at the current moment, where the task request information of the user includes the source node and destination node of the user. Based on the collected source node and destination node of the user, calculate the set of all feasible paths for each user to complete the task;

[0045] The source node represents the node where the data flow generated by the user's needs enters the network, while the node where the data flow flows out of the network is represented as the destination node. In order to meet the user's communication needs, it is necessary to establish a network path connecting the source node to the destination node in the network. The path contains a series of links and nodes;

[0046] According to the collected source nodes and destination nodes of the users, the logic of calculating all feasible path sets for each user to complete the task is: use the depth-first search algorithm to find a feasible path set that meets the user's communication needs from the source node to the destination node for each user in the communication network, and mark T as the set of all feasible paths for all users, where T = {P1, P2, ..., P n}, where P u ={p u1 , p u1 ,…,p uk}, P u represents the set of all feasible paths for user u, n represents the number of users with task requests, and k represents the number of feasible paths owned by user u. For example, if one of the paths is p 11={2, 4, 6, 8}, which means that nodes 2, 4, 6, 8 and links (2, 4), (4, 6), (6, 8) are activated to serve the network communication needs of user 1; the depth-first search of the network topology graph needs to first determine the source node, and after determining the source node, find all adjacent nodes that meet a certain condition with the source node, and then perform a depth-first search with each node as the source node until the destination node is found.

[0047] Step 2: Construct multiple groups of individual chromosomes. Each individual chromosome is composed of a gene encoded by randomly selecting a path from the feasible path set of each user. Each individual chromosome has n gene positions. The gene corresponding to each gene position is a path randomly selected from the feasible path set of the corresponding user. A population is created based on the obtained individual chromosomes. The population size is not less than 1. Select m individual chromosomes as the initial population.

[0048] Each path selected for each user is encoded as a gene, and an individual chromosome is formed according to the obtained gene. The individual chromosome is composed of n genes, where each feasible path for each user is a gene, i.e., pu1, pu2, ..., pu k are all optional values ​​of individual chromosome gene positions, and pu1, pu1, ..., Pu k These k genes are alleles of each other, and the individual chromosome S i It consists of n genes, each of which is randomly selected from the feasible path set P of the corresponding user. i A path selected from , i = 1, 2, ..., n;

[0049] Assume that an individual's chromosome S i ={1, 2, ..., 3}, which means that the first path in the feasible path set P1 is selected for user 1, the second path in the feasible path set P2 is selected for user 2, and the feasible path set P is selected for user n. n The third path in , and the n genes constituting the chromosome of individual Si are not alleles of each other;

[0050] The logic for creating a population based on the obtained individual chromosomes is: randomly select the feasible path set P of the corresponding user i A path selected in is used as a gene in one of the gene positions of the individual chromosome, and n genes are selected to fill in the gene positions of the corresponding individual chromosome to form a complete individual chromosome S i , a total of m individual chromosomes are formed, and the m individual chromosomes are selected to form an initial population Γ, where Γ={S1,S2,…,S m}, if multiple populations are set, set them as the initial population, and the number of individual chromosomes in each population should remain consistent at m.

[0051] Step 3: Calculate the fitness and selection probability of each individual chromosome in the initial population. According to the calculated fitness and selection probability, perform a selection operation, select two individuals from the current population, perform a crossover operation on the two selected individuals, and exchange the chromosomes of the two selected individuals from a certain gene position and the chromosome fragment after the gene position to obtain two crossover individual chromosomes;

[0052] The logic for calculating the fitness of each individual chromosome in the initial population and the probability of being selected is: the fitness of each individual chromosome in the initial population is calculated by the fitness function, and the formula based on the fitness function is:

[0053]

[0054] In the formula, f(S i ) represents the fitness of the i-th individual chromosome, delay(S i ) is the total network delay of the i-th individual chromosome, is the total energy consumption of the individual chromosome, a, b are the weight coefficients of the total network delay and the total energy consumption, a, b are both greater than 0 and a <b,W l is the bandwidth capacity of node l, wl is the load of node l; the solution to be selected needs to have a small total energy consumption and a low total network delay, so the fitness function is inversely proportional to the total network delay and the total energy consumption of the individual chromosome. The larger the total network delay of the individual chromosome and the total energy consumption of the individual chromosome, the lower its fitness. At the same time, w l is the load of node l, W l is the bandwidth capacity of node l, when W l <w l When , it indicates that the load of node l exceeds the bandwidth capacity, and the fitness of its individual chromosome is 0, that is, it is an invalid individual;

[0055] The load w of node l is l and the bandwidth capacity W of node l l The monitoring method is: Use network monitoring tools (such as SNMP, NetFlow, sFlow, etc.) to collect performance data of network devices (such as routers, switches) in real time or periodically. These tools can provide information about node load and bandwidth usage.

[0056] Total energy consumption of individual chromosomes The calculation is based on the formula:

[0057]

[0058] In the formula, Clink The sum of energy consumption of links activated for the selected path, C node The sum of energy consumption of nodes activated for the selected path;

[0059] The probability of being selected is calculated based on the formula:

[0060]

[0061] In the formula, f(S i ) represents the fitness of the i-th individual chromosome, pro(S i ) is the probability of the i-th individual chromosome being selected. The roulette method is used to select individuals. The system randomly generates a random number in the interval [0, 1]. According to the generated probability of selection, the random number is distributed in the selection interval of the individual chromosome to determine which individual chromosome to select in this round. The process of selecting individuals is based on the fitness of the individual. The greater the fitness of the individual, the greater the possibility of being selected, and vice versa. The possibility of being selected is smaller, and an individual may be selected multiple times;

[0062] The selection operation imitates the evolutionary theory of biologist Darwin. Individuals with better adaptability to nature will have a better chance of obtaining mating rights and passing on their good genes to their offspring. In each generation of genetic evolution, a part of the individuals need to be selected from the current population to evolve into the next generation of population.

[0063] The mainstream crossover operations now include: single-point crossover, multi-point crossover, etc. Among them, single-point crossover is to randomly select the same gene position on the chromosomes of the parent individuals as the crossover point, and exchange the gene fragments on the right side of the chromosome crossover point of the parents, so that two new individuals are obtained, each of which contains part of the genetic information of the parents.

[0064] A crossover operation is performed on the two selected individuals. The logic of the crossover operation is: a single-point crossover method is used to process the two individual chromosomes selected in the selection operation process, which are represented as and in represents the value of the jth gene of chromosome i of individual i, randomly generates a random integer in the interval [1, n], which is the crossover point on the chromosome, and selects individual S i and individual S j The yth gene position of the chromosome is used as the crossover point, where 1≤y≤n, that is, and Then S i and S j The chromosomes from and The two intersection points are cut off and S i and S j The chromosome segments composed of genes at and after the crossover point are exchanged to obtain two individual chromosomes S' after the crossover i and S′ j , respectively expressed as and The introduction of crossover operation can greatly increase the search range of the algorithm and promote rapid convergence of the algorithm. It is the core operation of the algorithm.

[0065] Step 4: Perform mutation operation on the two individual chromosomes after the crossover, mutate a gene on the two individual chromosomes after the crossover based on the mutation rate, obtain two new individual chromosomes, generate an iterative population, and place the two new individual chromosomes in the iterative population;

[0066] The logic behind the mutation operation on the chromosomes of two individuals after the crossover is: the mutation operation is performed on the two new individuals S′ generated by the crossover operation i and S′ j To perform the operation, the system first randomly generates an integer in the interval [1, n] to represent the chromosome S′ i A gene position on m The selected gene is mutated into one of its alleles, and the mutated gene is used to replace the gene at the gene position, thereby obtaining a new individual S″ i , expressed as is the variant allele, where the mutation probability R m The formula is:

[0067]

[0068] In the formula, is the maximum mutation probability, f avg 、f best and f worst Represent the average, highest and lowest fitness of the current population respectively, where the maximum mutation probability is set to 0.8, and the mutation probability R is adaptively and dynamically adjusted according to the average, highest and lowest fitness of the current population m , when the difference between the average fitness of the current population and the highest fitness of the current population becomes larger, the mutation probability R m As it gets bigger, the genes mutate to a greater extent, changing the fitness of individuals; the mutation operation can ensure and enrich the diversity of the population, that is, it can increase the search range of the algorithm and prevent it from falling into the local optimal solution, which will cause the evolution of the population to slow down or even stop.

[0069] Step 5: Perform selection, crossover and mutation operations on the initial population in a loop, and place the two newly generated individual chromosomes each time in the iteration population until the number of individual chromosomes in the iteration population reaches m-1. Copy the individual chromosome with the highest fitness in the initial population to the iteration population, clear the initial population, and transfer the m individual chromosomes in the iteration population to the initial population. At this time, the initial population contains a new generation of m individual chromosomes, and the individual chromosomes in the iteration population are cleared.

[0070] After the three major operations, two new individuals are obtained. Repeat these three operations to continuously generate new individuals until the number of individual chromosomes in the iterative population reaches m-1. Copy the individual chromosome with the highest fitness in the initial population to the iterative population. At this time, the iterative population contains m individual chromosomes of a new generation. The iterative population at this time is used as the initial population of the next generation. At the same time, the individuals in the initial population are cleared and enter the inheritance of the next generation population as a new iterative population. The above process is repeated for each generation of inheritance.

[0071] Step 6: Perform an iterative operation based on the initial population containing m individual chromosomes of the new generation, set the maximum number of iterations, and stop the iterative operation after reaching the maximum number of iterations. Select the individual with the highest fitness from the initial population generated by the last iterative operation as the optimal solution for communication network scheduling. The optimal solution is the path that completes the task request of the corresponding user with the least total energy consumption and network delay.

[0072] The iterative operation is to perform selection, crossover and mutation operations on the initial population in a cycle to generate an iterative population, and use the iterative population as a new initial population. When the set population size is greater than 1, a migration period G is set. According to the migration period G, a ring topology is used to migrate individuals in the population, that is, each population only exchanges individuals with adjacent populations. In each migration process, individuals with the top d% of fitness in each population are selected as migration candidate individuals, and the migration candidate individuals are replaced with the adjacent populations to form a new complete population.

[0073] The maximum number of iterations is set to 50 generations. After reaching the maximum number of iterations, the iteration operation is stopped. When the set population size is greater than 1, multiple populations evolve in parallel, and the migration cycle G is set at the same time. According to the migration cycle G, the individuals in the population are migrated using a ring topology, that is, each population only exchanges individuals with adjacent populations, such as the migration candidate individuals of population 1 migrate to population 2, the migration candidate individuals of population 2 migrate to population 3, and so on. The migration candidate individuals of the last population migrate to population 1. During each migration process, individuals with the top d% of fitness in each population are selected as migration candidate individuals, and the migration candidate individuals are replaced with the corresponding adjacent populations to form a new complete population. After the migration is completed, all populations continue to independently perform the iterative operation of the genetic algorithm to ensure that the migration operation does not affect the independence of the population and the normal operation of the genetic algorithm. The use of multiple populations in parallel evolution prevents local optimality, enhances the global search capability, and improves population diversity and optimization effects through individual migration information exchange.

[0074] See also Figure 2 The present invention also provides a communication network scheduling system based on a genetic algorithm, wherein the communication network scheduling system based on a genetic algorithm is used to execute the above-mentioned communication network scheduling method based on a genetic algorithm, comprising:

[0075] A feasible path generation module is used to collect task request information of all users at the current moment, wherein the task request information of the user includes the source node and the destination node of the user, and calculate all feasible path sets for each user to complete the task based on the collected source node and the destination node of the user;

[0076] The population establishment module is used to construct multiple groups of individual chromosomes. Each individual chromosome is composed of a gene encoded by a path selected from the feasible path set of each user. Each individual chromosome has n gene positions. The gene corresponding to each gene position is a path randomly selected from the feasible path set of the corresponding user. A population is created based on the obtained individual chromosomes. The number of populations is not less than 1, and m individual chromosomes are selected as the initial population.

[0077] A selection crossover operation module is used to calculate the fitness size and selection probability of each individual chromosome in the initial population, and perform a selection operation based on the calculated fitness size and selection probability, select two individuals from the current population, perform a crossover operation on the two selected individuals, and exchange the two selected individual chromosomes from a certain gene position and the chromosome fragment after the gene position to obtain two crossed individual chromosomes;

[0078] A mutation operation module is used to perform a mutation operation on the two individual chromosomes after the crossover, mutate a gene on the two individual chromosomes after the crossover based on the mutation rate, obtain two new individual chromosomes, generate an iterative population, and place the two new individual chromosomes in the iterative population;

[0079] The population iteration module is used to perform selection, crossover and mutation operations on the initial population in a cycle, and place the two newly obtained individual chromosomes each time in the iteration population until the number of individual chromosomes in the iteration population reaches m-1, copy the individual chromosome with the highest fitness in the initial population to the iteration population, clear the initial population, and transfer the m individual chromosomes in the iteration population to the initial population. At this time, the initial population contains a new generation of m individual chromosomes, and the individual chromosomes in the iteration population are cleared;

[0080] The optimal solution acquisition module is used to perform an iterative operation based on an initial population containing m individual chromosomes of a new generation, set a maximum number of iterations, stop the iterative operation after reaching the maximum number of iterations, and select the individual with the highest fitness from the initial population generated by the last iterative operation as the optimal solution for communication network scheduling. The optimal solution is a path that completes the task request of the corresponding user with the least total energy consumption and network delay.

[0081] The above formulas are all dimensionless and numerical calculations. The formula is a formula for the most recent real situation obtained by collecting a large amount of data and performing software simulation. The preset parameters in the formula are set by technicians in this field according to actual conditions.

[0082] The above embodiments may be implemented in whole or in part by software, hardware, firmware or any other combination thereof. When implemented by software, the above embodiments may be implemented in whole or in part in the form of a computer program product. Those skilled in the art may appreciate that the units and algorithm steps of each example described in conjunction with the embodiments disclosed herein may be implemented by electronic hardware, or a combination of computer software and electronic hardware. Whether these functions are performed by hardware or software methods depends on the specific application and design constraints of the technical solution.

[0083] The units described as separate components may or may not be physically separated, and the components shown as units may or may not be physical units, and may be located in one place or distributed on multiple network units. Some or all of the units may be selected according to actual needs to achieve the purpose of the solution of this embodiment.

[0084] The above description is only a specific implementation manner of the present application, but the protection scope of the present application is not limited thereto. Any technician familiar with the technical field can easily think of changes or substitutions within the technical scope disclosed in the present application, which should be included in the protection scope of the present application.

Claims

1. A communication network scheduling method based on genetic algorithm, characterized in that: The specific steps include: Collect the task request information of all users at the current moment. The task request information of the user includes the source node and destination node of the user. According to the collected source node and destination node of the user, calculate the set of all feasible paths for each user to complete the task: Construct multiple groups of individual chromosomes. Each individual chromosome is composed of a gene encoded by randomly selecting a path from the set of feasible paths for each user. Each individual chromosome has n gene positions. The gene corresponding to each gene position is a path randomly selected from the set of feasible paths for the corresponding user. Create a population based on the obtained individual chromosomes. The number of populations is not less than 1. Select m individual chromosomes as the initial population. Calculate the fitness and selection probability of each individual chromosome in the initial population, perform a selection operation based on the calculated fitness and selection probability, select two individuals from the current population, perform a crossover operation on the two selected individuals, exchange the two selected individual chromosomes from a certain gene position and the chromosome fragment after the gene position, and obtain two crossover individual chromosomes; Perform mutation operation on the two individual chromosomes after the crossover, mutate a gene on the two individual chromosomes after the crossover based on the adaptive mutation rate, obtain two new individual chromosomes, generate an iterative population, and place the two new individual chromosomes in the iterative population; The initial population is subjected to selection, crossover and mutation operations in a cycle, and the two newly obtained individual chromosomes are placed in the iteration population each time until the number of individual chromosomes in the iteration population reaches m-1. The individual chromosome with the highest fitness in the initial population is copied to the iteration population, the initial population is cleared, and the m individual chromosomes in the iteration population are transferred to the initial population. At this time, the initial population contains a new generation of m individual chromosomes, and the individual chromosomes in the iteration population are cleared. According to the initial population containing m individual chromosomes of the new generation, an iterative operation is performed, and the maximum number of iterations is set. After the maximum number of iterations is reached, the iterative operation is stopped, and the individual with the highest fitness is selected from the initial population generated by the last iterative operation as the optimal solution for communication network scheduling. The optimal solution is a path with the least total energy consumption and network delay to complete the task request of the corresponding user.

2. The communication network scheduling method based on genetic algorithm according to claim 1, characterized in that: According to the collected source nodes and destination nodes of the users, the logic of calculating all feasible path sets for each user to complete the task is: use the depth-first search algorithm to find a feasible path set that meets the user's communication needs from the source node to the destination node for each user in the communication network, and mark T as the set of all feasible paths for all users, where T = {P1, P2, ..., P n }, where P u ={p u1 , p u1 ,…,p uk }, P u represents the set of all feasible paths of user u, n represents the number of users with task requests, and k represents the number of feasible paths owned by user u.

3. The communication network scheduling method based on genetic algorithm according to claim 2, characterized in that: Each path selected for each user is encoded as a gene, and an individual chromosome is formed according to the obtained genes. The individual chromosome is composed of n genes, where each feasible path for each user is a gene, that is, p u1 , p u1 ,…,p uk are all optional values ​​of individual chromosome gene positions, and p u1 , p u1 ,…,p uk These k genes are alleles of each other, and the individual chromosome S i It consists of n genes, each of which is randomly selected from the feasible path set P of the corresponding user. i A path selected from , i = 1, 2, …, n.

4. The communication network scheduling method based on genetic algorithm according to claim 3, characterized in that: The logic for creating a population based on the obtained individual chromosomes is: randomly select the feasible path set P of the corresponding user i A path selected in is used as a gene in one of the gene positions of the individual chromosome, and n genes are selected to fill in the gene positions of the corresponding individual chromosome to form a complete individual chromosome S i , a total of m individual chromosomes are formed, and the m individual chromosomes are selected to form an initial population Γ, where Γ={S1,S2,…,S m }.

5. The communication network scheduling method based on genetic algorithm according to claim 4, characterized in that: The logic for calculating the fitness of each individual chromosome in the initial population and the probability of being selected is: the fitness of each individual chromosome in the initial population is calculated by the fitness function, and the formula based on the fitness function is: In the formula, f(S i ) represents the fitness of the i-th individual chromosome, delay(S i ) is the total network delay of the i-th individual chromosome, is the total energy consumption of the individual chromosome, a, b are the weight coefficients of the total network delay and the total energy consumption, a, b are both greater than 0 and a <b,W l is the bandwidth capacity of node l, w l is the load of node l; The probability of being selected is calculated based on the formula: In the formula, f(S i ) represents the fitness of the i-th individual chromosome, pro(S i ) is the probability of the i-th individual chromosome being selected. The roulette method is used to select individuals. The system randomly generates a random number in the interval [0, 1]. According to the generated probability of being selected, the random number is distributed in the selection interval of the corresponding individual chromosome to determine the individual chromosome to be selected in this round.

6. The communication network scheduling method based on genetic algorithm according to claim 5, characterized in that: A crossover operation is performed on the two selected individuals. The logic of the crossover operation is: a single-point crossover method is used to process the two individual chromosomes selected in the selection operation process, which are represented as and in represents the value of the jth gene of chromosome i of individual i, randomly generates a random integer in the interval [1, n], which is the crossover point on the chromosome, and selects individual S i and individual S j The yth gene position of the chromosome is used as the crossover point, where 1≤y≤n, that is, and Then S i and S j The chromosomes from and The two intersection points are cut off and S i and S j The chromosome segments composed of genes at and after the crossover point are exchanged to obtain two individual chromosomes S' after the crossover i and S′ j , respectively expressed as and 7. The communication network scheduling method based on genetic algorithm according to claim 6, characterized in that: The logic behind the mutation operation on the chromosomes of two individuals after the crossover is: the mutation operation is performed on the two new individuals S′ generated by the crossover operation i and S′ j To perform the operation, the system first randomly generates an integer in the interval [1, n] to represent the chromosome S′ i A gene position on m The selected gene is mutated into one of its alleles, and the mutated gene is used to replace the gene at the gene position, thereby obtaining a new individual S″ i , expressed as is the variant allele, where the mutation probability R m The formula is: In the formula, is the maximum mutation probability, f avg 、f best and f worst Respectively represent the average, maximum and minimum fitness of the current population.

8. The communication network scheduling method based on genetic algorithm according to claim 1, characterized in that: The iterative operation is to perform selection, crossover and mutation operations on the initial population in a cycle to generate an iterative population, and use the iterative population as a new initial population. When the set population size is greater than 1, a migration period G is set. According to the migration period G, a ring topology is used to migrate individuals in the population, that is, each population only exchanges individuals with adjacent populations. In each migration process, individuals with the top d% of fitness in each population are selected as migration candidate individuals, and the migration candidate individuals are replaced with the adjacent populations to form a new complete population.

9. A communication network scheduling system based on genetic algorithm, characterized in that: The communication network scheduling system based on genetic algorithm is used to execute the communication network scheduling method based on genetic algorithm according to any one of claims 1 to 8, comprising: A feasible path generation module is used to collect task request information of all users at the current moment, wherein the task request information of the user includes the source node and the destination node of the user, and calculate all feasible path sets for each user to complete the task based on the collected source node and the destination node of the user; The population establishment module is used to construct multiple groups of individual chromosomes. Each individual chromosome is composed of a gene encoded by a path selected from the feasible path set of each user. Each individual chromosome has n gene positions. The gene corresponding to each gene position is a path randomly selected from the feasible path set of the corresponding user. A population is created based on the obtained individual chromosomes. The number of populations is not less than 1, and m individual chromosomes are selected as the initial population. A selection crossover operation module is used to calculate the fitness size and selection probability of each individual chromosome in the initial population, and perform a selection operation based on the calculated fitness size and selection probability, select two individuals from the current population, perform a crossover operation on the two selected individuals, and exchange the two selected individual chromosomes from a certain gene position and the chromosome fragment after the gene position to obtain two crossed individual chromosomes; A mutation operation module is used to perform a mutation operation on the two individual chromosomes after the crossover, mutate a gene on the two individual chromosomes after the crossover based on the mutation rate, obtain two new individual chromosomes, generate an iterative population, and place the two new individual chromosomes in the iterative population; The population iteration module is used to perform selection, crossover and mutation operations on the initial population in a cycle, and place the two newly obtained individual chromosomes each time in the iteration population until the number of individual chromosomes in the iteration population reaches m-1, copy the individual chromosome with the highest fitness in the initial population to the iteration population, clear the initial population, and transfer the m individual chromosomes in the iteration population to the initial population. At this time, the initial population contains a new generation of m individual chromosomes, and the individual chromosomes in the iteration population are cleared; The optimal solution acquisition module is used to perform an iterative operation based on an initial population containing m individual chromosomes of a new generation, set a maximum number of iterations, stop the iterative operation after reaching the maximum number of iterations, and select the individual with the highest fitness from the initial population generated by the last iterative operation as the optimal solution for communication network scheduling. The optimal solution is a path that completes the task request of the corresponding user with the least total energy consumption and network delay.

Citation Information

Patent Citations

  • Communication network scheduling system, method and system

    CN116016711A

Cited By

  • Communication network scheduling method based on improved particle swarm optimization algorithm

    CN120880966A