Communication device, information processing method, and information processing program

Abstract

To improve the accuracy of determining the location of wireless tags. [Solution] The communication device of the embodiment includes an antenna, a drive unit, a first acquisition unit, an input unit, and a second acquisition unit. The drive unit moves the position of the antenna. The first acquisition unit acquires tag data for each radio tag based on the radio waves of each radio tag received by the antenna at multiple positions on the antenna. The input unit inputs the tag data for each radio tag at multiple positions on the antenna to a trained model. Based on the input of the tag data for each radio tag to the trained model, the second acquisition unit acquires data from the trained model indicating whether each radio tag falls within a first range or a second range.

Description

Technical Field

[0001] Embodiments of the present invention relate to a communication device, an information processing method, and an information processing program.

Background Art

[0002] There is a device that determines whether a wireless tag is within a predetermined range or outside the predetermined range by receiving radio waves transmitted from a wireless tag attached to an article with an antenna. Such a device moves an antenna to measure the phase of the wireless tag. The device determines whether the wireless tag is within a predetermined range or outside the predetermined range based on a phase difference that is the amount of change in the measured phase. The device needs to set a phase difference corresponding to the boundary of the predetermined range as a threshold value.

[0003] When a predetermined fixed threshold value is used in the determination process, the determination accuracy may decrease due to the influence of the environment such as disturbance.

Prior Art Documents

Patent Documents

[0004]

Patent Document 1

Summary of the Invention

Problems to be Solved by the Invention

[0005] The problem to be solved by the embodiments of the present invention is to provide a technology for improving the determination accuracy of the position of a wireless tag.

Means for Solving the Problems

[0006] The communication device of the embodiment includes an antenna, a drive unit, a first acquisition unit, an input unit, and a second acquisition unit. The drive unit moves the position of the antenna. The first acquisition unit acquires tag data for each radio tag based on the radio waves of each radio tag received by the antenna at multiple positions on the antenna. The input unit inputs the tag data for each radio tag at multiple positions on the antenna acquired by the first acquisition unit to a trained model. Based on the input of the tag data for each radio tag to the trained model by the input unit, the second acquisition unit acquires data from the trained model indicating whether each radio tag falls into a first range or a second range. The trained model is a model generated by machine learning based on training data that includes tag data for multiple radio tags at multiple positions on the antenna and data indicating whether each of the multiple radio tags falls into a first range or a second range. [Brief explanation of the drawing]

[0007] [Figure 1] Figure 1 is a block diagram showing an example of the configuration of a communication system according to an embodiment. [Figure 2] Figure 2 is a block diagram showing an example of the configuration of a reading device according to an embodiment. [Figure 3] Figure 3 shows an example of a data structure that constitutes measurement data according to the embodiment. [Figure 4] Figure 4 is a block diagram showing an example of the configuration of a drive device according to this embodiment. [Figure 5] Figure 5 is a schematic diagram illustrating the drive device according to the embodiment. [Figure 6] Figure 6 is a schematic diagram illustrating the first and second ranges according to the embodiment. [Figure 7] Figure 7 is a flowchart showing an example of the determination process by the processor of the reading device according to the embodiment. [Figure 8] Figure 8 shows an example of the arrangement of multiple wireless tags to be used for learning according to the embodiment. [Figure 9] Figure 9 is a graph showing an example of training tag data according to the embodiment. [Figure 10] Figure 10 is a graph showing another example of training tag data according to the embodiment. [Figure 11] Figure 11 is a graph showing yet another example of training tag data according to the embodiment. [Figure 12] Figure 12 is a flowchart showing an example of the process by which the processor of the reader according to the embodiment generates a trained model. [Modes for carrying out the invention]

[0008] The communication system according to the embodiment will be described below with reference to the drawings. Note that the scale of the parts in the drawings used in the description of the embodiment below may have been changed as appropriate. Also, for illustrative purposes, some components may be omitted from the drawings used in the description of the embodiment below.

[0009] Figure 1 is a block diagram showing an example of the configuration of a communication system 1 according to an embodiment. The communication system 1 includes a communication device 10 and multiple wireless tags 600 attached to multiple articles 500. Figure 1 shows one wireless tag 600 attached to one article 500, but the communication system 1 includes multiple wireless tags 600 attached to multiple articles 500.

[0010] The communication device 10 is a device that reads information from the wireless tag 600 and processes the read information. The communication device 10 can be used for inspection in a warehouse, but its applications are not limited to this. The communication device 10 includes a reader 100, a drive unit 200, an antenna 300, and a terminal 400.

[0011] The reader device 100 controls the drive unit 200 and the antenna 300 to read information from the wireless tag 600. An example configuration of the reader device 100 will be described later. The drive unit 200 is a device that moves the antenna 300. An example configuration of the drive unit 200 will be described later. Antenna 300 transmits and receives radio waves to and from the wireless tag 600. Antenna 300 converts the radio waves received from the wireless tag 600 into a high-frequency signal and outputs the high-frequency signal to the reader 100. The terminal 400 is a device that processes the information read from the wireless tag 600 by the reader 100. The terminal 400 is a PC (personal computer) or the like, but any device that can process information is acceptable and is not limited to this.

[0012] The article 500 is a commodity or the like. The wireless tag 600 is typically an RFID (radio frequency identification) tag. The wireless tag 600 may be other wireless tags. The wireless tag 600 is a passive wireless tag that operates using a predetermined radio wave transmitted from the antenna 300 as an energy source. The wireless tag 600 performs backscatter modulation on an unmodulated signal to transmit a signal including the information stored in the wireless tag 600. The information stored in the wireless tag 600 may include uniquely identifiable identification information. The information stored in the wireless tag 600 may include information regarding the article 500 to which the wireless tag 600 is attached.

[0013] The reader 100 will be described with reference to FIG. 2. FIG. 2 is a block diagram showing an example of the configuration of the reader 100. The reader 100 includes a processor 101, a ROM (read-only memory) 102, a RAM (random-access memory) 103, a first connection interface 104, a second connection interface 105, a high-frequency front-end unit 106, a digital amplitude modulation unit 107, a DA (digital to analog) conversion unit 108, an AD (analog to digital) conversion unit 109, a demodulation unit 110, and a storage device 111. Each unit included in the reader 100 is connected by a bus 112 or the like.

[0014] Processor 101 corresponds to the central part of a computer that performs operations such as calculations and controls necessary for the operation of the reading device 100. Processor 101 expands various programs stored in ROM 102 or the storage device 111, etc. into RAM 103. By executing the programs expanded in RAM 103, Processor 101 realizes each part described later and executes various operations.

[0015] Processor 101 is a CPU (central processing unit), MPU (micro processing unit), SoC (system on a chip), DSP (digital signal processor), GPU (graphics processing unit), ASIC (application specific integrated circuit), PLD (programmable logic device), or FPGA (field-programmable gate array), etc. Processor 101 may be a combination of a plurality of these.

[0016] ROM 102 corresponds to the main storage device of a computer centered on Processor 101. ROM 102 is a non-volatile memory used exclusively for reading data. ROM 102 stores the above programs. Also, ROM 102 stores data or various setting values, etc. used by Processor 101 when performing various processes.

[0017] RAM 103 corresponds to the main storage device of a computer centered on Processor 101. RAM 103 is a memory used for reading and writing data. RAM 103 is a work area that stores data temporarily used by Processor 101 when performing various processes.

[0018] The first connection interface 104 is an interface for the reading device 100 to communicate with the driving device 200.

[0019] The second connection interface 105 is an interface for the reader 100 to communicate with the terminal 400.

[0020] The high-frequency front-end unit 106 outputs a high-frequency signal to the antenna 300. The high-frequency front-end unit 106 receives a high-frequency signal from the antenna 300.

[0021] The digital amplitude modulation unit 107 is a circuit that adds information to be transmitted to the wireless tag 600 to the carrier wave transmitted to the wireless tag 600.

[0022] The DA conversion unit 108 is a circuit that converts a digital signal to an analog signal. The DA conversion unit 108 converts the digital signal modulated by the digital amplitude modulation unit 107 into an analog signal. The DA conversion unit 108 outputs a high-frequency signal to the antenna 300 via the high-frequency front-end unit 106.

[0023] The AD conversion unit 109 is a circuit that converts an analog signal into a digital signal. The AD conversion unit 109 converts the high-frequency signal input from the antenna 300 into a digital signal via the high-frequency front-end unit 106.

[0024] The demodulation unit 110 is a circuit that extracts various information from radio waves received from the wireless tag 600. For example, the demodulation unit 110 extracts a unique identification code stored in the wireless tag 600 from the digital signal converted by the AD conversion unit 109. Also, using known technology, when the antenna 300 receives the radio waves from the wireless tag 600, the demodulation unit 110 outputs tag data of the wireless tag 600 in time series from the digital signal converted by the AD conversion unit 109. The tag data is time series data based on the radio waves from the wireless tag 600 received by the antenna 300. The tag data includes phase data. The phase data is data indicating the phase of the radio waves from the wireless tag 600. The tag data includes received signal strength indicator (RSSI) data. The received signal strength data is data indicating the received signal strength from the wireless tag 600. The tag data may include both phase data and received signal strength data. Furthermore, each wireless tag 600 may store radio wave reception strength data in its own memory when it receives radio waves transmitted from the antenna 300. In this example, the demodulation unit 110 may extract the radio wave reception strength data stored in the wireless tag 600 in a time series from the digital signal converted by the AD conversion unit 109. The demodulation unit 110 is an example of a detection unit that detects the tag data of each wireless tag 600 in a time series based on the radio waves from each wireless tag 600.

[0025] The storage device 111 is a device composed of non-volatile memory for storing data and programs. The storage device 111 is composed of an HDD (Hard Disk Drive) or an SSD (Solid State Drive), but is not limited to these. The storage device 111 is an example of a storage unit.

[0026] The memory device 111 stores the measurement data 1111. The measurement data 1111 includes tag data for each wireless tag 600 to be determined at multiple locations on the antenna 300. The tag data for each wireless tag 600 to be determined is associated with each of the multiple locations on the antenna 300. The tag data included in the measurement data 1111 is data measured by the reader 100. The wireless tag 600 to be determined is the wireless tag to be determined as to whether its location falls within a first range or a second range. The tag data for the wireless tag 600 to be determined is also called determination tag data. The first range and the second range are different ranges that do not overlap with each other. For example, the first range and the second range are three-dimensional regions. Examples of the first range and the second range will be described later. The measurement data 1111 can be updated. An example of the configuration of the measurement data 1111 will be described later.

[0027] The memory device 111 stores the training data 1112. The training data 1112 is data used for machine learning. The training data 1112 includes tag data for multiple wireless tags 600 to be trained at multiple locations on the antenna 300. The tag data for each wireless tag 600 to be trained is associated with each of the multiple locations on the antenna 300. The tag data for the wireless tags 600 to be trained is data that has been measured in advance by the reader 100. The tag data for the wireless tags 600 to be trained is also called training tag data. The training data 1112 includes ground truth data indicating whether each of the multiple wireless tags 600 to be trained falls within a first range or a second range. The ground truth data can also be said to indicate whether each location of the multiple wireless tags 600 to be trained falls within a first range or a second range. The ground truth data is data entered by the user. The training data 1112 can be updated.

[0028] The memory device 111 stores the trained model 1113. The trained model 1113 is a model generated by machine learning based on the training data 1112. The term "generated" includes not only newly created models but also updated models. The trained model 1113 is used to determine whether the location of a target wireless tag 600 falls within a first range or a second range. The trained model 1113 outputs determination output data based on the input determination input data. The determination input data consists of determination tag data for each target wireless tag 600 at multiple locations on the antenna 300. The tag data for each target wireless tag 600 is associated with each of the multiple locations on the antenna 300. The determination output data indicates whether each target wireless tag 600 falls within a first range or a second range. Each target wireless tag 600 is associated with either the first range or the second range.

[0029] Bus 112 includes a control bus, an address bus, and a data bus, etc. Bus 112 transmits signals exchanged between the various parts of the reader 100.

[0030] The hardware configuration of the reading device 100 is not limited to the configuration described above. The reading device 100 may be modified or have its components omitted or changed as appropriate, and new components added as needed.

[0031] The various components implemented by processor 101 will now be described. The processor 101 implements the movement control unit 1011, the communication control unit 1012, the first acquisition unit 1013, the input unit 1014, the second acquisition unit 1015, the output unit 1016, and the model processing unit 1017. Each part implemented by the processor 101 can also be called a function. Each part implemented by the processor 101 can also be said to be implemented by a control unit including the processor 101, ROM 102, and RAM 103.

[0032] The movement control unit 1011 controls the movement of the antenna 300 by controlling the drive unit 200.

[0033] The communication control unit 1012 controls the start and end of radio wave transmission from the antenna 300.

[0034] The first acquisition unit 1013 acquires determination tag data for each wireless tag 600 to be determined, based on the radio waves of each wireless tag 600 received by the antenna 300, at multiple locations on the antenna 300.

[0035] The input unit 1014 inputs judgment input data to the trained model 1113. The judgment input data is judgment tag data for each wireless tag 600 to be judged at multiple locations of the antenna 300, which is acquired by the first acquisition unit 1013.

[0036] The second acquisition unit 1015 acquires judgment output data from the trained model 1113 based on the input of judgment input data to the trained model 1113 by the input unit 1014.

[0037] The output unit 1016 outputs the determination result to the terminal 400. The determination result includes data indicating whether each wireless tag 600 to be determined, which was acquired as determination output data by the second acquisition unit 1015, falls within the first range or the second range.

[0038] The model processing unit 1017 generates a trained model 1113.

[0039] Figure 3 shows an example of the data structure that makes up the measurement data 1111. The antenna 300 moves back and forth in one direction based on control by the drive unit 200. The scanning range of the antenna 300 is a unidirectional range from position 0, which corresponds to the home position, to position L. Position L can be set as appropriate.

[0040] The measurement data 1111 includes tag data for each wireless tag 600 to be determined, associated with each of the multiple positions of the antenna 300. For example, the multiple positions of the antenna 300 are positions at a fixed interval a between position 0 and position L. The value of the fixed interval a can be set as appropriate. The measurement data 1111 only needs to include tag data for each wireless tag 600 to be determined, associated with at least each of the positions at a fixed interval a between position 0 and position L. The measurement data 1111 may also include tag data for each wireless tag 600 to be determined, associated with positions other than those at a fixed interval a between position 0 and position L.

[0041] The drive unit 200 will be explained using Figures 4 and 5. Figure 4 is a block diagram showing an example of the configuration of the drive unit 200. The drive unit 200 includes a processor 201, ROM 202, RAM 203, connection interface 204, drive unit 205, and home position sensor 206. Each part of the drive unit 200 is connected by a bus 208, etc.

[0042] The processor 201 corresponds to the central part of the computer that performs calculations and control necessary for the operation of the drive unit 200. The processor 201 loads various programs stored in ROM 202, etc., into RAM 203. The processor 201 performs various operations by executing the programs loaded into RAM 203. The processor 201 can be a CPU, MPU, SoC, DSP, GPU, ASIC, PLD, or FPGA, etc. The processor 201 may be a combination of several of these.

[0043] ROM202 corresponds to the main memory of a computer centered around processor 201. ROM202 is a non-volatile memory used exclusively for reading data. ROM202 stores the above-mentioned program. ROM202 also stores data or various settings used by processor 201 in performing various processes.

[0044] RAM203 corresponds to the main memory of a computer centered around processor 201. RAM203 is memory used for reading and writing data. RAM203 is a work area that stores data temporarily used by processor 201 when performing various processes.

[0045] The connection interface 204 is an interface for the drive unit 200 to connect with the reader unit 100.

[0046] The drive unit 205 moves the antenna 300. For example, the drive unit 205 is a stepping motor.

[0047] The home position sensor 206 is a sensor that detects whether or not the moving stage 213, which will be described later, is in the home position.

[0048] Bus 208 includes a control bus, an address bus, and a data bus, etc. Bus 208 transmits signals exchanged between various parts of the drive unit 200.

[0049] Figure 5 is a schematic diagram illustrating the drive unit 200. The drive unit 200 includes a rotating shaft 211, a rail 212, and a moving stage 213.

[0050] As illustrated in Figure 5, the drive unit 200 and the antenna 300 are located below the counter base 700. The counter base 700 is a platform having a horizontal surface on which the article 500 with the wireless tag 600 attached is placed. The counter base 700 is an example of a mounting section. The counter base 700 may be included in the communication system 1 or the communication device 10.

[0051] The rotating shaft 211 transmits the driving force of the drive unit 205. Screw grooves are formed on the rotating shaft 211 and the rail 212. The screw grooves are opposite each other and connected. Therefore, when the drive unit 205 is driven to rotate, the rotating shaft 211 rotates and the rail 212 moves. A moving stage 213 on which the antenna 300 is mounted is attached to the rail 212.

[0052] The moving stage 213 is equipped with a ball screw nut, and moves horizontally when the rail 212 rotates due to the ball screw nut. That is, the moving stage 213 moves in the direction along the x-axis as shown in Figure 5. Also, if the rotation direction of the rail 212 is reversed, the moving stage 213 moves in the opposite direction. In this way, the drive device 200 moves the antenna 300 back and forth along the rail 212 in one direction along the x-axis.

[0053] The hardware configuration of the drive unit 200 is not limited to the configuration described above. The drive unit 200 may be modified or have the above-described components omitted or changed, and new components added as appropriate.

[0054] The first and second scopes will be explained below. Figure 6 is a schematic diagram illustrating the first range 81 and the second range 82, and is a plan view of the countertop 700 seen from above.

[0055] The first range 81 and the second range 82 are horizontally separated ranges. The first range 81 is the range set in the central part of the horizontal plane of the counter base 700. The second range 82 is the range set in the outer periphery of the horizontal plane of the counter base 700 and the range set horizontally outside the counter base 700. The second range 82 is set to surround the first range 81. In Figure 6, the second range 82 is set with a gap between it and the first range 81, but is not limited to this. The second range 82 may be adjacent to the first range 81.

[0056] The settings of the first range 81 and the second range 82 are not limited to these. The first range 81 may be a range set in the central part of the horizontal plane of the counter base 700, and the second range 82 may be a range set in the outer periphery of the horizontal plane of the counter base 700. The first range 81 may be a range set across the entire horizontal plane of the counter base 700, and the second range 82 may be a range set horizontally outside the counter base 700. The second range 82 is not limited to a range set to surround the first range 81.

[0057] The first range 81 and the second range 82 may be different ranges that do not overlap with each other, and are not limited to ranges separated horizontally. The first range 81 and the second range 82 may also be ranges separated vertically.

[0058] Next, the determination process performed by the processor 101 of the reader 100 configured as described above will be explained. The determination process is the process of acquiring data indicating whether each wireless tag 600 to be determined falls within the first range or the second range.

[0059] Figure 7 is a flowchart showing an example of the determination process performed by the processor 101 of the reading device 100. The processing procedure described below is merely an example, and each process may be modified as much as possible. Furthermore, depending on the embodiment, steps in the processing procedure described below may be omitted, replaced, or added as appropriate.

[0060] For example, suppose that the counter 700 has an item 500 on it that is to be read from the wireless tag 600. There may also be items in the vicinity of the counter 700 that are not to be read from the wireless tag 600.

[0061] The processor 101 of the reading device 100 may start the determination process based on the acquisition of a determination process start instruction entered by the user at the terminal 400.

[0062] The movement control unit 1011 controls the movement of the antenna 300 (ACT1). In ACT1, for example, the movement control unit 1011 transmits a movement instruction to the drive unit 200. The movement instruction is an instruction to move the antenna 300 in one direction from position 0, which corresponds to the home position, to position L.

[0063] The processor 201 of the drive unit 200 receives a movement instruction from the reader 100. Based on the movement instruction, the processor 201 uses the home position sensor 206 to determine whether the antenna 300 is in the home position. If the antenna 300 is not in the home position, the processor 201 controls the drive unit 205 to move the antenna 300 to the home position. Based on the control by the processor 201, the drive unit 205 moves the antenna 300 to the home position. The processor 201 controls the drive unit 205 to start moving the antenna 300 from position 0, which corresponds to the home position. Based on the control by the processor 201, the drive unit 205 starts moving the antenna 300 from position 0. The processor 201 controls the drive unit 205 to move the antenna 300 in one direction from position 0 to position L. Based on the control by the processor 201, the drive unit 205 moves the antenna 300 in one direction from position 0 to position L.

[0064] The communication control unit 1012 controls the start of radio wave transmission from the antenna 300 (ACT2). In ACT2, for example, the communication control unit 1012 controls the start of radio wave transmission from the antenna 300 based on the start of movement of the antenna 300 from position 0. The communication control unit 1012 may also control the start of radio wave transmission from the antenna 300 based on a movement start notification from the drive unit 200. The movement start notification may indicate that the antenna 300 has started moving from position 0. The antenna 300 starts transmitting radio waves to read the information stored in the radio tag 600.

[0065] The first acquisition unit 1013 acquires the determination tag data for each wireless tag 600 to be determined (ACT3). In ACT3, the first acquisition unit 1013 acquires the determination tag data for each wireless tag 600 to be determined detected by the demodulation unit 110. If the first acquisition unit 1013 acquires the determination tag data (ACT3, YES), the process transitions from ACT3 to ACT4. If the first acquisition unit 1013 does not acquire the determination tag data (ACT3, NO), the process transitions from ACT3 to ACT5.

[0066] The first acquisition unit 1013 stores the determination tag data in the storage device 111 based on the acquisition of determination tag data for each wireless tag 600 to be determined (ACT4).

[0067] The communication control unit 1012 determines whether the movement of the antenna 300 has finished (ACT5). In ACT5, for example, the communication control unit 1012 determines whether the movement of the antenna 300 from position 0 to position L has finished. The communication control unit 1012 may also determine that the movement of the antenna 300 has finished based on a movement completion notification from the drive unit 200. The movement completion notification may indicate that the movement of the antenna 300 has finished upon reaching position L. If the movement of the antenna 300 has finished (ACT5, YES), the process transitions from ACT5 to ACT6. If the movement of the antenna 300 has not finished (ACT5, NO), the process transitions from ACT5 to ACT3.

[0068] The first acquisition unit 1013 repeats the processes of ACT3 and ACT4 from the time the antenna 300 starts moving at position 0 until it finishes moving at position L.

[0069] In ACT3, the first acquisition unit 1013 acquires determination tag data for each wireless tag 600 to be determined at multiple locations on the antenna 300. The first acquisition unit 1013 acquires determination tag data for each wireless tag 600 to be determined at each of the fixed intervals a between at least position 0 and position L. The first acquisition unit 1013 can acquire the position of the antenna 300 in cooperation with the drive unit 200.

[0070] In ACT4, the first acquisition unit 1013 stores the determination tag data for each wireless tag 600 to be determined at multiple locations on the antenna 300 in the storage device 111. The first acquisition unit 1013 stores the determination tag data for each wireless tag 600 to be determined at each of the locations a at a fixed interval between at least location 0 and location L in the storage device 111.

[0071] The communication control unit 1012 controls the termination of radio wave transmission from the antenna 300 (ACT6). In ACT6, for example, the communication control unit 1012 controls the termination of radio wave transmission from the antenna 300 based on the termination of the movement of the antenna 300 from position 0 to position L. The antenna 300 terminates radio wave transmission for reading the information stored in the wireless tag 600.

[0072] The input unit 1014 inputs judgment input data to the trained model 1113 (ACT7). In ACT7, for example, the input unit 1014 acquires judgment input data based on measurement data 1111 stored in the memory device 111. The judgment input data is the judgment tag data for each wireless tag 600 to be judged at multiple positions of the antenna 300. The multiple positions of the antenna 300 in the judgment input data are the same as the multiple positions of the antenna 300 in the training data 1112. For example, the judgment input data is the judgment tag data for each wireless tag 600 to be judged at each of the positions a at a fixed interval between positions 0 and L. The input unit 1014 inputs the acquired judgment input data to the trained model 1113.

[0073] The second acquisition unit 1015 acquires judgment output data from the trained model 1113 based on the input of judgment input data to the trained model 1113 by the input unit 1014 (ACT8). The judgment output data includes data indicating whether the location of each wireless tag 600 to be judged falls within the first range or the second range.

[0074] The output unit 1016 outputs a determination result to the terminal 400, which includes the determination output data acquired by the second acquisition unit 1015. The determination result may include information stored in each wireless tag 600 that is the target of determination and read by the reader 100. The terminal 400 may process the information stored in each wireless tag 600 that is the target of determination, depending on whether each wireless tag 600 falls within the first range or the second range. The terminal 400 may process the information stored in each wireless tag 600 that falls within the first range. The terminal 400 does not have to process the information stored in each wireless tag 600 that falls within the second range.

[0075] This section describes an example of measuring training tag data included in the training data 1112 used to generate the trained model 1113. Figure 8 shows an example of the arrangement of multiple wireless tags 601-615 that are the learning targets, and is a plan view of the counter stand 700 seen from above.

[0076] The wireless tags 601-615 are positioned on a virtual plane including the horizontal plane of the counter base 700, parallel to the vertical direction in which the antenna 300 moves. The wireless tags 601-605 are positioned at different locations from each other, and are included in the first range 81. The wireless tags 601-605 are positioned sequentially, moving away from position 0. As the antenna 300 moves from position 0 to position L, it passes through the wireless tags 601-605 in order, corresponding to each wireless tag.

[0077] The radio tags 606-610 are positioned at different locations from each other, but within the second range 82. The radio tags 606-610 are arranged in order, approaching position 0. Antenna 300 moves away from radio tags 606-610 while moving from position 0 to position L.

[0078] The radio tags 611-615 are positioned at different locations from each other, but within the second range 82. The radio tags 611-615 are positioned in order away from position L. The antenna 300 moves closer to the radio tags 611-615 as it moves from position 0 to position L.

[0079] Note that the number and arrangement of wireless tags for multiple learning targets are not limited to the example shown in Figure 8. It is sufficient if some of the wireless tags for multiple learning targets are placed in the first range 81 and the remaining wireless tags for multiple learning targets are placed in the second range 82.

[0080] This section describes the training tag data for wireless tags 601-615 at multiple locations on antenna 300. Here, phase data is used as an example of training tag data.

[0081] Figure 9 is a graph showing an example of training tag data for wireless tags 601-605 at multiple locations on antenna 300. The horizontal axis represents the position of antenna 300. Position L is assumed to be 600 mm. The vertical axis represents the phase. The graph shows the phase for each of the wireless tags 601 to 605 at a fixed interval 'a' between position 0 and position L.

[0082] The phases of each of the wireless tags 601-605 change as the position of antenna 300 changes. This is because the distance between antenna 300 and each of the wireless tags 601-605 changes as antenna 300 moves. Regardless of the position of antenna 300, the phases of each of the wireless tags 601-605 are different. This is because the distance between antenna 300 and each of the wireless tags 601-605 is different.

[0083] Figure 10 is a graph showing an example of training tag data for wireless tags 606-610 at multiple locations on antenna 300. The horizontal axis represents the position of antenna 300. Position L is assumed to be 600 mm. The vertical axis represents the phase. The graph shows the phase for each of the wireless tags 606 to 610 at a fixed interval 'a' between position 0 and position L.

[0084] The phases of each wireless tag 606-610 change as the position of antenna 300 changes. Regardless of the position of antenna 300, the phases of each wireless tag 606-610 are different.

[0085] Figure 11 is a graph showing an example of training tag data for wireless tags 611-615 at multiple locations on antenna 300. The horizontal axis represents the position of antenna 300. Position L is assumed to be 600 mm. The vertical axis represents the phase. The graph shows the phase for each of the wireless tags 610 to 615 at a fixed interval 'a' between position 0 and position L.

[0086] The phases of each of the wireless tags 611-615 change as the position of antenna 300 changes. Regardless of the position of antenna 300, the phases of each of the wireless tags 611-615 are different.

[0087] We have explained the characteristics of phase data, and the same applies to the characteristics of radio wave reception strength data. The radio wave reception strength of each of the wireless tags 601 to 615 changes as the position of antenna 300 changes. This is because the distance between antenna 300 and each of the wireless tags 601 to 615 changes as antenna 300 moves. Regardless of the position of antenna 300, the radio wave reception strength of each of the wireless tags 601 to 615 will be different. This is because the distance between antenna 300 and each of the wireless tags 601 to 615 is different.

[0088] As described above, the processor 101 of the reading device 100 measures the learning tag data of multiple wireless tags 600 to be learned at multiple locations on the antenna 300. The processor 101 stores the measured learning tag data in the storage device 111. The learning tag data constitutes the learning data 1112.

[0089] Figure 12 is a flowchart showing an example of the process by which the processor 101 of the reading device 100 generates a trained model 1113. The processing procedure described below is merely an example, and each process may be modified as much as possible. Furthermore, depending on the embodiment, steps in the processing procedure described below may be omitted, replaced, or added as appropriate.

[0090] The model processing unit 1017 may start the process of generating a trained model 1113 at any time and create a new trained model 1113. The model processing unit 1017 may also start the process of generating a trained model 1113 at any time and update the trained model 1113.

[0091] The model processing unit 1017 acquires the training data 1112 (step S10). In step S10, the model processing unit 1017 acquires the training data 1112 from the memory device 111.

[0092] The model processing unit 1017 generates a trained model 1113 by machine learning based on the training data 1112 (step S11). In step S11, for example, the model processing unit 1017 trains the training data 1112 by machine learning. The model processing unit 1017 estimates the relationship between training tag data of multiple wireless tags 600 to be trained at multiple locations of the antenna 300 and ground truth data indicating whether each of the multiple wireless tags 600 to be trained falls into a first range or a second range. Based on the estimation, the model processing unit 1017 generates a trained model 1113. Machine learning is, but is not limited to, a neural network.

[0093] The training tag data of the wireless tag 600, whether phase data or radio wave reception strength data, changes depending on the distance between the antenna 300 and the training tag 600. The pattern of the training tag data of the training tag 600 at multiple positions on the antenna 300 is different for each position of the training tag 600. There may be a certain correlation between the training tag data of the training tag 600 at multiple positions on the antenna 300 and the position of the wireless tag.

[0094] The model processing unit 1017 saves the generated trained model 1113 to the memory device 111 (step S12).

[0095] According to this embodiment, the communication device can input tag data for each radio tag at multiple antenna locations into a trained model. The communication device can obtain data from the trained model indicating whether each radio tag falls within a first range or a second range. This allows the communication device to improve the accuracy of the data indicating whether each of the multiple radio tags falls within the first range or the second range. As a result, the communication device provides a technique to improve the accuracy of determining the location of radio tags.

[0096] The tag data may include phase data. The tag data may also include radio wave reception strength data. This allows the communication device to use the phase data or radio wave reception strength data to improve the accuracy of the data indicating whether each of the multiple radio tags falls within the first range or the second range.

[0097] The communication device can generate a trained model through machine learning based on training data. The training data may include tag data for multiple radio tags at multiple locations on the antenna. The training data may also include data indicating whether each of the multiple radio tags falls within the first range or the second range. This allows the communication device to generate a highly accurate trained model.

[0098] Modifications of this embodiment will now be described. An example has been described in which the processor 101 of the reading device 100 implements a model processing unit 1017 that generates a trained model 1113, but the example is not limited to this. The generation of the trained model 1113 may be implemented by a device other than the reading device 100.

[0099] An example has been described in which the storage device 111 of the reading device 100 stores the training data 1112 and the trained model 1113, but it is not limited to this. At least one of the training data 1112 and the trained model 1113 may be stored in a device other than the reading device 100.

[0100] An example has been described in which the processor 101 of the reader 100 acquires decision output data through software processing, but it is not limited to this example. The communication device 10 may include an inference unit using a trained model. In this example, the input unit 1014 of the processor 101 inputs decision input data to the inference unit. The reader 100 inputting decision input data to the trained model includes transmitting the decision input data from the reader 100 to the inference unit. The second acquisition unit 1015 of the processor 101 acquires decision output data from the trained model based on the input of decision input data to the trained model. The reader 100 acquiring decision output data from the trained model includes the reader 100 receiving decision output data from the inference unit.

[0101] The communication device may be implemented using multiple devices as described in the example above, or it may be implemented using a single device that integrates the functions of multiple devices. The reader, drive device, and antenna may be implemented using a single device that integrates their functions. The reader may be implemented using multiple devices with distributed functions.

[0102] The program may be transferred while stored in the device according to the embodiment, or it may be transferred without being stored in the device. In the latter case, the program may be transferred via a network, or it may be transferred while recorded on a recording medium. The recording medium is a non-temporary tangible medium. The recording medium is a computer-readable medium. The recording medium can be any medium that is capable of storing a program and is readable by a computer, such as a CD-ROM or memory card, and its form is not limited.

[0103] While several embodiments of the present invention have been described, these embodiments are presented as examples only and are not intended to limit the scope of the invention. These novel embodiments can be carried out in a variety of other forms, and various omissions, substitutions, and modifications can be made without departing from the spirit of the invention. These embodiments and their variations are included in the scope and spirit of the invention, as well as in the claims of the invention and its equivalents. [Explanation of Symbols]

[0104] 1...Communication system, 10...Communication device, 81...First range, 82...Second range, 100...Reader, 101...Processor, 102...ROM, 103...RAM, 104...First connection interface, 105...Second connection interface, 106...High-frequency front-end section, 107...Digital amplitude modulation section, 108...DA conversion section, 109...AD conversion section, 110...Demodulation section, 111...Storage device, 112...Bus, 200...Drive unit, 201...Processor, 202...ROM, 203...RAM, 204...Connection interface S, 205...Drive unit, 206...Home position sensor, 208...Bus, 211...Rotating axis, 212...Rail, 213...Moving stage, 300...Antenna, 400...Terminal, 500...Item, 600...Wireless tag, 601~615...Wireless tag, 700...Counter stand, 1011...Movement control unit, 1012...Communication control unit, 1013...First acquisition unit, 1014...Input unit, 1015...Second acquisition unit, 1016...Output unit, 1017...Model processing unit, 1111...Measurement data, 1112...Training data, 1113...Trained model.

Claims

1. Antenna and, A drive unit for moving the position of the aforementioned antenna, A first acquisition unit acquires tag data for each wireless tag based on the radio waves of each wireless tag received by the antenna at multiple locations on the antenna, An input unit that inputs tag data of each wireless tag at multiple locations of the antenna acquired by the first acquisition unit into a trained model, A second acquisition unit obtains data from the trained model indicating whether each wireless tag falls within a first range or a second range, based on the input of tag data for each wireless tag by the input unit. Equipped with, A communication device in which the trained model is a model generated by machine learning based on training data that includes tag data of multiple radio tags at multiple locations of the antenna and data indicating whether each of the multiple radio tags falls within the first range or the second range.

2. The communication device according to claim 1, wherein the tag data includes phase data.

3. The communication device according to claim 1 or 2, wherein the tag data includes radio wave reception strength data.

4. At multiple locations of the antenna, the tag data of each wireless tag is acquired based on the radio waves of each wireless tag received by the antenna, The trained model is input with tag data from each wireless tag at multiple locations of the antenna, Based on the input of tag data for each wireless tag to the trained model, data is obtained from the trained model indicating whether each wireless tag falls within a first range or a second range. Equipped with, Information processing method, wherein the trained model is a model generated by machine learning based on training data which includes tag data of multiple radio tags at multiple locations of the antenna and data indicating whether each of the multiple radio tags falls within the first range or the second range.

5. On the computer, A function to acquire tag data for each wireless tag based on the radio waves of each wireless tag received by the antenna at multiple positions of the antenna, The trained model has a function to input tag data for each wireless tag at multiple locations of the antenna, Based on the input of tag data for each wireless tag into the aforementioned trained model, a function is provided to obtain data from the trained model indicating whether each wireless tag falls within a first range or a second range. An information processing program for executing, An information processing program, wherein the trained model is a model generated by machine learning based on training data that includes tag data of multiple radio tags at multiple locations of the antenna and data indicating whether each of the multiple radio tags falls within the first range or the second range.

Images