Information processing device, information processing system, information processing method, program, and recording medium

Abstract

To suppress an increase in difficulty of optimization calculation related to a connection pattern.SOLUTION: An information processing device (1) comprises: a first acquisition unit (11) which acquires first data including a signal-to-interference-plus-noise ratio in each of a plurality of terminal devices when each terminal device is connected to each of N base stations; a first generation unit (12) which generates second data representing a cost function where variables defining the connection between the i-th terminal device among the plurality of terminal devices and defining connection with each of base stations of M (M<N) among the N base stations serve as arguments by referring to the first data; and a second acquisition unit (13) which acquires optimization results obtained with reference to the second data and related to connection patterns between the plurality of terminal devices and the N base stations.SELECTED DRAWING: Figure 1

Description

[Technical Field]

[0001] The present invention relates to an information processing device, an information processing system, an information processing method, a program, and a recording medium for performing optimization processing. [Background technology]

[0002] In a mobile communication network, a technique for optimizing connections between a plurality of terminal devices and a plurality of base stations is known. As an example, Patent Document 1 discloses a wireless communication method that uses DQN (Deep Q-Network) to determine combinations (connection patterns) of connections between a plurality of wireless base stations and wireless terminals. [Prior art documents] [Patent documents]

[0003] [Patent Document 1] JP 2023-10000 A Summary of the Invention [Problem to be solved by the invention]

[0004] In optimization techniques such as those disclosed in Patent Document 1, the difficulty of performing optimization calculations increases as the number of terminal devices and base stations increases. For example, if a configuration is adopted in which the optimization calculations are performed using classical computing, the increase in calculation costs (e.g., calculation time) becomes a problem, and if a configuration is adopted in which the optimization calculations are performed using quantum computing, the limitation on the number of quantum bits becomes a problem.

[0005] An object of one aspect of the present invention is to provide a technology that can suppress an increase in the difficulty of optimization calculations related to connection patterns even if the number of terminal devices and the number of base stations increases. [Means for solving the problem]

[0006] To solve the above problems, an information processing apparatus according to an aspect of the present invention includes a first acquisition unit that acquires first data including a signal-to-interference-noise ratio in each of a plurality of terminal devices when each of the plurality of terminal devices is connected to each of N (N is an integer of 2 or more) base stations; a first generation unit that generates second data representing a cost function that is a variable defining the connection between the i-th terminal device among the plurality of terminal devices and a base station, with variables defining the connection between each of M (M < N) base stations among the N base stations as arguments, by referring to the first data; and a second acquisition unit that acquires an optimization result obtained by referring to the second data and related to a connection pattern between the plurality of terminal devices and the N base stations.

[0007] To solve the above problems, an information processing system according to an aspect of the present invention is an information processing system including an information processing apparatus and a server apparatus. The information processing apparatus includes a first acquisition unit that acquires first data including a signal-to-interference-noise ratio in each of a plurality of terminal devices when each of the plurality of terminal devices is connected to each of N (N is an integer of 2 or more) base stations; a first generation unit that generates second data representing a cost function that is a variable defining the connection between the i-th terminal device among the plurality of terminal devices and a base station, with variables defining the connection between each of M (M < N) base stations among the N base stations as arguments, by referring to the first data; and a second acquisition unit that acquires an optimization result obtained by referring to the second data and related to a connection pattern between the plurality of terminal devices and the N base stations. The server apparatus includes an optimization execution unit that executes an optimization process related to a connection pattern between the plurality of terminal devices and the N base stations by referring to the second data, and a provision unit that provides the result of the optimization process to the information processing apparatus.

[0008] In order to solve the above problems, an information processing method according to an aspect of the present invention includes a first acquisition step of acquiring first data including a signal-to-interference-noise ratio in each of a plurality of terminal devices when each of the plurality of terminal devices is connected to each of N base stations (N is an integer of 2 or more), a first generation step of generating second data representing a cost function that is a variable defining the connection between the i-th terminal device among the plurality of terminal devices and the base station, with variables defining the connection between each of M base stations (M < N) among the N base stations as arguments, referring to the first data, and a second acquisition step of acquiring an optimization result obtained by referring to the second data and regarding an optimization result related to a connection pattern between the plurality of terminal devices and the N base stations.

[0009] Each aspect of the information processing apparatus according to the present invention may be implemented by a computer. In this case, a program for implementing the information processing apparatus by operating a computer as each part (software element) included in the information processing apparatus, and a computer-readable recording medium recording the same also fall within the scope of the present invention.

Effects of the Invention

[0010] According to one aspect of the present invention, even when the number of terminal devices and the number of base stations increase, it is possible to suppress an increase in the difficulty of optimization calculation related to the connection pattern.

Brief Description of the Drawings

[0011] [[ID=1十七]] [[ID=1十八]] [Figure 1] It is a block diagram showing the configuration of an information processing system according to an embodiment. [Figure 2] It is a diagram for explaining the processing by the information processing apparatus according to the embodiment. [Figure 3] It is a flowchart showing the flow of processing by the information processing system according to the embodiment. [Figure 4] It is a diagram for explaining the processing by the information processing system according to an application example. [Figure 5]FIG. 10 is a diagram for explaining processing by an information processing system according to an application example. [Figure 6] FIG. 10 is a diagram for explaining processing by an information processing system according to an application example. [Figure 7] FIG. 10 is a diagram for explaining processing by an information processing system according to an application example. [Figure 8] FIG. 10 is a diagram for explaining processing by an information processing system according to an application example. [Figure 9] FIG. 10 is a diagram for explaining processing by an information processing system according to an application example. DETAILED DESCRIPTION OF THE INVENTION

[0012] [Embodiment 1] The configuration of an information processing system 100 according to this embodiment will be described below with reference to Figures 1 to 3. The information processing system 100 is, in general terms, a system that optimizes connection patterns between each of a plurality of terminal devices arranged in a target area and each of a plurality of base stations arranged in the target area. For example, the example shown in the upper part of Figure 2 shows an example of connection patterns between 10 terminal devices (UE: User Equipment) and base stations (Base Stations) BS1 to BS3 arranged in three locations in a target area TA.

[0013] In connection pattern 1 in the upper part of Fig. 2, three UEs are connected to base station BS1, four UEs are connected to base station BS2, and three UEs are connected to base station BS3. In connection pattern 2, three UEs are connected to base station BS1, three UEs are connected to base station BS2, and four UEs are connected to base station BS3. In connection pattern 3, four UEs are connected to base station BS1, two UEs are connected to base station BS2, and four UEs are connected to base station BS3.

[0014] In this way, there are multiple connection patterns between each of the multiple terminal devices and each of the multiple base stations, and as the number of terminal devices and the number of base stations increases, the number of connection patterns also becomes enormous.

[0015] However, among these connection patterns, there are more preferable connection patterns and less preferable connection patterns. For example, if there are multiple base stations in a target area, a terminal device connected to a certain base station will receive interference signals from base stations other than the certain base station. Therefore, it is preferable to optimize the connection pattern between the terminal device and the base station in the target area, taking into account the influence of the interference signals.

[0016] The information processing system 100 according to this embodiment is a system capable of suitably executing the optimization process described above. In this embodiment, the optimization process may also be referred to as a process for solving an optimization problem. Furthermore, in this embodiment, the term "optimization process" refers to a process for deriving a solution using an update process that maximizes, minimizes, maximizes, or minimizes a predetermined cost function, rather than a process intended to completely match a theoretically optimal solution. Similarly, the term "optimization result" refers to a result (solution) derived using an update process that maximizes, minimizes, maximizes, or minimizes a predetermined cost function, rather than a process intended to completely match a theoretically optimal solution.

[0017] (Information processing system 100) The configuration of an information processing system 100 will be described with reference to Fig. 1. Fig. 1 is a block diagram showing the configuration of the information processing system 100 according to this embodiment. As shown in Fig. 1, the information processing system 100 includes an information processing device 1 and a server device 50 communicably connected to the information processing device 1 via a network N. Here, the specific configuration of the network N is not particularly limited, but as an example, a wireless LAN (Local Area Network), a wired LAN, a WAN (Wide Area Network), a public line network, a mobile data communication network, or a combination of these networks can be used.

[0018] (Server device 50) 1, the server device 50 includes an optimization execution unit 51 and a communication unit 52. The communication unit 52 receives data transmitted from the information processing device 1 and supplies the received data to the optimization execution unit 51. The communication unit 52 also transmits the optimization result calculated by the optimization execution unit 51 to the information processing device 1.

[0019] On the other hand, the optimization execution unit 51 is configured, for example, by any one of a quantum computer, a quantum simulator, and a classical computer, or a combination of these. For example, the optimization execution unit 51 executes an optimization process by referring to data provided from the information processing device 1 and received by the communication unit 52, and provides the result of the optimization process to the information processing device 1 via the communication unit 52.

[0020] Although a specific example of the optimization process executed by the optimization execution unit 51 does not limit this embodiment, as an example, Quantum annealing using quantum computers Simulated annealing using quantum simulators - Various optimization algorithms using classical computers The above configuration can be used.

[0021] 1 illustrates an example in which the server device 50 is a device separate from the information processing device 1, but this does not limit the present embodiment. At least some of the functions of the server device 50 (for example, the functions of the optimization execution unit 51) may be provided in the information processing device 1, and such a configuration is also included in the present embodiment.

[0022] (Information processing device 1) As shown in FIG. 1, the information processing device 1 includes a control unit 10, a storage unit 20, a communication unit 30, and an input / output unit 40.

[0023] (Communication unit 30) The communication unit 30 is an interface for transmitting and receiving data via the network N. As an example, the communication unit 30 transmits second data generated by a first generation unit 12 (described later) to the server device 50. The communication unit 30 also receives a result of the optimization process executed by an optimization execution unit 51 of the server device 50.

[0024] (Input / output section 40) The input / output unit 40 is an interface with an input device that accepts input of data and an output device that outputs data. Examples of input devices include, but are not limited to, a microphone, a camera, an eye-gaze input device, a keyboard, and a touchpad. Examples of output devices include, but are not limited to, a speaker and a liquid crystal display.

[0025] (Storage unit 20) The storage unit 20 stores various data referenced by the control unit 10 and various data derived by the control unit 10. As an example, as shown in FIG. 1, the storage unit 20 stores: First data D1 Second data D2 Optimization results OR Output data OUT Reference Data RD The first data D1 is data that is generated by a second generation unit 15 (to be described later) and acquired by a first acquisition unit 11 (to be described later). The first data D1 will be described in detail later.

[0026] On the other hand, the second data D2 is data that is generated by a first generation unit 12 (described later) and provided to the server device 50 via the communication unit 30. Details of the second data D2 will be described later. The optimization result OR is data that indicates the result of the optimization process executed by the optimization execution unit 51 of the server device 50, and is acquired by a second acquisition unit 13 (described later).

[0027] The output data OUT is data generated by an output data generation unit 14, which will be described later. The output data OUT will be described in detail later. The reference data RD is data that is referenced by a second generation unit 15, which will be described later, to generate first data D1. The reference data RD will be described in detail later.

[0028] (Control unit 10) As shown in FIG. 1, the control unit 10 includes a first acquisition unit 11, a first generation unit 12, a second acquisition unit 13, an output data generation unit 14, and a second generation unit 15.

[0029] (First acquisition unit 11) The first acquisition unit 11 acquires first data D1 including a signal-to-interference-plus-noise ratio of each of a plurality of terminal devices in a target area when each of the terminal devices is connected to each of N base stations (N is an integer of 2 or more). As an example, the first data D1 includes: When the i-th terminal device among the plurality of terminal devices is connected to base station 1 among the N base stations, the signal-to-interference-and-noise ratio S i,1 and, When the i-th terminal device among the plurality of terminal devices is connected to base station 2 among the N base stations, the signal-to-interference-and-noise ratio S i,2 and More specifically, if the number of the plurality of terminal devices is three, the first data D1 includes: When the first terminal device is connected to the base station 1, the signal-to-interference-and-noise ratio S 1,1 and, When the first terminal device is connected to the base station 2, the signal-to-interference-and-noise ratio S 1,2 and, When a second terminal device is connected to base station 1, the signal-to-interference-and-noise ratio S 2,1 and, When the second terminal device is connected to the base station 2, the signal-to-interference-and-noise ratio S 2,2 and, When a third terminal device is connected to base station 1, the signal-to-interference-and-noise ratio S 3,1 and, When a third terminal device is connected to base station 2, the signal-to-interference-and-noise ratio S 3,2 and Such first data D1 can be configured to be calculated (generated) in advance by a second generating unit 15, which will be described later, as an example, but this example does not limit the present embodiment.

[0030] The signal to interference and noise ratio is also expressed as SINR (Signal to Interference and Noise Ratio), and has a meaning as an index indicating how clearly a terminal device can receive data, for example. In this embodiment, the SINR is also expressed as DL-SINR (Down Link Signal to Interference and Noise Ratio). DL-SINR is defined as follows: Desired signal strength Interference signal strength Noise signal strength Using (DL-SINR) = (Desired signal strength) / {(Interference signal strength) + (Noise signal strength)} The signal strength is defined as follows: (Signal strength) = {(Transmit power) × (Transmit beam pattern)} / (Free space path loss) The DL-SINR may also be expressed as a logarithm, in which case the DL-SINR may take a positive or negative value.

[0031] The lower part of Figure 2 shows the SINR at each point in a target area when two base stations and multiple terminal devices are placed in the target area. More specifically, the lower left part of Figure 2 shows the SINR at each point in a target area of ​​1 km square. Two base stations with omnidirectional antennas; -Multiple terminal devices at 1m intervals The figure shows the SINR obtained when a terminal at each point connects to the base station so that the SINR is highest. The bottom right of Figure 2 shows the SINR obtained when a target area of ​​1 km square is Two base stations with directional antennas; -Multiple terminal devices at 1m intervals This figure shows the SINR obtained when a terminal at each location connects to the base station so that the SINR is highest. In this example, beamforming is performed in three directions from each base station.

[0032] In general, the SINR at each point may vary depending on which base station the point is connected to (in other words, which base station the desired signal is from). In this embodiment, the SINRs for each terminal device when connected to each base station are calculated in advance, and the calculated results are used as the signal-to-interference-and-noise ratio for each terminal device.

[0033] In this way, the signal-to-interference-and-noise ratio at each terminal device is ·Deployment of one or more base stations -Type of antenna at each base station - Placement of one or more terminal devices The data relating to the SINR such as those shown in the lower part of Fig. 2 may also be referred to as SINR data, SINR distribution, or SINR map.

[0034] (First generation unit 12) The first generation unit 12 generates second data D2 used in optimization processing related to connection patterns between terminal devices and base stations in a target area, the second data D2 expressing a cost function that defines the optimization problem, by referring to the first data D1. Here, the cost function is, for example, a variable (xi,j、 or x i (also denoted as such), and among the N base stations, a cost function having, as arguments, variables (x i,j、 or x i ) that define the connection with each of M (M < N) base stations.

[0035] Thus, in this embodiment, by formulating the above optimization problem · Variables (x i,j、 or x i ) that define the connection between the i-th terminal device and each of M (M < N) base stations among the N base stations the number of variables can be reduced as compared with the case of considering the connection with each of all N base stations. Note that the second data generated by the first generation unit 12 and specific examples of the cost function will be described later. Also, the cost function may be referred to as an objective function, a Hamiltonian, or the like, but such expressions do not limit this embodiment.

[0036] (Second acquisition unit 13) The second acquisition unit 13 acquires an optimization result obtained by referring to the second data, the optimization result regarding the connection pattern between the plurality of terminal devices and the N base stations. As an example, the second acquisition unit 13 acquires, via the communication unit 30, the optimization result by the optimization execution unit 51 of the server device 50 that refers to the second data.

[0037] (Output data generation unit 14) The output data generation unit 14 generates output data OUT by referring to the optimization result acquired by the second acquisition unit 13. As an example, the output data generation unit 14 uses, as the output data OUT, · Display data that visually shows the connection pattern between the plurality of terminal devices and the plurality of base stations in the target area indicated by the above optimization result, or · Data for input to a connection control device that controls the actual connection between the plurality of terminal devices and the plurality of base stations, the data representing the connection pattern indicated by the above optimization result Generate the like, and present the generated output data OUT to the user or provide it to the connection control device via the input / output unit 40 or the communication unit.

[0038] (Second generation unit 15) The second generation unit 15 generates the first data D1 for providing to the first acquisition unit 11 by referring to the reference data RD. Here, as an example, the reference data RD · Each position of the plurality of terminal devices in the target area where the N base stations are arranged is included. More specifically, the second generation unit 15 is for the target area · Arrangement of one or more base stations · Type of antenna at each base station · Arrangement of one or more terminal devices can be configured to generate the first data D1 by referring to the reference data RD including the above. Since specific examples of the first data D1 generated by the second generation unit 15 have been described above, the description is omitted here.

[0039] (Effect by information processing device 1) As described above, in the present embodiment, the cost function that defines the above optimization problem · Variables (x i,j、 or x i ) that define the connection between the i-th terminal device and each of M (M < N) base stations among the N base stations By formulating with, the number of variables can be reduced compared to the case of considering the connection with each of all N base stations. In other words, the scale of the variables required for the optimization process can be reduced. Therefore, according to the formulation of the cost function as described above, even if the number of terminal devices and the number of base stations increase, an increase in the difficulty of the optimization calculation regarding the connection pattern can be suppressed, so that a larger-scale optimization problem can be solved.

[0040] (Processing flow by information processing system 100) Next, an example of the flow of processing by the information processing system 100 will be described with reference to FIG.

[0041] (Step S10) First, in step S10, the second generating unit 15 generates the first data D1 with reference to the reference data RD. The specific processing by the second generating unit 15 has been described above, and therefore will not be described here.

[0042] (Step S11) Subsequently, in step S11, the first acquisition unit 11 acquires the first data D1 generated by the second generation unit 15. Specific examples of the first data D1 have been described above, and therefore will not be described here.

[0043] (Step S12) Next, in step S12, the first generator 12 generates second data D2 by referring to the first data D1. The second data D2 and the cost function referred to for generating the second data are as described above. Specific examples of the second data D2 and the cost function will be described later.

[0044] In this step, the first generation unit 12 also provides the generated second data D2 to the server device 50 via the communication unit 30.

[0045] (Step S21) Subsequently, in step S21, the optimization execution unit 51 included in the server device 50 acquires the second data D2 generated by the first generation unit 12, and executes an optimization process using the second data D2. Quantum annealing using quantum computers Simulated annealing using quantum simulators - Various optimization algorithms using classical computers Optimization processing is performed using

[0046] (Step S22) Subsequently, in step S22, the optimization execution unit 51 provides the result of the optimization process to the information processing apparatus 1 via the communication unit 52.

[0047] (Step S13) Subsequently, in step S13, the second acquisition unit 13 included in the information processing apparatus 1 acquires the result of the optimization process by the optimization execution unit 51.

[0048] (Step S14) Subsequently, in step S14, the output data generation unit 14 generates the output data OUT by referring to the result of the optimization process acquired by the second acquisition unit 13. Since the specific process by the output data generation unit 14 has been described above, the description is omitted here.

[0049] (Specific example of the process by the first generation unit 12) Hereinafter, a specific example of the process by the first generation unit 12 will be described. As described above, the first generation unit 12 refers to the first data D1, constructs a cost function, and generates the second data D2 representing the cost. Then, the second data D2 is used in the optimization process by the optimization execution unit 51.

[0050] First, a specific example of the cost function referred to by the first generation unit 12 will be described. As an example, the cost function

Equation

[0051] As an example, in a target area, there are terminal devices 1 to 3 (also referred to as terminals 1 to 3), and there are three or more base stations (corresponding to the above N≧3), For terminal 1, the base station with the highest SINR is base station 1, and the base station with the second highest SINR is base station 2. For terminal 2, the base station with the highest SINR is base station 2, and the base station with the second highest SINR is base station 1. For terminal 3, the base station with the highest SINR is base station 1, and the base station with the second highest SINR is base station 2. In this case, the variable x i For example, the values ​​of (i=1,2,3) are: [Table 1] Here, "base station" in Table 1 refers to the above-mentioned base station 1 or base station 2, and Table 1 is For terminal 1, x=1 is the connection to base station 1, and x=0 is the connection to base station 2. For terminal 2, x=1 is the connection to base station 2, and x=0 is the connection to base station 1. For terminal 3, x=1 is the connection to base station 1, and x=0 is the connection to base station 2. In the example shown in Table 1, terminals 1 to 3 are all connected to base station 1, for example.

[0052] Thus, in the example shown in Table 1, · The variable x1 that defines the connection regarding terminal 1 becomes x1 = 1 when the terminal 1 is connected to base station 1 (the "base station" in Table 1), · The variable x2 that defines the connection regarding terminal 2, for the terminal 2, since base station 1 is the base station with the second largest SINR, when the terminal 2 is connected to base station 1, x2 = 0, · The variable x3 that defines the connection regarding terminal 3 becomes x3 = 1 when the terminal 3 is connected to base station 1 (the "base station" in Table 1) is shown. The above variables x1 to x3 are an example of variables that define the connection with each of M (M < N, in the above example M = 2) base stations among N (in the above example N ≥ 3) base stations.

[0053] Thus, when M = 2, it can be seen that as a variable that defines the connection between the i-th terminal device and the base station, it is sufficient to consider the connection with substantially one base station. Therefore, according to the above configuration, compared with the case of using variables that define the connection between the i-th terminal device and each of multiple base stations, the number of variables for defining the cost function is significantly reduced.

[0054] Also, as shown in Equation 1, as the coefficient of the variable (x i ) that defines the connection between the i-th terminal device and the base station, the coefficient (S i,1 , S i,2 ) corresponding to the signal-to-interference-plus-noise ratio in the i-th terminal device is included. By including the above counting (S i,1 , S i,2 ) in the cost function, a suitable weight regarding the optimization problem can be set for the connection between the i-th terminal device and the base station.

[0055] Also, as shown in Equation 1, as a constraint term, the cost function is

Equation

[0056] In the constraint term shown in Equation 2, Ca indicates that the number of terminals connected to base station a is Ca.

[0057] (Cost Function According to Comparative Example) In order to more specifically explain the effect of the cost function according to this embodiment, a cost function according to a comparative example will also be explained. The cost function according to the comparative example is

number

[0058] Here, in the cost function according to the comparative example, the variable xi,a The total number of terminals in the target area is the product of the total number of terminals in the target area and the total number of base stations in the target area. Therefore, the more the number of base stations in the target area increases, the more the variable x i,a The total number of

[0059] On the other hand, the cost function according to the present exemplary embodiment is i and the variable x i In this case, it is sufficient to consider the connection with one base station. Therefore, even if the number of base stations in the target area increases, if the number of terminal devices remains the same, the variable x i The number of variables does not increase. Generally, a target area has a large number of terminal devices and a large number of base stations. Therefore, by using the cost function according to this embodiment, the number of variables can be significantly reduced. Furthermore, the calculation cost of the optimization calculation using the cost function can also be significantly reduced.

[0060] Furthermore, the cost function according to the comparative example includes a constraint term (the second term in Equation 3) indicating that each terminal is connected to one base station, but the cost function according to this embodiment does not require such a term because M=2. In other words, the constraint condition is obviously satisfied. Therefore, the calculation cost of the optimization process using the cost function according to this embodiment is reduced compared to when the cost function according to the comparative example is used.

[0061] (Additional notes on specific examples of cost functions) In the above example, a cost function in the case of M=2 is used as the cost function according to this embodiment, but this example does not limit this embodiment. For example, in the cost function shown in (Equation 2), ·S i,1 , S i,2 , S i,j generalized to Here, S i,jindicates the SINR in the i-th terminal device when the signal-to-interference-noise ratio with the i-th terminal device is the j-th largest among the base stations connected to the ji-th terminal device. Also, in the cost function shown in (Equation 2), · N1(a) and N2(a) are generalized to N j (a). Here, N j (a) is the set of terminal devices whose SINR becomes the j-th largest when connected to base station a. Even in the case of the above extension, the extended cost function is · Variables defining the connection between the i-th terminal device and each of M (M < N) base stations out of the N base stations Since it is formulated by, the number of variables can be reduced compared to the case of considering the connection with each of all N base stations.

[0062] (Specific Example of Processing by First Generation Unit 12 (Continued)) The description of the specific example of the processing by the first generation unit 12 will be continued. The first generation unit 12 generates second data D2 representing the cost function according to the present embodiment described above. As the data format of the second data D2, a data format acceptable to the optimization execution unit 51 may be adopted.

[0063] As an example, when the optimization execution unit 51 is configured to execute quantum annealing or simulated annealing, the first generation unit 12 may generate the second data D2 as data in a QUBO (Quadratic Unconstrained Binary Optimization) format representing the above cost function (Equation 2) or a data format equivalent thereto.

[0064] Also, when the optimization execution unit 51 is configured to execute various optimization algorithms using a classical computer, the first generation unit 12 may generate the second data D2 as data in a data format including various parameters referred to in the optimization algorithm and including various parameters representing the above cost function (Equation 2).

[0065] The second data D2 generated by the first generation unit 12 is provided to the optimization execution unit 51, which executes the optimization process. More specifically, the optimization execution unit 51 executes the following:

number

[0066] <Example> An example of the information processing system 100 according to this embodiment will be described below. In this example, D-Wave (registered trademark) provided by D-Wave Systems Inc. was used as the optimization execution unit 51. Furthermore, patterns 1 to 4 shown in FIGS. 4 to 5 were used as test patterns for terminal devices, antenna patterns, and beam patterns. BS1 to BS3 in FIGS. 4 to 5 indicate base stations. Furthermore, each point shown in "terminal device placement" in FIGS. 4 to 5 indicates a terminal device.

[0067] Pattern 1 is as shown in Figure 4. Terminal placement: uniformly random Antenna pattern of each base station: Omnidirectional (Isotropic) Beam pattern: isotropic In addition, as shown in Figure 4, pattern 2 is Terminal device placement: some sparsely populated Antenna pattern of each base station: Omnidirectional (Isotropic) Beam pattern: isotropic In addition, as shown in Figure 5, pattern 3 is Terminal placement: uniformly random Antenna pattern of each base station: Directional (Gaussian: (half-width 30 degrees, SLL-15dB)) Beam pattern: Non-isotropic (beamforming in three directions from each base station) In addition, as shown in Figure 5, pattern 4 is Terminal device placement: some sparsely populated Antenna pattern of each base station: Directional (Gaussian: (half-width 30 degrees, SLL-15dB)) Beam pattern: Non-isotropic (beamforming in three directions from each base station) It is as follows.

[0068] Also, in each pattern 1 to 4, Number of users: 30 Number of connectable base stations: 1 Number of cells: 3 cells / base station Number of frequencies: 1 frequency Capacity: Number of users / base stations = 10 · Placement (minimum distance between base stations): 30m -Type of propagation loss: Free space propagation loss (inversely proportional to the square of the distance) For each of these patterns 1 to 4, the above-mentioned cost function (Equation 2) was constructed, the above-mentioned second data D2 was generated in the QUBO format that expresses the cost function, and optimization processing was performed in D-Wave.

[0069] FIG. 6 is a graph comparing the number of bits used in D-Wave when using the cost function (Equation 2) according to this embodiment and the number of bits used in D-Wave when using the cost function (Equation 3) according to the comparative example. In FIG. 6, the "dotted line (simple)" indicates the graph according to the comparative example, and the "solid line (proposed)" indicates the graph according to this embodiment. As shown in FIG. 6, in this embodiment, the increase in the number of bits used when the number of terminals increases is more gradual than in the comparative example. Therefore, according to the configuration according to this embodiment, it is possible to solve larger-scale optimization problems without increasing computational resources compared to the configuration according to the comparative example.

[0070] The upper part of Fig. 7 is a graph comparing the relative error between this example and the comparative example for each of the above-mentioned patterns 1 to 4. In Fig. 7, "simple" indicates the graph related to the comparative example, and "proposed" indicates the graph related to this example. Here, the relative error is (Relative error) = {(calculation result SINR) - (full search solution SINR)} / (full search solution SINR) As shown in the upper part of Fig. 7, for all patterns, the relative error according to this example is smaller than that of the comparative example.

[0071] The lower part of Fig. 7 is a graph comparing the constraint fulfillment rate between the present embodiment and the comparative example for each of the above-mentioned patterns 1 to 4. Here, the constraint fulfillment rate is (Constraint satisfaction rate) = The percentage of times the constraint was satisfied out of 100 trials As shown in the lower part of Fig. 7, for three of the four patterns, the constraint fulfillment rate according to this embodiment is greater than that of the comparative example.

[0072] Moreover, the upper part of Fig. 8 is a graph comparing the relative error by quantum annealing (QA in the upper part of Fig. 8) with the relative error by simulated annealing (SA in the upper part of Fig. 8) for the above-mentioned pattern 1. In Fig. 8, "simple" indicates the graph related to the comparative example, and "proposed" indicates the graph related to this embodiment. As shown in the upper part of Fig. 8, with regard to both the relative error related to this embodiment and the relative error related to the comparative example, simulated annealing has a smaller relative error than quantum annealing.

[0073] The lower part of Fig. 8 is an enlarged graph of the relative error by simulated annealing in the upper part of Fig. 8. As shown in the lower part of Fig. 8, in simulated annealing, the configuration according to this example obtained a relative error of the same order of magnitude as the configuration according to the comparative example.

[0074] (Appendix 1) In the above examples, free space path loss was used as an example of path loss, but we will now add a bit more about this. Free space path loss (FSPL) is the loss that occurs when electromagnetic waves propagate through a space with no obstacles, and is a good approximation of the path loss in actual space. More specifically, free space path loss indicates the current state in which electromagnetic waves emitted from an antenna or wireless communication device are weakened while propagating through space,

number

[0075] (Appendix 2)

[0076] The processing by the information processing device 1 according to this embodiment is not limited to the above-described example. As an example, the information processing device 1 may be configured to sequentially acquire reference data RD including the locations of one or more terminal devices in a target area, sequentially generate first data D1 and second data D2 accordingly, and sequentially acquire corresponding optimization results. For example, 24 hours that make up a day may be divided into three-hour periods, and the above-described sequential processing may be performed for each time period.

[0077] Furthermore, in the above-described processing, the reference data RD may include map information for the target area, and the first data D1 may be generated by also referencing the map information.

[0078] In addition, the information processing device 1 may be configured to compare past reference data RD with current reference data RD, and if the two reference data are similar to each other to a predetermined degree or more, omit the optimization processing based on the current reference data RD by the optimization execution unit 51 and output the optimization results already obtained in the past.

[0079] (Appendix 3) In the above-described embodiment, an optimization process was performed using the cost function defined by Equation 2, but this configuration can also be compared with Comparative Example 2 below. Comparative Example 2 is an optimization method in which, as an example, a cumulative distribution function is calculated for each of connection patterns 1 to 3 shown in the upper part of Fig. 2, and the median value is maximized. The lower part of Fig. 9 shows an example of a cumulative distribution function referenced in the configuration according to Comparative Example 2. In fact, The median of the solution (SINR) obtained in the configuration according to this embodiment The median value of the solution (SINR) obtained in the configuration according to the above comparative example 2 When the median values ​​calculated for the configuration according to Comparative Example 2 were compared, the median value calculated for the configuration according to this embodiment was not greater than the median value calculated for the configuration according to this embodiment. Furthermore, it was found that the connection pattern calculated for the configuration according to this embodiment (which can also be expressed as the connection pattern with the largest sum of SINR) was always the connection pattern with the largest median SINR. For this reason, the configuration according to this embodiment is superior to the configuration according to Comparative Example 2 in that it Maximization of the median is also realized, and a better solution (larger median) can be obtained than in the configuration according to Comparative Example 2. This has the following advantages.

[0080] [Software implementation example] The functions of the information processing device 1 (hereinafter referred to as the "device") can be realized by a program that causes a computer to function as the device, and a program that causes a computer to function as each control block of the device (particularly each part included in the control unit 10).

[0081] In this case, the device includes a computer having at least one control device (e.g., a processor) and at least one storage device (e.g., a memory) as hardware for executing the program. The control device and storage device execute the program, thereby realizing the functions described in each of the above embodiments.

[0082] The above program may be recorded on one or more computer-readable recording media, rather than temporarily. This recording medium may or may not be provided in the above device. In the latter case, the above program may be supplied to the above device via any wired or wireless transmission medium.

[0083] In addition, part or all of the functions of each of the above control blocks can also be realized by a logic circuit. For example, an integrated circuit in which a logic circuit functioning as each of the above control blocks is formed is also included in the scope of the present invention. In addition to this, for example, it is also possible to realize the functions of each of the above control blocks by a quantum computer.

[0084] Also, each of the processes described in each of the above embodiments may be executed by AI (Artificial Intelligence). In this case, AI may operate in the above control device, or may operate in another device (for example, an edge computer or a cloud server, etc.).

[0085] (Supplementary Notes) At least the following configurations are described in this specification.

[0086] (Configuration 1) A first acquisition unit that acquires first data including a signal-to-interference-plus-noise ratio in each of a plurality of terminal devices when each of the plurality of terminal devices is connected to each of N (N is an integer of 2 or more) base stations, A second data that represents a cost function that is a variable defining the connection between the i-th terminal device and the base station among the plurality of terminal devices, and takes as an argument a variable defining the connection with each of M (M < N) base stations among the N base stations, and a first generation unit that generates the second data by referring to the first data, A second acquisition unit that acquires an optimization result obtained by referring to the second data and relates to an optimization result regarding a connection pattern between the plurality of terminal devices and the N base stations An information processing apparatus comprising the above.

[0087] (Configuration 2) wherein M is 2; The value of the variable that defines the connection between the i-th terminal device and the base station is The value is 1 when the terminal device is connected to the base station among the N base stations, which has the highest signal-to-interference-and-noise ratio with the i-th terminal device; It becomes 0 when the i-th terminal device is connected to the base station that has the second largest signal-to-interference-and-noise ratio among the N base stations. 2. The information processing device according to configuration 1.

[0088] (Configuration 3) The cost function is: The coefficient of the variable that defines the connection between the i-th terminal device and the base station includes a coefficient corresponding to the signal-to-interference-and-noise ratio in the i-th terminal device. 3. The information processing device according to configuration 1 or 2 (Configuration 4) The cost function is: It includes a constraint term defined using a set of terminal devices that will have the highest signal-to-interference-and-noise ratio when connected to a certain base station and a set of terminal devices that will have the second highest signal-to-interference-and-noise ratio when connected to the certain base station. 4. The information processing device according to any one of configurations 1 to 3.

[0089] (Configuration 5) The N base stations are located in a target area. The N base stations are located in a target area. 5. The information processing device according to any one of configurations 1 to 4.

[0090] (Configuration 6) An information processing system including an information processing device and a server device, The information processing device includes: a first acquisition unit that acquires first data including a signal-to-interference-plus-noise ratio in each of a plurality of terminal devices when the terminal devices are each connected to N base stations (N is an integer equal to or greater than 2); A second data representing a cost function that is a variable defining the connection between the i-th terminal device among the plurality of terminal devices and the base station, and taking as arguments variables defining the connection with each of M (M < N) base stations among the N base stations is generated by a first generation unit that refers to the first data. A second acquisition unit that acquires an optimization result obtained by referring to the second data, the optimization result regarding the connection pattern between the plurality of terminal devices and the N base stations. Comprising The server device An optimization execution unit that executes an optimization process regarding the connection pattern between the plurality of terminal devices and the N base stations by referring to the second data, A providing unit that provides the result of the optimization process to the information processing device. An information processing system comprising.

[0091] (Configuration 7) A first acquisition step of acquiring first data including the signal-to-interference-noise ratio in the terminal device when each of the plurality of terminal devices is connected to each of N (N is an integer of 2 or more) base stations, A first generation step of generating second data representing a cost function that is a variable defining the connection between the i-th terminal device among the plurality of terminal devices and the base station, and taking as arguments variables defining the connection with each of M (M < N) base stations among the N base stations, by referring to the first data. A second acquisition step of acquiring an optimization result obtained by referring to the second data, the optimization result regarding the connection pattern between the plurality of terminal devices and the N base stations. An information processing method including.

[0092] (Configuration 8) A program for causing a computer to function as the information processing device according to Configuration 1, the program for causing a computer to function as the first acquisition unit, the first generation unit, and the second acquisition unit.

[0093] (Configuration 9) A computer-readable recording medium having the program according to configuration 8 recorded thereon.

[0094] The present invention is not limited to the above-described embodiments, and various modifications are possible within the scope of the claims. Embodiments obtained by appropriately combining the technical means disclosed in different embodiments are also included in the technical scope of the present invention. [Explanation of symbols]

[0095] 1. Information processing device 11 First acquisition section 12 First generation part 13 Second acquisition section 14 Output data generation unit 15 Second generation part

Claims

1. a first acquisition unit that acquires first data including a signal-to-interference-plus-noise ratio in each of a plurality of terminal devices when the terminal devices are each connected to N base stations (N is an integer equal to or greater than 2); a first generation unit that generates, by referring to the first data, second data that expresses a cost function having as an argument a variable that defines a connection between an i-th terminal device among the plurality of terminal devices and a base station, the variable defining a connection with each of M (M<N) base stations among the N base stations; a second acquisition unit that acquires an optimization result obtained by referring to the second data, the optimization result relating to a connection pattern between the plurality of terminal devices and the N base stations; An information processing device comprising:

2. said M is 2; The value of the variable that defines the connection between the i-th terminal device and the base station is The value is 1 when the terminal device is connected to the base station among the N base stations, which has the highest signal-to-interference-and-noise ratio with the i-th terminal device; It becomes 0 when the i-th terminal device is connected to the base station that has the second largest signal-to-interference-and-noise ratio between the i-th terminal device and the N base stations. The information processing device according to claim 1 .

3. The cost function is: The coefficient of the variable that defines the connection between the i-th terminal device and the base station includes a coefficient corresponding to the signal-to-interference-and-noise ratio in the i-th terminal device. The information processing device according to claim 2 .

4. The cost function is: The constraint term is defined using a set of terminal devices that will have the highest signal-to-interference-and-noise ratio when connected to a certain base station, and a set of terminal devices that will have the second highest signal-to-interference-and-noise ratio when connected to the certain base station. The information processing device according to claim 3 .

5. The system further includes a second generating unit that generates the first data by referring to reference data including the positions of the plurality of terminal devices in a target area where the N base stations are located. The information processing device according to claim 1 .

6. An information processing system including an information processing device and a server device, The information processing device includes: a first acquisition unit that acquires first data including a signal-to-interference-plus-noise ratio in each of a plurality of terminal devices when the terminal devices are each connected to N base stations (N is an integer equal to or greater than 2); a first generation unit that generates, by referring to the first data, second data that expresses a cost function having as an argument a variable that defines a connection between an i-th terminal device among the plurality of terminal devices and a base station, the variable defining a connection with each of M (M<N) base stations among the N base stations; a second acquisition unit that acquires an optimization result obtained by referring to the second data, the optimization result relating to a connection pattern between the plurality of terminal devices and the N base stations; Equipped with The server device an optimization execution unit that refers to the second data and executes optimization processing regarding connection patterns between the plurality of terminal devices and the N base stations; a providing unit that provides a result of the optimization process to the information processing device; An information processing system comprising:

7. a first acquisition step of acquiring first data including a signal-to-interference-plus-noise ratio in each of a plurality of terminal devices when the terminal devices are connected to N base stations (N is an integer equal to or greater than 2); a first generation step of generating, by referring to the first data, second data expressing a cost function having as an argument a variable that defines a connection between an i-th terminal device among the plurality of terminal devices and a base station, the variable defining a connection between the i-th terminal device and each of M (M<N) base stations among the N base stations; a second obtaining step of obtaining an optimization result obtained by referring to the second data, the optimization result relating to a connection pattern between the plurality of terminal devices and the N base stations; An information processing method comprising:

8. 2. A program for causing a computer to function as the information processing device according to claim 1, the program causing a computer to function as the first acquisition unit, the first generation unit, and the second acquisition unit.

9. A computer-readable recording medium on which the program according to claim 8 is recorded.

Images