Position estimation system, position estimation device, position estimation method, and position estimation program

Abstract

Efficiently estimate the location of wireless terminals whose location is unknown. [Solution] A position estimation device according to one embodiment includes: an acquisition unit that acquires channel state information; a sound collection unit that collects sound waves from a plurality of areas including the installation location of the second wireless terminal to the base station; an arrival direction detection unit that detects the arrival direction of each sound wave with a predetermined resolution; a determination unit that determines the timing for extracting channel state information so as to maximize the number of areas corresponding to the arrival direction of the sound waves detected by the arrival direction detection unit, based on the arrival direction of each detected sound wave, the number of sound waves arriving in each arrival direction, and the installation location of the second wireless terminal; a sample extraction unit that extracts a portion of the acquired channel state information of each second wireless terminal and the first wireless terminal; and an estimation unit that estimates that the first wireless terminal is located at the position closest to the second wireless terminal which has the highest correlation with the channel state information of the first wireless terminal.

Description

Technical Field

[0001] The present invention relates to a technique for estimating the position of a wireless terminal with an unknown position using a wireless signal.

Background Art

[0002] There is known a technique for estimating the position of a wireless terminal with an unknown installation position using wireless communication technology (for example, Wi-Fi (registered trademark), etc.). For example, the correlation between the propagation path state information (channel state information: CSI) transmitted by a STA (Station) with a known installation position and the propagation path state information transmitted by a STA with an unknown existence position is determined for each STA with a known installation position, and it is determined that the STA exists in the vicinity of the installation position of the STA that transmitted the propagation path state information having the highest correlation with the propagation path state information transmitted by the STA with an unknown existence position (for example, see Patent Document 1).

Prior Art Documents

Patent Documents

[0003]

Patent Document 1

Summary of the Invention

Problems to be Solved by the Invention

[0004] However, in the prior art, when there is no change in the situation of the propagation path, the CSI to be used has little variation. For example, in a warehouse where there are few people, or a house where residents are out or asleep, the estimation accuracy of the installation position using CSI may decrease, and problems such as an increase in power consumption and consumption of computing resources may occur.

[0005] The present invention has been made in view of the above-described problems, and an object thereof is to provide a position estimation system, a position estimation device, a position estimation method, and a position estimation program that can efficiently estimate the position of a wireless terminal with an unknown existence position.

Means for Solving the Problems

[0006] A position estimation system according to one embodiment of the present invention comprises a first wireless terminal whose location is unknown, a plurality of second wireless terminals whose installation locations are known, and a base station that performs wireless communication with the first wireless terminal and each of the plurality of second wireless terminals, and in a position estimation system for estimating the location of the first wireless terminal, the base station, an acquisition unit that acquires channel state information indicating the state of propagation channels between each of the plurality of second wireless terminals and the first wireless terminal at predetermined time intervals, a sound collection unit that collects sound waves arriving toward the base station from each of a plurality of areas including the installation locations of the second wireless terminals in synchronization with the acquisition unit at predetermined time intervals, an arrival direction detection unit that detects the arrival direction of each sound wave collected by the sound collection unit with a predetermined resolution, and the sound waves detected by the arrival direction detection unit The device is characterized by comprising: a determination unit that determines the timing for extracting channel state information such that the number of corresponding sound wave directions detected by the direction of arrival detection unit and the region is maximized, based on the direction of arrival, the number of sound waves arriving in each direction of arrival, and the installation position of the second wireless terminal; a sample extraction unit that extracts a portion of the channel state information of each of the second wireless terminals and the first wireless terminal acquired by the acquisition unit as samples to be processed, based on the timing determined by the determination unit; and an estimation unit that estimates that the first wireless terminal is located at the position closest to the second wireless terminal with the highest correlation among the plurality of second wireless terminals extracted by the sample extraction unit, based on the channel state information of the first wireless terminal extracted by the sample extraction unit.

[0007] Furthermore, a position estimation device according to one embodiment of the present invention is a position estimation device for estimating the position of a first wireless terminal whose location is unknown, comprising: a base station; an acquisition unit that acquires channel state information indicating the state of propagation channels between each of a plurality of second wireless terminals whose installation locations are known and the first wireless terminal at predetermined time intervals; a sound collection unit that collects sound waves arriving toward the base station from each of a plurality of areas including the installation locations of the second wireless terminals at predetermined time intervals in synchronization with the acquisition unit; an arrival direction detection unit that detects the arrival direction of each sound wave collected by the sound collection unit with a predetermined resolution; the arrival direction of each sound wave detected by the arrival direction detection unit, the number of sound waves arriving for each arrival direction, and the second The wireless device is characterized by comprising: a determination unit that determines the timing for extracting channel state information such that the number of correspondences between the direction of arrival of sound waves detected by the direction of arrival detection unit and the region is maximized, based on the installation position of the wireless terminal; a sample extraction unit that extracts a portion of the channel state information of each of the second wireless terminals and the first wireless terminal acquired by the acquisition unit as samples to be processed, based on the timing determined by the determination unit; and an estimation unit that estimates that the first wireless terminal is located at the position closest to the second wireless terminal with the highest correlation among the plurality of second wireless terminals extracted by the sample extraction unit, based on the channel state information of the first wireless terminal extracted by the sample extraction unit.

[0008] Furthermore, a location estimation method according to one embodiment of the present invention is a location estimation method for estimating the location of a first wireless terminal whose location is unknown, comprising: an acquisition step of acquiring channel state information indicating the state of propagation channels between a base station, each of a plurality of second wireless terminals whose installation locations are known, and the first wireless terminal at predetermined time intervals; a sound collection step of collecting sound waves arriving toward the base station from each of a plurality of areas including the installation locations of the second wireless terminals at predetermined time intervals; an arrival direction detection step of detecting the arrival direction of each sound wave collected by the sound collection step with a predetermined resolution; and the arrival direction of each sound wave detected by the arrival direction detection step, the number of sound waves arriving for each arrival direction, and the installation location of the second wireless terminal. Based on the above, the method includes a determination step of determining the timing for extracting channel state information such that the number of correspondences between the direction of arrival of sound waves detected by the direction of arrival detection step and the region is maximized; a sample extraction step of extracting a portion of the channel state information of each of the second wireless terminals and the first wireless terminal acquired by the acquisition step as samples to be processed, based on the timing determined by the determination step; and an estimation step of estimating that the first wireless terminal is located at the position closest to the second wireless terminal with the highest correlation among the plurality of second wireless terminals extracted by the sample extraction step, based on the channel state information of the first wireless terminal extracted by the sample extraction step.

[0009] Furthermore, a position estimation program according to one embodiment of the present invention is a position estimation program executed by a position estimation device that estimates the position of a first wireless terminal whose location is unknown, comprising: a base station; an acquisition unit that acquires channel state information indicating the state of propagation channels between each of a plurality of second wireless terminals whose installation locations are known and the first wireless terminal at predetermined time intervals; a sound collection unit that collects sound waves arriving toward the base station from each of a plurality of areas including the installation locations of the second wireless terminals at predetermined time intervals in synchronization with the acquisition unit; an arrival direction detection unit that detects the arrival direction of each sound wave collected by the sound collection unit with a predetermined resolution; the arrival direction of each sound wave detected by the arrival direction detection unit, the number of sound waves arriving for each arrival direction, and the second wireless terminal The computer functions as each part of the position estimation device, which includes: a determination unit that determines the timing for extracting channel state information such that the number of correspondences between the direction of arrival of sound waves detected by the direction of arrival detection unit and the region is maximized based on the installation position of the wire terminal; a sample extraction unit that extracts a portion of the channel state information of each of the second wireless terminals and the first wireless terminal acquired by the acquisition unit as samples to be processed, based on the timing determined by the determination unit; and an estimation unit that estimates that the first wireless terminal is located at the position closest to the second wireless terminal with the highest correlation among the multiple second wireless terminals extracted by the sample extraction unit, based on the channel state information of the first wireless terminal extracted by the sample extraction unit. [Effects of the Invention]

[0010] According to the present invention, the location of a wireless terminal whose location is unknown can be efficiently estimated. [Brief explanation of the drawing]

[0011] [Figure 1] This figure schematically illustrates the outline of a position estimation system according to one embodiment. [Figure 2] This diagram illustrates the distribution of the direction of arrival of sound waves detected by a smart speaker. [Figure 3] This is a conceptual diagram illustrating the operation of the position estimation system. [Figure 4] This is a functional block diagram illustrating the functions of a position estimation device. [Figure 5] This is a diagram illustrating compressed CSI. [Figure 6] This diagram shows the pretreatment performed by the pretreatment unit. [Figure 7] This diagram illustrates the direction of arrival of sound waves and the number of arrivals for each direction of arrival of sound waves, as stored in the direction of arrival count storage unit. [Figure 8] This diagram illustrates the clustering performed by the first clustering unit. [Figure 9] This diagram illustrates the timing at which the decision-making unit should extract the determined CSI. [Figure 10] This figure shows an example of the hardware configuration of a position estimation device according to one embodiment. [Figure 11] This diagram schematically illustrates the overview of the position estimation system used in the comparative example. [Figure 12] This figure shows an overview of the first and second techniques by which the position estimation system 10 estimates the position of the first wireless terminal. [Modes for carrying out the invention]

[0012] In describing a location estimation system according to one embodiment, the background leading to the present invention will first be explained using Figures 11 and 12. Figure 11 is a schematic diagram illustrating the outline of a comparative example location estimation system 10. As shown in Figure 11, the comparative example location estimation system 10 includes, for example, a base station (AP) 20, a first wireless terminal 40, second wireless terminals 31-35, a smart speaker 50, and a location estimation device 60. Here, it is assumed that a person enters a room at time t1 and leaves at time t2.

[0013] The base station 20 performs wireless communication with each of the first wireless terminal 40 and the second wireless terminals 31 using channel state information (CSI) indicating the state of the propagation channel between the base station 20 and each of the first wireless terminal 40 and the second wireless terminals 31, for example, in accordance with IEEE802.11ac.

[0014] The location of the first wireless terminal 40 is unknown. The installation locations of the second wireless terminals 31 to 35 are known respectively. The smart speaker 50 has functions of, for example, detecting (monitoring) sound waves (sound) and detecting the direction of arrival of sound waves.

[0015] Then, the position estimation device 60 estimates the position of the first wireless terminal 40 whose existence location is unknown using each of the channel state information. FIG. 12 is a diagram showing an overview of a first technique and a second technique for the position estimation system 10 to estimate the position of the first wireless terminal 40.

[0016] [First Technical Example] For example, the position estimation device 60 estimates that the first wireless terminal 40 exists in the vicinity of the installation location of any one of the second wireless terminals 31 to 35 that transmitted the CSI having the highest correlation with the CSI transmitted by the first wireless terminal 40 whose existence location is unknown (see FIG. 12).

[0017] It is considered that the CSI fluctuates more greatly when people or objects that affect the propagation of radio waves move significantly. However, in the first technical example, all the CSI is acquired and used. For example, when using the CSI at a time when people or objects do not move much, such as at night in a house or a warehouse, the accuracy of estimating the position of the first wireless terminal 40 may deteriorate.

[0018] [Second Technical Example] In order to solve the above-described problem in the first technical example, in the second technical example, using the direction of arrival of the sound wave detected by the smart speaker 50, it is determined that the CSI has fluctuated greatly due to the movement of people or the like in the direction where the sound wave has arrived, and the position of the first wireless terminal 40 is estimated using only the CSI at the time when the sound was detected (see FIG. 12).

[0019] However, in the second technical example, all sound waves detected by the smart speaker 50 are subject to processing. The fluctuation of CSI is more affected the closer you are to the location where a person or other object has moved, and less affected the further away you are. Therefore, if there is a lot of movement or intrusion of people in a certain area, the fluctuation of CSI will be large in areas close to that location, and only the CSI that has fluctuated at that location will be acquired.

[0020] For example, when estimating the position of the first wireless terminal 40 by targeting the entire predetermined area, the position estimation device 60 needs to collect CSI data over a long period of time. Furthermore, the amount of CSI data to be processed becomes enormous, which means that creating training data takes a significant amount of time when using machine learning or similar methods. In addition, if the locations of the collected sound sources are biased, the accuracy of estimating the position of the first wireless terminal 40 may deteriorate.

[0021] Therefore, the position estimation device 60a according to one embodiment is configured to reduce the number of CSIs to be processed and to efficiently estimate the position of the first wireless terminal 40 whose location is unknown.

[0022] Figure 1 is a schematic diagram illustrating the outline of a position estimation system 10a according to one embodiment. Components that are substantially the same as those in the position estimation system 10 shown in Figure 11 are denoted by the same reference numerals.

[0023] As shown in Figure 1, a location estimation system 10a according to one embodiment includes, for example, a base station (AP) 20, a first wireless terminal 40, second wireless terminals 31-35, a smart speaker 50, and a location estimation device 60a.

[0024] For example, the second wireless terminals 31-33 are installed in room A, and their installation locations are known. The second wireless terminal 34 is installed in room B, and its installation location is known. The second wireless terminal 36 is installed in room C, and its installation location is known.

[0025] The second wireless terminal 35, base station 20, smart speaker 50, and position estimation device 60a are installed in room D, and their respective installation locations are known. Furthermore, the base station 20, smart speaker 50, and position estimation device 60a are considered to be located in substantially the same location. In addition, the smart speaker 50 may be included in and integrated with the position estimation device 60a.

[0026] Here, we assume that a person enters room A at time t1, moves to room B at time t2, moves to room C at time t3, and moves to room D at time t4.

[0027] Figure 2 illustrates the distribution of the direction of arrival of sound waves detected by the smart speaker 50. As shown in Figure 2, for example, the smart speaker 50 detects the time and direction of arrival of sound waves.

[0028] Figure 3 is a conceptual diagram showing an overview of the operation of the position estimation system 10a. The position estimation system 10a performs, for example, a clustering implementation stage that carries out clustering and a determination stage that estimates (determines) the position of the first wireless terminal 40.

[0029] In the clustering phase, the location estimation system 10a performs clustering based on the CSI used by the base station 20 and the second wireless terminals 31-35, and then performs the determination phase processing after the clustering is complete. The location estimation system 10a also uses the CSI between the base station 20 and the first wireless terminal 40, acquired at the same time, to estimate the location of the first wireless terminal 40 (the room in which the first wireless terminal 40 is installed).

[0030] As shown in Figure 3, the position estimation system 10a, for example, when a person moves between times t1 and t4, has the position estimation device 60a acquire the CSI used by the second wireless terminals 31 to 35 and the first wireless terminal 40 at predetermined time intervals (WS: Window Size).

[0031] The position estimation device 60a then classifies the CSIs according to the distribution of the sound wave's direction of arrival. Here, the position estimation device 60a classifies the CSIs, for example, into the number of rooms (4).

[0032] Next, the position estimation device 60a extracts, for example, N samples to be processed from each of the classified CSIs.

[0033] Next, the position estimation device 60a selects a predetermined number of CSIs from each of the four classified CSIs and performs CSI classification using only the selected CSIs.

[0034] Subsequently, the position estimation device 60a estimates that, for example, the first wireless terminal 40 is located at the position closest to one of the second wireless terminals 31-35 that has the highest correlation with the CSI of the first wireless terminal 40.

[0035] The second wireless terminals 31-35 are known to be installed in one of rooms A-D (they are associated with each other). In other words, the position estimation device 60a extracts CSIs in the direction of rooms A-D without bias, so it can estimate which room the first wireless terminal 40 is located in.

[0036] In this case, the direction of arrival θ of the sound wave takes a value such that 0 ≤ θ < 360 degrees (0 ≤ θ < 2π). Although θ = 0 and θ = 360 degrees are the same direction, their numerical values ​​are significantly different. Therefore, it is acceptable to set θ = 0 to the direction in which the sound wave arrives less frequently.

[0037] Next, a specific example of the configuration of the position estimation device 60a will be described. Figure 4 is a functional block diagram illustrating the functions of the position estimation device 60a. As shown in Figure 4, the position estimation device 60a includes, for example, an acquisition unit 61, a sound collection unit 500, an arrival direction detection unit 502, an arrival direction count storage unit 614, a first clustering unit 616, a determination unit 618, a sample extraction unit 620, a second clustering unit 622, and an estimation unit 624.

[0038] The acquisition unit 61 includes, for example, an antenna 600, a wireless interface (I / F) unit 602, a capture unit 604, a filtering unit 606, a CSI extraction unit 608, a preprocessing unit 610, and a CSI storage unit 612.

[0039] Antenna 600 is one or more antennas that transmit and receive wireless signals such as CSI and data. Wireless I / F unit 602 is an interface that connects antenna 600 and capture unit 604.

[0040] The capture unit 604 captures receivable wireless frames and outputs them to the filtering unit 606.

[0041] The filtering unit 606, for example, selects only Compressed Beamforming Report frames containing CSI transmitted to the base station 20 from pre-specified first wireless terminal 40 and second wireless terminals 31-35, and outputs them to the CSI extraction unit 608.

[0042] For example, if the position estimation system 10a is a wireless LAN system compliant with IEEE 802.11ac, the Compressed Beamforming Report frame does not contain the CSI directly, but rather a compressed CSI following a predetermined procedure. Figure 5 illustrates the compressed CSI. [Reference] H. Yu and T. Kim, “Beamforming transmission in IEEE 802.11ac under time-varying channels,” The Scientific World J., vol. 2014, pp. 1-11, Jul. 2014, article ID 920937.

[0043] The CSI extraction unit 608 extracts the compressed CSI from the frame captured by the capture unit 604 and outputs it to the preprocessing unit 610.

[0044] The preprocessing unit 610 performs preprocessing on the CSI data extracted by the CSI extraction unit 608. Note that there is a CSI for each subcarrier. Therefore, when the base station 20 has 3 antennas and the second wireless terminals 31-35 and the first wireless terminal 40 each have 2 antennas, there are 6 compressed CSIs, as shown in Figure 5. In other words, there are 6 × the number of subcarriers of CSIs.

[0045] Figure 6 shows the preprocessing (vector conversion of CSI) performed by the preprocessing unit 610. For example, since the compressed CSI is angular information as shown in Figure 5, the preprocessing unit 610 performs a preprocessing step of converting from polar coordinates to Cartesian coordinates, as shown in Figure 6.

[0046] Specifically, the preprocessor 610 converts the angle φ or ψ from the x-axis on the circumference of a circle with radius 1 into corresponding xy coordinates. As mentioned above, angles of 0 and 2π are the same, but they are numerically discontinuous and therefore unsuitable as input for machine learning. Therefore, the preprocessor 610 converts the angle φ or ψ to maintain continuity so that it can be treated as a numerical value on the xy plane.

[0047] The CSI storage unit 612 stores the CSI preprocessed by the preprocessing unit 610 for each WS (Window Size).

[0048] In other words, the acquisition unit 61 acquires and stores channel status information indicating the state of the propagation channels between the base station 20 and each of the second wireless terminals 31-35 and the first wireless terminal 40 at predetermined time intervals.

[0049] Furthermore, in the example shown in Figure 4, the position estimation device 60a is equipped with a sound collection unit 500 and an approach direction detection unit 502, but if a smart speaker 50 is connected, the sound collection unit 500 and the approach direction detection unit 502 do not need to be included.

[0050] The sound collection unit 500 collects sound waves arriving toward the base station 20 from each of a plurality of areas (in this case, any of rooms A to D) including the installation location of at least one of the second wireless terminals 31 to 35 at predetermined time intervals in synchronization with the acquisition unit 61, and outputs them to the direction of arrival detection unit 502.

[0051] The direction of arrival detection unit 502 detects the direction of arrival of each sound wave collected by the sound collection unit 500 with a predetermined resolution (for example, a resolution of 10 degrees: see Figure 7) and outputs it to the direction of arrival count storage unit 614.

[0052] The arrival direction count storage unit 614 stores the arrival direction of the sound wave detected by the arrival direction detection unit 502 (for example, with a resolution of 10 degrees) and the number of arrivals for each arrival direction of the sound wave, for each WS (Window Size). Figure 7 is an example of the arrival direction of the sound wave and the number of arrivals for each arrival direction of the sound wave stored by the arrival direction count storage unit 614.

[0053] The first clustering unit 616 performs clustering based on the sound wave arrival directions accumulated by the arrival direction count accumulation unit 614 and the number of arrivals for each sound wave arrival direction. The first clustering unit 616 may remove an arrival direction if the total number of sound wave arrivals is 0 or less than or equal to a predetermined threshold.

[0054] For example, the first clustering unit 616 uses a vector of the number of sound wave arrivals for each angle of arrival direction to perform clustering using k-means, x-means, spectral clustering, (Baysian) Gaussian mixture models clustering, etc., as shown in Figure 8.

[0055] The number of clusters in this case can be, for example, the number of rooms whose direction of arrival needs to be identified. For example, in the table on the far left of Figure 7, the vector is (0,0,···,10,20,···,0).

[0056] In this case, angles between 280 and 290 degrees are assigned a value of 10, and angles between 290 and 300 degrees are assigned a value of 20. Also, in this case, 0 degrees and 360 degrees are treated as different values ​​even though they are in the same direction, so the following preprocessing may be performed.

[0057] Regarding the angle, an offset value may be given such that the direction from which sound waves do not arrive is set to 0. For example, in the configuration shown in Figure 1, it is expected that no sound waves will come from the 45-degree direction, so 45 degrees may be subtracted from the angle. • The direction of arrival of all sound waves may be counted, and an offset may be applied such that the direction with the fewest number of sound wave arrivals within a certain angle (e.g., a 45-degree range) is set to 0.

[0058] The first clustering unit 616 may reduce the number of dimensions using methods such as independent component analysis (PCA) or t-SNE (t-distribution stochastic nearest neighbor analysis) before performing clustering.

[0059] Figure 8 illustrates the clustering performed by the first clustering unit 616. The first clustering unit 616, for example, plots the number of occurrences of angles n to n+10 degrees on the horizontal axis and the number of occurrences of angles m to m+10 degrees on the vertical axis, and creates clusters for time t1 to t2, time t2 to t3, time t3 to t4, and time t4 onwards.

[0060] Furthermore, the first clustering unit 616 is not limited to two-dimensional clusters, but may also create multi-dimensional clusters. Also, if the number of dimensions is reduced by principal component analysis or the like, the axes will represent something other than angles, and the number of axes will also be reduced. In addition, the first clustering unit 616 may perform clustering using features where the horizontal axis is angle and the vertical axis is the probability density for each angle.

[0061] Based on the direction of arrival of each sound wave detected by the direction of arrival count storage unit 614, the number of sound waves arriving in each direction, and the installation positions of the second wireless terminals 31 to 35, the determination unit 618 determines the timing for extracting the CSI such that the number of corresponding sound wave directions detected by the direction of arrival count storage unit 614 and the aforementioned regions (i.e., rooms A to D) is maximized. Figure 9 is a diagram illustrating the timing for extracting the CSI determined by the determination unit 618.

[0062] Based on the timing determined by the decision unit 618, the sample extraction unit 620 extracts a portion of the CSI of each of the second wireless terminals 31-35 and the first wireless terminal 40 acquired by the acquisition unit 61 as samples to be processed, and outputs them to the second clustering unit 622.

[0063] For example, the sample extraction unit 620 extracts only the CSIs corresponding to each timing (the time determined by the determination unit 618) shown in Figure 9 from each cluster shown in Figure 8. In other words, the sample extraction unit 620 extracts samples from all the CSIs acquired by the acquisition unit 61 that show as little bias as possible in the changes in radio wave propagation conditions.

[0064] The sample extraction unit 620 extracts, for example, one or more samples from each cluster. The number of samples may be predetermined, or it may be 1 / N (where N is a predetermined value). Furthermore, the sample extraction unit 620 may use the smallest or largest number among the above two options as the sample. In addition, the sample extraction unit 620 may randomly select and extract samples from each class, extract the sample closest to the center, or select and extract samples from the center and periphery of the cluster.

[0065] The second clustering unit 622 performs clustering of CSIs using the timing determined by the determination unit 618 with respect to the incoming direction, and outputs the result to the estimation unit 624.

[0066] The estimation unit 624 estimates, for example, that the first wireless terminal 40 is located at the position closest to the second wireless terminal with the highest correlation among the second wireless terminals 31-35 extracted by the sample extraction unit 620, based on the CSI of the first wireless terminal 40 extracted by the sample extraction unit 620.

[0067] Alternatively, the estimation unit 624 may use the CSI of the first wireless terminal 40 at the time determined by the determination unit 618 to determine which region (room) the feature belongs to, and then determine which of the nearest second wireless terminals 31 to 35 is located therein to estimate the location of the first wireless terminal 40.

[0068] The position estimation device 60a may use the CSI as is, or it may calculate and use temporal statistics (for example, the standard deviation of the change, the difference between the maximum and minimum values, the difference between the 25th and 75th percentile values).

[0069] Thus, the position estimation system 10a according to one embodiment can reduce the number of CSIs used and efficiently estimate the position of a wireless terminal whose location is unknown.

[0070] Furthermore, each function of the position estimation device 60a may be partially or entirely composed of hardware such as a PLD (Programmable Logic Device) or FPGA (Field Programmable Gate Array), or it may be composed of a program executed by a processor such as a CPU.

[0071] For example, the position estimation device 60a can be implemented using a computer and a program, and the program can be recorded on a storage medium or provided via a network. Furthermore, each function of the position estimation device 60a may also be possessed by other devices of the position estimation system 10a.

[0072] Figure 10 shows an example of the hardware configuration of a position estimation device 60a according to one embodiment. As shown in Figure 10, for example, the position estimation device 60a has an input unit 70, an output unit 71, a communication unit 72, a CPU 73, a memory 74, and an HDD 75 connected via a bus 76, and is equipped with computer functions. The position estimation device 60a is also configured to be able to input and output data to and from a computer-readable storage medium 77.

[0073] The input unit 70 is, for example, a keyboard and mouse. The output unit 71 is, for example, a display device such as a display that outputs images. The communication unit 72 is, for example, a wireless network interface and may also have the function of an output unit that outputs data to the outside.

[0074] The CPU 73 controls each component of the position estimation device 60a and performs predetermined processing. The memory 74 and HDD 75 are storage units that store data, etc.

[0075] The storage medium 77 is capable of storing programs and the like that which execute the functions of the position estimation device 60a. Note that the architecture of the position estimation device 60a is not limited to the example shown in Figure 10. [Explanation of Symbols]

[0076] 10,10a...Position estimation system, 20...Base station, 31-35...Second wireless terminal, 40...First wireless terminal, 50...Smart speaker, 60,60a...Position estimation device, 61...Acquisition unit, 70...Input unit, 71...Output unit, 72...Communication unit, 73...CPU, 74...Memory, 75...HDD, 76...Bus, 77...Storage medium, 500...Sound collection unit, 502... • Direction detection unit, 600... Antenna, 602... Wireless I / F unit, 604... Capture unit, 606... Filtering unit, 608... CSI extraction unit, 610... Preprocessing unit, 612... CSI storage unit, 614... Direction count storage unit, 616... First clustering unit, 618... Determination unit, 620... Sample extraction unit, 622... Second clustering unit, 624... Estimation unit

Claims

1. The location of the first wireless terminal is unknown, Multiple second wireless terminals whose installation locations are known, A base station that communicates wirelessly with the first wireless terminal and each of the multiple second wireless terminals. Equipped with, In a location estimation system for estimating the location of the first wireless terminal, The base station and an acquisition unit that acquires channel status information indicating the state of propagation channels between each of the multiple second wireless terminals and the first wireless terminal at predetermined time intervals, A sound collection unit collects sound waves arriving toward the base station from each of a plurality of areas including the installation location of the second wireless terminal at predetermined time intervals in synchronization with the acquisition unit, An arrival direction detection unit detects the direction of arrival of each sound wave collected by the sound collection unit with a predetermined resolution, A determination unit determines the timing for extracting channel state information based on the direction of arrival of each sound wave detected by the direction of arrival detection unit, the number of sound waves arriving in each direction, and the installation position of the second wireless terminal, such that the number of corresponding regions between the direction of arrival of the sound waves detected by the direction of arrival detection unit is maximized. A sample extraction unit extracts a portion of the channel state information of each of the second wireless terminals and the first wireless terminal acquired by the acquisition unit, based on the timing determined by the determination unit, as samples to be processed. Based on the channel state information of the first wireless terminal extracted by the sample extraction unit, the estimation unit estimates that the first wireless terminal is located closest to the second wireless terminal with the highest correlation among the multiple second wireless terminals extracted by the sample extraction unit. A position estimation system characterized by having the following features.

2. In a position estimation device that estimates the location of a first wireless terminal whose location is unknown, A base station and an acquisition unit that acquires channel status information indicating the state of the propagation channel between each of the multiple second wireless terminals whose installation locations are known and the first wireless terminal at predetermined time intervals, A sound collection unit collects sound waves arriving toward the base station from each of a plurality of areas including the installation location of the second wireless terminal at predetermined time intervals in synchronization with the acquisition unit, An arrival direction detection unit detects the direction of arrival of each sound wave collected by the sound collection unit with a predetermined resolution, A determination unit determines the timing for extracting channel state information based on the direction of arrival of each sound wave detected by the direction of arrival detection unit, the number of sound waves arriving in each direction, and the installation position of the second wireless terminal, such that the number of corresponding regions between the direction of arrival of the sound waves detected by the direction of arrival detection unit is maximized. A sample extraction unit extracts a portion of the channel state information of each of the second wireless terminals and the first wireless terminal acquired by the acquisition unit, based on the timing determined by the determination unit, as samples to be processed. Based on the channel state information of the first wireless terminal extracted by the sample extraction unit, the estimation unit estimates that the first wireless terminal is located closest to the second wireless terminal with the highest correlation among the multiple second wireless terminals extracted by the sample extraction unit. A position estimation device characterized by having the following features.

3. In a location estimation method for estimating the location of a first wireless terminal whose location is unknown, An acquisition step of acquiring channel status information indicating the state of the propagation channel between the base station and each of the multiple second wireless terminals whose installation locations are known, and the first wireless terminal, at predetermined time intervals; A sound collection step of collecting sound waves arriving toward the base station from each of a plurality of areas including the installation location of the second wireless terminal at predetermined time intervals, An arrival direction detection step is performed to detect the direction of arrival of each sound wave collected by the sound collection step with a predetermined resolution, A determination step determines the timing for extracting channel state information, based on the direction of arrival of each sound wave detected in the direction of arrival detection step, the number of sound waves arriving in each direction, and the installation position of the second wireless terminal, such that the number of corresponding sound wave directions detected in the direction of arrival detection step and the region is maximized. A sample extraction step is performed to extract a portion of the channel status information of each of the second wireless terminals and the first wireless terminal acquired in the acquisition step as samples to be processed, based on the timing determined in the determination step. An estimation step is performed to estimate that the first wireless terminal is located at the position closest to the second wireless terminal, which has the highest correlation among the multiple second wireless terminals extracted in the sample extraction step, based on the channel state information of the first wireless terminal extracted in the sample extraction step. A method for estimating a position, characterized by including the following:

4. In a position estimation program executed by a position estimation device that estimates the location of a first wireless terminal whose location is unknown, A base station and an acquisition unit that acquires channel status information indicating the state of the propagation channel between each of the multiple second wireless terminals whose installation locations are known and the first wireless terminal at predetermined time intervals, A sound collection unit collects sound waves arriving toward the base station from each of a plurality of areas including the installation location of the second wireless terminal at predetermined time intervals in synchronization with the acquisition unit, An arrival direction detection unit detects the direction of arrival of each sound wave collected by the sound collection unit with a predetermined resolution, A determination unit determines the timing for extracting channel state information based on the direction of arrival of each sound wave detected by the direction of arrival detection unit, the number of sound waves arriving in each direction, and the installation position of the second wireless terminal, such that the number of corresponding regions between the direction of arrival of the sound waves detected by the direction of arrival detection unit is maximized. A sample extraction unit extracts a portion of the channel state information of each of the second wireless terminals and the first wireless terminal acquired by the acquisition unit, based on the timing determined by the determination unit, as samples to be processed. Based on the channel state information of the first wireless terminal extracted by the sample extraction unit, the estimation unit estimates that the first wireless terminal is located closest to the second wireless terminal with the highest correlation among the multiple second wireless terminals extracted by the sample extraction unit. A position estimation program for causing a computer to function as one of the parts of the position estimation device having the above-mentioned features.

Images