Position estimation system

Abstract

To provide a position estimation system capable of estimating a position of a radio device quickly and accurately.SOLUTION: A position estimation system comprises acquisition means, first estimation means, and second estimation means. The acquisition means acquires first communication state data between a first radio device and each of a plurality of second radio devices. The first estimation means estimates a position of the first radio device by using the first communication state data. The second estimation means estimates the position of the first radio device by using an output of a learned model when the first communication state data is input to the learned model. The first communication state data is acquired at frequencies that differ depending on channels. Processing by the first or second estimation means is selected on the basis of the first communication state data. The learned model is generated by using learned data including second communication state data.SELECTED DRAWING: Figure 4

Description

[Technical Field]

[0001] FIELD OF THE INVENTION Embodiments of the present invention relate to a position estimation system. [Background technology]

[0002] In recent years, the position of a wireless device having a function of performing wireless communication has been estimated based on the reception strength of a signal (radio wave) observed between the wireless device and another wireless device, for example.

[0003] Specifically, if the wireless device is a personal computer (hereinafter referred to as PC) or the like, estimating the location of the PC (employee) is thought to be useful in cases such as determining the location of people in an office building and performing efficient air conditioning control.

[0004] However, there is a demand for the location of the wireless device to be estimated quickly and with high accuracy. [Prior art documents] [Patent documents]

[0005] [Patent Document 1] Patent No. 6459362 Summary of the Invention [Problem to be solved by the invention]

[0006] Therefore, an object of the present invention is to provide a position estimation system that can estimate the position of a wireless device quickly and with high accuracy. [Means for solving the problem]

[0007] A position estimation system according to an embodiment includes an acquisition means, a first estimation means, and a second estimation means. The acquisition means acquires first communication status data related to communication status between a first wireless device and each of a plurality of second wireless devices. The first estimation means estimates the position of the first wireless device using the acquired first communication status data. The second estimation means estimates the position of the first wireless device using the output of a trained model when the acquired first communication status data is input to the trained model. The first communication status data is acquired at different frequencies depending on channels connecting the first wireless device and each of the plurality of second wireless devices. The processing by the first estimation means or the processing by the second estimation means is selected based on the first communication status data. The trained model is generated using training data including second communication status data related to communication status between the first wireless device and each of the plurality of second wireless devices when the first wireless device is located at a predetermined position. [Brief explanation of the drawings]

[0008] [Figure 1] FIG. 1 is a diagram showing an example of the configuration of a position estimation system according to an embodiment. [Figure 2] FIG. 2 is a block diagram showing an example of the functional configuration of a terminal device. [Figure 3] FIG. 2 is a block diagram showing an example of the functional configuration of a base station. [Figure 4] FIG. 2 is a block diagram showing an example of the functional configuration of a position estimation device. [Figure 5] FIG. 1 is a diagram showing an example of a system configuration of a position estimation device. [Figure 6] 10 is a flowchart showing an example of a processing procedure of a position estimation device. [Figure 7] FIG. 10 is a diagram for explaining an example of a method for acquiring reception strength. [Figure 8] FIG. 10 is a diagram for explaining another example of a method for acquiring reception strength. [Figure 9] FIG. 10 is a diagram showing yet another example of a method for acquiring reception strength. [Figure 10] FIG. 10 is a diagram showing yet another example of a method for acquiring reception strength. [Figure 11] 5 is a diagram for explaining the accuracy of the position of the terminal device estimated by executing the first estimation process. FIG. [Figure 12] FIG. 10 is a diagram for explaining an overview of a trained model used in the second estimation process. [Figure 13] FIG. 10 is a diagram for explaining an overview of a trained model used in the second estimation process. DETAILED DESCRIPTION OF THE INVENTION

[0009] Hereinafter, embodiments will be described with reference to the drawings. 1 shows an example of the configuration of a position estimation system according to this embodiment. The position estimation system 1 shown in FIG.

[0010] The terminal device 10 is an example of a wireless device having a function of performing wireless communication, and includes, for example, a notebook personal computer (notebook PC) used by a user (such as a corporate employee) who can move within a predetermined space such as an office building, but may also be other electronic devices such as a smartphone, or a stationary device such as a desktop personal computer. In this embodiment, a situation is assumed in which the terminal device 10 moves within a predetermined space together with the user.

[0011] The plurality of base stations 20 are examples of wireless devices having the function of performing wireless communication, and are arranged in a predetermined space such as the above-mentioned office building. The terminal device 10 performs wireless communication with each of the plurality of base stations 20 via a communication channel.

[0012] Here, as described above, grasping the location of the terminal device 10 used by a user moving within a specified space (for example, a company employee moving within an office building) is equivalent to grasping the location (people flow) of people within the space, and by utilizing data regarding the location (people flow) of the people, efficient air conditioning control, etc. within the space can be realized.

[0013] In this case, considering that the communication status between the terminal device 10 and each of the plurality of base stations 20 (hereinafter referred to as the communication status at the base station 20) depends on the distance between the terminal device 10 and the base station 20, it is considered possible to estimate the location of the terminal device 10 based on the communication status at each of the plurality of base stations 20. Note that, in order to estimate the location of the terminal device 10 with high accuracy, it is preferable to use the communication status at a larger number of base stations 20.

[0014] However, the base station 20 connected to the terminal device 10 via a communication channel is switched according to the movement of the terminal device 10, and the terminal device 10 basically performs wireless communication with only one base station 20. Therefore, in order to grasp the communication status of a base station 20 other than the base station 20 that performs wireless communication with the terminal device 10 via a communication channel, scanning with an observation channel different from the communication channel (off-channel scanning) is required.

[0015] On the other hand, because essential communications cannot be performed during scanning on observation channels, scanning on these channels too frequently can degrade network performance. For this reason, it is desirable to scan observation channels less frequently (i.e., to reduce the proportion of time spent scanning on observation channels).

[0016] However, if scanning of observation channels is performed infrequently as described above, it takes time to grasp the communication situations at a sufficient number of base stations 20 to enable highly accurate estimation of the position of the terminal device 10, making it impossible to speed up position estimation of the terminal device 10. Furthermore, if the position of the terminal device 10 is estimated in a state in which the communication situations at a sufficient number of base stations 20 to enable highly accurate estimation of the position of the terminal device 10 are not grasped, the accuracy of the position estimation may decrease.

[0017] Therefore, the position estimation device 30 in this embodiment provides a mechanism for estimating the position of the terminal device 10 quickly and with high accuracy based on communication status information indicating the communication status between the terminal device 10 and each of multiple base stations 20.

[0018] Note that the communication status information in this embodiment will be described as being the reception strength of packets (signals) observed between the terminal device 10 and each of the plurality of base stations 20, such as a received signal strength indicator (RSSI).

[0019] In this embodiment, the terminal device 10 and each of the base stations 20 perform wireless communication based on the wireless LAN defined in IEEE 802.11. In this case, the terminal device 10 corresponds to a station (STA), and each of the base stations 20 corresponds to an access point (AP).

[0020] IEEE 802.11k defines a standard for scanning observation channels other than the communication channel used by the base station 20 for wireless communication in order to know the radio wave usage status. This allows, for example, multiple other base stations 20 to observe communication between the terminal device 10 and a specific base station 20 (i.e., to grasp the reception strength at the multiple other base stations 20).

[0021] In this embodiment, it is assumed that a plurality of base stations 20 are arranged (installed) at a relatively high density in a predetermined space, such as an office building, and that even if a terminal device 10 (or a user using the terminal device 10) moves within the predetermined space, the reception strength at each of the plurality of base stations 20 can be measured using a communication channel and an observation channel. Furthermore, as described above, the reception strength at a base station 20 (i.e., a portion of the plurality of base stations 20) that performs wireless communication with the terminal device 10 via a communication channel can be measured (acquired) relatively frequently using the communication channel. On the other hand, the reception strength at a base station 20 other than the base station 20 that performs wireless communication with the terminal device 10 via a communication channel (i.e., the other portion of the plurality of base stations 20) is measured (acquired) less frequently than the communication channel using an observation channel.

[0022] Fig. 2 is a block diagram showing an example of the functional configuration of the terminal device 10 shown in Fig. 1. As shown in Fig. 2, the terminal device 10 includes a transmitting / receiving unit 11 and a control unit 12.

[0023] The transmitting / receiving unit 11 is a functional unit for transmitting and receiving various packets to and from each of the plurality of base stations 20. The control unit 12 is a functional unit for controlling the operation of the terminal device .

[0024] Fig. 3 is a block diagram showing an example of the functional configuration of the base station 20 shown in Fig. 1. As shown in Fig. 3, the base station 20 includes a transceiver 21, a control unit 22, and an output unit 23.

[0025] The transmitter / receiver 21 is a functional unit for transmitting and receiving various packets to and from the terminal device 10. The controller 22 is a functional unit for controlling the operation of the base station 20. The output unit 23 is a functional unit for outputting the reception strength of packets observed using the above-mentioned communication channel or observation channel (reception strength at the base station 20).

[0026] Fig. 4 is a block diagram showing an example of the functional configuration of the position estimation device 30 shown in Fig. 1. As shown in Fig. 4, the position estimation device 30 includes an acquisition unit 31, a selection unit 32, a first estimation unit 33, a second estimation unit 34, and a machine learning unit 35.

[0027] The position estimation device 30 is connected to each of the plurality of base stations 20 so as to perform wired or wireless communication with the base stations 20. The acquisition unit 31 acquires, for example, the reception strength at the base station 20 output from each of the plurality of base stations 20 (from an output unit 23 included in the base station 20). Note that the frequency at which the reception strength is output from each of the plurality of base stations 20 (i.e., the frequency at which the reception strength is acquired by the acquisition unit 31) differs depending on whether the reception strength is the reception strength of packets observed using a communication channel or the reception strength of packets observed using an observation channel.

[0028] The selector 32 selects the first or second estimation process based on the reception strength acquired by the acquirer 31.

[0029] The first estimation unit 33 executes the first estimation process when the first estimation process is selected by the selection unit 32. The first estimation process corresponds to a process of estimating the position of the terminal device 10 using the reception strength acquired by the acquisition unit 31.

[0030] The second estimation unit 34 executes the second estimation process when the selection unit 32 selects the second estimation process. The second estimation unit 34 holds a trained model, which will be described later. The second estimation process corresponds to a process of estimating the position of the terminal device 10 using the output of the trained model when the reception strength acquired by the acquisition unit 31 is input to the trained model.

[0031] The machine learning unit 35 uses training data including, for example, the reception strength at each of multiple base stations 20 measured when the terminal device 10 is in the same location to train the above-mentioned trained model (i.e., generate the trained model).

[0032] Fig. 5 shows an example of the system configuration of the position estimation device 30 shown in Fig. 4. The position estimation device 30 includes a CPU 301, a nonvolatile memory 302, a RAM 303, a communication device 304, and the like.

[0033] The CPU 301 is a processor for controlling the operation of various components within the position estimation device 30. The CPU 301 may be a single processor or may be configured with multiple processors. The CPU 301 executes various programs loaded from the non-volatile memory 302 to the RAM 303. The programs executed by the CPU 301 include an operating system (OS) and various application programs.

[0034] The nonvolatile memory 302 is a storage medium used as an auxiliary storage device. The RAM 303 is a storage medium used as a main storage device. Although only the nonvolatile memory 302 and the RAM 303 are shown in Fig. 5, the position estimation device 30 may include other storage devices such as a hard disk drive (HDD) and a solid state drive (SSD).

[0035] The communication device 304 is a device configured to perform wired or wireless communication.

[0036] 4 may be realized by causing the CPU 301 (i.e., the computer of the position estimation device 30) to execute a predetermined program, that is, by software. This program may be stored in a computer-readable storage medium and distributed, or may be downloaded to the position estimation device 30 via a network. Some or all of the units 31 to 35 may be realized by dedicated hardware or the like, or by a combination of software and hardware.

[0037] An example of the processing procedure of the position estimation device 30 in this embodiment will be described below with reference to the flowchart of FIG.

[0038] First, the acquisition unit 31 acquires the reception strength at each of the plurality of base stations 20 (step S1).

[0039] In this embodiment, the terminal device 10 is movable within a predetermined space, and the terminal device 10 performs wireless communication via a communication channel with one base station 20 (hereinafter referred to as the first base station 20) among the multiple base stations 20 depending on the location of the terminal device 10. In this case, the reception strength at the first base station 20 is measured frequently using the communication channel.

[0040] Furthermore, the reception frequency at each of the plurality of base stations 20 other than the first base station 20 (hereinafter referred to as second base station 20) is measured at a low frequency using an observation channel different from the communication channel.

[0041] That is, in step S1, the reception strength at the first base station 20 and the reception strength at each of the plurality of second base stations 20 are acquired at different frequencies (timings). In this way, the reception strength at each of the plurality of base stations 20 (the first base station 20 and the plurality of second base stations 20) acquired in step S1 is stored (accumulated) inside the position estimation device 30 each time the reception strength is acquired, in association with the time when the reception strength was acquired and an identifier for identifying the base station 20.

[0042] In an environment such as an office building, which is assumed as the predetermined space in this embodiment, a plurality of base stations 20 (first and second base stations 20) are arranged (installed) at a relatively high density. Therefore, in such an environment, even if the frequency of acquiring the reception strength of each second base station 20 is low, it can be expected that the intervals at which the reception strength of any of the second base stations 20 is acquired (a packet is observed in any of the observation channels) will be short.

[0043] The above-mentioned reception strength is acquired by a method unique to the manufacturer of the plurality of base stations 20 (APs), a method based on SNMP (Simple Network Management Protocol), or the like.

[0044] The following describes a method for acquiring the reception strength of packets observed in the above-mentioned observation channel. Here, we will explain using the above-mentioned IEEE 802.11 wireless LAN as an example, and assume that the standard for scanning observation channels is defined in IEEE 802.11k.

[0045] FIG. 7 is a diagram for explaining an example of a method for acquiring the reception intensity using the above-mentioned observation channel.

[0046] In Figure 7, the second base station 20 observes packets between the terminal device 10 and the first base station 20 by scanning an observation channel (off-channel) 402, which is different from a communication channel (on-channel) 401 through which the first base station 20 performs wireless communication with the terminal device 10, for a short period of time, and measures the reception strength of the packets (reception strength at the second base station 20).

[0047] The reception strength at the second base station 20 measured in this manner is output from the second base station 20 to the position estimation device 30, and is acquired by the position estimation device 30 (acquisition unit 31).

[0048] It should be noted that the second base station 20 can perform wireless communication with terminal devices other than the terminal device 10 via the communication channel 403, for example, during times other than when scanning the observation channel 402.

[0049] Although an example of a method for acquiring reception strength using an observation channel has been described above with reference to Fig. 7, the reception strength may also be acquired based on an instruction from the first base station 20 (AP) to the terminal device 10 (STA). In this case, the terminal device 10 operates to measure reception strength in response to a Beacon Measurement Request transmitted from the first base station 20, and return the measured reception strength as a Beacon Measurement Report.

[0050] Fig. 8 shows another example of a method for acquiring reception strength using an observation channel. Note that this example assumes that the preset reception strength measurement mode (Measurement Mode) is active-all-ch or active-ch-rpt.

[0051] In FIG. 8, first, the first base station 20 instructs (transmits) a Beacon Measurement Request to the terminal device 10 via the communication channel 401.

[0052] Next, the terminal device 10 transmits a Probe Request to the second base station 20 on the observation channel (all or designated channels) 402, thereby observing (receiving) a Probe Response transmitted from the second base station 20.

[0053] The terminal device 10 measures the reception strength of the thus observed Probe Response (packet) as the reception strength at the second base station 20, and transmits the measured reception strength to the first base station 20 as a Beacon Measurement Report.

[0054] The reception strength thus transmitted from the terminal device 10 to the first base station 20 is output from the first base station 20 to the position estimation device 30, and is acquired by the position estimation device 30 (acquisition unit 31).

[0055] 9 shows another example of a method for acquiring the received signal strength using an observation channel, where the Measurement Mode is assumed to be Passive.

[0056] In FIG. 9, first, the first base station 20 instructs (transmits) a Beacon Measurement Request to the terminal device 10 via the communication channel 401.

[0057] Next, the terminal device 10 performs passive scanning on the observation channels (all or specified channels) 402, thereby observing (receiving) Beacons periodically transmitted from the second base station 20, for example.

[0058] The terminal device 10 measures the reception strength of the Beacon (packet) observed in this way as the reception strength at the second base station 20, and transmits the observed reception strength to the first base station 20 as a Beacon Measurement Report.

[0059] The reception strength thus transmitted from the terminal device 10 to the first base station 20 is output from the first base station 20 to the position estimation device 30, and is acquired by the position estimation device 30 (acquisition unit 31).

[0060] In the example shown in FIG. 9, unlike the example shown in FIG. 8, the terminal device 10 can observe the Beacon transmitted from the second base station 20 without transmitting a Probe Request.

[0061] Here, the explanation has been given with reference to Figures 8 and 9, but for example, in a case where the terminal device 10 manages a table that holds (data on) the reception strength at each of multiple base stations 20 (first and second base stations 20) that have already been measured using communication channels and observation channels, as shown in Figure 10, the terminal device 10 may operate so as to transmit the reception strength held in the table to the first base station 20 as a Beacon Measurement Report in response to a Beacon Measurement Request sent from the first base station 20 (i.e., to compile the reception strength results stored in the terminal device 10 and return them to the second base station 20).

[0062] In this way, the reception strength transmitted from the terminal device 10 to the first base station 20 is output from the first base station 20 to the position estimation device 30, and is acquired by the position estimation device 30 (acquisition unit 31).

[0063] It should be noted that the operation described with reference to FIG. 10 is performed when the Measurement Mode is beacon-table.

[0064] Although the position estimation device 30 (acquisition unit 31) in this embodiment can acquire the reception strength at each of the plurality of base stations 20 as described above, the acquisition of the reception strength may be realized by other methods.

[0065] Next, the position estimation device 30 executes a process of estimating the position of the terminal device 10 using the reception strength at each of the plurality of base stations 20, for example, at a predetermined timing.

[0066] One possible method of estimating location using reception strength is to apply triangulation using reception strength measured (observed) at three points, but by increasing the number of measurement points for reception strength, it is possible to improve the accuracy of location estimation (i.e., reduce the error term).

[0067] Here, the reception strength at each of the multiple second base stations 20 other than the first base station 20 that performs wireless communication with the terminal device 10 via the communication channel can be obtained by scanning the observation channel (observing packets in the observation channel) as described above.

[0068] However, in this embodiment, for convenience, one terminal device 10 is described, but it is assumed that other terminal devices other than the terminal device 10 exist in a specific space such as the office building mentioned above, and each of the multiple second base stations 20 performs wireless communication with the other terminal devices via a communication channel.

[0069] In this case, the second base station 20 cannot perform scanning on the observation channel while simultaneously performing wireless communication via the communication channel. Therefore, if the time spent scanning on the observation channel increases, the time spent by the second base station 20 performing wireless communication via the communication channel will decrease, which may result in a decrease in network performance.

[0070] For this reason, it is preferable to make the ratio of the time spent scanning the observation channel to the time spent performing wireless communication via the communication channel as small as possible, but in such a case, the frequency with which reception strength is obtained by scanning the observation channel will decrease.

[0071] In other words, when estimating the position of the terminal device 10 using the reception strength at each of multiple base stations 20 (reception strength at multiple points), it may take some time to obtain reception strength at a sufficient number of base stations 20 (measurement points) to estimate the position of the terminal device 10 with high accuracy.

[0072] Therefore, in this embodiment, the selection unit 32 holds a condition that is set in advance based on, for example, whether the reception strength has been acquired to such an extent that the position of the terminal device 10 can be estimated with high accuracy. The estimation process selection condition held in the selection unit 32 corresponds to a condition (hereinafter referred to as the estimation process selection condition) for selecting the first estimation process (first estimation unit 33) or the second estimation process (second estimation unit 34), and the selection unit 32 determines whether the estimation process selection condition is satisfied (step S2).

[0073] An example of the estimation process selection condition used in the process of step S2 will be described below.

[0074] First, as described above, in order to estimate the location of the terminal device 10 with high accuracy, it is considered necessary for the terminal device 10 to be in a predetermined location during a period in which the reception strength at a sufficient number of base stations 20 is expected to be measured (acquired).

[0075] For this reason, the above-mentioned estimation process selection conditions include, for example, a condition (hereinafter referred to as the first condition) that "the terminal device 10 is at a predetermined position for a predetermined period of time (i.e., the time from when the terminal device 10 moved to the present is equal to or longer than a certain time)." Note that "the terminal device 10 is at a predetermined position" includes a case where the movement distance is short enough that the terminal device 10 can be considered not to be moving.

[0076] Here, assuming that the terminal device 10 is moving within a specified space (e.g., an office building), the reception strength (i.e., the reception strength of the packet observed in the communication channel) at the first base station 20 that performs wireless communication with the terminal device 10 via the communication channel varies depending on the movement of the terminal device 10 (i.e., the distance between the terminal device 10 and the first base station 20).

[0077] According to this, the timing at which the terminal device 10 has moved (i.e., whether or not the terminal device 10 has moved) can be determined based on whether the change in the reception strength of packets observed in the communication channel among the reception strengths stored inside the position estimation device 30 is greater than a predetermined value (i.e., the reception strength has increased or decreased by more than a certain value).

[0078] The timing at which the terminal device 10 has moved may be determined based on whether the amount of change in the reception strength observed in a predetermined observation channel is equal to or greater than a certain value.

[0079] Furthermore, when the terminal device 10 (for example, a notebook PC) is started up in a predetermined space, the reception strength is acquired immediately after the terminal device 10 is started up (that is, the reception strength at the position of the terminal device 10 is newly measured). This situation in which the reception strength is newly measured can be considered the same as when the terminal device 10 has moved.

[0080] For this reason, if the interval between acquisition of the reception strength of packets observed on the same channel (communication channel or observation channel) among the reception strengths stored inside the position estimation device 30 is equal to or longer than a certain time, it may be determined that this is the timing when the terminal device 10 has moved (started up).

[0081] Here, the case where the terminal device 10 is started up has been described, but the same applies to a case where an already started terminal device 10 enters a predetermined space, etc.

[0082] Furthermore, when the terminal device 10 moves within a predetermined space, the base station 20 to which the terminal device 10 is connected via a communication channel (that is, which performs wireless communication) may change (be switched).

[0083] According to this, the timing at which the terminal device 10 has moved may be determined based on, for example, a change in the base station 20 (AP) to which the terminal device 10 is connected via a communication channel.

[0084] Furthermore, in order to estimate the position of the terminal device 10 with high accuracy, it is preferable that the reception strengths of a sufficient number of base stations 20 be acquired.

[0085] For this reason, the estimation process selection conditions described above may include, for example, a condition that "the number of base stations 20 whose reception strength has been acquired is equal to or greater than a certain number" (hereinafter referred to as a second condition). Note that the number of base stations 20 whose reception strength has been acquired can be acquired by referring to the reception strength stored inside the position estimation device 30 and counting the number of base stations 20 whose reception strength has been acquired during the period from the time when the terminal device 10 moved to the present.

[0086] The estimation process selection condition may be at least one of the first and second conditions, or may be a condition obtained by partially modifying the first or second condition.

[0087] Furthermore, the estimation process selection condition may be a condition that combines the first and second conditions. Specifically, according to the estimation process selection condition that combines the first and second conditions, it is possible to determine whether or not the reception strengths of a sufficient number of base stations 20 have been acquired while the terminal device 10 is in a predetermined position.

[0088] In addition, the estimation process selection conditions in this embodiment may be conditions set based on viewpoints such as whether the reception strength has been acquired to the extent that the position of the terminal device 10 can be estimated with high accuracy, as described above, and are not limited to those described here.

[0089] If it is determined that the estimation process selection condition is met (YES in step S2), the selector 32 selects the first estimation process (first estimation unit 33) (step S3).

[0090] After the process of step S3 is executed, the first estimation unit 33 executes the first estimation process (step S4).

[0091] On the other hand, if it is determined that the estimation process selection condition is not satisfied (NO in step S2), the selector 32 selects the second estimation process (second estimation unit 34) (step S5).

[0092] After the process of step S5 is executed, the second estimation unit 34 executes the second estimation process in step S4.

[0093] According to the processing shown in FIG. 6 described above, for example, when the terminal device 10 continues to stay in the same position for a certain period of time or more and the reception strength at a sufficient number of base stations 20 is acquired (observed), the first estimation processing is executed, and when the reception strength at a sufficient number of base stations 20 is not acquired (observed), for example, immediately after the terminal device 10 moves, the second estimation processing is executed (i.e., the first and second estimation processing are switched based on the estimation processing selection conditions).

[0094] In this embodiment, it is possible to select the first or second estimation process (switch between the first and second estimation processes) based on the reception strength at each of the multiple base stations 20 stored inside the position estimation device 30. However, when the reception strength as described in FIG. 10 is acquired in step S1, for example, the acquired reception strength may include reception strength measured before the timing when the terminal device 10 moved in the past. In this case, in this embodiment, since the second estimation process is selected when the terminal device 10 moves as described above (that is, the first estimation process is switched to the second estimation process), it is possible to store information indicating the timing when the estimation process was switched inside the position estimation device 30, and to use only reception strength acquired after that timing (that is, reception strength after the terminal device 10 moved) when selecting the first or second estimation process. Here, the reception strength used when selecting the first or second estimation process has been described, but the same applies to reception strength used in the first and second estimation processes described later.

[0095] The first and second estimation processes in this embodiment will be described below. First, the first estimation process is executed when the terminal device 10 has remained in the same position for a sufficiently long time as described above, and the reception strengths of, for example, all base stations 20 located in the vicinity of the position (i.e., the base stations 20 near the terminal device 10) have been acquired. In the first estimation process, of the reception strengths stored inside the position estimation device 30 (i.e., the reception strengths acquired by the acquisition unit 31), the reception strengths acquired from the time the terminal device 10 moved to the present are used.

[0096] Generally, the relationship between the received signal strength (RSSI) and distance is expressed by the following formula (1): For simplicity, only the received signal strength in the 2.4 GHz band is considered here.

number

[0097] Here, i in equation (1) is an index assigned to each base station 20 (AP), and rssi i is the reception strength at the base station 20 whose index is i (hereinafter referred to as base station 20-i), and x i ,y i ,z i indicates the position (x, y, z coordinate values) of the base station 20-i. Furthermore, x, y, z in formula (1) indicate the position of the terminal device 10. Furthermore, A in formula (1) is the reception strength when the distance between the terminal device 10 and the base station 20-i is 1 m, and differs for each terminal device 10. Note that A is the reception strength when the distance between the terminal device 10 and the base station 20-i is sufficiently short, so it can be said to correspond to the transmission strength of the terminal device 10. Furthermore, n in formula (1) i is the attenuation constant, and in ideal free space, n i = 2, where n i is known to vary in a complex manner depending on the surrounding conditions in an indoor environment and the positions of the terminal device 10 and the base station 20-i, and in most cases takes a value between 2 and 6.

[0098] In this embodiment, the location of each of the plurality of base stations 20 is assumed to be stored (registered) in advance, and the location of the base station 20-i is represented by x i ,y i ,z i is known. In addition, if the terminal device 10 is a notebook PC and is used by placing it on a desk or the like, the position of the terminal device 10 in the height direction can be expected to be about 1 m, so z is also known. Furthermore, n i Assuming that all of the variables have the same value n, there are four unknown variables in equation (1): A, n, x, and y.

[0099] In this case, the unknown variables A, n, x, and y can be obtained by solving the minimization problem expressed by the following equation (2).

number

[0100] d in equation (2) rssi,i is the RSSI i (i.e., the reception strength at the base station 20-i) represents the distance between the terminal device 10 and the base station 20-i.

[0101] In this embodiment, x, y found by solving such a minimization problem and the above-mentioned known z (that is, x, y, z) correspond to the position of the terminal device 10, which is the estimation result of the first estimation process.

[0102] When solving the minimization problem expressed by the above equation (2), a constraint may be added, as necessary, so that the terminal device 10 is located within a predetermined space (inside an office building). Furthermore, A may be set to a fixed value, taking advantage of the knowledge that the transmission power of the terminal device 10 (for example, a notebook PC) is generally the same.

[0103] Here, Figure 11 is a box-and-whisker plot in which the vertical axis represents the error between the position (estimated value) of the terminal device 10 estimated by executing the first estimation process and the actual position (correct value) of the terminal device 10, and the horizontal axis represents the number of base stations 20 (receiving strength at) used in the first estimation process.

[0104] FIG. 6 shows that as the number of base stations 20 increases, the error between the estimated value and the correct value decreases, and it can be seen that the estimation accuracy of the position of the terminal device 10 improves.

[0105] As described above, in the first estimation process, when the terminal device 10 is sufficiently stationary and the received signal strength indicators (RSSI) from many base stations 20 are acquired, it can be said that relatively highly accurate position estimation can be performed.

[0106] Note that the first estimation process described here is just one example, and other processes may be executed as long as they estimate the location of the terminal device 10 using the reception strength at a sufficient number of base stations 20 acquired during the period when the terminal device 10 is in a specified location.

[0107] Incidentally, as described above, A used to estimate the position of the terminal device 10 in the first estimation process is the reception strength when the distance between the terminal device 10 and the base station 20-i is 1 m, and corresponds to the transmission strength of the signal transmitted from the terminal device 10. In the present embodiment, this A may be an unknown constant, but since A is roughly determined depending on the type of the terminal device 10, such as a notebook PC or a smartphone, A may be determined depending on the type of the terminal device 10 and applied to equation (2). In this way, A is set based on the type of the terminal device 10, and the optimization problem can be solved with fewer unknowns (the solution space is reduced, resulting in higher accuracy of estimation).

[0108] The type of terminal device 10 can be estimated, for example, from the vendor ID included in the MAC address used in wireless communication or information on whether mobile communication is possible included in the Probe Request from the terminal device 10 (i.e., information on the type of terminal device 10).

[0109] In addition, although the first estimation process has been described here as being performed only on the reception strength in the 2.4 GHz band, propagation characteristics differ depending on the frequency band of the radio waves used in wireless communication. Wireless LANs mainly use two frequency bands, the 2.4 GHz band and the 5 GHz band. In this case, the first estimation process may be performed separately for the 2.4 GHz band and the 5 GHz band. While the first estimation process has been described here, the same applies to the second estimation process described below.

[0110] Furthermore, when each of the multiple base stations 20 is equipped with multiple antennas, the multiple antennas equipped in the same base station 20 have different characteristics depending on the orientation and position of the antennas. Therefore, for example, if one base station 20 is equipped with first and second antennas and packets are observed using each of the first and second antennas, the reception strength of the packet observed using the first antenna and the reception strength of the packet observed using the second antenna may be distinguished (treated as separate reception strengths) in the first estimation process. Although the first estimation process has been described here, the same applies to the second estimation process described below.

[0111] Furthermore, when using Beacon Measurement Report as described above, the terminal device 10 measures the reception strength, but if the transmission strength of each base station 20 (the transmission strength of the signal transmitted from that base station 20) is known, the value obtained by subtracting the reception strength from the transmission strength (i.e., the difference between the transmission strength and the reception strength) becomes the radio wave attenuation term. In other words, when such a value obtained by subtracting the reception strength from the transmission strength is used, it becomes possible to perform the first estimation process using an equation that does not require A.

[0112] In the above formula (2), it has been described that the position of the base station 20-i (or a plurality of base stations 20 including the base station 20-i) is known, but the position (information indicating the position) of each of the base stations 20 may be acquired based on SNMP or the like from the base station 20 via, for example, a wireless LAN controller (wireless device). Also, when the base station 20 supports an LCI (Location Configuration Information) request (i.e., a base station 20 that supports LCI is used), the position estimation device 30 may acquire the position of the base station 20 by the LCI request.

[0113] Furthermore, if it is not possible to obtain the positions of each of the multiple base stations 20 as described above, it is possible to use positions input (instructed) by a user who visually confirms, for example, the installation drawings for the base stations 20 or the positions of the base stations 20 actually placed within a specified space.

[0114] However, there are cases where it is not possible to arrange (install) the base station 20 according to the above-described installation drawings, and even if a location based on the drawings is input, the location may differ from the actual location of the base station 20. In this case, for example, packets may be transmitted and received between multiple base stations 20 to measure the reception strength at each of the multiple base stations 20, and the location of a specific base station 20 may be estimated using the reception strength, and the input location may be corrected based on the estimated location as described above (i.e., the input location may be revised to the estimated location). Note that when estimating the location of the base station 20 using the reception strength, for example, a process similar to the first estimation process described above may be executed.

[0115] Next, the second estimation process will be described. As described above, the first estimation process is executed when the reception strengths of a sufficient number of base stations 20 have been acquired while the terminal device 10 is at a predetermined position (i.e., remaining at the same position), whereas the second estimation process is executed when the reception strengths of a sufficient number of base stations 20 have not been acquired. Note that, in the second estimation process, as in the first estimation process, of the reception strengths stored inside the position estimation device 30 (i.e., the reception strengths acquired by the acquisition unit 31), the reception strengths acquired during the time from when the terminal device 10 moved to the present are used.

[0116] Here, the reception strength used in the second estimation process is not sufficient to estimate the position of the terminal device 10 with high accuracy, but if a process similar to the first estimation process is performed using only such reception strength, the accuracy of the estimation result of the position of the terminal device 10 is considered to be low.

[0117] Therefore, in the second estimation process, the position of the terminal device 10 is estimated using a trained model generated by applying a known machine learning algorithm such as a neural network.

[0118] Hereinafter, an overview of the trained model used in the second estimation process will be described with reference to FIGS. 12 and 13.

[0119] First, for example, before the location estimation system 1 is put into operation, the terminal device 10 is kept stationary (i.e., at a predetermined position) for a sufficient period of time at the same position in a predetermined space, and the reception strengths of all base stations 20 in the vicinity of the terminal device 10 are acquired and accumulated.

[0120] In this case, in this embodiment, by using the training data including the reception strength and correct value at all base stations 20 accumulated as described above (i.e., learning the training data), a trained model is generated (constructed) that outputs the correct value when the reception strength is input.

[0121] In this embodiment, the correct value is the reception strength at all base stations 20 in the vicinity of the terminal device 10, which is accumulated as described above.

[0122] FIG. 12 conceptually shows a trained model that is trained using the reception intensities at all base stations 20 as described above.

[0123] In FIG. 12, it is assumed that N base stations 20 are arranged in a predetermined space, and the reception strength at base station 20-i (i=1, 2, . . . , j, . . . , N) among the N base stations 20 is calculated as AP i RSSI, whether the relevant reception strength was acquired or not is checked by the AP. i It is expressed as used. i When used=1, it indicates that the reception strength at the base station 20-i has been acquired, and i When used=1, it indicates that the reception strength at the base station 20-i has not been acquired.

[0124] For example, when the reception strength at the base station 20-i is acquired multiple times while the terminal device 10 is at a predetermined position from the same position, the reception strength at the base station 20-i included in the learning data may be a reception strength that has been subjected to preprocessing, such as calculating the average value of the reception strength acquired multiple times.

[0125] That is, FIG. 12 shows a learning phase in which a trained model (neural network) is constructed that can output the reception strengths (estimated values) at all base stations 20-1 to 20-N when the reception strengths at all base stations 20-1 to 20-N are input.

[0126] Furthermore, assuming that the reception strengths at all base stations 20 are accumulated as described above, in this embodiment, by using training data including the reception strengths at some of the base stations 20 among all the base stations 20 (i.e., learning the reception strengths at some of the base stations 20), a trained model is generated that outputs the reception strengths at all base stations 20 (i.e., correct values) when the reception strengths are input.

[0127] The reception strengths at some base stations 20 used as learning data may be extracted intentionally or randomly from the reception strengths at all base stations 20.

[0128] FIG. 13 conceptually illustrates a trained model in which training is performed using the reception strength at some base stations 20 as described above.

[0129] In Figure 13, AP j used=-1, which indicates that the reception strength at base station 20-j is not used as learning data.

[0130] That is, Figure 13 shows a learning phase in which a trained model (neural network) is constructed that can output the reception strength (estimated value) at all base stations 20-1 to 20-N when the reception strength at some base stations 20 out of all base stations 20-1 to 20-N, for example excluding at least base station 20-j, is input.

[0131] By performing the learning described above in Figures 12 and 13, a trained model is generated that can output the reception strength at all base stations 20 even if only the reception strength at a limited number of base stations 20 is input.

[0132] Here, it has been explained that learning is performed using the reception strength obtained when the terminal device 10 is at a predetermined position (stationary at the same position), but by repeating the learning for each position of the terminal device 10, it is possible to generate a highly accurate trained model.

[0133] In addition, the process of generating a trained model in this embodiment (i.e., the process related to learning the trained model) is assumed to be performed by the machine learning unit 35 included in the position estimation device 30, but may also be performed outside the position estimation device 30.

[0134] As described above, when the second estimation process in this embodiment is executed, the reception strengths at a sufficient number of base stations 20 are not acquired (i.e., only the reception strengths at some of the base stations 20 are acquired), but the second estimation unit 34 can acquire the reception strengths at all of the base stations 20 output from the trained model by inputting the reception strengths at such some of the base stations 20 into the trained model.

[0135] Since the reception strengths at all base stations 20 acquired in this manner correspond to the reception strengths at a sufficient number of base stations 20 acquired when the terminal device 10 is located at a specified position, the second estimation unit 34 can estimate the position of the terminal device 10 with high accuracy by performing a process similar to the first estimation process described above using the acquired reception strengths (i.e., the output of the trained model).

[0136] In this embodiment, a trained model has been described that is generated to output the received power at all base stations 20 by inputting the received power at some base stations 20, but the trained model may also be generated to output the position (estimated value) of the terminal device 10 by inputting, for example, the received power at some base stations 20.

[0137] Such a trained model is generated by learning training data including the reception strength at all base stations 20 acquired and accumulated as described above, the reception strength at some of the base stations 20 among all the base stations 20, and the position (correct value) of the terminal device 10 when the reception strength was measured.

[0138] In addition, the position of the terminal device 10 used as the correct value in learning the trained model that outputs the position of the terminal device 10 may be the position of the terminal device 10 estimated by executing a first estimation process using the reception strength at all base stations 20 (i.e., the estimation result in the first estimation process), or it may be an actual measurement value measured using various sensors or a position input by the user, etc.

[0139] As described above, the position estimation system 1 (position estimation device 30) according to this embodiment includes a first estimation unit 33 that estimates the position of the terminal device 10 using the reception strength (first communication status data relating to the communication status between a first wireless device moving in the space and each of a plurality of second wireless devices arranged in the space) at each of a plurality of base stations 20 arranged in a predetermined space in which the terminal device 10 moves, and a second estimation unit 34 that estimates the position of the terminal device 10 using the output of a trained model when the reception strength is input to the trained model.

[0140] In this case, the reception strength at each of the plurality of base stations 20 is acquired at different frequencies depending on the channel connecting the terminal device 10 and each of the plurality of base stations 20. In addition, the processing by the first estimation unit 33 (first estimation processing) or the processing by the second estimation unit 34 (second estimation processing) is selected based on the reception strength at each of the plurality of base stations 20.

[0141] Furthermore, the trained model is generated by using training data including the reception strength at each of the multiple base stations 20 (second communication status data related to the communication status between the terminal device 10 and each of the multiple base stations 20) acquired (measured) while the terminal device 10 is at a predetermined position (same position). The trained model may be generated by learning such training data so as to output the reception strength at each of the multiple base stations 20 (e.g., all of the base stations 20) upon input of the reception strength at at least some of the multiple base stations 20. The trained model may also be generated by learning training data including the reception strength at each of the multiple base stations 20 and an estimation result by a first estimation process using the reception strength, so as to output the position (estimated value) of the terminal device 10 upon input of the reception strength at at least some of the multiple base stations 20.

[0142] In this embodiment, the reception strength at the first base station 20 (i.e., the first communication status data relating to the communication status between the terminal device 10 and some of the multiple base stations 20) is measured (acquired) frequently using a communication channel. Also, in this embodiment, the reception strength at the second base station 20 (i.e., the first communication status data relating to the communication status between the terminal device 10 and other parts of the multiple base stations 20) is measured (acquired) less frequently using an observation channel.

[0143] In this embodiment, the above-described configuration makes it possible to estimate the position of the terminal device 10 quickly and with high accuracy.

[0144] Specifically, in this embodiment, for example, when a reception strength sufficient to estimate the position of the terminal device 10 with high accuracy is acquired, the first estimation process is selected, and the first estimation process is executed to estimate the position of the terminal device 10 with high accuracy.

[0145] On the other hand, in this embodiment, if reception strength sufficient to estimate the position of the terminal device 10 with high accuracy (i.e., reception strength at a sufficient number of base stations 20) is not obtained, the second estimation process is selected.

[0146] This second estimation process is executed without waiting until the reception strengths of a sufficient number of base stations 20 are acquired, and therefore, the process of estimating the position of the terminal device 10 can be performed at high speed.

[0147] Furthermore, in the second estimation process, a trained model generated by using the above-mentioned learning data (for example, a trained model that can output the reception strength at each of all base stations 20 by inputting the reception strength at some base stations 20) is used, thereby making it possible to achieve the same level of accuracy in estimating the position of the terminal device 10 as in the first estimation process using the reception strength at a sufficient number of base stations 20.

[0148] For convenience, the present embodiment has been mainly described in terms of the case where there is one terminal device 10. However, in an environment where multiple terminal devices 10 move within a given space, the reception strength at the base station 20 and an identifier for identifying the terminal device 10 may be acquired for each terminal device 10, and the position of each of the multiple terminal devices 10 may be estimated based on the reception strength acquired for each terminal device 10.

[0149] Furthermore, the received signal strength acquired as described above is stored in the position estimation system 1 (for example, inside the position estimation device 30) in association with the time at which the received signal strength was acquired. With this configuration, it becomes possible to select the first or second estimation process using the time at which the received signal strength was acquired.

[0150] Specifically, the second estimation process may be selected when the amount of change in the reception strength at at least one base station 20 (for example, the first base station 20) among the plurality of base stations 20 is equal to or greater than a predetermined value. Also, the second estimation process may be selected when the interval at which the reception strength is acquired at at least one base station 20 (for example, the first base station 20) among the plurality of base stations 20 is equal to or greater than a predetermined value. Furthermore, the second estimation process may be selected when the base station 20 connected to the terminal device 10 via a communication channel is changed. With this configuration, it becomes possible to determine that the terminal device 10 has moved and switch from the first estimation process to the second estimation process.

[0151] Furthermore, when the number of base stations 20 from which reception strength has been acquired while the terminal device 10 is in a predetermined position is less than a predetermined value, the second estimation process may be selected. With this configuration, when reception strengths from the number of base stations 20 required for the first estimation process have not been acquired, it is possible to estimate the position of the terminal device 10 with high accuracy in the second estimation process.

[0152] Although the first or second estimation process is selected using the time when the reception strength was acquired in the above description, the time when the reception strength was acquired is also used when the first or second estimation process is executed. Specifically, the first or second estimation process is executed using the reception strength acquired while the terminal device 10 is in a predetermined position (from the time when the terminal device 10 moved to the present).

[0153] Furthermore, in the first estimation process, the location of the terminal device 10 may be estimated using the transmission strength of a packet (signal) transmitted from the terminal device 10. Note that this transmission strength can be acquired based on the type of the terminal device 10 (for example, a notebook PC, a smartphone, etc.). With this configuration, it is expected that the accuracy of estimating the location of the terminal device 10 will be improved.

[0154] In the first estimation process, the location of the terminal device 10 may be estimated based on the frequency band of the channel. With this configuration, it is possible to estimate the location of the terminal device 10 taking into consideration that propagation characteristics vary depending on the frequency of the channel.

[0155] Furthermore, in the first estimation process, if each of the multiple base stations 20 is equipped with multiple antennas, the reception strength acquired for each of the antennas may be used. With this configuration, when the base station 20 (AP) equipped with multiple antennas is installed, it is possible to improve the estimation accuracy of the location of the terminal device 10 by using information on the antenna used to observe packets for measuring the reception strength (i.e., distinguishing between the reception strengths of packets observed using different antennas).

[0156] Furthermore, in the first estimation process, the position of the terminal device 10 may be estimated using the transmission strength of packets (signals) transmitted from a plurality of base stations 20. With such a configuration, it is possible to reduce the number of unknowns in the optimization problem expressed by the above-mentioned equation (2).

[0157] Furthermore, in the first estimation process, the position of the terminal device 10 is estimated using the positions of each of the multiple base stations 20, but the positions of each of the multiple base stations 20 may be acquired from the base station 20, or may be stored in advance inside the position estimation device 30, for example. Furthermore, the positions of each of the multiple base stations 20 may be estimated based on the reception strength measured between the base stations 20, for example.

[0158] In the first estimation process of this embodiment, the position of the terminal device 10 is estimated by applying the reception strength at each of the multiple base stations 20 to a minimization problem (a calculation formula related to radio wave propagation) expressed by, for example, equation (2), but calculation formulas other than those described in this embodiment may also be used.

[0159] Although the first estimation process has been described here, the description of the first estimation process can also be applied to processing similar to the first estimation process that is performed using the output of the trained model in the second estimation process.

[0160] In this embodiment, the communication status data is described as being received power (data), but the communication status data may be any data relating to the communication status between the terminal device 10 and each of the multiple base stations 20, and the communication status data may be, for example, a bit error rate.

[0161] In addition, in the present embodiment, the position estimation device 30 has been described as including the units 31 to 35 shown in Fig. 4, but at least some of the units 31 to 35 may be arranged outside the position estimation device 30. Furthermore, the position estimation device 30 may be realized by a plurality of devices, and the units 31 to 35 may be arranged in different devices.

[0162] Furthermore, although the present embodiment has been described as estimating the position of a terminal device 10 such as a notebook PC used by a user moving within a predetermined space, the present embodiment can be applied to any case in which the position of a wireless device placed within the space is estimated, and can also be applied when the terminal device 10 is a stationary device. For example, if the terminal device 10 is a desktop personal computer, it may be difficult to obtain sufficient radio wave strength for position estimation in the communication channel immediately after startup. In such a situation, position estimation may be performed by scanning an observation channel, and then scanning the communication channel when sufficient radio wave strength is obtained in the communication channel, thereby performing position estimation. Furthermore, a configuration may be adopted in which the position of a predetermined base station 20, for example, is estimated.

[0163] Although several embodiments of the present invention have been described, these embodiments are presented as examples and are not intended to limit the scope of the invention. These embodiments can be implemented in various other forms, and various omissions, substitutions, and modifications can be made without departing from the spirit of the invention. These embodiments and their modifications are included within the scope and spirit of the invention, as well as within the scope of the invention described in the claims and their equivalents.

[0164] The following additional notes are provided regarding the above-described embodiment. [1] an acquisition means for acquiring first communication status data relating to communication status between the first wireless device and each of the plurality of second wireless devices; a first estimation means for estimating a position of the first wireless device using the acquired first communication status data; a second estimation means for estimating a position of the first wireless device using an output of a trained model when the acquired first communication situation data is input to the trained model; and Equipped with the first communication status data is acquired at different frequencies depending on channels connecting the first wireless device and each of the plurality of second wireless devices; the process by the first estimation means or the process by the second estimation means is selected based on the first communication status data; The trained model is generated by using training data including second communication status data relating to communication status between the first wireless device and each of the plurality of second wireless devices when the first wireless device is in a predetermined position. Location estimation system. [2] first communication status data relating to a communication status between the first wireless device and some of the plurality of second wireless devices is measured at a high frequency using a communication channel for performing wireless communication with the first wireless device; First communication status data relating to a communication status between the first wireless device and other units of the plurality of second wireless devices is measured at a low frequency using an observation channel different from the communication channel. [1] The location estimation system according to claim 1. [3] The position estimation system according to [1] or [2], wherein the first and second communication status data include reception strengths of signals observed between the first wireless device and each of a plurality of base stations (20). [4] the first wireless device includes a terminal device used by a user; The second radio device includes a base station. The position estimation system according to any one of [1] to [3]. [5] When a plurality of first wireless devices are moving, the acquisition means acquires the first communication status data for each of the first wireless devices; The first and second estimation means estimate the position of each of the plurality of first wireless devices based on first communication status data acquired for each of the first wireless devices. The position estimation system according to any one of [1] to [4]. [6] The position estimation system according to [3], wherein the reception strength included in the first communication status data is stored in the position estimation system in association with the time at which the reception strength was acquired. [7] A location estimation system according to [6], wherein processing by the second estimation means is selected when the change in the reception strength of the signal observed between at least one of the plurality of second wireless devices is greater than or equal to a predetermined value. [8] [6] A location estimation system according to [6], wherein processing by the second estimation means is selected when the acquisition interval of the reception strength of the signal observed between at least one of the plurality of second wireless devices is equal to or greater than a predetermined value. [9] The position estimation system according to [6], wherein when a second wireless device connected to the first wireless device via the communication channel is changed, processing by the second estimation means is selected.

[10] A location estimation system as described in [6], wherein if the number of second wireless devices from which the reception strength has been acquired when the first wireless device is located at a predetermined position is less than a predetermined value, processing by the second estimation means is selected.

[11] the first wireless device is mobile; The position estimation system according to [6], wherein the first and second estimation means estimate the position of the first wireless device using reception strength included in first communication status data acquired after the first wireless device has moved.

[12] the first and second estimation means estimate a position of the first wireless device using a transmission strength of a signal transmitted from the first wireless device; The transmission strength is obtained based on information about the type of the first wireless device. [3] The location estimation system according to claim 1.

[13] The position estimation system according to any one of [1] to

[12] , wherein the first and second estimation means estimate the position of the first wireless device based on the frequency band of the channel.

[14] The location estimation system according to any one of [1] to

[13] , wherein the acquisition means acquires the first communication status data for each of the plurality of second wireless devices when each of the plurality of second wireless devices is equipped with a plurality of antennas.

[15] The position estimation system according to any one of [1] to

[14] , wherein the first and second estimation means estimate the position of the first wireless device using transmission strengths of signals transmitted from the plurality of second wireless devices.

[16] each of the plurality of second wireless devices instructs the first wireless device to scan for signals, and acquires the reception strength of the signals from the first wireless device as a result of the scan; The acquisition means acquires the reception strength from each of the plurality of second wireless devices. [3] The location estimation system according to claim 1.

[17] The position estimation system according to any one of [1] to

[16] , wherein the trained model is generated by learning learning data including the second communication status data, so as to output fourth communication status data regarding the communication status between the first wireless device and each of the plurality of second wireless devices by inputting third communication status data regarding the communication status between the first wireless device and at least some of the plurality of second wireless devices.

[18] The position estimation system according to any one of [1] to

[16] , wherein the trained model is generated by learning training data including the second communication status data and the estimation result of the position of the first wireless device by the first estimation means using the second communication status data, and is configured to output the position of the first wireless device by inputting third communication status data regarding the communication status between the first wireless device and at least some of the plurality of second wireless devices.

[19] The position estimation system according to any one of [1] to

[18] , wherein the first and second estimation means estimate the position of the first radio device using the position of the second radio device obtained from each of the plurality of second radio devices.

[20] The position estimation system according to any one of [1] to

[18] , wherein the first and second estimation means estimate the position of the first radio device using the positions of each of the plurality of second radio devices that are stored in advance. [twenty one] The position estimation system according to any one of [1] to

[18] , wherein the first and second estimation means estimate the position of the first radio device using the positions of each of the second radio devices estimated based on fifth communication status data indicating the communication status between the second radio devices. [twenty two] The position estimation system according to [3], wherein the first estimation means estimates the position of the first wireless device by applying the reception strength included in the acquired first communication status data to a calculation formula related to radio wave propagation. [Explanation of symbols]

[0165] 1...location estimation system, 10...terminal device (first wireless device), 11...transmitter / receiver unit, 12...control unit, 20...base station (second wireless device), 21...transmitter / receiver unit, 22...control unit, 23...output unit, 30...location estimation device, 31...acquisition unit, 32...selection unit, 33...first estimation unit, 34...second estimation unit, 35...machine learning unit, 301...CPU, 302...non-volatile memory, 303...RAM, 304...communication device.

Claims

1. an acquisition means for acquiring first communication status data relating to communication status between the first wireless device and each of the plurality of second wireless devices; a first estimation means for estimating a position of the first wireless device using the acquired first communication status data; a second estimation means for estimating a position of the first wireless device using an output of a trained model when the acquired first communication status data is input to the trained model; and Equipped with the first communication status data is acquired at different frequencies depending on channels connecting the first wireless device and each of the plurality of second wireless devices; the process by the first estimation means or the process by the second estimation means is selected based on the first communication status data; The trained model is generated by using training data including second communication status data relating to communication status between the first wireless device and each of the plurality of second wireless devices when the first wireless device is located at a predetermined position. Location estimation system.

2. first communication status data relating to a communication status between the first wireless device and some of the plurality of second wireless devices is measured at a high frequency using a communication channel for performing wireless communication with the first wireless device; First communication status data relating to a communication status between the first wireless device and other units of the plurality of second wireless devices is measured at a low frequency using an observation channel different from the communication channel. The location estimation system of claim 1 .

3. 3. The position estimation system according to claim 2, wherein the first and second communication status data include reception strengths of signals observed between the first wireless device and each of a plurality of base stations.

4. the first wireless device includes a terminal device used by a user; The second wireless device includes a base station. The location estimation system of claim 3 .

5. When a plurality of first wireless devices are moving, the acquisition means acquires the first communication status data for each of the first wireless devices; The first and second estimation means estimate the position of each of the plurality of first wireless devices based on first communication status data acquired for each of the first wireless devices. The location estimation system of claim 1 .

6. 4. The position estimation system according to claim 3, wherein the reception strength included in the first communication status data is stored in the position estimation system in association with the time at which the reception strength was acquired.

7. 7. The position estimation system according to claim 6, wherein processing by the second estimation means is selected when a change in the reception strength of a signal observed between at least one of the plurality of second wireless devices is greater than or equal to a predetermined value.

8. 7. The position estimation system according to claim 6, wherein processing by the second estimation means is selected when the interval between acquisitions of the reception strength of a signal observed between at least one of the plurality of second wireless devices is equal to or greater than a predetermined value.

9. 7. The position estimation system according to claim 6, wherein when a second wireless device connected to the first wireless device via the communication channel is changed, processing by the second estimation means is selected.

10. 7. A position estimation system as described in claim 6, wherein when the number of second wireless devices from which the reception strength has been acquired when the first wireless device is located at a predetermined position is less than a predetermined value, processing by the second estimation means is selected.

11. the first wireless device is mobile; 7. The position estimation system according to claim 6, wherein the first and second estimation means estimate the position of the first wireless device using reception strength included in first communication status data acquired after the first wireless device has moved.

12. the first and second estimation means estimate a position of the first wireless device using a transmission strength of a signal transmitted from the first wireless device; The transmission strength is obtained based on information about the type of the first wireless device. The location estimation system of claim 3 .

13. 2. The position estimation system according to claim 1, wherein the first and second estimation means estimate the position of the first wireless device based on the frequency band of the channel.

14. 2. The position estimation system according to claim 1, wherein, when each of the plurality of second wireless devices is equipped with a plurality of antennas, the acquisition means acquires the first communication status data for each of the antennas.

15. 2. The position estimation system according to claim 1, wherein the first and second estimation means estimate the position of the first wireless device using transmission strengths of signals transmitted from the plurality of second wireless devices.

16. each of the plurality of second wireless devices instructs the first wireless device to scan for signals, and acquires the reception strength of the signals from the first wireless device as a result of the scan; The acquisition means acquires the reception strength from each of the plurality of second wireless devices. The location estimation system of claim 3 .

17. The position estimation system of claim 1, wherein the trained model is generated by learning learning data including the second communication status data, so that when third communication status data regarding the communication status between the first wireless device and at least some of the plurality of second wireless devices is input, fourth communication status data regarding the communication status between the first wireless device and each of the plurality of second wireless devices is output.

18. The position estimation system of claim 1, wherein the trained model is generated by learning learning data including the second communication status data and the estimation result of the position of the first wireless device by the first estimation means using the second communication status data, so as to output the position of the first wireless device when third communication status data regarding the communication status between the first wireless device and at least some of the plurality of second wireless devices is input.

19. 2. The position estimation system according to claim 1, wherein the first and second estimation means estimate the position of the first wireless device using the position of the second wireless device acquired from each of the plurality of second wireless devices.

20. 2. The position estimation system according to claim 1, wherein the first and second estimation means estimate the position of the first wireless device using the positions of the plurality of second wireless devices that are stored in advance.

21. 2. The position estimation system according to claim 1, wherein the first and second estimation means estimate the position of the first wireless device using the positions of each of the second wireless devices estimated based on fifth communication status data indicating communication status between the second wireless devices.

22. 4. The position estimation system according to claim 3, wherein the first estimation means estimates the position of the first wireless device by applying the reception strength included in the acquired first communication status data to a calculation formula related to radio wave propagation.

Images