Multi-objective optimization 5G base station layout optimization method

Through the multi-objective optimization genetic algorithm method, the layout of 5G base stations is optimized, and the problem of coverage and cost balance of 5G base stations in complex environments is solved, signal coverage and number of base stations are optimized, and construction costs are reduced.

CN119012211APending Publication Date: 2024-11-22陕西智引科技有限公司 +1
View PDF 0 Cites 4 Cited by

Patent Information

Application Number
CN202410944738.7
Authority / Receiving Office
CN · China
Patent Type
Applications(China)
Current Assignee / Owner
Filing Date
2024-07-15
Publication Date
2024-11-22

AI Technical Summary

Technical Problem

It is difficult for 5G communication base stations to achieve a balance between effective coverage and low construction costs in changing and complex environments, and traditional planning methods are difficult to cope with the increase in cost and signal propagation blocking, interference, penetration and other problems caused by the increase in base station density.

Method used

The multi-objective optimization genetic algorithm method is adopted to optimize the base station layout to reduce the number of base stations and improve coverage by setting the coordinate set of base station locations, the coverage radius calculation and the fitness function, and combining the Monte Carlo principle and Pareto domination relationship.

Benefits of technology

While ensuring signal coverage, the number of base stations is reduced, the cost of infrastructure is reduced, and efficient and accurate base station layout design solutions are provided, suitable for 5G communication networks in complex environments.

✦ Generated by Eureka AI based on patent content.

Smart Images

  • Figure CN119012211A_ABST
    Figure CN119012211A_ABST
Patent Text Reader

Abstract

The invention provides a 5G base station layout optimization method based on multi-objective optimization. The method comprises the following steps: acquiring spatial data, establishing a coordinate system, and setting an alternative base station point coordinate set; converting a base station layout optimization problem into a problem suitable for a genetic algorithm; calculating a coverage rate and setting a fitness function; a multi-objective optimization genetic algorithm is adopted for solving, and the method comprises the steps of determining population related parameters and initializing populations, conducting crossover and mutation operation on the operation populations, calculating fitness values and coverage rates after the populations are combined, conducting non-dominated sorting, selecting new populations according to non-dominated levels and crowding degree values, and selecting the new populations according to the non-dominated levels and the crowding degree values. And judging whether iteration is terminated or not and outputting a site selection scheme set, and finally decoding an optimal solution set and selecting a final scheme. According to the invention, the number of the base stations can be reduced while the signal coverage rate is ensured, the construction cost is reduced, and the problem of layout and site selection of the 5G communication base stations is efficiently solved.
Need to check novelty before this filing date? Find Prior Art

Description

Technical Field

[0001] The present invention belongs to the field of mobile communication technology, and specifically relates to a multi-objective optimized 5G base station layout optimization method. Background Art

[0002] 5G communication technology meets the signal propagation and speed requirements of various emerging application scenarios. However, the characteristics of millimeter waves make them more susceptible to loss when encountering blockage, interference, penetration and attenuation, which makes the effective coverage of 5G communication base stations extremely limited. Due to the ultra-dense heterogeneous network characteristics of 5G communication base stations, the obstruction of millimeter wave signal propagation by buildings and the constraints of other related technology development, the layout of its base stations is a critical step. Especially in the case of changing and complex environments, the distance between base stations will be greatly reduced. Massive and dense scale will become an important feature of 5G base station deployment. The increase in base station density will inevitably lead to a corresponding increase in costs. In the spatial layout optimization design of 5G communication base stations, we face a major challenge: how to ensure that base stations provide high-speed, low-latency and high-capacity services while maintaining effective coverage in changing and complex environments and reducing construction costs. Due to the large number of 5G base stations and the diversity of coverage, traditional manual planning methods are no longer competent. Therefore, the current research focus has shifted to the use of intelligent optimization decision-making technology and spatial information technology, combined with advanced methods such as genetic algorithms, to fine-tune the spatial location layout of base stations.

[0003] Specifically, the goal of this optimization design is to pursue the optimal spatial coverage and excellent communication service quality based on the spatial information of the deployable locations, taking into full account the signal transmission characteristics of the base station and the limitations of the deployment environment, while reasonably controlling the number of base stations. This problem is essentially a spatial node selection optimization decision problem with multiple constraints. The crossover mutation mechanism of the genetic algorithm can flexibly respond to the needs of selective layout of 5G communication base stations in multiple different locations, ensuring that the base station layout plan achieves the best effect in terms of signal coverage. Summary of the invention

[0004] In order to achieve the purpose of ensuring signal propagation coverage while reducing the number of base stations as much as possible, the present invention provides a multi-objective optimization method for 5G base station layout, including the following steps:

[0005] S1. Obtain the spatial data of the 5G base station construction planning area and establish a coordinate system such as Figure 1 As shown, based on the sampling method and data model, a coordinate set C of candidate base station points is set, the size of which is n, that is, the set C = {C1(X c1 ,Y c1 ),C2(X c2 ,Y c2 ),...,Cn (X cn ,Y cn )};

[0006] S2. Convert the base station layout optimization problem into an optimization problem suitable for genetic algorithms. The base station selection can be regarded as a dynamic programming model based on solving the 0-1 knapsack problem. The number of base stations selected by the device is n, and the position state of each base station is X. i (i=1,2,...,n), where X i =1 means to select this location as the base station, X i =0 means no selection;

[0007] S3. The coverage radius R of a single base station has been confirmed. The coverage rate can be calculated based on the Monte Carlo principle. A set of m points within the layout space is randomly selected.

[0008] K={K1(X k1 ,Y k1 ),K2(X k2 ,Y k2 ),...,K n (X km ,Y km )}

[0009] Calculate each point K to state X i =1, the distance L from the base station i , when there is L i When ≤R, it means that the point is within the coverage range. The ratio of the number of points m1 to m within the coverage range is the coverage rate.

[0010] S4. Set the fitness function as the percentage of the selected base stations, calculate the coding fitness of the individuals that meet the coverage rate, and the number of individuals with position states of 0 in the population of size n, n0, then the individual fitness value

[0011] S5. Use a multi-objective optimization genetic algorithm and the Pareto dominance relationship between objective functions to evolve and select a set of mutually non-dominated solutions to solve the optimal layout of 5G base stations.

[0012] The step S5 uses a genetic algorithm to solve the optimal layout of 5G base stations, and the steps are as follows:

[0013] S51, determine the population size, crossover mutation probability and maximum number of iterations, and initialize the population;

[0014] S52, copying an operation population, and performing a crossover operation on the individuals therein according to the crossover probability;

[0015] S53, performing mutation operation on the individuals of the new population obtained in step S52 according to the mutation probability;

[0016] S54, merge the operation population with the original population to generate a new generation population of size 2N;

[0017] S55. Calculate the fitness value and coverage of each individual in the new generation population and perform non-dominated sorting. If individual P optimizes the target coverage P cov If the size of the fitness value Fitness is not less than that of individual Q, and is greater than that of individual Q in at least one goal, then individual P is said to dominate individual Q, otherwise individual P is said to be dominated by individual Q. If the above conditions are not met between individuals, the relationship between the two is called non-dominance, and then the order is distinguished according to the number of individuals dominated by the individual. The more the number of dominated individuals, the smaller the non-dominance level.

[0018] S56. Add the new population layer by layer from small to large according to the non-dominated layer, until one layer cannot completely accommodate the next generation, calculate the crowding value of all individuals in this layer, and use the crowding value as an indicator to add them one by one into the next generation population from large to small until the next generation population size is N;

[0019] S57, the method for calculating the crowding value is to calculate the perimeter of the largest rectangle that includes individual i but does not include other individuals according to the position of the specified individual i in the current level. The addition of crowding can make the solutions in the solution set as dispersed as possible to maintain the diversity of the population;

[0020] S58, judging whether to terminate the iteration according to the termination condition, and outputting a collection of site selection schemes after the termination condition is met;

[0021] S59, decode the Pareto optimal solution set output by step S58, and select the final solution according to the actual situation and needs on site.

[0022] As a preference, in S51, the initial population size is set to N, and the crossover operation probability is set to P. cro , the probability of mutation operation is P mut , the maximum number of iterations is G, the position state of the random base station is X i (i=1,2,...,n) is 0 or 1, generating a set X N ={X1,X2,...,X n} is an individual, and the process is repeated N times to generate the initial population.

[0023] Preferably, the specific steps of step S52 are:

[0024] S521, first randomly generate a floating point number P1 between 0 and 1 for each gene of the replicated operating population individual to determine whether the individual participates in the crossover operation; if P1 <P cro , then randomly select another individual in the new population for crossover operation;

[0025] S522. Next, for each pair of individuals undergoing crossover, two natural numbers r1 and r2 between 0 and n are randomly generated as the crossover start and end points, and two new offspring individuals are generated from the partial genes from the r1th to the r2th positions in the exchanged individuals;

[0026] S523, storing the new individuals after the crossover operation into the operation population;

[0027] S524. Finally, the remaining individuals that have not undergone the crossover operation are stored in the operation population.

[0028] Preferably, the specific steps of step S53 are:

[0029] S531, randomly generate a floating point number P2 between 0 and 1 for each gene of the operating population individual to determine whether each individual participates in the mutation. If P1 <P mut , then perform mutation operation, otherwise skip mutation operation;

[0030] S532, for the individual to be mutated, randomly generate r natural numbers between 0 and n, and reverse the 0-1 codes at the corresponding positions in the individual;

[0031] S533, storing the new individuals after the mutation operation into the operation population;

[0032] S534. Store the remaining individuals that have not undergone mutation operation into the operation population.

[0033] Preferably, the step S55 calculates the coverage of a single individual and fitness value Then, non-dominated sorting is performed according to the coverage rate and fitness value to determine the dominance level of each individual; each individual in the population has two parameters c i and S i , where c i is the number of other individuals in the population that dominate individual i, S i is the set of individuals dominated by individual i.

[0034] Preferably, step S56 selects all individuals from top to bottom according to the dominance hierarchy and places them into a new population until there is a layer that cannot completely accommodate the next generation, calculates the crowding values ​​of all individuals in this layer, and uses the crowding value as an indicator to place them one by one into the next generation population from large to small until the next generation population size is N.

[0035] Preferably, the step S57 is based on the individual N i The corresponding coordinates C i (X i ,Y i ), obtain the coordinates C of the two adjacent individuals before and after it i-1 (X i-1 ,Y i-1 ) and C i+1 (X i+1 ,Y i+1 ), calculated with C i-1 and C i+1 The semi-perimeter L of the rectangle with the opposite diagonal i =|X i-1 -X i+1 |+|Y i-1 -Y i+1 | is the crowding degree. The individuals at the two ends are considered to have a crowding degree of positive infinity and are given priority in entering the next generation.

[0036] Preferably, the termination condition of step S58 is whether the algorithm has converged or whether the algorithm has reached the maximum number of iterations. If any one of the conditions is met, the iteration is terminated and the individual with the highest fitness value is output as the site selection plan; otherwise, the execution continues until the termination condition is met.

[0037] Compared with the prior art, the advantages and positive effects of the present invention are:

[0038] 1. The base station layout has been optimized, which can reduce the number of base stations as much as possible while ensuring signal coverage. Based on theoretical research, the coverage area optimization based on genetic algorithm is applied to the base station layout to ensure product application quality and reduce infrastructure construction costs.

[0039] 2. The research on coverage area optimization problem is applicable to the entire base station site selection design. It can provide design solutions or reference solutions with high efficiency and quality, and can solve practical problems quickly and accurately, which has great practical significance and value. BRIEF DESCRIPTION OF THE DRAWINGS

[0040] In order to more clearly illustrate the technical solutions of the embodiments of the present invention, the following briefly introduces the drawings required for use in the description of the embodiments. Obviously, the drawings described below are some embodiments of the present invention. For ordinary technicians in this field, other drawings can be obtained based on these drawings without creative work.

[0041] Figure 1 It is a schematic diagram of the planning area coordinate system;

[0042] Figure 2 This is a schematic diagram of congestion calculation;

[0043] Figure 3 It is a flow chart of a multi-objective optimization method for 5G base station layout optimization;

[0044] Figure 4 This is a multi-objective optimization algorithm flow chart in a multi-objective optimization method for 5G base station layout optimization; DETAILED DESCRIPTION

[0045] In order to more clearly understand the above-mentioned purpose, features and advantages of the present invention, the present invention is further described below in conjunction with the accompanying drawings and embodiments. It should be noted that the embodiments of the present application and the features in the embodiments can be combined with each other without conflict.

[0046] In the following description, many specific details are set forth to facilitate a full understanding of the present invention. However, the present invention may also be implemented in other ways than those described herein. Therefore, the present invention is not limited to the specific embodiments of the following disclosure.

[0047] Examples, such as Figure 3 , the multi-objective optimization 5G base station layout optimization method of the embodiment of the present invention first studies the spatial discretization modeling of the area, and generates a set of candidate points for 5G communication base stations based on the spatial discretization method. Since the area to be deployed of the base station in the real world is a continuous space, in order to meet the requirements of the simulated base station signal coverage range and reduce the search space of the optimization algorithm, it is necessary to discretize the area to be deployed of the base station using the spatial discretization modeling method. The continuous planning space is coordinateized, and a point set is generated in the area to be deployed of the base station using the spatial discretization method. The base station candidate points are expressed in the form of coordinates. The point set is used as the candidate point set C for base station site selection, and the number of candidate points contained in the point set is n.

[0048] Further chromosome encoding is performed to encode each site selection scheme of the 5G communication base station into an individual of the genetic algorithm. Each candidate point location state corresponds to a gene, and the location state of each base station is X i (i=1,2,...,n), where X i =1 means to select this location as the base station, X i =0 means no selection; an individual represents a site selection plan with a length of n, that is, each plan contains n sites.

[0049] Then, the coverage rate is confirmed and the fitness function is set. Before the genetic algorithm is used, the acceptable coverage rate of the final solution needs to be confirmed. Set the coverage rate P according to the actual requirements. cov ; When the coverage radius R of a single base station is confirmed, the coverage rate can be calculated according to the Monte Carlo principle. A set of m points K is randomly generated within the layout space = {K1(X k1 ,Y k1),K2(X k2 ,Y k2 ),...,K m (X km ,Y km )}, calculate each point K m Go to position state X in set C i =1 point C i (X ci ,Y ci ) When there is L i When ≤R, it means that the point is within the coverage range. The ratio of the number of points m1 to m within the coverage range is the coverage rate. When there is a |X between two points km -X ci |>R or |Y km -Y ci |>R, then L>R; only individuals that meet this coverage can participate in the evolutionary iteration of the genetic algorithm. Set the fitness function to the position state X i =The proportion of base stations with a position state of 0, that is, the proportion of 0 in each individual code, the number of individuals with a length of n whose position state is 0 n0, then the individual fitness value

[0050] Considering that genetic algorithms may conflict with each other when solving multi-objective optimization problems, that is, the optimization of one of the objectives is at the expense of other objectives. Therefore, the present invention proposes to use the Pareto dominance relationship between objective functions to evolutionarily select a set of solutions that do not dominate each other. The Pareto solution set obtained in this way can contain more effective information to facilitate further decision-making based on actual conditions. The following concepts are about the Pareto optimal solution set: minimize the two feasible solutions x in the multi-objective problem. a and x b For example, for All have f i (x a ) <f i (x b ), then x a dominatex b ; In the feasible solution set, if x a is not dominated by any other solution, then x a is the Pareto optimal solution; the set of all Pareto optimal solutions is called the Pareto optimal solution set. The specific algorithm operation process is as follows Figure 4 , initialize the population, set the size of the initial individual population to N, and the crossover probability P cro , mutation probability P mut , individual length n, maximum number of iterations G; position state of random base station X i(i=1,2,...,n), generating body X N ={X1,X2,...,X n}, repeat the process to generate N individuals to initialize the population; copy an operation population and perform crossover and mutation operations on it; crossover operation, according to the probability P cro Perform a crossover operation on the individuals in the operating population, and randomly generate a floating point number P1 between 0 and 1 for each individual to determine whether the individual participates in the crossover operation; if P1 <P cro , then randomly select another individual in the new population for crossover operation; perform multi-point crossover on each pair of individuals in crossover operation, randomly generate two natural numbers r1 and r2 between 0 and n as the starting and ending points of crossover, and generate two new offspring individuals from the partial genes from the r1th to r2th positions in the exchanged individuals; the new individuals after the crossover operation replace the original individuals and are stored in the operating population; mutation operation, randomly generate a floating point number P2 between 0 and 1 for each individual in the operating population to determine whether each individual participates in mutation; if P1 <P mut , then mutation operation is performed, otherwise mutation operation is skipped; for the individuals to be mutated, r natural numbers between 0 and n are randomly generated, and the 0-1 codes of the corresponding positions in the individuals are reversed; the mutated new individuals replace the original individuals and are stored in the operation population; the original population and the operation population are merged to generate a new generation population of size 2N; the fitness value Fitness and coverage rate P of each individual in the new generation population are calculated cov And perform non-dominated sorting. Each individual in the population has two parameters c i and S i , where c i is the number of other individuals in the population that dominate individual i, S i is the set of individuals dominated by individual i. The specific calculation process is as follows: First, find all individuals in the population that are not dominated by any other individual, that is, c i The individuals with a value of 0 are stored in the set F1, and F1 is the first non-dominated level of the current population; then, according to each individual in F1, its dominating set S is examined separately. i The individuals included, let S i The dominant quantity c of each individual in iSubtract 1, if the result is 0, put the corresponding individual into the set F2, then F2 is the second non-dominated level of the current population; repeat the above operation on the current level until the entire population is graded; put the new population layer by layer from small to large according to the non-dominated level, until there is a layer that cannot completely put the next generation, calculate the crowding value of all individuals in this layer, and use the crowding value as an indicator to put them into the next generation population one by one from large to small until the next generation population size is N; the crowding value is calculated according to the position of the specified individual i in the current level, and the perimeter of the largest rectangle that contains individual i but does not contain other individuals. The addition of crowding can make the solutions in the solution set as dispersed as possible and maintain the diversity of the population. For example, individual N i The corresponding coordinates C i (X i ,Y i ), obtain the coordinates C of the two adjacent individuals before and after it i-1 (X i-1 ,Y i-1 ) and C i+1 (X i+1 ,Y i+1 ), calculated with C i-1 and C i+1 The semi-perimeter L of the rectangle with the opposite diagonal i =|X i-1 -X i+1 |+|Y i-1 -Y i+1 | is the crowding degree. The individuals at the two ends are considered to have a crowding value of positive infinity and are given priority to enter the next generation. The termination condition is set to determine whether the algorithm has reached the maximum number of iterations. If it is met, the iteration will be terminated and the current population will be output as the Pareto optimal solution set. Otherwise, the execution will continue until the termination condition is met.

[0051] Finally, by decoding the Pareto optimal solution set, multiple deployment plans can be obtained, and the final plan is determined based on the actual situation and needs on site.

[0052] The above description is only a preferred embodiment of the present invention and does not limit the present invention in other forms. Any technician familiar with the profession may use the technical content disclosed above to change or modify it into an equivalent embodiment with equivalent changes and apply it to other fields. However, any simple modification, equivalent change and modification made to the above embodiment based on the technical essence of the present invention without departing from the content of the technical solution of the present invention still falls within the protection scope of the technical solution of the present invention.

Claims

1. A multi-objective optimization method for 5G base station layout optimization, characterized in that: The following steps are involved: S1. Obtain the spatial data of the 5G base station construction planning area and establish a coordinate system as shown in Figure 1. Based on the sampling method and data model, set the coordinate set C of the candidate base station points, with a size of n, that is, set C = {C1(X c1 ,Y c1 ),C2(X c2 ,Y c2 ),...,C n (X cn ,Y cn )}; S2. Convert the base station layout optimization problem into an optimization problem suitable for genetic algorithms. The base station selection can be regarded as a dynamic programming model based on solving the 0-1 knapsack problem. The number of base stations selected by the device is n, and the position state of each base station is X. i (i=1,2,...,n), where X i =1 means to select this location as the base station, X i =0 means no selection; S3. The coverage radius R of a single base station has been confirmed. The coverage rate can be calculated based on the Monte Carlo principle. A set of m points within the layout space is randomly selected. K={K1(X k1 ,Y k1 ),K2(X k2 ,Y k2 ),...,K n (X km ,Y km )} Calculate each point K to state X i =1, the distance L from the base station i , when there is L i When ≤R, it means that the point is within the coverage range. The ratio of the number of points m1 to m within the coverage range is the coverage rate. S4. Set the fitness function as the percentage of the selected base stations, calculate the coding fitness of the individuals that meet the coverage rate, and the number of individuals with position states of 0 in the population of size n, n0, then the individual fitness value S5. Use a multi-objective optimization genetic algorithm and the Pareto dominance relationship between objective functions to evolve and select a set of mutually non-dominated solutions to solve the optimal layout of 5G base stations. The step S5 uses a genetic algorithm to solve the optimal layout of 5G base stations, and the steps are as follows: S51, determine the population size, crossover mutation probability and maximum number of iterations, and initialize the population; S52, copying an operation population, and performing a crossover operation on the individuals therein according to the crossover probability; S53, performing mutation operation on the individuals of the new population obtained in step S52 according to the mutation probability; S54, merge the operation population with the original population to generate a new generation population of size 2N; S55. Calculate the fitness value and coverage of each individual in the new generation population and perform non-dominated sorting. If individual P optimizes the target coverage P cov If the size of the fitness value Fitness is not less than that of individual Q, and is greater than that of individual Q in at least one goal, then individual P is said to dominate individual Q, otherwise individual P is said to be dominated by individual Q. If the above conditions are not met between individuals, the relationship between the two is called non-dominance, and then the order is distinguished according to the number of individuals dominated by the individual. The more the number of dominated individuals, the smaller the non-dominance level. S56. Add the new population layer by layer from small to large according to the non-dominated layer, until one layer cannot completely accommodate the next generation, calculate the crowding value of all individuals in this layer, and use the crowding value as an indicator to add them one by one into the next generation population from large to small until the next generation population size is N; S57, the method for calculating the crowding value is to calculate the perimeter of the largest rectangle that includes individual i but does not include other individuals according to the position of the specified individual i in the current level. The addition of crowding can make the solutions in the solution set as dispersed as possible to maintain the diversity of the population; S58, judging whether to terminate the iteration according to the termination condition, and outputting a collection of site selection schemes after the termination condition is met; S59, decode the Pareto optimal solution set output by step S58, and select the final solution according to the actual situation and needs on site.

2. According to a multi-objective optimization method for 5G base station layout optimization according to claim 1, it is characterized in that: In S51, the initial population size is set to N, and the crossover operation probability is set to P cro , the probability of mutation operation is P mut , the maximum number of iterations is G, the position state of the random base station is X i (i=1,2,...,n) is 0 or 1, generating a set X N ={X1,X2,...,X n } is an individual, and the process is repeated N times to generate the initial population.

3. According to a multi-objective optimization method for 5G base station layout optimization according to claim 1, it is characterized in that: The specific steps of step S52 are: S521, first randomly generate a floating point number P1 between 0 and 1 for each gene of the replicated operating population individual to determine whether the individual participates in the crossover operation; if P1 <P cro , then randomly select another individual in the new population for crossover operation; S522. Next, for each pair of individuals undergoing crossover, two natural numbers r1 and r2 between 0 and n are randomly generated as the crossover start and end points, and two new offspring individuals are generated from the partial genes from the r1th to the r2th positions in the exchanged individuals; S523, storing the new individuals after the crossover operation into the operation population; S524. Finally, the remaining individuals that have not undergone the crossover operation are stored in the operation population.

4. According to a multi-objective optimization method for 5G base station layout optimization according to claim 1, it is characterized in that: The specific steps of step S53 are: S531, randomly generate a floating point number P2 between 0 and 1 for each gene of the operating population individual to determine whether each individual participates in the mutation. If P1 <P mut , then perform mutation operation, otherwise skip mutation operation; S532, for the individual to be mutated, randomly generate r natural numbers between 0 and n, and reverse the 0-1 codes at the corresponding positions in the individual; S533, storing the new individuals after the mutation operation into the operation population; S534. Store the remaining individuals that have not undergone mutation operation into the operation population.

5. According to a multi-objective optimization method for 5G base station layout optimization according to claim 1, it is characterized in that: The step S55 calculates the coverage rate of a single individual and fitness value Then, non-dominated sorting is performed according to the coverage rate and fitness value to determine the dominance level of each individual; each individual in the population has two parameters c i and S i , where c i is the number of other individuals in the population that dominate individual i, S i is the set of individuals dominated by individual i.

6. The multi-objective optimization method for 5G base station layout optimization according to claim 1 is characterized in that: The step S56 selects all individuals from top to bottom according to the dominance level and puts them into the new population until there is a layer that cannot be fully placed in the next generation. The crowding value of all individuals in this layer is calculated, and the crowding value is used as an indicator to put them into the next generation population one by one from large to small until the size of the next generation population is N.

7. The multi-objective optimization method for 5G base station layout optimization according to claim 1 is characterized in that: The step S57 is based on the individual N i The corresponding coordinates C i (X i ,Y i ), obtain the coordinates C of the two adjacent individuals before and after it i-1 (X i-1 ,Y i-1 ) and C i+1 (X i+1 ,Y i+1 ), calculated with C i-1 and C i+1 The semi-perimeter L of the rectangle with the opposite diagonal i =|X i-1 -X i+1 |+|Y i-1 -Y i+1 | is the crowding degree. The individuals at the two ends are considered to have a crowding degree of positive infinity and are given priority in entering the next generation.

8. The multi-objective optimization method for 5G base station layout optimization according to claim 1, characterized in that: The step S58 sets the termination condition as whether the algorithm has converged or whether the algorithm has reached the maximum number of iterations. If any of the conditions is met, the iteration is terminated and the individual with the highest fitness value is output as the site selection plan; otherwise, the execution continues until the termination condition is met.

Citation Information

Cited By

  • Wireless communication site planning method based on multi-target genetic algorithm

    CN119907008A

  • Method, device and equipment for determining site selection of power conversion cabinet and medium

    CN120450768A

  • Sonar buoy array layout optimization method based on multi-objective evolution and electronic equipment

    CN122113329A

  • A Sonar Buoy Array Layout Optimization Method and Electronic Equipment Based on Multi-Objective Evolution

    CN122113329B