Station placement design method, station placement design program, station placement design device, and wireless communication system construction method

Abstract

Provided are a station placement design method, a station placement design program, a station placement design device, and a wireless communication system construction method that enable station placement design satisfying target quality with a smaller amount of calculation. In the station placement design method, a population including a plurality of individuals is generated as an initial population. Subsequently, a generation change for updating the plurality of individuals of the population using a method of a genetic algorithm is performed on the basis of the evaluation using an objective function for each individual of the population. Each individual has genetic information representing whether or not a wireless base station 4 is installed at each installation candidate point 13. Subsequently, by selecting an individual from the population after the generation change has been performed a plurality of times, a station placement design pattern corresponding to the individual is obtained. The objective function is calculated on the basis of the wireless communication quality of a wireless terminal 2 when the wireless base station 4 is installed at the installation candidate point 13. The initial population includes individuals generated so as to have genetic information obtained by a local preliminary search method, and individuals in which the genetic information is randomly set.

Description

Site design method, site design program, site design device, and method for constructing a wireless communication system

[0001] This disclosure relates to a site design method, a site design program, a site design device, and a method for constructing a wireless communication system.

[0002] There is a limit to the range of radio waves that can reach from a wireless base station. Furthermore, there is a limit to the number of wireless terminals that a single wireless base station can accommodate. Therefore, if the number of wireless base stations is insufficient, problems arise such as inadequate coverage and insufficient capacity for wireless terminals. On the other hand, if the number of wireless base stations is excessive, it becomes inefficient due to increased costs for the base station equipment itself. For these reasons, when determining the placement of wireless base stations, it is necessary to place a sufficient number of base stations in appropriate locations, and various base station placement design methods, such as those disclosed in Non-Patent Document 1, are being considered.

[0003] Furthermore, site design is a combinatorial optimization problem, and various methods have been proposed for solving general combinatorial optimization problems. For example, the evolutionary algorithm shown in Non-Patent Document 2 is proposed as a population-based metaheuristic optimization algorithm inspired by evolutionary mechanisms such as reproduction, mutation, genetic modification, natural selection, and survival of the fittest.

[0004] Toshiro Nakahira, Daisuke Murayama, Satoshi Takaya, Kenichi Kawamura, Takayoshi Moriyama, "Multi-Wireless Area Design Method Based on Communication Capacity and Base Station Costs," IEICE Technical Review, IEICE General Convention, B-5-97, Mar. 2022. Shota Yagami, Sho Kuwajima, "Evolutionary Algorithms," [online], April 2014, [Retrieved September 6, 2024], Internet <URL: http: / / mikilab.doshisha.ac.jp / dia / monthly / monthly2014 / mlm152 / syagami / syagami.pdf>

[0005] In genetic algorithms such as those disclosed in Non-Patent Document 2, the search is generally performed by proceeding through generations from a randomly generated initial population. However, if the individuals included in the random initial population differ significantly from the optimal solution, many generations may be required to reach the optimal solution, resulting in a large amount of computational power being needed for convergence.

[0006] This disclosure relates to solving these problems. This disclosure provides a site design method, a site design program, and a site design device that enable site design that meets target quality with less computational effort, as well as a method for constructing a wireless communication system using the same.

[0007] The station placement design method relating to this disclosure involves a computer setting conditions for a design area, including the placement of one or more wireless terminals and one or more candidate installation points; generating a population containing multiple individuals as an initial population such that each of the multiple individuals has genetic information that encodes a station placement design pattern by representing the placement or non-placement of a wireless base station at each of the one or more candidate installation points in a binary manner; performing generational changes to update the multiple individuals in the population using a genetic algorithm method based on an evaluation using an objective function for each of the multiple individuals in the population; and after performing the generational changes multiple times, the multiple individuals in the population A location design method that performs the following: obtaining a location design pattern represented by the genetic information of individuals selected from a plurality of individuals based on pre-set criteria, wherein the objective function is calculated based on the wireless communication quality of one or more wireless terminals when a wireless base station is installed at each of the installation candidate points that have been designated as installation locations according to the location design pattern, and the initial population includes one or more individuals generated such that each has the genetic information representing the location design pattern obtained by a local preliminary search method, and one or more individuals whose genetic information is randomly set.

[0008] The station placement design program described herein involves: setting conditions for a design area on a computer, including the placement of one or more wireless terminals and one or more candidate installation points; generating a population containing multiple individuals as an initial population such that each of the multiple individuals has genetic information that encodes a station placement design pattern by representing the placement or non-placement of a wireless base station at each of the one or more candidate installation points in a binary manner; performing generational changes to update the multiple individuals in the population using a genetic algorithm method based on an evaluation using an objective function for each of the multiple individuals in the population; and after performing the generational changes multiple times, updating the multiple individuals in the population. A location design program that performs the following: obtains a location design pattern represented by the genetic information of individuals selected from a number of individuals based on pre-set criteria, wherein the objective function is calculated based on the wireless communication quality of one or more wireless terminals when a wireless base station is installed at each of the installation candidate points that have been selected for installation according to the location design pattern, and the initial population includes one or more individuals generated such that each has the genetic information representing the location design pattern obtained by a local preliminary search method, and one or more individuals whose genetic information is randomly set.

[0009] The station placement design device according to this disclosure includes: a setting unit that sets conditions for the area to be designed, including the placement of one or more wireless terminals and the placement of one or more candidate installation points; a generation unit that generates a population including multiple individuals as an initial population such that each of the multiple individuals has genetic information in which a station placement design pattern is encoded by representing the placement or non-placement of a wireless base station at each of the one or more candidate installation points in a binary manner; an update unit that performs generational changes to update the multiple individuals included in the population using a genetic algorithm method based on an evaluation using an objective function for each of the multiple individuals included in the population; and after performing the generational changes multiple times, the population included The system includes a selection unit that obtains a site placement design pattern represented by the genetic information of individuals selected from the plurality of individuals based on pre-set criteria, wherein the objective function is calculated based on the wireless communication quality of one or more wireless terminals when a wireless base station is installed at each of the installation candidate points that have been selected for installation according to the site placement design pattern, and the initial population includes one or more individuals generated such that each has the genetic information representing the site placement design pattern obtained by a local preliminary search method, and one or more individuals whose genetic information is randomly set.

[0010] The method for constructing a wireless communication system relating to this disclosure includes obtaining a site design pattern by having a computer execute the above-described site design method, and installing a wireless base station in the area to be designed at a location corresponding to one or more candidate installation points that have been designated as installation points according to the obtained site design pattern.

[0011] According to the site design method, site design program, or site design device, or method for constructing a wireless communication system using the same, site design that meets target quality can be performed with less computation.

[0012] This is a configuration diagram of a wireless communication system according to Embodiment 1. This is a configuration diagram of a site design device according to Embodiment 1. This is a diagram illustrating an example of site design for a wireless communication system according to Embodiment 1. This is a block diagram showing an example of the functions of the site design device according to Embodiment 1. This is a flowchart showing an example of the operation of the site design device according to Embodiment 1. This is a flowchart showing an example of the operation of the site design device according to Embodiment 1. This is a flowchart showing an example of the operation of the site design device according to Embodiment 1. This is a flowchart showing an example of the operation of the site design device according to Embodiment 1.

[0013] Embodiments of this disclosure will be described with reference to the attached drawings. In each drawing, the same or corresponding parts are denoted by the same reference numerals, and redundant explanations are simplified or omitted as appropriate. This disclosure is not limited to the following embodiments, and any combination of embodiments, any modification of any component of each embodiment, or any omission of any component of each embodiment is permitted without departing from the spirit of this disclosure.

[0014] Embodiment 1. Figure 1 is a configuration diagram of the wireless communication system 1 according to Embodiment 1.

[0015] The wireless communication system 1 is a system that provides wireless communication functionality to one or more wireless terminals 2 in the applicable area. Each wireless terminal 2 is, for example, an information processing terminal device capable of wireless communication. Each wireless terminal 2 may be, for example, a general-purpose information processing device such as a tablet computer, smartphone, or smartwatch, or it may be a vehicle, robot, mobility device, or other device equipped with a wireless communication module. The wireless terminal 2 may be a device that moves around the target area, or it may be a device that is fixedly installed in the target area.

[0016] In the area where the wireless communication system 1 is applied, there is one or more shielding objects 3. In this example, there are multiple shielding objects 3 in the area. The shielding objects 3 are objects that block the wireless signals from the wireless communication system 1. The shielding objects 3 may be fixed objects such as buildings, or moving objects such as vehicles.

[0017] The wireless communication system 1 comprises one or more wireless base stations 4. In this example, the wireless communication system 1 comprises a plurality of wireless base stations 4. Each wireless base station 4 is located in an area to which the wireless communication system 1 is applied. Each wireless base station 4 is equipped with the function of transmitting and receiving wireless signals of the wireless communication system 1. The wireless communication system 1 may comprise a plurality of wireless base stations 4 that correspond to different wireless communication methods. Each wireless base station 4 is configured to communicate wirelessly or wired with equipment inside or outside the wireless communication system 1. Each wireless base station 4 provides wireless communication functionality to a wireless terminal 2 by transmitting and receiving wireless signals to and from the wireless terminal 2.

[0018] The wireless communication system 1 is constructed by actually installing one or more wireless base stations 4 in locations determined by a pre-determined site placement design within the target area. Here, the process of constructing the wireless communication system 1 so that it can operate is sometimes referred to as manufacturing the wireless communication system 1. The site placement design for the wireless communication system 1 is performed, for example, using a site placement design device 5 not shown in Figure 1.

[0019] Figure 2 is a diagram showing the configuration of the site location design device 5 according to Embodiment 1.

[0020] The site design device 5 is, for example, a computer system consisting of one or more server devices, or a device including such a system. Here, a computer system consisting of one or more devices may be simply referred to as a computer. When the site design device 5 is composed of multiple server devices, these multiple server devices may be located in different locations. Some or all of the functions of the site design device 5 may be implemented, for example, by a virtual machine on a cloud service, or by processing or storage resources on a cloud service. The site design device 5 includes a communication unit 6, an external input / output unit 7, a processing unit 8, and a database 9.

[0021] The communication unit 6 communicates with the input / output device 10 located outside the site design device 5 via wired or wireless connection. The input / output device 10 accepts information input from the operator and displays the input information and the information output from the site design device 5 on a display.

[0022] The external input / output unit 7 is an interface between the communication unit 6 and the processing unit 8, and the communication unit 6 inputs information received from the input / output device 10 to the processing unit 8. The external input / output unit 7 also outputs information obtained from calculation processing by the processing unit 8 to the communication unit 6 and transmits it to the input / output device 10 via the communication unit 6.

[0023] The processing unit 8 comprises a processor 11 and a memory 12. The processor 11 is typically a CPU. The memory 12 stores programs that can be executed by the processor 11. The processor 11 is, for example, a CPU, arithmetic unit, microprocessor, or microcomputer. The memory 12 is, for example, a non-volatile or volatile semiconductor memory such as RAM, ROM, flash memory, EPROM, and EEPROM, or a magnetic disk, flexible disk, optical disk, compact disk, minidisc, or DVD. The memory 12 stores programs, for example, as software or firmware. The site design device 5 then performs pre-set processing by having the processor 11 execute the programs stored in the memory 12, and realizes each function as a result of the cooperation between hardware and software. Programs can be stored on computer-readable storage media. Programs can also be provided via a communication network. A program may be a program package that includes multiple subprograms, modules, or libraries. A program is sometimes called a program product. Each function of the site design device 5 may be implemented by a processing circuit. Alternatively, some or all of the functions of the site design device 5 may be implemented together by a processing circuit. Furthermore, the processing circuit may be implemented as, for example, a single circuit, a composite circuit, a programmed processor, a parallel programmed processor, an ASIC, or an FPGA, or a combination thereof.

[0024] Database 9 stores information such as calculation conditions or calculation results in site design. Information such as calculation conditions stored in database 9 can be read from database 9 by the processing unit 8 during site design. Various types of information, such as calculation results in site design, can be written to database 9 by the processing unit 8. Various types of information in site design can be written to database 9 by the operator operating the input / output device 10. Information stored in database 9 can be read from database 9 by the operator operating the input / output device 10. Information read from database 9 can be displayed on the input / output device 10 or transmitted externally via a communication network.

[0025] Figure 3 illustrates an example of a site placement design for a wireless communication system 1 according to Embodiment 1.

[0026] In site placement design, for example, the area to which the wireless communication system 1 is applied is designated as the design area, and the number and placement of wireless base stations 4 within the area are determined. The wireless base stations 4 are installed in part or all of one or more candidate installation points 13 located within the area. The candidate installation points 13 are set to locations where wireless base stations 4 can be installed, depending on the area. The candidate installation points 13 are identified, for example, by their location within the area. In site placement design, for example, a site placement design pattern is determined that indicates, among the candidate installation points 13 within the area, at which wireless base stations 4 are installed and at which candidate installation points 13 are not.

[0027] In site placement design, for example, the communication quality when wireless base stations 4 are installed at some or all of the candidate installation points 13 is evaluated. The evaluation of communication quality is performed by simulation of a wireless communication system 1 virtually constructed on the site placement design device 5 for the area under design. At this time, one or more wireless terminals 2 are placed in the virtual area. One or more wireless terminals 2 may be placed, for example, on a grid within the area, randomly, based on previously observed placements, or by other methods.

[0028] Figure 4 is a block diagram showing an example of the functions of the site location design device 5 according to Embodiment 1.

[0029] In this example, the site placement design device 5 performs site placement design for the wireless communication system 1 by solving a combinatorial optimization problem using a genetic algorithm. The site placement design device 5 comprises a setting unit 14, a generation unit 15, an update unit 16, and a selection unit 17 as functional units. The functions of each functional unit of the site placement design device 5 are realized, for example, as calculation processing in the processing unit 8.

[0030] The setting unit 14 is a part equipped with a function for setting site design conditions for the area to be designed. Setting site design conditions includes setting the size of a virtual area on the site design device 5, the arrangement of shielding objects 3 within the area, and other calculation conditions. The virtual area on the site design device 5 is set, for example, based on a model of the actual area to be designed. Setting site design conditions includes placing one or more wireless terminals 2 in the virtual area on the site design device 5. Setting site design conditions also includes placing one or more candidate installation points 13 in the virtual area on the site design device 5. The setting unit 14 sets site design conditions, for example, based on information input by the operator to the input / output device 10. The operator inputs, for example, the arrangement of one or more wireless terminals 2 and the arrangement of one or more candidate installation points 13 within the area to the site design device 5 via the input / output device 10.

[0031] The generation unit 15 is a part equipped with the function of generating a population containing multiple individuals as an initial population as a processing of a genetic algorithm. Each individual included in the population has genetic information that represents the corresponding base station design pattern. The genetic information is information that encodes the base station design pattern. In the genetic information of this example, the base station design pattern is encoded by representing the installation or non-installation of the wireless base station 4 at each candidate installation point 13 with a binary value such as "1" or "0". For example, for numbered candidate installation points 13, the genetic information that encodes a base station design pattern in which a wireless base station 4 is installed at the first candidate installation point 13, not installed at the second candidate installation point 13, and a wireless base station 4 is installed at the third candidate installation point 13 is represented as "101...".

[0032] The generation unit 15 generates multiple individuals with genetic information corresponding to the location design conditions set by the setting unit 14, for example, as follows, and the population containing the multiple generated individuals becomes the initial population.

[0033] The generation unit 15 calculates one or more base station design patterns using a local preliminary search method. The preliminary search method is a method that requires less computation than a genetic algorithm. The preliminary search method includes a greedy method. The preliminary search method includes a search method that makes the evaluation function local, and a search method that makes the search range of the solution local. The preliminary search method includes, for example, a method based on a greedy method as disclosed in Non-Patent Literature 1. Here, in base station design, a base station design pattern is searched for, i.e., whether or not each radio base station 4 is necessary to be installed, so that the overall wireless communication quality of the area to be designed is high relative to the number of radio base stations 4 installed or the equipment cost of the entire area to be designed. In contrast, the greedy method searches for a single radio base station in such a way that the wireless communication quality per unit or per equipment cost is high, i.e., the radio base stations 4 are installed in order from the radio base station 4 that can accommodate the most radio terminals 2. Thus, in the greedy method, the evaluation function is localized as for a single radio base station 4, and the search is performed.

[0034] The generation unit 15 includes an individual having genetic information representing a station placement design pattern obtained by a preliminary search method in the initial population. Also, there may be cases where a plurality of station placement design patterns are obtained by the preliminary search method. In this case, the generation unit 15 may, for example, select a predetermined number of station placement design patterns in the order of good evaluation values in the preliminary search method, and include one or more individuals having genetic information representing the selected station placement design patterns in the initial population. Also, the generation unit 15 may, for example, randomly select one or more patterns from a predetermined number of station placement design patterns in the order of good evaluation values in the preliminary search method, and include an individual having genetic information representing the selected station placement design patterns in the initial population.

[0035] Also, the generation unit 15 may generate a station placement design pattern in which a predetermined number of installation candidate points 13 randomly selected in one or more station placement design patterns obtained by the preliminary search method are additionally in the installed state. The generation unit 15 may include one or more individuals having genetic information representing the station placement design patterns generated in this way in the initial population.

[0036] The generation unit 15, for example, generates a plurality of individuals having random genetic information according to the condition setting of the area design set by the setting unit 14, and includes the one or more generated individuals in the initial population. At this time, the generation unit 15 may uniformly and randomly set each bit representing the genetic information. Also, the generation unit 15 may randomly set each bit representing the genetic information according to the number of installed radio base stations 4 in the station placement design pattern obtained by the preliminary search method. When a plurality of station placement design patterns are obtained by the preliminary search method, the generation unit 15 may randomly set each bit representing the genetic information according to, for example, the average value of the number of installed radio base stations 4. The generation unit 15 may uniformly and randomly set each bit representing the genetic information with a probability corresponding to the ratio of the number of installed radio base stations 4 to the total number of installation candidate points 13 in the station placement design pattern obtained by the preliminary search method. The generation unit 15 adds an individual having genetic information representing the station placement design pattern obtained by the preliminary search method, etc. to the initial population, and repeatedly generates individuals having random genetic information until the number of individuals included in the initial population reaches the required number, thereby forming the initial population.

[0037] As a result, the initial population includes one or more individuals having genetic information representing a placement design pattern obtained by a local preliminary search method, and one or more individuals having randomly set genetic information.

[0038] The update unit 16 is a part equipped with a function of performing generation replacement processing on a population including a plurality of individuals such as the initial population generated by the generation unit 15. The generation replacement processing is a process of updating a plurality of individuals included in the population using the method of a genetic algorithm. The generation replacement processing is performed based on an evaluation using a preset objective function for each individual included in the current generation population. The objective function is calculated based on, for example, an index representing the performance of wireless communication such as a coverage rate. The generation replacement processing includes, for example, processes such as elitism preservation, tournament selection, uniform crossover, and mutation.

[0039] Elitism preservation is a process of directly inheriting an individual with a good evaluation value by the objective function from the current generation population to the next generation population. Here, the evaluation value by the objective function is a better value when it is larger in the case of an optimization problem of maximizing the objective function, and a better value when it is smaller in the case of an optimization problem of minimizing the objective function. In elitism preservation, for example, one or more individuals are inherited to the next generation population in the order of good evaluation values by the objective function. As a result, the individual with the best evaluation value in each generation, that is, the evaluation value of the optimal solution, is updated in a better direction.

[0040] Tournament selection is a process of selecting an individual with a good evaluation value by the objective function from among a group of individuals randomly selected from the current generation population. Uniform crossover is a process of randomly swapping each element represented by the binary value of genetic information for two individuals selected by tournament selection or the like from the current generation population and inheriting it to the next generation population. Mutation is a process of randomly converting genetic information, for example, by bit flipping or swapping of a bit string, for an individual selected from the current generation population or an individual that has undergone a process such as uniform crossover, and inheriting it to the next generation population. As a result, the genetic information of the next generation population diversifies.

[0041] The update unit 16 performs elite preservation and then repeatedly performs processes such as tournament selection, uniform crossover, and mutation until the required number of individuals are included in the next generation population, thereby forming the next generation population. The update unit 16 may also perform generational changes using other genetic algorithms, such as roulette selection, ranking selection, two-point crossover, or other processes. The update unit 16 then repeatedly performs generational changes on the updated population. For example, the update unit 16 performs generational changes for a predetermined number of generations.

[0042] The selection unit 17 is a part equipped with the function of selecting one of several individuals included in the latest generation of the population after multiple generations have been performed on the population. The selection unit 17 selects one individual from the population based on, for example, a predetermined criterion. The selection unit 17 selects, for example, the individual with the best evaluation value according to the objective function from among the individuals included in the latest generation of the population. The selection unit 17 obtains a localization design pattern represented by the genetic information of the selected individual as the result of localization design.

[0043] The operator obtains and refers to the site design results obtained by the selection unit 17 from the site design device 5 via the input / output device 10.

[0044] Next, the objective function used in the site placement design device 5 will be explained. The objective function is a function that returns an evaluation value for each individual that possesses genetic information. The objective function is calculated using the wireless communication quality of the wireless terminal 2 when the wireless base station 4 is installed at the candidate installation point 13 based on the site placement design pattern represented by the genetic information of the target individual. The objective function is also calculated using a penalty corresponding to the number of wireless base stations 4 installed in the site placement design pattern. The objective function can be expressed, for example, by the following equation (1).

[0045]

[0046] Here, the first term of Equation (1) represents the wireless communication quality, and the second term represents the penalty according to the number of installed wireless base stations 4. The coefficient α and the coefficient μ are positive weights that determine the strength of the influence of the wireless communication quality and the penalty on the evaluation value of the objective function. The subscript j, which is an element of the set J, corresponds to each wireless terminal 2. The number of elements |J| of the set J is the number of wireless terminals 2 arranged in the area to be designed in the condition setting of the setting unit 14. The subscript n, which is an element of the set N, corresponds to each station placement design pattern. The number of elements of the set N is the total number of possible station placement design patterns for one or more installation candidate points 13 arranged in the area to be designed in the condition setting of the setting unit 14. The goal achievement degree z ｊ，ｎ is a value indicating whether the maximum reception power at the j-th wireless terminal 2 achieves the target value in the station placement design pattern n. The goal achievement degree z ｊ，ｎ is represented, for example, by the following Equation (2).

[0047]

[0048] The maximum reception power p ｊ，ｎ represents the maximum reception power at the j-th wireless terminal 2 in the station placement design pattern n. The target reception power p ｔ１ is the target value of the reception strength of the wireless signal from the wireless base station 4 at each wireless terminal 2. The target reception power p ｔ１ is set, for example, so that wireless communication at the wireless terminal 2 can be achieved with sufficient quality. The maximum reception power p ｊ，ｎ is the maximum value over the wireless base stations 4 of the reception power r ｊ，ｋ . Here, the set K ｎ The subscript k, which is an element of, corresponds to each wireless base station 4 arranged in the area to be designed in the station placement design pattern n. The set K ｎ The number of elements |K ｎ | is the number of wireless base stations 4 arranged in the area to be designed in the station placement design pattern n. The reception power r ｊ，ｋ is the reception strength at the j-th wireless terminal 2 of the wireless signal transmitted from the k-th wireless base station 4. The reception power r ｊ，ｋThis is calculated, for example, based on the signal strength of the radio signal transmitted by the radio base station 4, the distance between the radio base station 4 and the radio terminal 2, and the presence or absence of an obstruction 3 between the radio base station 4 and the radio terminal 2. Received power r ｊ，ｋ This may be calculated by considering, for example, the reflection of radio signals between the radio base station 4 and the radio terminal 2. The degree of target achievement z for the j-th radio terminal 2. ｊ，ｎ The value of is the maximum received power p ｊ，ｎ Target received power p ｔ１ If this condition is met, the result is 1. Therefore, the first term of the objective function represented by equation (1) is the value obtained by multiplying the ratio of wireless terminals 2, of which one or more wireless terminals 2 placed in the area to be designed in the setting unit 14 have achieved the target value for maximum received power in the station placement design pattern n, by the coefficient α. In other words, if this value is sufficiently large, it can be said that the wireless communication quality in the station placement design pattern n is good.

[0049] The penalty P is, for example, the number of wireless base stations 4 installed when following the site placement design pattern n | K ｎ Using |, it can be expressed by the following equation (3).

[0050]

[0051] Thus, the evaluation value of the objective function represented by equation (1) is obtained by subtracting from the first term, which represents the wireless communication quality in the site placement design pattern n, the value obtained by multiplying the penalty P corresponding to the number of wireless base stations 4 by the coefficient μ. For this reason, the site placement design of the wireless communication system 1 becomes a problem of maximizing the objective function represented by the following equation (4), under the conditions of equations (2) and (3).

[0052]

[0053] Furthermore, the penalty P in the objective function does not have to be simply the number of wireless base stations 4 installed. For example, considering the equipment costs of the wireless base stations 4 themselves, which differ depending on the model of wireless base station 4 to be installed at the candidate installation point 13, the penalty may be the number of wireless base stations 4 installed weighted according to the equipment cost, or the total value of the equipment costs.

[0054] Next, we will explain an example of the operation of the site location design device 5 using Figures 5 to 7. Figures 5 to 7 are flowcharts illustrating an example of the operation of the site location design device 5 according to Embodiment 1.

[0055] Figure 5 shows an example of the overall processing in the site design device 5 during site design.

[0056] In step S1, the setting unit 14 places the wireless terminal 2 and candidate installation points 13 in the area to be designed, based on information input by the operator to the input / output device 10, as a condition setting for site placement design. After that, the site placement design device 5 proceeds to step S2.

[0057] In step S2, the setting unit 14 calculates the received power at each wireless terminal 2 for the wireless signals transmitted by each wireless base station 4, based on the arrangement of wireless terminals 2 and candidate installation points 13 in the area to be designed, as well as the arrangement of shielding objects 3. After that, the site design device 5 proceeds to step S3.

[0058] In step S3, the site design device 5 searches for a site design pattern that optimizes the objective function using a genetic algorithm. In this example, the site design device 5 searches for a site design pattern that maximizes the objective function. After storing the searched site design pattern as a design result in a database 9 or the like, the processing of the site design device 5 is completed.

[0059] Figure 6 shows an example of the processing performed by the site design device 5 when searching for site design patterns using a genetic algorithm.

[0060] In step S31, the generation unit 15 generates an initial population that includes at least one or more individuals having genetic information representing a location design pattern obtained by a local preliminary search method, and one or more individuals having randomly set genetic information. After that, the location design device 5 proceeds to step S32.

[0061] In step S32, the update unit 16 calculates an evaluation value using the objective function for each individual included in the group. After that, the site design device 5 proceeds to step S33.

[0062] In step S33, the update unit 16 performs generational change processing using the calculated evaluation values. In this example, as part of the generational change processing, the update unit 16 performs elite preservation processing, passing on individuals with good evaluation values ​​based on the objective function to the next generation. The update unit 16 also performs processing to pass on individuals that have undergone tournament selection, uniform crossover, and mutation in sequence to the next generation as part of the generational change processing. After the required number of individuals are included in the next generation's population through the generational change processing, the processing of the location design device 5 proceeds to step S34.

[0063] In step S34, the update unit 16 determines whether the current group's generation count has reached a predetermined end generation count. If the end generation count has not been reached, the site design device 5 proceeds to step S32. On the other hand, if the end generation count has been reached, the site design device 5 proceeds to step S35.

[0064] In step S35, the selection unit 17 calculates an evaluation value using the objective function for each individual included in the latest generation of the population. The selection unit 17 selects the individual with the best evaluation value using the objective function from among the multiple individuals included in the population. In this example, the selection unit 17 selects the individual with the largest evaluation value using the objective function. The selection unit 17 obtains a location design pattern represented by the genetic information of the selected individual as the result of location design. After that, the location design device 5's processing for searching for a location design pattern is completed.

[0065] Figure 7 shows an example of the processing for the preliminary search method performed by the generation unit 15 when generating the initial population. In this example, the generation unit 15 uses a greedy algorithm as the preliminary search method. The generation unit 15 may, for example, obtain one or more location design patterns by performing the processing shown in Figure 7 one or more times using the preliminary search method.

[0066] In step S311, the generation unit 15 performs an initialization process. In the initialization process, the generation unit 15 sets the status of all installation candidate points 13 to "not installed". Also, in the initialization process, the generation unit 15 sets the status of all wireless terminals 2 to "not occupied," that is, the wireless base station 4 to which they will connect has not been determined. After that, the site design device 5 proceeds to step S312.

[0067] In step S312, the generation unit 15 determines whether there are unoccupied wireless terminals 2 and whether there are unoccupied installation candidate points 13 corresponding to wireless base stations 4 capable of accommodating wireless terminals 2. Whether or not the wireless base station 4 can accommodate wireless terminals 2 is determined based, for example, on the wireless transmission rate of the wireless terminals 2 per number of wireless terminals 2 accommodated by the wireless base station 4. The wireless transmission rate of the wireless terminals 2 is calculated, for example, based on the received power of the wireless terminals 2. If the wireless transmission rate per number of terminals does not exceed a preset threshold, the generation unit 15 determines that the wireless base station 4 can accommodate wireless terminals 2. On the other hand, if the wireless transmission rate per number of terminals exceeds the threshold, the generation unit 15 determines that the wireless base station 4 cannot accommodate wireless terminals 2. The generation unit 15 may also determine whether or not the wireless base station 4 can accommodate wireless terminals 2 by other methods. If there is an unoccupied wireless terminal 2 and there is a candidate installation site 13 that is not yet installed and corresponds to a wireless base station 4 capable of accommodating the wireless terminal 2, the site placement design device 5 proceeds to step S313. On the other hand, if there is no unoccupied wireless terminal 2, or if there is no candidate installation site 13 that is not yet installed and corresponds to a wireless base station 4 capable of accommodating the wireless terminal 2, the site placement design device 5 terminates its processing for the preliminary search method.

[0068] In step S313, the generation unit 15 selects an installation candidate point 13 from among the uninstalled installation candidate points 13 that corresponds to a wireless base station 4 with a large number of wireless terminals 2 that can be accommodated per unit. The generation unit 15 may also select an installation candidate point 13 from among the uninstalled installation candidate points 13 that corresponds to a wireless base station 4 with a large number of wireless terminals 2 that can be accommodated per unit or per unit of equipment cost. Alternatively, the generation unit 15 may select an installation candidate point 13 that corresponds to the wireless base station 4 with the largest number of wireless terminals 2 that can be accommodated per unit or per unit of equipment cost, or it may randomly select one wireless base station 4 from a predetermined number of installation candidate points 13 in descending order of the number of wireless terminals 2 that can be accommodated in the corresponding wireless base station 4. When the generation unit 15 randomly selects an installation candidate point 13, it may select an installation candidate point 13 that corresponds to a wireless base station 4 with a larger number of wireless terminals 2 that can be accommodated with a higher probability. The generation unit 15 sets the selected installation candidate point 13 to an installed state. After that, the site design device 5 proceeds to step S314.

[0069] In step S314, the generation unit 15 determines whether there are any unaccommodated wireless terminals 2 that can be accommodated at the wireless base station 4 corresponding to the selected installation candidate point 13. If there are wireless terminals 2 that satisfy the conditions, the site design device 5 proceeds to step S315. On the other hand, if there are no wireless terminals 2 that satisfy the conditions, the site design device 5 proceeds to step S312.

[0070] In step S315, the generation unit 15 selects a wireless terminal 2 with high received power from among the unaccommodated wireless terminals 2 that can be accommodated at the wireless base station 4 corresponding to the selected installation candidate point 13. The generation unit 15 may select the wireless terminal 2 with the highest received power among the wireless terminals 2 that satisfy the conditions, or it may randomly select one wireless terminal 2 from a predetermined number of wireless terminals 2 in descending order of received power. When the generation unit 15 randomly selects a wireless terminal 2, it may select a wireless terminal 2 with higher received power with a higher probability. After that, the processing of the site design device 5 proceeds to step S316.

[0071] In step S316, the generation unit 15 determines whether the selected wireless terminal 2 can be accommodated at the wireless base station 4 corresponding to the selected installation candidate point 13. For example, the generation unit 15 determines that the selected wireless terminal 2 can be accommodated if the value obtained by dividing the wireless transmission rate of the selected wireless terminal 2 by the number of wireless terminals 2 already accommodated at the wireless base station 4 corresponding to the selected installation candidate point 13 plus 1 is equal to or greater than a preset threshold. If the generation unit 15 determines that the selected wireless terminal 2 can be accommodated, the site design device 5 proceeds to step S317. On the other hand, if the generation unit 15 determines that the selected wireless terminal 2 cannot be accommodated, the site design device 5 proceeds to step S312.

[0072] In step S317, the generation unit 15 accommodates the selected wireless terminal 2 at the wireless base station 4 corresponding to the selected candidate installation point 13. After that, the site design device 5 proceeds to step S314.

[0073] As described above, the site placement design method according to Embodiment 1 is performed by a site placement design device 5, which is a computer. The site placement design method includes setting conditions for the area to be designed. The condition setting includes the placement of one or more wireless terminals 2 and the placement of one or more candidate installation points 13. The site placement design method includes generating a population containing multiple individuals as an initial population. Each individual in the population has genetic information in which a site placement design pattern is encoded by representing the installation or non-installation of a wireless base station 4 at each candidate installation point 13 in binary. The site placement design method includes performing generational changes to update multiple individuals in the population using a genetic algorithm method based on an evaluation using an objective function for each individual in the population. After performing generational changes multiple times, the site placement design method includes obtaining a site placement design pattern represented by the genetic information of individuals selected from multiple individuals in the population based on predetermined criteria. The initial population includes at least one or more individuals having genetic information representing a site placement design pattern obtained by a local preliminary search method, and one or more individuals having randomly set genetic information.

[0074] This configuration ensures that the initial population includes individuals with genetic information representing a base station design pattern that is at least close to a local optimum and where the number of base stations 4 installed does not deviate significantly from the optimal solution. This reduces the number of generations required to reach the optimal solution and the computational cost until convergence. Consequently, base station design that satisfies the target quality can be achieved with less computation. In contrast, base station design methods that use only local search, such as greedy algorithms, can find a solution with less computation even if there are many candidate installation points 13, but the solution tends to fall into a local optimum. Also, methods that solve the optimization problem by exhaustive search for candidate installation points 13 have the problem that the computational cost becomes enormous and computationally difficult as the number of candidate installation points 13 increases. In contrast, in the base station design method according to Embodiment 1, the diversity of genetic information in the genetic algorithm makes it easier to obtain a global optimum than in the case of local search. Furthermore, even when the number of candidate installation points 13 is large, the increase in computational cost is suppressed compared to the case of exhaustive search. Furthermore, the wireless communication system 1 is constructed by actually installing wireless base stations 4 in locations specified by the site placement design pattern obtained by the site placement design device 5, etc., within the target area. This makes it possible to obtain the wireless communication system 1, which is designed to meet the target quality, more efficiently.

[0075] Furthermore, the candidate installation points 13 may be identified not only by their location within the area, but also by the wireless communication method of the wireless base station 4 to be installed. For the same location within the area, for example, candidate installation points 13 for a wireless base station 4 using a first wireless communication method and candidate installation points 13 for a wireless base station 4 using a second wireless communication method may be set separately. The first wireless communication method is, for example, a wireless communication method using radio signals in the 5.2 GHz band. The second wireless communication method is, for example, a wireless communication method using radio signals in the 28 GHz band. In addition, the candidate installation points 13 may be identified not only by their location within the area, but also by the type of antenna or the direction of the antenna of the wireless base station 4. For example, for the same location within the area, candidate installation points 13 for a wireless base station 4 with its antenna pointed in the first direction and candidate installation points 13 for a wireless base station 4 with its antenna pointed in the second direction may be set separately.

[0076] Furthermore, while the site design method was explained using an example objective function that evaluates wireless communication quality based on the received power at the wireless terminal 2, other objective functions that evaluate wireless communication quality may also be used in the site design method. The objective function may evaluate wireless communication quality based on indicators such as the signal-to-interference and noise ratio or throughput. Estimates of these indicators are calculated, for example, based on the received power at the wireless terminal 2.

[0077] Furthermore, the generation unit 15 may use methods other than the greedy algorithm as a local preliminary search method. For example, the generation unit 15 may use a preliminary search method that repeatedly selects local neighbors of the selected genetic information that yield a better evaluation value using the objective function, using randomly or appropriately selected genetic information as initial values. In this case, the local neighbors of the genetic information represented by a bit string may be, for example, a bit string whose Hamming distance to the genetic information is shorter than a preset threshold.

[0078] The site design method, site design program, site design device, and construction method relating to this disclosure are applicable to wireless communication systems.

[0079] 1. Wireless communication system, 2. Wireless terminal, 3. Shielding device, 4. Wireless base station, 5. Site design device, 6. Communication unit, 7. External input / output unit, 8. Processing unit, 9. Database, 10. Input / output device, 11. Processor, 12. Memory, 13. Installation candidate points, 14. Setting unit, 15. Generation unit, 16. Update unit, 17. Selection unit

Claims

1. A base station design method comprising: a computer setting conditions for a design area including the placement of one or more wireless terminals and the placement of one or more candidate installation points; generating a population of multiple individuals as an initial population such that each of the multiple individuals has genetic information that encodes a base station design pattern by representing the installation or non-installation of a wireless base station at each of the one or more candidate installation points in a binary manner; performing generational changes to update the multiple individuals in the population using a genetic algorithm method based on an evaluation using an objective function for each of the multiple individuals in the population; and, after performing the generational changes multiple times, obtaining a base station design pattern represented by the genetic information of individuals selected from the multiple individuals in the population based on pre-set criteria, wherein the objective function is calculated based on the wireless communication quality of one or more wireless terminals when a wireless base station is installed at each of the candidate installation points that are selected for installation in the base station design pattern, and the initial population is A location design method comprising: one or more individuals generated such that each possesses the genetic information representing a location design pattern obtained by a local preliminary search method; and one or more individuals whose genetic information is randomly set.

2. A site design program that causes a computer to perform the following: setting conditions for a design area, including the placement of one or more wireless terminals and the placement of one or more candidate installation points; generating a population of multiple individuals as an initial population such that each of the multiple individuals has genetic information that encodes a site design pattern by representing the installation or non-installation of a wireless base station at each of the one or more candidate installation points in a binary manner; performing generational changes to update the multiple individuals in the population using a genetic algorithm method based on an evaluation using an objective function for each of the multiple individuals in the population; and, after performing the generational changes multiple times, obtaining a site design pattern represented by the genetic information of individuals selected from the multiple individuals in the population based on pre-set criteria, wherein the objective function is calculated based on the wireless communication quality of one or more wireless terminals when a wireless base station is installed at each of the candidate installation points that are selected for installation in the site design pattern, and the initial population is A location design program comprising: one or more individuals generated such that each possesses the genetic information representing a location design pattern obtained by a local preliminary search method; and one or more individuals whose genetic information is randomly set.

3. The system comprises: a setting unit that sets conditions for the area to be designed, including the placement of one or more wireless terminals and the placement of one or more candidate installation points; a generation unit that generates a population including multiple individuals as an initial population such that each of the multiple individuals has genetic information that encodes a base station design pattern by representing the installation or non-installation of a wireless base station at each of the one or more candidate installation points in a binary manner; an update unit that performs generational changes to update the multiple individuals included in the population using a genetic algorithm method based on an evaluation using an objective function for each of the multiple individuals included in the population; and a selection unit that, after performing the generational changes multiple times, obtains a base station design pattern represented by the genetic information of individuals selected from the multiple individuals included in the population based on pre-set criteria, wherein the objective function is calculated based on the wireless communication quality of one or more wireless terminals when a wireless base station is installed at each of the candidate installation points that are selected for installation in the base station design pattern, among at least one or more candidate installation points, and the initial population is A location design device comprising: one or more individuals generated such that each individual possesses the genetic information representing a location design pattern obtained by a local preliminary search method; and one or more individuals whose genetic information is randomly set.

4. A method for constructing a wireless communication system, comprising: having a computer execute the site location design method described in claim 1 to obtain a site location design pattern; and installing a wireless base station in the area to be designed at a location corresponding to one or more candidate installation points that have been designated as installation points in the obtained site location design pattern.

Images