A site selection method, apparatus and storage medium

Abstract

The application discloses a site selection method and device and a storage medium. A first candidate population corresponding to M to-be-sited objects is determined, and the first candidate population includes N candidate solution vectors. The candidate solution vectors correspond to candidate addresses of at least one to-be-sited object. For each candidate solution vector in the first candidate population, the fitness of the candidate solution vector is determined based on a crowded distance, a cost and a blind area loss corresponding to the candidate solution vector. The first candidate population is adjusted based on the fitness of each candidate solution vector in the first candidate population, and a new first candidate population is obtained. If the number of adjustment times of the first candidate population is less than an adjustment threshold, the fitness of each candidate solution vector in the new first candidate population is continuously determined, and the new first candidate population is adjusted based on the fitness, until the number of adjustment times of the first candidate population is equal to the adjustment threshold. A target solution vector is determined from the first candidate population, and the to-be-sited objects can be reasonably deployed.

Description

TECHNICAL FIELD

[0001] The present application relates to the computer technical field, and particularly relates to a site selection method, device and storage medium. BACKGROUND

[0002] In recent years, with the rapid development of intelligent optimization decision technology, scholars at home and abroad apply some bionics intelligent optimization algorithms, such as genetic algorithm, immune algorithm and particle swarm algorithm, to the process of base station site selection. Base station site selection based on intelligent optimization algorithms causes the base station distribution to be too dense, and the coverage area to have many overlaps, resulting in a large waste of base station resources. SUMMARY

[0003] The present application provides a site selection method, device and storage medium, which can avoid site selection overlap of to-be-sited objects and reasonably deploy to-be-sited objects.

[0004] The technical solution of the present application is implemented as follows:

[0005] The present application provides a site selection method, which comprises the following steps:

[0006] determining a first candidate population corresponding to M to-be-sited objects, wherein the first candidate population comprises N candidate solution vectors; the candidate solution vectors correspond to candidate addresses of the at least one to-be-sited object; wherein M is a positive integer, and N is a positive integer;

[0007] for each candidate solution vector in the first candidate population, determining the fitness of the candidate solution vector based on the congestion distance, cost and blind area loss corresponding to the candidate solution vector;

[0008] adjusting the first candidate population based on the fitness of each candidate solution vector in the first candidate population to obtain a new first candidate population;

[0009] if the number of adjustment times of the first candidate population is less than an adjustment threshold, continuing to determine the fitness of each candidate solution vector in the new first candidate population, and adjusting the new first candidate population based on the fitness until the number of adjustment times of the first candidate population is equal to the adjustment threshold;

[0010] determining a target solution vector from the first candidate population.

[0011] In the above solution, before the step of determining the fitness of the candidate solution vector based on the congestion distance, cost and blind area loss corresponding to the candidate solution vector, the method further comprises the following steps:

[0012] determine a total area of the blind area according to a total area of the deployment area and a coverage area of the to-be-sited object corresponding to the first candidate population; the deployment area is used to represent a to-be-deployed area corresponding to the to-be-sited object;

[0013] determine the blind area loss based on the total area of the blind area.

[0014] In the foregoing solution, before the fitness of the candidate solution vector is determined based on the congestion distance, the cost, and the blind area loss corresponding to the candidate solution vector, the method further includes:

[0015] determine a distance between the first to-be-sited object and each second to-be-sited object in the M to-be-sited objects, and add the distances between the first to-be-sited object and each second to-be-sited object to obtain a distance sum corresponding to the first to-be-sited object; the second to-be-sited object is a to-be-sited object other than the first to-be-sited object in the M to-be-sited objects;

[0016] add the distance sums corresponding to the M to-be-sited objects to obtain the congestion distance.

[0017] In the foregoing solution, the fitness of the candidate solution vector is determined based on the congestion distance, the cost, and the blind area loss corresponding to the candidate solution vector, including:

[0018] perform weighted summation on the cost and the blind area loss to obtain a temporary fitness;

[0019] perform weighted summation on the temporary fitness and a reciprocal value of the congestion distance to obtain the fitness of the candidate solution vector.

[0020] In the foregoing solution, the first candidate population is adjusted based on the fitness of each candidate solution vector in the first candidate population to obtain a new first candidate population, including:

[0021] for each candidate solution vector in the first candidate population, determine an adjusted candidate solution vector corresponding to the candidate solution vector according to a step factor and a first adjustment step;

[0022] adjust N candidate solution vectors in the first candidate population and the N adjusted candidate solution vectors based on the fitness of each candidate solution vector in the first candidate population and the fitness of each adjusted candidate solution vector to obtain N new candidate solution vectors;

[0023] the new first candidate population is composed of the N new candidate solution vectors.

[0024] In the foregoing scheme, the adjustment candidate solution vector corresponding to each candidate solution vector in the first candidate solution population is determined according to the step factor and the first adjustment step, and the method comprises the following steps of:

[0025] The step factor is multiplied by the first adjustment step to obtain a second adjustment step.

[0026] For each candidate solution vector of the first candidate solution population, the candidate solution vector is summed with the second adjustment step to obtain the adjustment candidate solution vector corresponding to the candidate solution vector.

[0027] In the foregoing scheme, the method further comprises the following steps of:

[0028] If the adjustment candidate solution vector is not in the first deployment area, the adjustment candidate solution vector is discarded, a summation result is obtained by summing the adjustment candidate solution vector with a random number, and the adjustment candidate solution vector is updated based on the summation result.

[0029] In the foregoing scheme, before the N new candidate solution vectors are obtained by adjusting the N candidate solution vectors in the first candidate solution population and the N adjustment candidate solution vectors based on the fitness of each candidate solution vector in the first candidate solution population and the fitness of each adjustment candidate solution vector, the method further comprises the following steps of:

[0030] For each adjustment candidate solution vector, the fitness of the adjustment candidate solution vector is determined based on the congestion distance corresponding to the adjustment candidate solution vector, the cost, and the blind area loss.

[0031] In the foregoing scheme, the N new candidate solution vectors are obtained by adjusting the N candidate solution vectors in the first candidate solution population and the N adjustment candidate solution vectors based on the fitness of each candidate solution vector in the first candidate solution population and the fitness of each adjustment candidate solution vector, and the method comprises the following steps of:

[0032] The N candidate solution vectors in the first candidate solution population and the N adjustment candidate solution vectors are sorted according to the fitness of each candidate solution vector in the first candidate solution population and the fitness of each adjustment candidate solution vector, and N candidate solution vectors with minimum fitness in the sorting result are determined.

[0033] The N candidate solution vectors with minimum fitness are determined as the N new candidate solution vectors.

[0034] Embodiments of the present application provide a site selection device, which comprises a determination unit and an adjustment unit, and the determination unit is configured to determine a first candidate solution population, and the adjustment unit is configured to adjust each candidate solution vector in the first candidate solution population to obtain an adjustment candidate solution vector corresponding to the candidate solution vector.

[0035] The determining unit is configured to determine a first candidate population corresponding to the M candidate site objects, the first candidate population comprising N candidate solution vectors, the candidate solution vectors corresponding to candidate addresses of the at least one candidate site object, wherein M is a positive integer, and N is a positive integer; determine, for each candidate solution vector in the first candidate population, a fitness of the candidate solution vector based on a congestion distance, a cost, and a blind area loss corresponding to the candidate solution vector; continue to determine the fitness of each candidate solution vector in a new first candidate population when a number of adjustments of the first candidate population is less than an adjustment threshold; and determine a target solution vector from the first candidate population.

[0036] The adjusting unit is configured to adjust the first candidate population based on the fitness of each candidate solution vector in the first candidate population to obtain a new first candidate population; and adjust the new first candidate population based on the fitness until the number of adjustments of the first candidate population is equal to the adjustment threshold.

[0037] Embodiments of the present application provide a site selection device, which comprises:

[0038] A memory configured to store executable data instructions.

[0039] A processor configured to execute the executable instructions stored in the memory, and when the executable instructions are executed, the processor performs the site selection method.

[0040] Embodiments of the present application provide a readable storage medium storing executable instructions, and when the executable instructions are executed by one or more processors, the processor performs the site selection method.

[0041] Embodiments of the present application provide a batch site selection method and device and a readable storage medium, which determine a first candidate population corresponding to M candidate site objects, the first candidate population comprising N candidate solution vectors, the candidate solution vectors corresponding to candidate addresses of the at least one candidate site object, wherein M is a positive integer, and N is a positive integer; determine, for each candidate solution vector in the first candidate population, a fitness of the candidate solution vector based on a congestion distance, a cost, and a blind area loss corresponding to the candidate solution vector; adjust the first candidate population based on the fitness of each candidate solution vector in the first candidate population to obtain a new first candidate population; continue to determine the fitness of each candidate solution vector in the new first candidate population when the number of adjustments of the first candidate population is less than an adjustment threshold; adjust the new first candidate population based on the fitness until the number of adjustments of the first candidate population is equal to the adjustment threshold; and determine a target solution vector from the first candidate population. The above method considers the congestion distance factor in the site selection process, so that the site selection result can avoid the site selection overlap of the candidate site objects and reasonably deploy the candidate site objects. BRIEF DESCRIPTION OF DRAWINGS

[0042] Figure 1 An optional flowchart of the site selection method provided by an embodiment of the present application Figure One ;

[0043] Figure 2 An optional flowchart of the site selection method provided by an embodiment of the present application Figure Two ;

[0044] Figure 3 An optional flowchart of the site selection method provided by an embodiment of the present application Figure Three ;

[0045] Figure 4 An optional flowchart of the site selection method provided by an embodiment of the present application Figure Four ;

[0046] Figure 5 An optional flowchart of the site selection method provided by an embodiment of the present application Figure Five ;

[0047] Figure 6 A structural diagram of a site selection device provided by an embodiment of the present application Figure One ;

[0048] Figure 7 A structural diagram of a site selection device provided by an embodiment of the present application Figure Two . DETAILED DESCRIPTION

[0049] In order to make the purpose, technical solutions and advantages of the present application clearer, the present application will be described in further detail below with reference to the drawings, and the described embodiments should not be regarded as limiting the present application, and all other embodiments obtained by those skilled in the art without making creative efforts fall within the scope of protection of the present application.

[0050] An embodiment of the present application provides a site selection method, which is applied to an electronic device. The function realized by the method can be realized by a processor in the electronic device calling program code, and of course the program code can be saved in a computer storage medium, so the electronic device at least includes a processor and a storage medium.

[0051] The electronic device can be any electronic device with information processing capability, in an embodiment, the electronic device can be a smart terminal, for example, a notebook or other mobile terminal with communication capability. In another embodiment, the electronic device can also be a terminal device with computing function which is inconvenient to move, such as a desktop computer, a desktop computer, a server, etc.

[0052] Of course, embodiments of the present application are not limited to being provided as methods and hardware, but can also have various implementation manners, for example, being provided as a storage medium (storing instructions for executing the transmission control method provided by the embodiments of the present application).

[0053] Figure 1 is an optional flowchart of the site selection method provided by the embodiments of the present application Figure One As shown in the figure, the method comprises: Figure 1

[0054] S101, determining a first candidate population corresponding to M candidate site objects, the first candidate population comprising N candidate solution vectors; the candidate solution vector corresponding to a candidate address of at least one candidate site object; wherein M is a positive integer, and N is a positive integer.

[0055] In the embodiments of the present application, the scenario suitable for initializing the first candidate population corresponding to the candidate site object is applicable.

[0056] In the site selection method provided by the embodiments of the present application, the candidate site object can be a base station, a warehouse, a store, etc., and the embodiments of the present application do not limit it.

[0057] The candidate solution vector comprises at least one candidate site object. For one candidate site object, the candidate site object can be represented by the parameters of the candidate site object, wherein the parameters of the candidate site object comprise information representing one candidate site object, such as position information, coverage area, etc. Taking the candidate site object as a base station as an example, the parameters of the candidate site object can comprise position information, antenna model, 4th Generation Mobile Communication Technology (4G, 4th Generation Mobile Communication Technology) base station sharing situation, etc.; wherein the position information can comprise longitude and latitude, and the longitude and latitude of the candidate site object are used to represent the candidate address of the candidate site object. Taking the candidate site object as a warehouse as an example, the parameters of the candidate site object can comprise position information, coverage range, etc. The first candidate population is used to represent the optimized population.

[0058] In the embodiments of the present application, the electronic device first determines a first candidate population corresponding to M candidate site objects, wherein the first candidate population comprises N candidate solution vectors; the candidate solution vector corresponding to a candidate address of at least one candidate site object; wherein M is a positive integer, and N is a positive integer.

[0059] It should be noted that when the candidate site object is a base station, the base station can be a 5th Generation Mobile Communication Technology (5G, 5th Generation Mobile Communication Technology) communication base station, and the embodiments of the present application do not limit it.

[0060] ​Exemplarily, the to-be-sited object is a base station, the parameters of the base station are longitude, latitude, antenna model, and 4G base station sharing, and then one to-be-sited object can be represented as a vector x=(x1, x2, x3, x4); wherein x1 and x2 represent the longitude and latitude of the to-be-sited object respectively; x3 represents the antenna model, and x3 e {0, 1}, 0 represents that the antenna model is type I antenna, which represents an antenna corresponding to a new 5G base station, and 1 represents that the antenna model is type II antenna, which represents an antenna corresponding to a 4G base station; x4 represents the 4G base station sharing, and x4 e {0, 1}. For example, x=(100, 39, 1, 1) represents that the to-be-sited object has a longitude position of 100°, a latitude position of 39°, a type II antenna, and shares a 4G base station. One to-be-sited object is represented as a vector x=(x1, x2, x3, x4), and then a set of m to-be-sited objects in the to-be-deployed area is a solution X of the optimal sited problem, that is, a candidate solution vector, which can be represented as follows:

[0061]

[0062] Randomly initialize to generate n Xs to form a to-be-optimized population, that is, a first candidate population Y, wherein Y=(X1, X2,..., Xn). n

[0063] In S102, for each candidate solution vector in the first candidate population, the fitness of the candidate solution vector is determined based on the corresponding congestion distance, cost, and blind area loss of the candidate solution vector.

[0064] In the embodiments of the present application, the scenario applicable to determining the fitness of the candidate solution vector.

[0065] In the sited method provided in the embodiments of the present application, the fitness of the candidate solution vector is determined based on the corresponding congestion distance, cost, and blind area loss of the candidate solution vector, which can be embodied as determining the fitness of the candidate solution vector by using the congestion distance, the cost function, and the coverage loss objective function. The fitness of the candidate solution vector is used to represent the rationality of deploying and sited the to-be-sited object according to the candidate solution vector.

[0066] In the embodiments of the present application, for each candidate solution vector in the first candidate population, the electronic device determines the fitness of the candidate solution vector based on the corresponding congestion distance, cost, and blind area loss of the candidate solution vector.

[0067] Exemplarily, for each candidate solution vector in the first candidate population, the fitness of each candidate solution vector is determined according to the corresponding congestion distance, cost function value, and coverage loss objective function value of each candidate solution vector.

[0068] ​It should be noted that, in the case of base stations as the objects to be sited in this application embodiment, the cost function for deploying a 5G network can consider the cost of building a new 5G base station, as well as the cost of the antennas corresponding to the new 5G base station and the antennas corresponding to the 4G base station. Coverage determines the performance of the 5G network; therefore, the site selection method provided in this application embodiment can also consider coverage maximization factors. For ease of calculation, the coverage loss objective function value can be determined based on the total area of ​​the area to be deployed, the sum of the areas covered by the objects to be sited, and the area of ​​the overlapping coverage areas between any two base stations. Furthermore, if the base stations are too close together, there will be significant overlap in coverage areas, leading to resource waste and affecting the coverage of the base station in the entire area to be deployed. Based on this, the site selection method provided in this application embodiment can reflect the redundancy density of a base station through the congestion distance between base stations.

[0069] It is understood that when determining the fitness of each candidate solution vector in the embodiments of this application, the crowding distance of the object to be located is taken into account, so as to ensure that there is not much overlap in the coverage area, which helps to improve the coverage of the object to be located in the whole area, reduce the overlap of the coverage areas of different objects to be located, and reduce the waste of resources caused by the overlap of the coverage areas of different objects to be located.

[0070] S103. Adjust the first candidate population based on the fitness of each candidate solution vector in the first candidate population to obtain a new first candidate population.

[0071] In this embodiment, it is applicable to the scenario of updating the first candidate population to obtain a new first candidate population.

[0072] In this embodiment of the application, the electronic device adjusts the first candidate population based on the fitness of each candidate solution vector in the first candidate population to obtain a new first candidate population.

[0073] S104. If the number of adjustments to the first candidate population is less than the adjustment threshold, continue to determine the fitness of each candidate solution vector in the new first candidate population, and adjust the new first candidate population based on the fitness until the number of adjustments to the first candidate population is equal to the adjustment threshold.

[0074] In this embodiment, the application is applicable to the scenario of continuously iteratively updating the first candidate population.

[0075] In this embodiment of the application, if the number of adjustments to the first candidate population is less than the adjustment threshold, the electronic device continues to determine the fitness of each candidate solution vector in the new first candidate population, and adjusts the new first candidate population based on the fitness until the number of adjustments to the first candidate population is equal to the adjustment threshold.

[0076] It should be noted that the threshold value can be adjusted according to the actual situation in this application embodiment, and this application embodiment does not impose any restrictions.

[0077] S105. Determine the target solution vector from the first candidate population.

[0078] In this embodiment, the optimal solution vector is determined when the number of adjustments equals the adjustment threshold.

[0079] In this embodiment of the application, the electronic device determines the target solution vector from the first candidate population.

[0080] For example, the target solution vector in the addressing method provided in the application embodiment is used to characterize the optimal solution vector, that is, the parameters of each object to be addressed corresponding to the target solution vector are the optimal parameters.

[0081] In some embodiments of this application, Figure 2 This is an optional flowchart illustrating the location selection method provided in the embodiments of this application. Figure Two ,like Figure 2 As shown, steps S106 and S107 are included before step S102:

[0082] S106. Determine the total blind zone area based on the total area of ​​the deployment area and the coverage area of ​​the candidate object corresponding to the first candidate population; the deployment area is used to characterize the deployment area corresponding to the candidate object.

[0083] This application embodiment is applicable to scenarios where the total blind area of ​​the deployment region is determined.

[0084] In this embodiment of the application, the electronic device determines the total blind area based on the total area of ​​the deployment area and the coverage area of ​​the candidate object corresponding to the first candidate population; wherein, the deployment area is used to characterize the deployment area corresponding to the candidate object.

[0085] For example, the effective area covered by the candidate objects is obtained by subtracting the area of ​​the overlapping area between each pair of candidate objects from the sum of the areas covered by the candidate objects; then the total blind zone area is obtained by subtracting the effective area covered by the candidate objects from the total area of ​​the deployment area.

[0086] S107. Determine blind zone loss based on the total blind zone area.

[0087] In this embodiment, the application is applicable to scenarios where the blind zone loss of candidate solution vectors is determined.

[0088] In this embodiment, the electronic device determines the blind zone loss based on the total blind zone area.

[0089] For example, in embodiments of this application, a coverage loss objective function can be used to determine the blind zone loss, wherein the coverage loss objective function f2(X) can be defined as formula (1):

[0090]

[0091] Where C2 represents the loss caused by each square kilometer of coverage blind spot; S represents the total area of ​​the deployment area; mπr 2 The sum of the areas covered by m objects to be located; s ij Let m be the area of ​​the overlapping region covered by the m objects to be located, i being greater than or equal to 1 and less than or equal to m; j being greater than or equal to 1 and less than or equal to m. In other words, it is calculated based on the sum of the areas covered by the m objects to be located, mπr. 2 Subtract the area of ​​the overlapping region between each pair of candidate sites. Obtain the effective area covered by the object to be located. Then, subtract the effective area covered by the target object from the total area S of the deployment area. Obtain the total area of ​​the blind spot Then, multiply C2 by the total area of ​​the blind zone based on the loss caused by each square kilometer of coverage blind zone. Losses due to blind spots.

[0092] In some embodiments of this application, Figure 3 This is an optional flowchart illustrating the location selection method provided in the embodiments of this application. Figure Three ,like Figure 3 As shown, steps S108 and S109 are included before step S102. (See below:)

[0093] S108. Take each of the M candidate objects as the first candidate object, determine the distance between the first candidate object and the second candidate object, and add the distances between the first candidate object and each of the second candidate objects to obtain the sum of the distances corresponding to the first candidate object; the second candidate object is the candidate object other than the first candidate object among the M candidate objects.

[0094] In this embodiment of the application, it is applicable to the scenario of calculating the sum of distances corresponding to the first object to be located.

[0095] In the addressing method provided in the embodiments of this application, the second addressable object is the addressable object other than the first addressable object among M addressable objects.

[0096] In this embodiment of the application, the electronic device takes each of the M candidate objects as a first candidate object, determines the distance between the first candidate object and the second candidate object, and adds the distances between the first candidate object and each of the second candidate objects to obtain the distance sum corresponding to the first candidate object.

[0097] For example, object x among M objects to be addressed p and object x q Distance d(x) p ,x q It can be defined as formula (2):

[0098]

[0099] If the first object to be selected is object x p Then object x p The sum of distances d(x) to M candidate objects p It can be defined as formula (3):

[0100]

[0101] It should be noted that, for ease of implementation, in this embodiment of the application, while ensuring consistent calculation results, the sum of distances between the first candidate object and the M candidate objects can be used to represent the distance to the first candidate object.

[0102] S109. Add the distances corresponding to each of the M candidate objects to obtain the congestion distance.

[0103] In this embodiment, the application is applicable to scenarios where the crowding distance of candidate solution vectors is obtained.

[0104] In this embodiment, the electronic device sums the distances corresponding to each of the M candidate objects to obtain the congestion distance.

[0105] For example, if the first candidate site is base station x p The first candidate object x p The corresponding distance is represented as Then the crowding distance d(X) of the candidate solution vector X can be defined by formula (4):

[0106]

[0107] Where, x p It is any one of the M candidate objects.

[0108] It is understood that, by providing a method for calculating the crowding distance of candidate solution vectors in this embodiment of the application, a basis is provided for subsequently determining the fitness of candidate solution vectors based on the crowding distance.

[0109] In some embodiments of this application, Figure 4 This is an optional flowchart illustrating the method provided in the embodiments of this application. Figure Four ,like Figure 4 As shown, step S102 can be implemented through S110 and S111, which will be explained in conjunction with each step.

[0110] S110. The temporary fitness is obtained by weighted summation of cost and blind spot loss.

[0111] This application embodiment is applicable to scenarios where temporary fitness is determined.

[0112] In the location selection method provided in this application embodiment, the blind zone loss can be determined based on the coverage loss objective function; the cost in the location selection method provided in this application embodiment can be determined based on the cost function.

[0113] In this embodiment, the electronic device obtains a temporary fitness by weighted summation of cost and blind spot loss.

[0114] For example, if the target site is a base station, based on the characteristics of 5G communication networks, 5G networks will not quickly and completely replace 4G networks. The current approach is to deploy 5G networks gradually. To reduce the cost of 5G networks, it is preferable to utilize existing 4G base stations. When using 4G base stations, only the cost of the antenna needs to be considered. Therefore, the cost function f1(X) for 5G base station construction can be defined as formula (5):

[0115] f1(X)=hC1+m1E1+m2E2 Formula (5);

[0116] Where h represents the number of 5G base stations to be built; C1 represents the cost of building a new 5G base station; m i E represents the total number of type i antennas used. i The cost of each type i antenna; the total number of base stations required for the planned area is m = m1 + m2. The coverage loss objective function in this embodiment can be defined as:

[0117]

[0118] Where C2 represents the loss caused by each square kilometer of coverage blind spot; S represents the total area of ​​the deployment area; mπr 2 Let s be the sum of the areas covered by m base stations; ijLet be the area of ​​the overlapping coverage region of base station i and base station j, where i is greater than or equal to 1 and less than or equal to m; j is greater than or equal to 1 and less than or equal to m. By combining the two functions using the weighting method, the temporary fitness function f(X) to be optimized can be defined as formula (6):

[0119] f(X)=λ1f1+λ2f2 Formula (6);

[0120] Where λ1 is the weighting coefficient of the cost function; λ2 is the weighting coefficient of the coverage loss objective function; and λ1+λ2=1. The temporary fitness function is used to determine the temporary fitness of the candidate solution vector.

[0121] It should be noted that, in this embodiment, the cost function and coverage loss objective function can be combined using a weighting method to obtain a temporary fitness function, and the temporary fitness of the candidate solution vector can be determined based on the temporary fitness function. The weighting system can be adjusted according to the influence of cost and blind zone loss factors on the site selection method; this embodiment does not impose any limitations on this adjustment.

[0122] S111. The fitness of the candidate solution vector is obtained by weighted summation of the temporary fitness and the reciprocal of the crowding distance.

[0123] This application embodiment is applicable to scenarios where the fitness of candidate solution vectors is determined.

[0124] In this embodiment, the electronic device obtains the fitness of the candidate solution vector by weighted summation of the temporary fitness and the reciprocal of the congestion distance.

[0125] For example, when the temporary utility function of the candidate solution vector X is expressed as f(X) = λ1f1 + λ2f2, the crowding distance is expressed as: The fitness function g(X) of the candidate solution vector can then be defined as Equation (7):

[0126]

[0127] ε represents the weight coefficient of the crowding distance, which ranges from [0, 1]. The fitness function values ​​of n candidate solution vectors can be calculated based on this fitness function.

[0128] It should be noted that when the electronic device obtains the fitness of the candidate solution vector by weighted summation based on the reciprocal of the temporary fitness and the congestion distance, the weight coefficients can be adjusted according to the proportion of the temporary fitness and the congestion distance. This application embodiment does not impose any restrictions.

[0129] It is understood that when determining the fitness of each candidate solution vector in the embodiments of this application, the crowding distance of the object to be located is taken into account, so as to ensure that there is not much overlap in the coverage area, which helps to improve the coverage of the object to be located in the whole area, reduce the overlap of the coverage areas of different objects to be located, and reduce the loss caused by the overlap of the coverage areas of different objects to be located.

[0130] In some embodiments of this application, Figure 5 This is an optional flowchart illustrating the location selection method provided in the embodiments of this application. Figure Five ,like Figure 5 As shown, step S103 can be implemented through S112-S114, which will be explained in conjunction with each step.

[0131] S112. For each candidate solution vector in the first candidate population, determine the adjusted candidate solution vector corresponding to the candidate solution vector based on the step size factor and the first adjustment step size.

[0132] In this embodiment, the application is applicable to the scenario of determining the adjusted candidate solution vector corresponding to the candidate solution vector.

[0133] In the location selection method provided in this application embodiment, the step size factor is a variable that can be adjusted according to different stages of the iteration.

[0134] In this embodiment of the application, for each candidate solution vector in the first candidate population, the electronic device determines the adjusted candidate solution vector corresponding to the candidate solution vector based on the step size factor and the first adjustment step size.

[0135] In some embodiments of this application, the electronic device multiplies the step size factor and the first adjustment step size to obtain the second adjustment step size; for each candidate solution vector of the first candidate population, the electronic device sums the candidate solution vector with the second adjustment step size to obtain the adjusted candidate solution vector corresponding to the candidate solution vector.

[0136] For example, each candidate solution vector in the first candidate population can be represented as X. i (k) will adjust the step size factor α and the first adjustment step size. Multiply to obtain the second adjustment step size Then, the candidate solution vector X i (k) and the second adjustment step size Summing yields the adjusted candidate solution vector X corresponding to the candidate solution vector. i (k+1) can be defined by formula (8):

[0137]

[0138] Among them, X bestdenoted as the optimal candidate solution vector of the first candidate population, μ and v are two random numbers following a normal distribution; β is a coefficient and is usually taken as 1.5.

[0139] In the standard Cuckoo algorithm, after obtaining the adjusted candidate solution vectors, a portion of the adjusted candidate solution vectors are discarded with a certain probability, and a random walk is used to mutate and generate new adjusted candidate solution vectors, which can be defined by formula (9):

[0140]

[0141] Where, p a This is the detection probability, with a default value of 0.25; X r1 With X r2 These are two candidate solution vectors randomly selected from the first candidate population. Finally, if the new adjusted candidate solution vector U... i Better than adjusting the candidate solution vector X i (k+1), then the new adjusted candidate solution vector U i Will replace X i (k+1); otherwise X i (k+1) remains unchanged.

[0142] It should be noted that, in order to improve the performance of the Cuckoo algorithm in this embodiment, the step size factor is sufficiently large in the early stage of the algorithm iteration to increase the diversity of the candidate solution vector group. However, the step size factor gradually decreases in the later stage of the iteration, so that the candidate solution vectors can be better fine-tuned near the optimal solution vector to find a better solution vector. The value of the step size factor changes dynamically with the number of iterations, and the step size factor α(k) can be defined by formula (10):

[0143]

[0144] Where, α max K max and α min The specific implementation details are determined based on the actual circumstances, and the embodiments in this application are not limited thereto.

[0145] It is understood that the step size factor in this application embodiment is adjusted from large to small to ensure that the addressing method provided in this application embodiment can find a better solution vector.

[0146] S113. Based on the fitness of each candidate solution vector in the first candidate population and the fitness of each adjusted candidate solution vector, adjust the N candidate solution vectors and N adjusted candidate solution vectors in the first candidate population to obtain N new candidate solution vectors.

[0147] In this embodiment, the application is applicable to scenarios that generate new candidate solution vectors.

[0148] In this embodiment of the application, the electronic device sorts the N candidate solution vectors and N adjusted candidate solution vectors in the first candidate population based on the fitness of each candidate solution vector and the fitness of each adjusted candidate solution vector in the first candidate population, and obtains N new candidate solution vectors.

[0149] It should be noted that, in this embodiment of the application, after calculating the fitness of each candidate solution vector and the fitness of each adjusted candidate solution vector in the first candidate population, the N candidate solution vectors and the N adjusted candidate solution vectors in the first candidate population are merged to form a temporary population of size 2N; the fitness of each candidate solution vector in the temporary population is determined according to the fitness of each candidate solution vector in the first candidate population and the fitness of each adjusted candidate solution vector; the candidate solution vectors in the temporary population are sorted according to the fitness of each candidate solution vector in the temporary population to obtain the sorting result; and the optimal N candidate solution vectors are selected from the sorting result as new candidate solution vectors.

[0150] S114. Based on N new candidate solution vectors, a new first candidate population is formed.

[0151] In this embodiment of the application, the scenario of determining a new first candidate population is applicable.

[0152] In this embodiment of the application, the electronic device constructs a new first candidate population based on N new candidate solution vectors.

[0153] It is understood that, in the embodiments of this application, a new first candidate population is formed based on N new candidate solution vectors, so that the location selection method provided in the embodiments of this application can perform the next iteration population update based on the new first candidate population.

[0154] In some embodiments of this application, the addressing method further includes step S115, as follows:

[0155] S115. If the candidate solution vector is not in the first deployment area, discard the candidate solution vector and sum it with a random number to obtain the summation result, and update the candidate solution vector based on the summation result.

[0156] In this embodiment, the proposed solution is applicable to scenarios where the adjusted candidate solution vector is not located within the first deployment area, and the adjusted candidate solution vector is discarded and a new adjusted candidate solution vector is generated.

[0157] In this embodiment of the application, if the candidate solution vector is not in the first deployment area, the electronic device discards the candidate solution vector and sums it with a random number to obtain a summation result, and updates the candidate solution vector based on the summation result.

[0158] For example, if an adjusted candidate solution vector is generated during the iteration process but is not within the specified region, a new adjusted candidate solution vector X is generated. i (k+1) can be defined by formula (11):

[0159] X i (k+1)=X i (k)+rand·(X r1 -X r2 ) formula (11);

[0160] Where rand is a random number in the range [0, 1]; X r1 With X r2 These are two candidate solution vectors randomly selected from the first candidate population; each candidate solution vector in the first candidate population can be represented as X. i (k).

[0161] It should be noted that the first deployment area can be the deployment area corresponding to the object to be located, and this application embodiment does not impose any restrictions.

[0162] Understandably, for adjustment solution vectors generated during the iteration process that are outside the specified area—that is, when the latitude and longitude corresponding to the candidate adjustment solution vectors exceed the boundary of the area to be covered—the candidate solution vectors are definitely not up to standard. Therefore, there's no need to calculate the corresponding fitness; the invalid adjustment solution vectors are directly discarded, and new adjustment solution vectors are generated, thereby improving the optimization speed of the location selection method. Furthermore, in practical applications, there may be areas where it's inconvenient to establish the target location; in such cases, this method can be used to directly exclude candidate solution vectors within the prohibited areas.

[0163] In some embodiments of this application, step S116 is included before step S113, as follows:

[0164] S116. For each adjustment candidate solution vector, determine the fitness of the adjustment candidate solution vector based on the congestion distance, cost, and blind zone loss corresponding to the adjustment candidate solution vector.

[0165] In this embodiment, the application is applicable to scenarios where the fitness of the candidate solution vector is calculated and adjusted.

[0166] In the embodiments of this application, for each adjustment candidate solution vector, the electronic device determines the fitness of the adjustment candidate solution vector based on the congestion distance, cost and blind spot loss corresponding to the adjustment candidate solution vector.

[0167] It should be noted that the method for determining the fitness of the adjusted candidate solution vector based on the crowding distance, cost, and blind zone loss corresponding to the adjusted candidate solution vector in this embodiment is the same as the method for calculating the fitness of each candidate solution vector in the first candidate population.

[0168] In some embodiments of this application, step S113 can be implemented by S117 and S118, which will be described in conjunction with each step.

[0169] S117. Based on the fitness of each candidate solution vector in the first candidate population and the fitness of each adjusted candidate solution vector, sort the N candidate solution vectors and N adjusted candidate solution vectors in the first candidate population, and determine the N candidate solution vectors with the smallest fitness in the sorting results.

[0170] In this embodiment, the application is applicable to the scenario of determining new candidate solution vectors.

[0171] In this embodiment of the application, the electronic device sorts the N candidate solution vectors and N adjustment candidate solution vectors in the first candidate population according to the fitness of each candidate solution vector in the first candidate population and the fitness of each adjustment candidate solution vector, and determines the N candidate solution vectors with the smallest fitness in the sorting result.

[0172] It should be noted that, in this embodiment of the application, after calculating the fitness of each candidate solution vector and the fitness of each adjusted candidate solution vector in the first candidate population, the N candidate solution vectors and the N adjusted candidate solution vectors in the first candidate population are merged to form a temporary population of size 2N; the fitness of each candidate solution vector in the temporary population is determined according to the fitness of each candidate solution vector in the first candidate population and the fitness of each adjusted candidate solution vector; and the candidate solution vectors in the temporary population are sorted in descending or ascending order according to the fitness of each candidate solution vector in the temporary population to obtain the sorting result, and the N candidate solution vectors with the smallest fitness are selected from the sorting result.

[0173] S118. Determine N new candidate solution vectors from the N candidate solution vectors with the smallest fitness.

[0174] In this embodiment, the application is applicable to the scenario of determining new candidate solution vectors.

[0175] In this embodiment of the application, the electronic device determines N new candidate solution vectors from the N candidate solution vectors with the smallest fitness.

[0176] The following will describe an exemplary application of the embodiments of this application in a real-world application scenario, wherein the object to be located is a base station to be located.

[0177] 5G is the latest generation of broadband mobile communication technology, characterized by high speed, low latency, and massive connectivity, serving as the network infrastructure for realizing the interconnection of humans, machines, and things. Because 5G communication uses millimeter waves with shorter wavelengths, it is generally accepted that the maximum distance between base stations should not exceed 200 meters. Dense network deployment makes the overall network topology more complex, further increasing the difficulty of base station site selection and placing higher demands on base station construction. Site selection is a crucial step in communication base station construction; proper site selection not only ensures the smooth progress of base station construction but also guarantees the security and stability of base station operation.

[0178] Base station site selection optimization is the process of configuring and optimizing the network architecture, frequency, and various parameters. It ensures that the system achieves maximum capacity and good operational quality under certain investment and appropriate interference. In essence, the 5G communication base station site selection problem is a constrained nonlinear optimization problem.

[0179] Base station site selection is typically handled by third-party companies specializing in network planning and optimization services. These companies use software to identify areas with weak signals and, after measurement, select approximate candidate base station locations. Each candidate site is then inspected on-site, and signal strength is predicted, coverage analyzed, and signal strength predicted. Through repeated local adjustments, the final site location that meets the overall network requirements is determined. This method is not only resource-intensive but also lacks accuracy. In recent years, with the rapid development of intelligent optimization decision-making technology, domestic and international scholars have applied biomimetic intelligent optimization algorithms, such as genetic algorithms, immune algorithms, and particle swarm optimization, to the base station site selection process. However, base station site selection based on intelligent optimization algorithms suffers from issues such as overly dense base station distribution and significant overlap in coverage areas, leading to substantial waste of base station resources.

[0180] Based on this, in the embodiments of this application, within the area to be deployed, the electronic device can adjust the first candidate population based on the fitness obtained from the crowding distance to obtain a new first candidate population, and iteratively update the first candidate population according to the new first candidate population until the number of iterations equals the iteration threshold; when the number of iterations equals the iteration threshold, the optimal solution vector in the location problem is determined from the first candidate population.

[0181] The site selection method provided in this application includes the following steps:

[0182] S1. Randomly generate a first candidate population Y of size n in the area where the base station is to be deployed, set the initial iteration number k = 0, and initialize the system parameters.

[0183] S2. Calculate the fitness of each candidate solution vector in the first candidate population based on the base station construction cost, blind zone loss, and congestion distance of the base station.

[0184] S3. Adjust the first candidate population based on the fitness of each candidate solution vector in the first candidate population to obtain a new first candidate population of size n.

[0185] S4. If the number of adjustments to the first candidate population is equal to the adjustment threshold, stop the iteration process and record the optimal candidate solution vector; otherwise, set the iteration number k = k + 1, update the first candidate population to a new first candidate population, and return to S2 to continue the iteration.

[0186] For S1, a first candidate population Y of size n is randomly generated within the area of ​​the base station to be deployed. The initial iteration number k = 0 is set, and the system parameters are initialized:

[0187] If there are m potential sites within the deployment area, and the parameters of each potential site are longitude, latitude, antenna type, and 4G base station sharing, then a potential site can be represented as a vector x = (x1, x2, x3, x4); where x1 and x2 represent the longitude and latitude of the potential site, respectively; x3 represents the antenna type, and x3 ∈ {0, 1}, where 0 represents a Type I antenna (corresponding to the new 5G base station) and 1 represents a Type II antenna (corresponding to the 4G base station); x4 represents the 4G base station sharing, and x4 ∈ {0, 1}. When a potential site is represented as a vector x = (x1, x2, x3, x4), the set of m potential sites within the deployment area constitutes a solution X to the optimal site selection problem, i.e., a candidate solution vector, which can be represented as follows:

[0188]

[0189] Randomly initialize n X values ​​to form the population to be optimized, i.e., the first candidate population Y, where Y = (X1, X2, ..., Xn). n ).

[0190] The initial iteration count k = 0 is set, and the system parameters are initialized. The system parameters include: the location and coverage radius of existing 4G base stations, the coverage radius of the 5G antenna, the area boundary value, and the weight coefficients of the fitness function. Specifically, the location of existing 4G base stations represents the longitude and latitude of existing 4G base stations within the area to be deployed; the coverage radius represents the signal coverage radius of existing 4G base stations, for example, a coverage radius of 1000m; the coverage radius of the 5G antenna represents the signal coverage radius of the 5G antenna, for example, a coverage radius of 300m; the area boundary value characterizes the range of the area to be deployed, for example, a radius of 20km * 20km corresponds to a longitude range of 110° to 111° and a latitude range of 30° to 31°; the weight coefficients of the fitness function include at least: λ1 and λ2 in the temporary fitness function and ε in the fitness function. For example, λ1 is set to 0.3, and λ2 is set to 0.7 to ensure λ1 + λ2 = 1; ε is set to 0.7 to ensure ε is within the range [0, 1].

[0191] For S2, the fitness of each candidate solution vector in the first candidate population is calculated based on the base station construction cost, blind zone loss, and base station congestion distance:

[0192] Based on the characteristics of 5G communication networks, 5G networks will not quickly and completely replace 4G networks; the current approach is to deploy 5G networks gradually. To reduce the cost of 5G networks, priority is given to utilizing existing 4G base stations. When using 4G base stations, only the cost of the antenna needs to be considered. Therefore, the cost function for 5G base station construction can be expressed as: f1(X) = hC1 + m1E1 + m2E2. Where h is the number of 5G base stations to be built; C1 is the cost of building a new 5G base station; m... i E represents the total number of type i antennas used. i The cost of each type i antenna; the total number of base stations required for the planned area is m = m1 + m2. In this embodiment, the blind zone loss can be determined according to the coverage loss objective function, which can be defined as: Where C2 represents the loss caused by each square kilometer of coverage blind spot; S represents the total area of ​​the deployment area; mπr 2 Let s be the sum of the areas covered by m base stations; ij Let be the area of ​​the overlapping coverage region between base station i and base station j, where i is greater than or equal to 1 and less than or equal to m; j is greater than or equal to 1 and less than or equal to m. Combining the two functions using a weighted approach, we obtain the temporary fitness function to be optimized: f(X) = λ1f1 + λ2f2. Here, λ1 is the weight coefficient of the cost function; λ2 is the weight coefficient of the coverage loss objective function; and λ1 + λ2 = 1. The temporary fitness function is used to determine the temporary fitness of the candidate solution vector.

[0193] In addition, base station x among the m candidate base stations p and base station x q The distance can be expressed in Euclidean distance as: Base station x p The sum of the distances to the M candidate base stations can be expressed as: If base station x p The corresponding distance is represented as The crowding distance of the candidate solution vector X can then be expressed as: Where, x p It is any one of the m objects to be addressed.

[0194] When the temporary utility function of the candidate solution vector X is expressed as f(X) = λ1f1 + λ2f2, the crowding distance is expressed as: The fitness function for the candidate solution vector can then be determined as: ε represents the weight coefficient of the crowding distance, which ranges from [0, 1]. The fitness function values ​​of n candidate solution vectors can be calculated based on this fitness function.

[0195] S3 adjusts the first candidate population based on the fitness of each candidate solution vector in the first candidate population to obtain a new first candidate population of size n, which can be achieved through the following steps:

[0196] S31. For each candidate solution vector in the first candidate population, determine the adjusted candidate solution vector corresponding to the candidate solution vector based on the step size factor and the first adjustment step size.

[0197] For example, each candidate solution vector in the first candidate population can be represented as X. i (k) will adjust the step size factor α and the first adjustment step size. Multiply to obtain the second adjustment step size Then, the candidate solution vector X i (k) and the second adjustment step size Summing yields the adjusted candidate solution vector X corresponding to the candidate solution vector. i (k+1) can be represented as The specific formula can be expressed as: Among them, X best denoted as the optimal candidate solution vector of the first candidate population, μ and v are two random numbers following a normal distribution; β is a coefficient and is usually taken as 1.5.

[0198] In the standard Cuckoo algorithm, after obtaining the adjusted candidate solution vectors, a portion of the adjusted candidate solution vectors are discarded with a certain probability, and a random walk is used to mutate and generate new adjusted candidate solution vectors. The specific formula is as follows:

[0199]

[0200] Where, p a This is the detection probability, with a default value of 0.25; X r1 With X r2 These are two candidate solution vectors randomly selected from the first candidate population. Finally, if the new adjusted candidate solution vector U... i Better than adjusting the candidate solution vector X i (k+1), then the new adjusted candidate solution vector U i Will replace X i (k+1); otherwise X i (k+1) remains unchanged.

[0201] In the standard Cuckoo algorithm, the step size factor is a fixed value and cannot be adjusted according to different stages of the iteration. In this embodiment, to improve the performance of the Cuckoo algorithm, the step size factor should be large enough in the early stages of the iteration to increase the diversity of the candidate solution vector set. However, in the later stages of the iteration, the step size factor should be gradually decreased, so that the candidate solution vectors can be better fine-tuned near the optimal solution vector to find a better solution vector. The value of the step size factor changes dynamically with the number of iterations, as follows: Where, α max K max and α min It will be determined based on the specific circumstances of implementation.

[0202] Furthermore, for any adjusted candidate solution vectors generated during the iteration process that are outside the specified region, a new adjusted candidate solution vector X is generated. i (k+1), the calculation formula can be as follows: X i (k+1)=X i (k)+rand·(X r1 -X r2 ), where rand is a random number in the range [0, 1]; X r1 With X r2 These are two candidate solution vectors randomly selected from the first candidate population.

[0203] S32. For each adjustment candidate solution vector, determine the fitness of the adjustment candidate solution vector based on the congestion distance, cost, and blind zone loss corresponding to the adjustment candidate solution vector.

[0204] It should be noted that, for each adjustment candidate solution vector, the electronic device determines the fitness of the adjustment candidate solution vector based on the corresponding congestion distance, cost, and blind zone loss using the same method as calculating the fitness of each candidate solution vector in the first candidate population, i.e., using a fitness function. Determine the fitness of the candidate solution vectors.

[0205] S33. Based on the fitness of each candidate solution vector in the first candidate population and the fitness of each adjusted candidate solution vector, adjust the n candidate solution vectors and n adjusted candidate solution vectors in the first candidate population to obtain n new candidate solution vectors, and construct a new first candidate population based on the n new candidate solution vectors.

[0206] For example, after calculating the fitness of each candidate solution vector in the first candidate population and the fitness of each adjusted candidate solution vector, the electronic device sets the n candidate solution vectors X1(k), X2(k), ..., X in the first candidate population to... n (k) and n adjusted candidate solution vectors X1(k+1), X2(k+1), ..., X n (k+1) are merged to form a temporary population of size 2n (X1(k), X2(k), ..., X n (k),X1(k+1),X2(k+1),...,X n (k+1)); Based on the fitness of each candidate solution vector in the first candidate population and the fitness of each adjusted candidate solution vector, determine the fitness of each candidate solution vector in the temporary population. Then, sort the candidate solution vectors in the temporary population according to their fitness (either from largest to smallest or smallest to largest) to obtain the sorting result. Select the n candidate solution vectors with the smallest fitness from the sorting result as new candidate solution vectors, and form a new first candidate population from these n new candidate solution vectors. Simultaneously, update the optimal candidate solution vector in the first candidate population by taking the candidate solution vector with the smallest fitness among the n new candidate solution vectors as the new optimal candidate solution vector X. best .

[0207] For step S4, if the number of adjustments to the first candidate population equals the adjustment threshold, then stop the iteration process and record the optimal candidate solution vector; otherwise, set the iteration number k = k + 1, update the first candidate population to the new first candidate population, and return to step S2 to continue the iteration.

[0208] For example, if the number of adjustments to the first candidate population equals the adjustment threshold, then the iteration process stops and the optimal candidate solution vector X is recorded. best If the number of adjustments to the first candidate population is less than the adjustment threshold, then update the iteration count k = k + 1, update the first candidate population to a new first candidate population, and return to step 2 to continue iterating.

[0209] It is understood that when determining the fitness of each candidate solution vector in this application embodiment, the crowding distance of the objects to be located is considered, thereby ensuring that there is not much overlap in the coverage area. This helps to improve the coverage rate of the objects to be located in the entire area, reduce the overlap of the coverage areas of different objects to be located, and reduce the losses caused by the overlap of the coverage areas of different objects to be located. In addition, for the adjusted solution vectors generated during the iteration process that are not within the specified area, the invalid adjusted solution vectors are directly discarded and new adjusted solution vectors are regenerated, thereby improving the optimization speed of the location selection method.

[0210] Based on the location selection method described in the above embodiments, this application also provides a location selection device. Figure 6 A schematic diagram of the structure of an address selection device provided in this application embodiment. Figure One The location selection device 1 includes: a determining unit 10 and an adjusting unit 11; wherein,

[0211] The determining unit 10 is configured to determine a first candidate population corresponding to M objects to be located, the first candidate population including N candidate solution vectors; the candidate solution vectors correspond to candidate addresses of the at least one object to be located; wherein M is a positive integer and N is a positive integer; it is also configured to determine the fitness of each candidate solution vector in the first candidate population based on the congestion distance, cost and blind zone loss corresponding to the candidate solution vector; it is also configured to continue to determine the fitness of each candidate solution vector in the new first candidate population if the number of adjustments to the first candidate population is less than the adjustment threshold; and it is also configured to determine the target solution vector from the first candidate population.

[0212] The adjustment unit 11 is used to adjust the first candidate population based on the fitness of each candidate solution vector in the first candidate population to obtain a new first candidate population; it is also used to adjust the new first candidate population based on the fitness until the number of adjustments to the first candidate population is equal to the adjustment threshold.

[0213] In some embodiments of this application, the determining unit is further configured to determine the total blind zone area based on the total area of ​​the deployment area and the coverage area of ​​the candidate object corresponding to the first candidate population; the deployment area is used to characterize the deployment area corresponding to the candidate object; and is further configured to determine the blind zone loss based on the total blind zone area.

[0214] In some embodiments of this application, the apparatus further includes an addition unit; wherein,

[0215] The determining unit is further configured to take each of the M candidate objects as a first candidate object and determine the distance between the first candidate object and the second candidate object;

[0216] The summing unit is used to add the distances between the first candidate object and each of the second candidate objects to obtain the sum of distances corresponding to the first candidate object; the second candidate object is the candidate object other than the first candidate object among the M candidate objects; it is also used to add the sum of distances corresponding to each of the M candidate objects to obtain the congestion distance.

[0217] In some embodiments of this application, the apparatus further includes a weighted summation unit; wherein,

[0218] The weighted summation unit is used to perform a weighted summation of the cost and the blind spot loss to obtain a temporary fitness; it is also used to perform a weighted summation of the temporary fitness and the reciprocal of the congestion distance to obtain the fitness of the candidate solution vector.

[0219] In some embodiments of this application, the apparatus further includes constituent units; wherein,

[0220] The determining unit is further configured to, for each candidate solution vector in the first candidate population, determine the adjusted candidate solution vector corresponding to the candidate solution vector based on the step size factor and the first adjustment step size;

[0221] The adjustment unit is further configured to adjust N candidate solution vectors and N adjusted candidate solution vectors in the first candidate population based on the fitness of each candidate solution vector in the first candidate population and the fitness of each adjusted candidate solution vector, to obtain N new candidate solution vectors.

[0222] The constitutive unit is used to construct the new first candidate population based on the N new candidate solution vectors.

[0223] In some embodiments of this application, the apparatus further includes a multiplication unit and a summation unit; wherein,

[0224] The multiplication unit is used to multiply the step size factor and the first adjustment step size to obtain the second adjustment step size;

[0225] The summation unit is used to sum the candidate solution vector with the second adjustment step size for each candidate solution vector in the first candidate population to obtain the adjusted candidate solution vector corresponding to the candidate solution vector.

[0226] In some embodiments of this application, the apparatus further includes a discard unit and an update unit; wherein,

[0227] The discard unit is used to discard the adjusted candidate solution vector if the adjusted candidate solution vector is not in the first deployment area.

[0228] The summation unit is also used to sum the candidate solution vector and the random number to obtain a summation result;

[0229] The update unit is used to update the adjusted candidate solution vector based on the summation result.

[0230] In some embodiments of this application, the determining unit is further configured to determine the fitness of each adjusted candidate solution vector based on the congestion distance, cost, and blind zone loss corresponding to the adjusted candidate solution vector.

[0231] In some embodiments of this application, the apparatus further includes a sorting unit; wherein,

[0232] The sorting unit is used to sort the N candidate solution vectors and the N adjusted candidate solution vectors in the first candidate population according to the fitness of each candidate solution vector in the first candidate population and the fitness of each adjusted candidate solution vector.

[0233] The determining unit is further configured to determine the N candidate solution vectors with the smallest fitness in the sorting results; and to determine the N candidate solution vectors with the smallest fitness as the N new candidate solution vectors.

[0234] Based on the location selection method described in the above embodiments, this application also provides a location selection device. Figure 7 A schematic diagram of the structure of an address selection device provided in this application embodiment. Figure Two ,like Figure 7 As shown, the device 7 includes a processor 701 and a memory 702; the memory 702 stores one or more programs executable by the processor, and when one or more programs are executed, the processor 701 executes any of the addressing methods described in the previous embodiments.

[0235] Based on the addressing method of the above embodiments, this application also provides a readable storage medium that stores one or more programs that can be executed by one or more processors. When the programs are executed by the processors, they implement the addressing method as described in the embodiments of this disclosure.

[0236] Those skilled in the art will understand that embodiments of this application can be provided as methods, systems, or computer program products. Therefore, this application can take the form of hardware embodiments, software embodiments, or embodiments combining software and hardware aspects. Furthermore, this application can take the form of a computer program product implemented on one or more computer-usable storage media (including, but not limited to, disk storage and optical storage) containing computer-usable program code.

[0237] This application is described with reference to flowchart illustrations and / or block diagrams of methods, apparatus (systems), and computer program products according to embodiments of this application. It will be understood that each block of the flowchart illustrations and / or block diagrams, and combinations of blocks in the flowchart illustrations and / or block diagrams, can be implemented by computer program instructions. These computer program instructions can be provided to a processor of a general-purpose computer, special-purpose computer, embedded processor, or other programmable data processing apparatus to produce a machine, such that the instructions, which execute via the processor of the computer or other programmable data processing apparatus, generate instructions for implementing the flowchart... Figure One One or more processes and / or boxes Figure One A device that provides the functions specified in one or more boxes.

[0238] These computer program instructions may also be stored in a computer-readable storage medium that can direct a computer or other programmable data processing device to function in a particular manner, such that the instructions stored in the computer-readable storage medium produce an article of manufacture including instruction means, which are implemented in a process Figure One One or more processes and / or boxes Figure One The function specified in one or more boxes.

[0239] These computer program instructions may also be loaded onto a computer or other programmable data processing equipment to cause a series of operational steps to be performed on the computer or other programmable equipment to produce a computer-implemented process, thereby providing instructions that execute on the computer or other programmable equipment for implementing the process. Figure One One or more processes and / or boxes Figure One The steps of the function specified in one or more boxes.

[0240] The above description is merely a preferred embodiment of this application and is not intended to limit the scope of protection of this application.

Claims

1. A site selection method, characterized in that, The method includes: A first candidate population is determined corresponding to M objects to be addressed, the first candidate population includes N candidate solution vectors; the candidate solution vectors correspond to the candidate addresses of the at least one object to be addressed; where M is a positive integer and N is a positive integer; Each of the M candidate sites is designated as a first candidate site. The distance between the first candidate site and the second candidate site is determined. The distances between the first candidate site and each of the second candidate sites are added together to obtain the sum of the distances corresponding to the first candidate site. The second candidate site is any candidate site other than the first candidate site among the M candidate sites. The congestion distance is obtained by summing the distances corresponding to each of the M candidate objects; For each candidate solution vector in the first candidate population, the fitness of the candidate solution vector is determined based on the crowding distance, cost, and blind zone loss corresponding to the candidate solution vector; The first candidate population is adjusted based on the fitness of each candidate solution vector in the first candidate population to obtain a new first candidate population. If the number of adjustments to the first candidate population is less than the adjustment threshold, the fitness of each candidate solution vector in the new first candidate population is determined, and the new first candidate population is adjusted based on the fitness until the number of adjustments to the first candidate population is equal to the adjustment threshold. The target solution vector is determined from the first candidate population.

2. The method according to claim 1, characterized in that, Before determining the fitness of the candidate solution vector based on the crowding distance, cost, and blind zone loss corresponding to the candidate solution vector, the method further includes: The total blind zone area is determined based on the total area of ​​the deployment area and the coverage area of ​​the candidate objects corresponding to the first candidate population; the deployment area is used to characterize the deployment area corresponding to the candidate objects. The blind zone loss is determined based on the total area of ​​the blind zone.

3. The method according to claim 1, characterized in that, The step of determining the fitness of the candidate solution vector based on the congestion distance, cost, and blind zone loss corresponding to the candidate solution vector includes: The temporary fitness is obtained by weighted summation of the cost and the blind spot loss; The fitness of the candidate solution vector is obtained by weighted summation of the temporary fitness and the reciprocal of the crowding distance.

4. The method according to claim 1, characterized in that, The step of adjusting the first candidate population based on the fitness of each candidate solution vector in the first candidate population to obtain a new first candidate population includes: For each candidate solution vector in the first candidate population, the adjusted candidate solution vector corresponding to the candidate solution vector is determined according to the step size factor and the first adjustment step size; Based on the fitness of each candidate solution vector in the first candidate population and the fitness of each adjusted candidate solution vector, the N candidate solution vectors in the first candidate population and the N adjusted candidate solution vectors are adjusted to obtain N new candidate solution vectors. The new first candidate population is formed based on the N new candidate solution vectors.

5. The method according to claim 4, characterized in that, The step of determining the adjusted candidate solution vector corresponding to each candidate solution vector in the first candidate population based on the step size factor and the first adjustment step size includes: The second adjustment step size is obtained by multiplying the step size factor by the first adjustment step size; For each candidate solution vector in the first candidate population, the candidate solution vector is summed with the second adjustment step size to obtain the adjusted candidate solution vector corresponding to the candidate solution vector.

6. The method according to claim 5, characterized in that, The method further includes: If the adjusted candidate solution vector is not within the first deployment area, discard the adjusted candidate solution vector and sum it with a random number to obtain a summation result, and update the adjusted candidate solution vector based on the summation result.

7. The method according to claim 4, characterized in that, Before adjusting the N candidate solution vectors and the N adjusted candidate solution vectors in the first candidate population based on the fitness of each candidate solution vector in the first candidate population and the fitness of each adjusted candidate solution vector to obtain N new candidate solution vectors, the method further includes: For each adjusted candidate solution vector, the fitness of the adjusted candidate solution vector is determined based on the congestion distance, cost, and blind zone loss corresponding to the adjusted candidate solution vector.

8. The method according to claim 4, characterized in that, The step of adjusting N candidate solution vectors in the first candidate population and N adjusted candidate solution vectors based on the fitness of each candidate solution vector in the first candidate population and the fitness of each adjusted candidate solution vector, to obtain N new candidate solution vectors, includes: Based on the fitness of each candidate solution vector in the first candidate population and the fitness of each adjusted candidate solution vector, the N candidate solution vectors and the N adjusted candidate solution vectors in the first candidate population are sorted, and the N candidate solution vectors with the smallest fitness in the sorting results are determined. The N candidate solution vectors with the smallest fitness are determined as the N new candidate solution vectors.

9. A location selection device, characterized in that, include: The unit consists of a determination unit, an addition unit, and an adjustment unit; among which, The determining unit is configured to determine a first candidate population corresponding to M candidate objects, the first candidate population including N candidate solution vectors; the candidate solution vectors correspond to candidate addresses of the at least one candidate object; wherein M is a positive integer and N is a positive integer; it is also configured to determine the fitness of each candidate solution vector in the first candidate population based on the congestion distance, cost, and blind zone loss corresponding to the candidate solution vector; it is also configured to continue to determine the fitness of each candidate solution vector in a new first candidate population if the number of adjustments to the first candidate population is less than the adjustment threshold; it is also configured to determine a target solution vector from the first candidate population; and it is also configured to take each of the M candidate objects as a first candidate object and determine the distance between the first candidate object and a second candidate object. The summation unit is used to add the distances between the first candidate object and each of the second candidate objects to obtain the sum of distances corresponding to the first candidate object; the second candidate objects are candidate objects other than the first candidate object among the M candidate objects; the sum of distances corresponding to each of the M candidate objects is added to obtain the congestion distance; The adjustment unit is configured to adjust the first candidate population based on the fitness of each candidate solution vector in the first candidate population to obtain a new first candidate population; and to adjust the new first candidate population based on the fitness until the number of adjustments to the first candidate population is equal to the adjustment threshold.

10. An addressing device, characterized in that, include: Memory, used to store executable data instructions; A processor for executing executable instructions stored in the memory to implement the addressing method as described in any one of claims 1 to 9.

11. A readable storage medium, characterized in that, It stores executable instructions for causing the processor to execute, thereby implementing the addressing method according to any one of claims 1 to 8.

Images