Authentication system, learning device and method thereof, inference device and method thereof, and storage medium

Abstract

An authentication system, a learning device and a method and a program thereof, and a reasoning device and a method and a program thereof are provided, which suppress a deviation in a distance from a reading device when a user is authenticated. In an authentication system (1), a generation section (10) generates a learned model by corresponding a reception strength of an interrogation signal received from a reading device (4) by a portable terminal (3) disposed at a position at a distance of an authentication distance from the reading device (4) to an output, and corresponding a transmission strength of the interrogation signal and terminal information of the portable terminal (3) to an input. The terminal information includes at least any of information of a state and a category of the portable terminal (3). An estimation section (11) obtains, as an estimation result using the learned model, a reception strength of an interrogation signal received when the portable terminal (3) of a user is located at a position at a distance of an authentication distance from the reading device (4), from a transmission strength of the interrogation signal and data including the terminal information of the portable terminal (3).

Description

Technical Field

[0001] This disclosure relates to authentication systems, learning devices, reasoning devices, learning methods, reasoning methods, learning programs, and reasoning programs. Background Technology

[0002] Patent Document 1 discloses an example of a security system. The security system includes a reader installed near the entrance / exit of a target area and a portable terminal carried by the user. The reader sends an inquiry signal to the portable terminal via a wireless signal, requesting identification information. Upon receiving the inquiry signal, the portable terminal sends identification information to the reader via a wireless signal. The user is authenticated based on the identification information received by the reader.

[0003] Existing technical documents

[0004] Patent documents

[0005] Patent Document 1: Japanese Patent Application Publication No. 2006-352703 Summary of the Invention

[0006] The problem the invention aims to solve

[0007] In Patent Document 1, the received strength of the wireless signal at a portable terminal located at a certain distance from the reading device varies depending on the state or type of the portable terminal. Therefore, the distance between the user and the reading device at the time of authentication is inconsistent.

[0008] This disclosure was made to solve such a problem. This disclosure provides an authentication system, as well as a learning device, inference device, learning method, inference method, learning program, and inference program, that suppresses deviations in the distance between the user and the reading device during authentication.

[0009] means for solving problems

[0010] The learning device of the authentication system disclosed herein includes: an acquisition unit that acquires learning data; and a generation unit that uses the learning data acquired by the acquisition unit to generate a learning completion model. The learning data includes: the transmission strength of a first interrogation signal, which is a wireless signal, transmitted by a reading device located at an authentication location; the reception strength of the first interrogation signal received by a first portable terminal positioned at a predetermined authentication distance from the reading device; and first terminal information, which includes at least any information among status and category associated with the first portable terminal. The learning completion model takes the transmission strength of a second interrogation signal, which is a wireless signal, transmitted by the reading device and the second terminal information as input, and outputs the reception strength of the second interrogation signal received by a second portable terminal located at the authentication distance from the reading device. The second terminal information includes at least any information among status and category associated with the second portable terminal held by a user.

[0011] The inference apparatus of the authentication system disclosed herein includes: an acquisition unit that acquires operational data; and an estimation unit that uses a learned model to obtain an estimation result based on the operational data acquired by the acquisition unit. The learned model is a model pre-generated through learning in which the reception strength of a first portable terminal receiving a first interrogation signal (a wireless signal) sent by a reading device located at the authentication location is correlated with the output, and the transmission strength of the first interrogation signal sent by the reading device and first terminal information are correlated with the input. The first portable terminal is located at a predetermined distance from the reading device. At the authentication distance location, the first terminal information includes at least any information of the status and category associated with the first portable terminal, and the operation data includes: the transmission strength of the second interrogation signal as a wireless signal transmitted by the reading device; and second terminal information, which includes at least any information of the status and category associated with the second portable terminal that receives the second interrogation signal from the reading device, the second portable terminal being held by a user, and the estimation unit estimates the reception strength of the second interrogation signal received by the second portable terminal at a location at a distance from the reading device equal to the authentication distance, as the estimation result.

[0012] The learning method of the authentication system disclosed herein includes: an acquisition step, in which a computer acquires learning data; and a generation step, in which the computer uses the learning data acquired in the acquisition step to generate a learning-completed model, the learning data including: the transmission strength of a first interrogation signal as a wireless signal transmitted by a reading device located at an authentication location; the reception strength of the first interrogation signal received by a first portable terminal configured at a location at a predetermined authentication distance from the reading device; and first terminal information, which includes at least any information among status and category associated with the first portable terminal, the learning-completed model taking the transmission strength of a second interrogation signal as a wireless signal transmitted by the reading device and the second terminal information as input, and taking the reception strength of the second interrogation signal received by a second portable terminal at a location at the authentication distance from the reading device as output, wherein the second terminal information includes at least any information among status and category associated with the second portable terminal held by the user.

[0013] The inference method of the authentication system disclosed herein includes: an acquisition step, in which a computer acquires runtime data; and an estimation step, in which the computer uses a learned model to obtain an estimation result based on the runtime data acquired in the acquisition step. The learned model is a model pre-generated through learning in which the received strength of a first query signal (a wireless signal) sent by a reading device located at the authentication location to a first portable terminal is correlated with its output, and the transmission strength of the first query signal sent by the reading device and information about the first terminal are correlated with the input. The first portable terminal is located at a predetermined distance from the reading device. At a location at a predetermined authentication distance, the first terminal information includes at least any information of the status and category associated with the first portable terminal, and the operational data includes: the transmission strength of a second interrogation signal as a wireless signal transmitted by the reading device; and second terminal information, which includes at least any information of the status and category associated with a second portable terminal that receives the second interrogation signal from the reading device, the second portable terminal being held by a user, and in the estimation step, the reception strength of the second interrogation signal received by the second portable terminal at a location at a distance equal to the authentication distance from the reading device is estimated as the estimation result.

[0014] The learning procedure of the authentication system disclosed herein causes a computer to perform: an acquisition step, acquiring learning data; and a generation step, using the learning data acquired in the acquisition step to generate a completed learning model, the learning data including: the transmission strength of a first interrogation signal as a wireless signal sent by a reading device located at the authentication location; the reception strength of the first interrogation signal received by a first portable terminal configured at a location at a predetermined authentication distance from the reading device; and first terminal information, which includes at least any information of state and category associated with the first portable terminal, the completed learning model taking the transmission strength of a second interrogation signal as a wireless signal sent by the reading device and the second terminal information as input, and taking the reception strength of the second interrogation signal received by a second portable terminal at a location at the authentication distance from the reading device as output, wherein the second terminal information includes at least any information of state and category associated with the second portable terminal held by the user.

[0015] The inference procedure of the authentication system disclosed herein causes a computer to perform: an acquisition step, acquiring runtime data; and an estimation step, using a learned model, obtaining an estimation result based on the runtime data acquired in the acquisition step. The learned model is a model pre-generated through learning in which the received strength and output of a first interrogation signal (a wireless signal) sent by a reading device located at the authentication location to a first portable terminal are correlated, and the transmission strength of the first interrogation signal sent by the reading device and information about the first terminal are correlated with the input, wherein the first portable terminal is located at a predetermined distance from the reading device. At the authentication distance location, the first terminal information includes at least any information of the status and category associated with the first portable terminal, the operational data includes: the transmission strength of the second interrogation signal as a wireless signal transmitted by the reading device; and second terminal information, which includes at least any information of the status and category associated with a second portable terminal that receives the second interrogation signal from the reading device, the second portable terminal being held by a user, and in the estimation step, the reception strength of the second interrogation signal received by the second portable terminal at a location at a distance from the reading device equal to the authentication distance is estimated as the estimation result.

[0016] The authentication system disclosed herein includes: an acquisition unit that acquires learning data; a generation unit that uses the learning data acquired by the acquisition unit to generate a learning completion model; and an estimation unit that uses the learning completion model to obtain an estimation result based on operational data acquired by the acquisition unit. The learning data includes: the transmission strength of a first interrogation signal, which is a wireless signal, transmitted by a reading device located at an authentication location; the reception strength of the first interrogation signal received by a first portable terminal positioned at a predetermined authentication distance from the reading device; and first terminal information, which includes at least any one of status and category information related to the first portable terminal. The generation unit transmits the first portable terminal... The received strength of the first interrogation signal is correlated with the output, and the transmission strength of the first interrogation signal sent by the reading device and the first terminal information are correlated with the input to generate the learned model. The operational data includes: the transmission strength of the second interrogation signal sent by the reading device as a wireless signal; and second terminal information, which includes at least any information of the state and category associated with a second portable terminal that receives the second interrogation signal from the reading device, which is held by a user. The estimation unit estimates the received strength of the second interrogation signal when the second portable terminal is located at a distance from the reading device equal to the authentication distance, as the estimation result.

[0017] The effects of the invention

[0018] The authentication system or its learning device, reasoning device, learning method, reasoning method, learning program, or reasoning program disclosed herein can suppress deviations in the distance between the user and the reading device when the user is authenticated. Attached Figure Description

[0019] Figure 1 This is a structural diagram of the authentication system in Implementation Method 1.

[0020] Figure 2 This diagram illustrates an example of the actions during the learning phase of the authentication system in Implementation Method 1.

[0021] Figure 3 This diagram illustrates an example of the actions taken during the operation phase of the authentication system in Implementation Method 1.

[0022] Figure 4 This is a flowchart illustrating an example of the actions during the operation phase of the authentication system in Implementation 1.

[0023] Figure 5 This is a hardware structure diagram of the main parts of the authentication system in Implementation Method 1. Detailed Implementation

[0024] The embodiments of the present disclosure will be described with reference to the accompanying drawings. In the drawings, identical or equivalent parts are labeled with the same reference numerals, and repetitive descriptions are omitted or simplified as appropriate. Furthermore, the subject matter of this disclosure is not limited to the following embodiments; any structural elements of the embodiments can be modified or omitted without departing from the spirit of this disclosure.

[0025] Implementation method 1.

[0026] Figure 1 This is a structural diagram of the authentication system 1 according to implementation method 1.

[0027] Authentication system 1 is a system for authenticating users. In this example, authentication system 1 is applied to an access control system. The access control system manages the entry and exit of users relative to an access control area. The access control area is, for example, an area inside or outside a building. At the boundary of the access control area, there is a passageway through which users can enter the area. The passageway is, for example, the main entrance of a building or the entrance to a room inside the building. The location containing the passageway is an example of an authentication location. A door equipped with an electronic lock 2 is installed at the passageway. The user of authentication system 1 is, for example, a person who can enter the access control area. The user of authentication system 1 carries a portable terminal 3. The portable terminal 3 is equipped with the function of communicating with wireless signals. The wireless signal is, for example, a wireless communication signal based on electromagnetic waves. The wireless signal is, for example, a wireless PAN (Personal Area Network) or UWB (Ultra Wide Band) signal. The wireless signal may be, for example, a wireless communication signal based on the IEEE 802.11 standard or the IEEE 802.15.1 standard or their successors, related standards, alternative standards or similar standards. The portable terminal 3 maintains identification information for identifying the user holding the portable terminal 3. The identification information may be, for example, a unique ID number (ID: Identifier) ​​assigned to the user. The portable terminal 3 may be, for example, a portable, general-purpose information terminal device such as a smartphone, or a dedicated device such as a wireless tag. The portable terminal 3 may be owned by the user, or it may be an internal device of the authentication system 1 distributed to the user from an administrator of the authentication system 1.

[0028] The authentication system 1 includes a reader 4 and a controller 5. The reader 4 is connected to the controller 5. The controller 5 is connected to the electronic lock 2. The reader 4 is equipped with the function of wirelessly communicating with a portable terminal 3. The reader 4 is installed at the authentication location. For example, the reader 4 is installed on a wall adjacent to the entrance / exit management area of ​​the authentication location.

[0029] An example of the authentication process in authentication system 1 will be explained. The reading device 4 transmits an inquiry signal as a wireless signal. When the portable terminal 3 receives the inquiry signal with a reception strength exceeding a preset reception threshold, it transmits a response signal as a wireless signal to the reading device 4. In this example, the portable terminal 3 does not transmit a response signal if it does not receive the inquiry signal or if the reception strength of the received inquiry signal does not exceed the reception threshold. The response signal contains the identification information held by the portable terminal 3. When the reading device 4 receives the response signal from the portable terminal 3, it notifies the controller 5 of the identification information contained in the response signal. The controller 5 determines, for example, whether the user identified by the notified identification information is an authenticable person by comparing it with a preset list of authenticable users. Here, an authenticable person is, for example, someone who can enter and exit the access control area. When the user is authenticable, the controller 5 authenticates the user. At this time, the controller 5 unlocks the electronic lock 2. The user holding the portable terminal 3 opens the door after the electronic lock 2 has been unlocked and enters the access control area.

[0030] Here, even if the transmission strength of the wireless signal sent by the reading device 4 is fixed, the reception strength of the wireless signal of the portable terminal 3 at a fixed distance from the reading device 4 will vary depending on the state or type of the portable terminal 3 and the setting environment of the reading device 4. Therefore, the conditions for the authentication action in the authentication system 1 are adjusted by the adjustment device 6 of the authentication system 1, so that the authentication action of the user holding the portable terminal 3 is performed when the user is located at a fixed authentication distance from the reading device 4. Here, the authentication distance is, for example, a distance preset in the authentication system 1. Furthermore, in this example, information about the setting environment of the reading device 4 is collected by the environmental information collection device 7 of the authentication system 1. The environmental information collection device 7 may be installed, for example, around the reading device 4 or inside the reading device 4.

[0031] The adjustment device 6 may be, for example, one or more server computers. Some or all of the functions of the adjustment device 6 may be mounted on one or more devices located at the application authentication system 1 or remotely configured thereon. Some or all of the functions of the adjustment device 6 may also be mounted on hardware shared with other devices. Some or all of the functions of the adjustment device 6 may be installed on a virtual machine on a cloud service, or installed using storage or processing resources on a cloud service. The adjustment device 6 communicates, for example, with multiple internal devices constituting the adjustment device 6 via a communication network of the Internet, and with external devices of the adjustment device 6, including the portable terminal 3, the reading device 4, and the environmental information collection device 7. The adjustment device 6, for example, reads a program recorded on a recording medium and performs actions according to the program, thereby executing various functions. The recording medium for recording the program may be, for example, a recording medium built into an internal device constituting the adjustment device 6 or a recording medium connected to that internal device, or a recording medium whose information is read through an external device communicating with the adjustment device 6. The program may be a software package containing multiple software programs that operate on the multiple internal devices constituting the adjustment device 6.

[0032] The adjustment device 6 uses a learned model to adjust the conditions for authentication actions. The learned model is generated, for example, through machine learning methods such as supervised learning. The adjustment device 6 is an example of a learning device that generates a learned model. Furthermore, the adjustment device 6 is an example of an inference device that uses the learned model for estimation. The adjustment device 6 has multiple operation phases, including a learning phase and an operation phase. In the learning phase, the adjustment device 6 generates the learned model. The learning phase occurs, for example, during the setup, maintenance, or replacement of the reading device 4 or its peripheral equipment. Switching between the operation phases of the adjustment device 6 can be performed, for example, by an operator performing setup work. In the operation phase, the adjustment device 6 uses the learned model to adjust the conditions for authentication actions. In the operation phase, the authentication system 1 performs authentication actions for users entering and exiting the management area.

[0033] The adjustment device 6 includes an acquisition unit 8, a recording unit 9, a generation unit 10, an estimation unit 11, and a correction unit 12.

[0034] The acquisition unit 8 is a component equipped with the function of acquiring data. The acquisition unit 8 acquires data from the portable terminal 3, the reading device 4, and the environmental information collection device 7, etc. During the learning phase, the acquisition unit 8 acquires learning data. This learning data is used to generate the learned model. During the operation phase, the acquisition unit 8 acquires operation data. This operation data is used for estimation using the learned model.

[0035] During the learning phase, the acquisition unit 8 acquires learning data as follows: The operator switches the operation phase of the adjustment device 6 to the learning phase. The operator positions the portable terminal 3 held by the operator at a distance from the reading device 4 that is the authentication distance. This portable terminal 3 is an example of a first portable terminal. Furthermore, the operator's placement of the portable terminal 3 to the designated location may include placing the portable terminal 3 at the designated location and then leaving the location, or it may include the operator being at the designated location while holding the portable terminal 3. Here, the operator can switch the operation mode corresponding to the learning phase of the adjustment device 6 using the portable terminal 3. The operator, for example, notifies the adjustment device 6 that the placement of the portable terminal 3 is complete via the portable terminal 3. The reading device 4 sends an inquiry signal. At this time, the reading device 4 notifies the acquisition unit 8 of the transmission strength of the inquiry signal. This inquiry signal is an example of a first inquiry signal. Then, the portable terminal 3 receives the inquiry signal. At this time, the portable terminal 3 notifies the acquisition unit 8 of the information on the received strength of the interrogation signal and the terminal information of the portable terminal 3. This terminal information is an example of first terminal information. The terminal information includes at least any information in the state and category of the portable terminal 3. The information on the state of the portable terminal 3 may also include information on the environment in which the portable terminal 3 is located. In addition, the environmental information collection device 7 notifies the acquisition unit 8 of the setting environment of the reading device 4 at the time when the reading device 4 sends the interrogation signal. This environmental information is an example of first environmental information. For example, the environmental information collection device 7 may also notify the acquisition unit 8 of the setting environment information when it receives a notification of the transmission of the interrogation signal from the reading device 4. The acquisition unit 8 acquires learning data including the transmission strength of the interrogation signal, the received strength of the interrogation signal, the terminal information of the portable terminal 3, and the environmental information of the reading device 4, for example. In addition, when the acquisition unit 8 acquires information from different devices as learning data, for example, the acquisition unit 8 may process the information acquired at close intervals as a set of learning data, or each device may notify the acquisition unit 8 of information according to a request for learning data. For example, after the operator has completed the acquisition of the learning data, the operator will switch the adjustment device 6 from the action stage to the operation stage.

[0036] During operation, the acquisition unit 8 acquires operational data as follows: A user holds a portable terminal 3 and approaches the reading device 4. This portable terminal 3 is an example of a second portable terminal. At this time, the portable terminal 3 notifies the acquisition unit 8 of its terminal information. This terminal information is an example of second terminal information. The portable terminal 3 may notify the acquisition unit 8 of terminal information at any time, when the terminal information changes, or when a request is made from the acquisition unit 8. The reading device 4 periodically sends an interrogation signal, for example. At this time, the reading device 4 notifies the acquisition unit 8 of the transmission strength of the interrogation signal. The reading device 4 may notify the acquisition unit 8 of the transmission strength each time an interrogation signal is sent, when the transmission strength changes, or when a periodic request is made from the acquisition unit 8. This interrogation signal is an example of a second interrogation signal. Furthermore, the environmental information collection device 7 notifies the acquisition unit 8 of the setting environment of the reading device 4 at the time the reading device 4 sent the interrogation signal. This environmental information is an example of the second type of environmental information. The acquisition unit 8 acquires, for example, operational data including the transmission strength of the interrogation signal, terminal information of the portable terminal 3, and environmental information of the reading device 4. Furthermore, when the acquisition unit 8 acquires information from different devices as operational data, for example, the acquisition unit 8 can process information acquired at close intervals as a set of operational data, or each device can notify the other of information based on a request for operational data from the acquisition unit 8.

[0037] Recording unit 9 is a part equipped with the function of recording data. Recording unit 9 records the data acquired by acquisition unit 8. Recording unit 9, for example, accumulates and records learning data and operation data. Recording unit 9 stores transmission strength DB13 (Database), reception strength DB14, environmental information DB15, and terminal information DB16. Transmission strength DB13 is a database that stores transmission strength information from learning data and operation data. Reception strength DB14 is a database that stores reception strength information from learning data. Environmental information DB15 is a database that stores environmental information from learning data and operation data. Terminal information DB16 is a database that stores terminal information from learning data and operation data.

[0038] The generation unit 10 is the part that carries the function of generating a learned model. The generation unit 10 generates a learned model based on multiple learning data accumulated and recorded by the recording unit 9. At this time, the generation unit 10 corresponds the transmission strength of the query signal, the terminal information of the portable terminal 3, and the environmental information of the reading device 4 to the input, and corresponds the reception strength of the query signal to the output, thereby generating the learned model.

[0039] The estimation unit 11 is equipped with the function of estimating based on the learned model. The estimation unit 11 performs estimation, for example, using the learned model generated by the generation unit 10. The estimation unit 11 continuously acquires operational data obtained by the acquisition unit 8 from the recording unit 9 and inputs it into the learned model. The learned model accepts inputs of the transmission strength of the query signal, terminal information of the portable terminal 3, and environmental information of the reading device 4, and outputs the reception strength of the query signal. Using the output of the learned model, the estimation unit 11 estimates the reception strength of the query signal received by the portable terminal 3 held by the user at a position at a distance equal to the authentication distance from the reading device 4, and uses this as the estimation result.

[0040] The correction unit 12 is equipped with the following function: using the estimation result of the estimation unit 11, it corrects the conditions for the authentication operation in the authentication system 1, so that the authentication operation of the user holding the portable terminal 3 is performed when the user is located at a position at a fixed authentication distance from the reading device 4. The correction unit 12 includes a receiving correction unit 17 and a transmitting correction unit 18.

[0041] The reception correction unit 17 corrects the conditions for authentication by correcting the reception threshold of the portable terminal 3 held by the user. For example, when the reception strength of the interrogation signal obtained as an estimation result increases, the reception correction unit 17 increases the reception threshold accordingly, thereby correcting the condition to keep the range of the portable terminal 3 receiving the interrogation signal with a reception strength exceeding the reception threshold fixed. Conversely, when the reception strength of the interrogation signal obtained as an estimation result decreases, the reception correction unit 17 decreases the reception threshold accordingly, thereby correcting the condition to keep the range of the portable terminal 3 receiving the interrogation signal with a reception strength exceeding the reception threshold fixed. The reception correction unit 17 notifies the portable terminal 3 of the corrected reception threshold or the corrected value of the reception threshold. Based on the notified reception threshold information, the portable terminal 3 performs processing such as determining whether to send a response signal.

[0042] The transmission correction unit 18 corrects the conditions for the authentication operation by correcting the transmission strength of the interrogation signal sent by the reading device 4. For example, when the received strength of the interrogation signal obtained as an estimation result increases, the transmission correction unit 18 correspondingly decreases the transmission strength, thereby correcting the situation so that the range in which the portable terminal 3 receives the interrogation signal with a received strength exceeding a received threshold is kept fixed. Furthermore, for example, when the received strength of the interrogation signal obtained as an estimation result decreases, the transmission correction unit 18 correspondingly increases the transmission strength, thereby correcting the situation so that the range in which the portable terminal 3 receives the interrogation signal with a received strength exceeding a received threshold is kept fixed. The transmission correction unit 18, for example, notifies the reading device 4 of the corrected transmission strength or the corrected value of the transmission strength. The reading device 4 adjusts the signal strength of the interrogation signal to be transmitted based on the notified transmission strength information.

[0043] Next, use Figures 2 to 4 Examples of actions in authentication system 1 are explained.

[0044] Figure 2 This diagram illustrates an example of the actions taken during the learning phase of the authentication system 1 in Implementation Method 1.

[0045] During the learning phase, the operator positions the portable terminal 3, held at a distance from the reading device 4 equal to the authentication distance. The portable terminal 3 notifies the acquisition unit 8 of the received strength of the interrogation signal sent by the reading device 4, located at the authentication distance. At this time, the portable terminal 3 collects information about its status. This information includes, for example, the orientation of the portable terminal 3, its holding position, the geomagnetic field at its location, and the electromagnetic field at its location. The orientation information indicates the angle at which the portable terminal 3 is facing the reading device 4. This orientation information is collected in the portable terminal 3, for example, based on the direction of arrival of the interrogation signal sent from the reading signal. The holding position information indicates how the user holds the portable terminal 3. Holding positions include, for example, a breast pocket, a back pocket, a handbag, or a backpack. The holding position is estimated, for example, using a model learned in advance through machine learning and based on the measurement values ​​of the accelerometer sensor mounted on the portable terminal 3. The operator can also hold the portable terminal 3 while walking within a range that does not change in distance from the reading device 4, allowing the portable terminal 3 to estimate the holding position. Alternatively, the holding position information can be selected and input by the operator into the portable terminal 3. Geomagnetic information at the location of the portable terminal 3 is collected, for example, based on the measurement values ​​of the magnetic sensor mounted on the portable terminal 3. Electromagnetic wave information at the location of the portable terminal 3 is collected, for example, based on the signal strength of the wireless signals transmitted and received via electromagnetic waves by the wireless communication module mounted on the portable terminal 3. The portable terminal 3 notifies the acquisition unit 8 of the collected information on the status of the portable terminal 3 and the pre-recorded information on the type of the portable terminal 3 as terminal information. The type information of the portable terminal 3 may include, for example, the model information of the portable terminal 3 itself or the model information of the wireless communication module mounted on the portable terminal 3.

[0046] When the reading device 4 sends an interrogation signal, the environmental information collection device 7 collects information about the environment in which the reading device 4 is installed. This environmental information includes, for example, information about structures surrounding the reading device 4, geomagnetic information at the authentication location, and electromagnetic wave information at the authentication location. Information about structures surrounding the reading device 4 is input into the environmental information collection device 7 by operators during the installation, maintenance, replacement, or removal of the reading device 4 or the structure. This information includes information about the positional relationship between the reading device 4 and the structure. This positional relationship information includes, for example, information about the distance from and direction relative to the reading device 4 of any metal-containing devices located near the reading device 4. This positional relationship information also includes, for example, information about the height of the reading device 4 from the ceiling, the height of the reading device 4 from the ground, or the width of the passage from the reading device 4 to the opposite wall. Information about structures surrounding the reading device 4 may include, for example, the shape, thickness, or material of a door located at an access point adjacent to the reading device 4, or the wall on which the reading device 4 is located. Geomagnetic information at the authentication location is collected, for example, based on measurements from a magnetic sensor mounted on the environmental information collection device 7. Electromagnetic wave information at the authentication location is collected, for example, based on the signal strength of wireless signals transmitted and received via electromagnetic waves by a wireless communication module mounted on the environmental information collection device 7. The environmental information collection device 7 notifies the acquisition unit 8 of the collected environmental information as environmental information.

[0047] The acquisition unit 8 acquires the notified information as learning data. Additionally, the operator can cause the acquisition unit 8 to acquire multiple learning data sets, enabling it to obtain learning data containing different terminal or environmental information. In this case, the operator may, for example, change the posture or holding position of the portable terminal 3, use a different model of the portable terminal 3, activate equipment that generates electromagnetic waves or fields at the authentication location, or change the configuration of structures surrounding the reading device 4. These multiple learning data sets are then recorded as a learning dataset in the recording unit 9.

[0048] Then, the generation unit 10 uses a learning dataset containing multiple learning data to generate a fully learned model.

[0049] Figure 3 This diagram illustrates an example of the actions taken during the operation phase of the authentication system 1 in Implementation Method 1.

[0050] During the operation phase, the user holds the portable terminal 3 and approaches the reading device 4 at the authentication location. At this time, the portable terminal 3 collects information about its status. This information is collected during the operation phase in the same way as during the learning phase. The portable terminal 3 then notifies the acquisition unit 8 of the collected status information and pre-recorded information about its category as terminal information.

[0051] The reading device 4 sends an inquiry signal. At this time, the reading device 4 notifies the acquisition unit 8 of the transmission strength of the inquiry signal.

[0052] When the reading device 4 sends an interrogation signal, the environmental information collection device 7 collects information about the setting environment of the reading device 4. During the operation phase, the setting environment information of the reading device 4 is collected in the same way as during the learning phase. The environmental information collection device 7 notifies the acquisition unit 8 of the collected setting environment information as environmental information.

[0053] The information obtained by the acquisition department 8 is used as operational data.

[0054] The estimation unit 11 inputs the learned model generated by the generation unit 10 and the operation data acquired by the acquisition unit 8 into it, thereby obtaining the reception strength of the interrogation signal received at a position where the distance between the portable terminal 3 held by the user and the reading device 4 is the authentication distance, as an estimation result. The correction unit 12 uses the estimation result of the estimation unit 11 to correct the reception threshold of the portable terminal 3 or the transmission strength of the interrogation signal, etc.

[0055] When the portable terminal 3 receives an interrogation signal with a reception strength exceeding the calibrated reception threshold, it transmits a response signal as a wireless signal to the reading device 4. The controller 5, which is notified from the reading device 4 of the identification information contained in the response signal, determines whether the user identified by the identification information can be authenticated. Thus, the authentication process is performed when the user holding the portable terminal 3 is located at a fixed authentication distance from the reading device 4.

[0056] Figure 4 This is a flowchart illustrating an example of the actions during the operation phase of the authentication system 1 in Implementation 1.

[0057] In step S1, the estimation unit 11 reads in the learned model. For example, the estimation unit 11 reads in the learned model from the generation unit 10. Then, the processing of the adjustment device 6 proceeds to step S2.

[0058] In step S2, the adjustment device 6 determines whether the acquisition unit 8 has acquired the operating time data. If the determination result is no, the processing of the adjustment device 6 returns to step S2. On the other hand, if the determination result is yes, the recording unit 9 records the operating time data acquired by the acquisition unit 8. Then, the processing of the adjustment device 6 proceeds to step S3.

[0059] In step S3, the estimation unit 11 inputs the operational data acquired by the acquisition unit 8 into the learned model, thereby obtaining the received strength of the interrogation signal received by the user's portable terminal 3 at a position at a distance equal to the authentication distance from the reading device 4 as an estimation result. Then, the processing of the adjustment device 6 proceeds to step S4.

[0060] In step S4, the correction unit 12 determines whether the environmental information contained in the newly acquired operating data by the acquisition unit 8 has changed compared to the environmental information contained in the previously acquired newly acquired operating data by the acquisition unit 8. If the determination result is yes, the processing of the adjustment device 6 proceeds to step S5. On the other hand, if the determination result is no, the processing of the adjustment device 6 proceeds to step S6. Furthermore, when the environmental information includes continuous values ​​such as the intensity of an electromagnetic field, the correction unit 12 may also determine that the environmental information has not changed if the change in the continuous value does not exceed a preset range.

[0061] In step S5, the transmission correction unit 18 corrects the transmission strength of the interrogation signal transmitted by the reading device 4 based on the estimation result of the estimation unit 11. Furthermore, the transmission correction unit 18 can retain the transmission strength correction even if the corrected value of the transmission strength is less than a preset range. Then, the processing of the adjustment device 6 proceeds to step S6.

[0062] In step S6, the receiving correction unit 17 corrects the receiving threshold of the portable terminal 3 whose terminal information is included in the operation data based on the estimation result of the estimation unit 11. Furthermore, the receiving correction unit 17 may retain the receiving threshold correction even if the corrected value is less than a preset range. Here, since the transmission strength was corrected in step S5, the receiving correction unit 17 can also correct the receiving threshold of other portable terminals 3 whose terminal information is included in the operation data, such that authentication is performed when these other portable terminals 3 are located at a fixed authentication distance from the reading device 4. This other portable terminal 3 is an example of a third portable terminal. Then, the processing of the adjustment device 6 proceeds to step S2.

[0063] As explained above, the authentication system 1 of Embodiment 1 includes an acquisition unit 8, a generation unit 10, and an estimation unit 11. The acquisition unit 8 acquires learning data and operational data. The learning data includes: the transmission strength of an interrogation signal, which is a wireless signal, sent by the reading device 4; the reception strength of the interrogation signal received by a portable terminal 3 located at an authentication distance from the reading device 4; and terminal information of the portable terminal 3. The terminal information of the portable terminal 3 includes at least any information from the portable terminal 3's state and category. The generation unit 10 generates a learning-completed model using the learning data. At this time, the generation unit 10 correlates the reception strength of the interrogation signal received by the portable terminal 3 with the output, and correlates the transmission strength of the interrogation signal sent by the reading device 4 and the terminal information of the portable terminal 3 with the input. The estimation unit 11 uses the learning-completed model to obtain an estimation result based on the operational data. The operational data includes: the transmission strength of the interrogation signal, which is a wireless signal, sent by the reading device 4; and the terminal information of the portable terminal 3 held by the user that receives the interrogation signal from the reading device 4. The estimation unit 11 obtains the reception strength of the interrogation signal received by the portable terminal 3 when it is located at a distance from the reading device 4 that is the authentication distance as the estimation result.

[0064] With this structure, the received signal strength of the portable terminal 3 located at a fixed distance from the reading device 4 can be estimated based on the state or type of the portable terminal 3. This suppresses deviations in the distance between the user and the reading device 4 during authentication. Therefore, users moving while holding the portable terminal 3 can receive authentication smoothly and stably. It also prevents passageway congestion caused by authentication failures even when the user approaches the reading device 4, thus further improving user convenience. Furthermore, by suppressing authentication before the user approaches the reading device 4, it further enhances security in areas where the authentication system 1 is applied.

[0065] Furthermore, the learned model incorporates environmental information representing the setup environment of the reading device 4 into the input. This allows for the estimation of the received wireless signal strength of the portable terminal 3 located at a fixed distance from the reading device 4 based on the authentication device's setup environment. Consequently, deviations in the distance between the user and the reading device 4 during authentication are effectively suppressed.

[0066] Furthermore, the information about the installation environment of the reading device 4 includes information about the positional relationship between the reading device 4 and structures surrounding it. Therefore, the received signal strength of the portable terminal 3 located at a fixed distance from the reading device 4 can be estimated based on the reflection, absorption, and other effects of the structures on the wireless signal. Thus, deviations in the distance from the reading device 4 during user authentication are effectively suppressed.

[0067] Furthermore, the information about the environment in which the reading device 4 is set includes information about electromagnetic waves or electromagnetic fields at the authentication location or the location of the portable terminal 3. This allows for the estimation of the received wireless signal strength of the portable terminal 3, located at a fixed distance from the reading device 4, based on the influence of external electromagnetic waves or electromagnetic fields. Therefore, deviations in the distance between the user and the reading device 4 during authentication are effectively suppressed. Alternatively, if there is a difference in the accuracy of the geomagnetic field measurement in the portable terminal 3 compared to the accuracy of the geomagnetic field measurement in the environmental information collection device 7, information with higher accuracy can be used. For example, the environmental information collection device 7 can be fixedly installed; therefore, it may sometimes be equipped with a magnetometry module with higher accuracy than the movable portable terminal 3. In this case, the authentication system 1 can also uniformly use the measurement value of the environmental information collection device 7 as the geomagnetic field value. For example, if the change in geomagnetic field between the location of the portable terminal 3 and the authentication location is small based on location, the measurement value of the environmental information collection device 7 can also be used as information about the electromagnetic field at the location of the portable terminal 3.

[0068] Furthermore, the terminal information includes information about the posture or holding position of the portable terminal 3. Therefore, the received strength of the wireless signal of the portable terminal 3 at a fixed distance from the reading device 4 can be estimated based on the influence of the state in which the portable terminal 3 is held. For example, when the transmit / receive characteristics of the antenna for transmitting and receiving wireless signals are anisotropic, the influence of differences in the direction of arrival of the wireless signal is considered. Furthermore, the influence of the user's own body, clothing, or possessions on the absorption of the wireless signal is also considered. Therefore, deviations in the distance from the reading device 4 when the user is authenticated are effectively suppressed.

[0069] Furthermore, the terminal information includes the model information of the portable terminal 3 or the model information of the wireless communication module equipped in the portable terminal 3. Therefore, the received signal strength of the portable terminal 3 located at a fixed distance from the reading device 4 can be estimated based on the characteristics of each model. Thus, deviations in the distance from the reading device 4 during user authentication are effectively suppressed.

[0070] Furthermore, when the user's portable terminal 3 receives an interrogation signal with a reception strength exceeding the reception threshold, it sends a response signal as a wireless signal to the reading device 4. The authentication system 1 includes a correction unit 12. The correction unit 12 uses the estimation result from the estimation unit 11 to correct the reception threshold of the portable terminal 3 or the transmission strength of the interrogation signal sent by the reading device 4, so that the portable terminal 3 sends the response signal at a fixed authentication distance from the reading device 4. This correction fixes the positional relationship between the portable terminal 3 sending the response signal and the reading device 4, thus effectively suppressing deviations in the distance between the user and the reading device 4 during authentication.

[0071] Furthermore, the transmission correction unit 18 retains the correction of the transmission strength of the interrogation signal sent by the reading device 4 if the information of the acquired setting environment remains unchanged. If the transmission strength of the interrogation signal sent by the reading device 4 is corrected, the authentication conditions of multiple portable terminals 3 that communicate wirelessly with the reading device 4 will be affected. When the necessity for correction is low, the correction of the transmission strength with a large impact range is retained. Therefore, even when there are multiple portable terminals 3 around the reading device 4, the authentication conditions of each portable terminal 3 can be made more stable.

[0072] Furthermore, while the transmission correction unit 18 corrects the transmission strength of the interrogation signal sent by the reading device 4, the reception correction unit 17 corrects the reception threshold of other portable terminals 3 besides the portable terminal 3 corresponding to the terminal information contained in the acquired operational data. For example, if the authentication conditions of other portable terminals 3 whose terminal information is not included in the operational data are affected by the correction of the transmission strength of the interrogation signal, the reception correction unit 17 corrects the reception threshold in a way that offsets the impact. For example, when the correction of increasing the transmission strength is performed, the reception correction unit 17 increases the reception threshold of the other portable terminal 3; when the correction of decreasing the transmission strength is performed, the reception correction unit 17 decreases the reception threshold of the other portable terminal 3. As a result, even when there are multiple portable terminals 3 around the reading device 4, the authentication conditions of each portable terminal 3 can be made more stable.

[0073] Furthermore, the completed learning model can also be a model obtained by combining multiple models. For example, the completed learning model may include a completed learning sub-model. The completed learning sub-model takes terminal information of the portable terminal 3 and environmental information of the reading device 4 as input, and outputs response information, which represents the relationship between the transmission strength of the interrogation signal sent by the reading device 4 and the reception strength of the interrogation signal received by the portable terminal 3 at a distance equal to the authentication distance from the reading device 4. The response information is, for example, the ratio of transmission strength to reception strength. The generation unit 10 calculates the response information based on the transmission strength and reception strength contained in the learning data. The generation unit 10 corresponds the calculated response information to the output and corresponds the terminal information and environmental information contained in the learning data to the input, thereby generating the completed learning sub-model using methods such as machine learning. The estimation unit 11 inputs the terminal information and environmental information contained in the operating data into the completed learning sub-model, thereby obtaining the response information as the output. The estimation unit 11 uses the response information output by the learned sub-model and the transmission strength included in the data during operation to obtain the reception strength of the interrogation signal received by the portable terminal 3 at a position at a distance equal to the authentication distance from the reading device 4 as an estimation result. The estimation unit 11 may, for example, multiply the response information, which is the intensity ratio, by the transmission strength to obtain the estimation result. Therefore, the types of data required for learning the learned sub-model through machine learning or the like are reduced, thus enabling more efficient generation of the learned model. Furthermore, in the generation unit 10, the part that processes the response information and the part that generates the learned sub-model can be separately mounted on separate hardware. Similarly, in the estimation unit 11, the part that outputs the response information from the learned sub-model and the part that obtains the estimation result using the response information and the transmission strength can be separately mounted on separate hardware.

[0074] Furthermore, in authentication system 1, the authentication distance can be set individually for each user. In this case, the receiving correction unit 17 corrects the receiving threshold accordingly for the user holding the portable terminal 3. For example, during the learning phase, the operator can position the portable terminal 3 at multiple authentication distances so that the acquisition unit 8 acquires learning data. The learning completion model can operate with the authentication distance as a parameter, or multiple learning completion models can be generated for each authentication distance. For example, the information acquired by the acquisition unit 8, such as terminal information, can include the user's identification information or attribute information representing the user's attributes. In this case, for example, the user's authentication distance is determined based on the identification information or attribute information. Here, the attribute information may include, for example, information such as the user's age group, or information such as whether they use a cane or wheelchair. The recording unit 9 can also record the user's authentication distance corresponding to the attribute information or identification information. The estimation unit 11 can also set or select the learning completion model based on the user's authentication distance.

[0075] Furthermore, some of the functions of the adjustment device 6, such as the acquisition unit 8, can also be mounted on other devices such as the portable terminal 3, the reading device 4, and the environmental information collection device 7.

[0076] Furthermore, the generation unit 10 can also combine the learning datasets obtained from multiple reading devices 4 to generate a learned model. The generation unit 10 can also output the generated learned model to an external device or the like. In addition, the estimation unit 11 can also read in the learned model generated by the external device or the like and use it for estimation.

[0077] Next, use Figure 5 An example of the hardware structure of authentication system 1 is given.

[0078] Figure 5 This is a hardware structure diagram of the main part of the authentication system 1 in Implementation Method 1.

[0079] The functions of the authentication system 1 can be implemented by a processing circuit. The processing circuit has at least one processor 100a and at least one memory 100b. The processing circuit may also have at least one dedicated hardware 200 together with the processor 100a and the memory 100b, or it may have at least one dedicated hardware 200 in place of the processor 100a and the memory 100b.

[0080] When the processing circuit includes a processor 100a and a memory 100b, the functions of the authentication system 1 are implemented by software, firmware, or a combination of software and firmware. At least one of the software and firmware is described as a program. The program is stored in the memory 100b. The processor 100a implements the functions of the authentication system 1 by reading and executing the program stored in the memory 100b.

[0081] The processor 100a is also called a CPU (Central Processing Unit), processing device, arithmetic device, microprocessor, microcomputer, or DSP. The memory 100b is composed of, for example, non-volatile or volatile semiconductor memories such as RAM, ROM, flash memory, EPROM, and EEPROM.

[0082] When the processing circuit has dedicated hardware 200, the processing circuit is implemented, for example, by a single circuit, a composite circuit, a programmable processor, a parallel programmable processor, an ASIC, an FPGA, or a combination thereof.

[0083] Each function of authentication system 1 can be implemented separately by processing circuitry. Alternatively, each function of authentication system 1 can be implemented uniformly by processing circuitry. Regarding each function of authentication system 1, some can be implemented by dedicated hardware 200, and others by software or firmware. Thus, the processing circuitry implements each function of authentication system 1 through dedicated hardware 200, software, firmware, or a combination thereof.

[0084] In summary, the structures that can be adopted by the technology disclosed herein include the following structures shown as appendices.

[0085] (Postscript 1)

[0086] A learning device for an authentication system, wherein,

[0087] The learning device of the authentication system includes:

[0088] The acquisition department acquires learning data; and

[0089] The generation unit uses the learning data acquired by the acquisition unit to generate a learned model.

[0090] The learning data includes:

[0091] The transmission strength of the first interrogation signal, which is a wireless signal, sent by the reading device located at the authentication location;

[0092] The reception strength of the first interrogation signal received by a first portable terminal positioned at a predetermined authentication distance from the reading device; and

[0093] The first terminal information includes at least any of the information regarding the status and category associated with the first portable terminal.

[0094] The learned model takes the transmission strength of the second interrogation signal, which is a wireless signal sent by the reading device, and the second terminal information as input, and takes the reception strength of the second interrogation signal received by the second portable terminal when it is located at a distance of the authentication distance from the reading device as output. The second terminal information includes at least any information of the status and category associated with the second portable terminal held by the user.

[0095] (Postscript 2)

[0096] The learning device of the authentication system according to Appendix 1, wherein...

[0097] The learned model includes a learned sub-model.

[0098] The learned sub-model takes the second terminal information as input and outputs the following response information, wherein the response information represents the relationship between the transmission strength of the second interrogation signal sent by the reading device and the reception strength of the second interrogation signal received by the second portable terminal.

[0099] The generation unit corresponds the following response information to the output and the first terminal information to the input, thereby generating the learned sub-model. The response information represents the relationship between the transmission strength of the first interrogation signal sent by the reading device and the reception strength of the first interrogation signal received by the first portable terminal.

[0100] (Note 3)

[0101] The learning device of the authentication system according to Appendix 1 or 2, wherein,

[0102] The learning data includes first environmental information, which represents the setting environment of the reading device at the time the first query signal was sent.

[0103] The input to the learned model includes second environmental information, which represents the setting environment of the reading device at the time when the second query signal was sent.

[0104] (Postscript 4)

[0105] According to the learning device of the authentication system described in Appendix 3, wherein...

[0106] The information about the environment in which the reading device is set includes information about the positional relationship between the reading device and the structures located around the reading device.

[0107] (Note 5)

[0108] The learning device of the authentication system according to Appendix 3 or 4, wherein...

[0109] When the reading device transmits wireless signals via electromagnetic waves, the information about the environment in which the reading device is set up includes information about the electromagnetic waves or electromagnetic fields at the authentication location.

[0110] (Note 6)

[0111] The learning device of the authentication system according to any one of Appendices 1 to 5, wherein,

[0112] The first terminal information includes information about the posture of the first portable terminal relative to the reading device.

[0113] The second terminal information includes information about the orientation of the second portable terminal relative to the reading device.

[0114] (Note 7)

[0115] The learning device of the authentication system according to any one of Appendices 1 to 6, wherein,

[0116] The first terminal information includes information about the holding location of the first portable terminal by its owner.

[0117] The second terminal information includes information about the holding location of the second portable terminal by its owner.

[0118] (Note 8)

[0119] The learning device of the authentication system according to any one of Appendices 1 to 7, wherein,

[0120] The first terminal information includes at least any one of the following: the model information of the first portable terminal and the model information of the wireless communication module mounted on the first portable terminal.

[0121] The second terminal information includes at least any one of the following: the model information of the second portable terminal and the model information of the wireless communication module mounted on the second portable terminal.

[0122] (Note 9)

[0123] The learning device of the authentication system according to any one of Appendices 1 to 8, wherein,

[0124] When the reading device transmits wireless signals via electromagnetic waves, the first terminal information includes information about the electromagnetic waves or electromagnetic fields at the location of the first portable terminal, and the second terminal information includes information about the electromagnetic waves or electromagnetic fields at the location of the second portable terminal.

[0125] (Postscript 10)

[0126] An inference device for an authentication system, wherein,

[0127] The inference device of the authentication system includes:

[0128] The acquisition department acquires operational data; and

[0129] The estimation unit uses the learned model to obtain an estimation result based on the operational data acquired by the acquisition unit.

[0130] The learned model is pre-generated through the following learning process: When the first portable terminal receives a first interrogation signal (a wireless signal) from a reading device located at an authentication location, the received strength is correlated with the output; and the transmission strength of the first interrogation signal from the reading device and the first terminal information are correlated with the input. The first portable terminal is located at a predetermined authentication distance from the reading device, and the first terminal information includes at least any information related to the state and category of the first portable terminal.

[0131] The operational data includes:

[0132] The transmission strength of the second interrogation signal, which is a wireless signal, sent by the reading device; and

[0133] The second terminal information includes at least any of the following information: status and category, associated with a second portable terminal that receives the second interrogation signal from the reading device and is held by a user.

[0134] The estimation unit estimates the reception strength of the second interrogation signal received by the second portable terminal when it is located at a position at a distance from the reading device equal to the authentication distance, and uses this as the estimation result.

[0135] (Postscript 11)

[0136] According to the inference device of the authentication system described in Appendix 10, wherein,

[0137] The learned model includes a learned sub-model.

[0138] The learned sub-model is a pre-generated model through the following learning process, in which the following response information is associated with the output, and the first terminal information is associated with the input, wherein the response information represents the relationship between the transmission strength of the first interrogation signal sent by the reading device and the reception strength of the first interrogation signal received by the first portable terminal.

[0139] The estimation unit obtains the estimation result using the following response information and the transmission strength of the second interrogation signal sent by the reading device, wherein the response information is the information output by the learned sub-model with the second terminal information as input, and the response information indicates the relationship between the transmission strength of the second interrogation signal sent by the reading device and the reception strength of the second interrogation signal received by the second portable terminal when it is located at a distance of the authentication distance from the reading device.

[0140] (Postscript 12)

[0141] According to the inference device of the authentication system described in Appendix 10 or 11, wherein,

[0142] The learned model is a model generated during the learning process by mapping the first environmental information to the input. This first environmental information represents the setting environment information of the reading device at the time when the first query signal was sent.

[0143] The operational data includes second environmental information, which indicates the setting environment of the reading device at the time the second interrogation signal was sent.

[0144] (Postscript 13)

[0145] According to the inference device of the authentication system described in Appendix 12, wherein,

[0146] The information about the environment in which the reading device is set includes information about the positional relationship between the reading device and the structures located around the reading device.

[0147] (Postscript 14)

[0148] According to the inference device of the authentication system described in Appendix 12 or 13, wherein,

[0149] When the reading device transmits wireless signals via electromagnetic waves, the information about the environment in which the reading device is set up includes information about the electromagnetic waves or electromagnetic fields at the authentication location.

[0150] (Postscript 15)

[0151] The inference device of the authentication system according to any one of Appendices 10 to 14, wherein,

[0152] The first terminal information includes information about the posture of the first portable terminal relative to the reading device.

[0153] The second terminal information includes information about the orientation of the second portable terminal relative to the reading device.

[0154] (Postscript 16)

[0155] The inference device of the authentication system according to any one of Appendices 10 to 15, wherein...

[0156] The first terminal information includes information about the holding location of the first portable terminal by its owner.

[0157] The second terminal information includes information about the holding location of the second portable terminal by its owner.

[0158] (Postscript 17)

[0159] The inference device of the authentication system according to any one of Appendices 10 to 16, wherein,

[0160] The first terminal information includes at least any one of the following: the model information of the first portable terminal and the model information of the wireless communication module mounted on the first portable terminal.

[0161] The second terminal information includes at least any one of the following: the model information of the second portable terminal and the model information of the wireless communication module mounted on the second portable terminal.

[0162] (Postscript 18)

[0163] The inference device of the authentication system according to any one of Appendices 10 to 17, wherein...

[0164] When the reading device transmits wireless signals via electromagnetic waves, the first terminal information includes information about the electromagnetic waves or electromagnetic fields at the location of the first portable terminal, and the second terminal information includes information about the electromagnetic waves or electromagnetic fields at the location of the second portable terminal.

[0165] (Postscript 19)

[0166] The inference device of the authentication system according to any one of Appendices 10 to 18, wherein...

[0167] When the second portable terminal receives the second interrogation signal with a reception strength exceeding a preset reception threshold, and the second portable terminal sends a response signal as a wireless signal to the reading device, the inference device of the authentication system includes a reception correction unit. This reception correction unit uses the estimation result of the estimation unit to correct the reception threshold of the second portable terminal, so that the second portable terminal sends the response signal at a position at a distance from the reading device equal to the authentication distance.

[0168] (Postscript 20)

[0169] According to the inference device of the authentication system described in Appendix 19, wherein...

[0170] When the authentication distance is set according to each user holding the second portable terminal, the reception correction unit corrects the reception threshold accordingly for each user holding the second portable terminal.

[0171] (Postscript 21)

[0172] The inference device of the authentication system according to any one of Appendices 10 to 18, wherein...

[0173] When the second portable terminal receives the second interrogation signal with a reception strength exceeding a preset reception threshold, and the second portable terminal sends a response signal as a wireless signal to the reading device, the inference device of the authentication system includes a transmission correction unit. This transmission correction unit uses the estimation result of the estimation unit to correct the transmission strength of the second interrogation signal sent by the reading device, so that the second portable terminal sends the response signal at a position at a distance from the reading device equal to the authentication distance.

[0174] (Postscript 22)

[0175] According to the inference device of the authentication system described in Appendix 12, wherein,

[0176] When the second portable terminal receives the second interrogation signal with a reception strength exceeding a preset reception threshold, and the second portable terminal sends a response signal as a wireless signal to the reading device, the inference device of the authentication system includes a transmission correction unit. This transmission correction unit uses the estimation result of the estimation unit to correct the transmission strength of the second interrogation signal sent by the reading device, so that the second portable terminal sends the response signal at a position at a distance from the reading device equal to the authentication distance.

[0177] If the information about the setting environment of the reading device represented by the second environmental information contained in the operating data has not changed relative to the information about the setting environment of the reading device before the acquisition unit acquired the operating data, the transmission correction unit retains the correction of the transmission strength of the second interrogation signal sent by the reading device.

[0178] (Postscript 23)

[0179] According to the inference device of the authentication system described in Appendix 21 or 22, wherein,

[0180] The inference device of the authentication system includes a reception correction unit, which uses the estimation result of the estimation unit to correct the reception threshold of the second portable terminal, so that the second portable terminal sends the response signal at a position at a distance from the reading device equal to the authentication distance.

[0181] While the transmission correction unit corrects the transmission strength of the second interrogation signal sent by the reading device, the reception correction unit corrects the reception threshold of the third portable terminal other than the second portable terminal.

[0182] (Postscript 24)

[0183] A learning method for an authentication system, wherein,

[0184] The learning method of the authentication system includes:

[0185] The steps involve the computer acquiring learning data; and

[0186] In the generation step, the computer uses the learning data obtained in the acquisition step to generate a learned model.

[0187] The learning data includes:

[0188] The transmission strength of the first interrogation signal, which is a wireless signal, sent by the reading device located at the authentication location;

[0189] The reception strength of the first interrogation signal received by a first portable terminal positioned at a predetermined authentication distance from the reading device; and

[0190] The first terminal information includes at least any of the information regarding the status and category associated with the first portable terminal.

[0191] The learned model takes the transmission strength of the second interrogation signal, which is a wireless signal sent by the reading device, and the second terminal information as input, and takes the reception strength of the second interrogation signal received by the second portable terminal when it is located at a distance of the authentication distance from the reading device as output. The second terminal information includes at least any information from the state and category associated with the second portable terminal held by the user.

[0192] (Postscript 25)

[0193] A reasoning method for an authentication system, wherein,

[0194] The reasoning method of the authentication system has the following features:

[0195] The acquisition process involves the computer acquiring runtime data; and

[0196] In the estimation step, the computer uses the learned model to obtain an estimation result based on the runtime data obtained in the acquisition step.

[0197] The learned model is pre-generated through the following learning process: When the first portable terminal receives a first interrogation signal (a wireless signal) from a reading device located at an authentication location, the received strength is correlated with the output; and the transmission strength of the first interrogation signal from the reading device and the first terminal information are correlated with the input. The first portable terminal is located at a predetermined authentication distance from the reading device, and the first terminal information includes at least any information related to the state and category of the first portable terminal.

[0198] The operational data includes:

[0199] The transmission strength of the second interrogation signal, which is a wireless signal, sent by the reading device; and

[0200] The second terminal information includes at least any of the following information: status and category, associated with a second portable terminal that receives the second interrogation signal from the reading device and is held by a user.

[0201] In the estimation step, the reception strength of the second interrogation signal received by the second portable terminal when it is located at a distance from the reading device equal to the authentication distance is estimated as the estimation result.

[0202] (Postscript 26)

[0203] A learning procedure for an authentication system, wherein,

[0204] The learning process of the authentication system is executed by the computer:

[0205] The steps to obtain learning data; and

[0206] The generation step uses the learning data obtained in the acquisition step to generate the learned model.

[0207] The learning data includes:

[0208] The transmission strength of the first interrogation signal, which is a wireless signal, sent by the reading device located at the authentication location;

[0209] The reception strength of the first interrogation signal received by a first portable terminal positioned at a predetermined authentication distance from the reading device; and

[0210] The first terminal information includes at least any of the information regarding the status and category associated with the first portable terminal.

[0211] The learned model takes the transmission strength of the second interrogation signal, which is a wireless signal sent by the reading device, and the second terminal information as input, and takes the reception strength of the second interrogation signal received by the second portable terminal when it is located at a distance of the authentication distance from the reading device as output. The second terminal information includes at least any information from the state and category associated with the second portable terminal held by the user.

[0212] (Postscript 27)

[0213] An inference procedure for an authentication system, wherein,

[0214] The reasoning program of the authentication system causes the computer to execute:

[0215] Obtaining steps, obtaining runtime data; and

[0216] The estimation step involves using the learned model to obtain an estimation result based on the operational data acquired in the acquisition step.

[0217] The learned model is pre-generated through the following learning process: When the first portable terminal receives a first interrogation signal (a wireless signal) from a reading device located at an authentication location, the received strength is correlated with the output; and the transmission strength of the first interrogation signal from the reading device and the first terminal information are correlated with the input. The first portable terminal is located at a predetermined authentication distance from the reading device, and the first terminal information includes at least any information related to the state and category of the first portable terminal.

[0218] The operational data includes:

[0219] The transmission strength of the second interrogation signal, which is a wireless signal, sent by the reading device; and

[0220] The second terminal information includes at least any of the following information: status and category, associated with a second portable terminal that receives the second interrogation signal from the reading device and is held by a user.

[0221] In the estimation step, the reception strength of the second interrogation signal received by the second portable terminal when it is located at a distance from the reading device equal to the authentication distance is estimated as the estimation result.

[0222] (Postscript 28)

[0223] An authentication system, wherein,

[0224] The authentication system has the following features:

[0225] The acquisition department acquires learning data;

[0226] The generation unit uses the learning data acquired by the acquisition unit to generate a learned model; and

[0227] The estimation unit uses the learned model to obtain an estimation result based on the operational data acquired by the acquisition unit.

[0228] The learning data includes:

[0229] The transmission strength of the first interrogation signal, which is a wireless signal, sent by the reading device located at the authentication location;

[0230] The reception strength of the first interrogation signal received by a first portable terminal positioned at a predetermined authentication distance from the reading device; and

[0231] The first terminal information includes at least any of the information regarding the status and category associated with the first portable terminal.

[0232] The generation unit corresponds the received strength of the first interrogation signal received by the first portable terminal with the output, and corresponds the transmitted strength of the first interrogation signal sent by the reading device and the first terminal information with the input to generate the learned model.

[0233] The operational data includes:

[0234] The transmission strength of the second interrogation signal, which is a wireless signal, sent by the reading device; and

[0235] The second terminal information includes at least any of the following information: status and category, associated with a second portable terminal that receives the second interrogation signal from the reading device and is held by a user.

[0236] The estimation unit estimates the reception strength of the second interrogation signal received when the second portable terminal is located at a position at a distance from the reading device equal to the authentication distance, and uses this as the estimation result.

[0237] Explanation of reference numerals in the attached figures

[0238] 1. Authentication system; 2. Electronic lock; 3. Portable terminal; 4. Reading device; 5. Controller; 6. Adjustment device; 7. Environmental information collection device; 8. Acquisition unit; 9. Recording unit; 10. Generation unit; 11. Estimation unit; 12. Correction unit; 13. Transmit strength (DB); 14. Receive strength (DB); 15. Environmental information (DB); 16. Terminal information (DB); 17. Receive correction unit; 18. Transmit correction unit; 100a. Processor; 100b. Memory; 200. Dedicated hardware.

Claims

1. A learning device for an authentication system, wherein, The learning device of the authentication system includes: The acquisition department acquires learning data; and The generation unit uses the learning data acquired by the acquisition unit to generate a learned model. The learning data includes: The transmission strength of the first interrogation signal, which is a wireless signal, sent by the reading device located at the authentication location; The reception strength of the first interrogation signal received by the first portable terminal, which is positioned at a location at a preset authentication distance from the reading device; as well as The first terminal information includes at least any of the information regarding the status and category associated with the first portable terminal. The learned model takes the transmission strength of the second interrogation signal, which is a wireless signal sent by the reading device, and the second terminal information as input, and takes the reception strength of the second interrogation signal received by the second portable terminal when it is located at a distance of the authentication distance from the reading device as output. The second terminal information includes at least any information of the status and category associated with the second portable terminal held by the user.

2. The learning device of the authentication system according to claim 1, wherein, The learned model includes a learned sub-model. The learned sub-model takes the second terminal information as input and outputs the following response information, wherein the response information represents the relationship between the transmission strength of the second interrogation signal sent by the reading device and the reception strength of the second interrogation signal received by the second portable terminal. The generation unit corresponds the following response information to the output and the first terminal information to the input, thereby generating the learned sub-model. The response information represents the relationship between the transmission strength of the first interrogation signal sent by the reading device and the reception strength of the first interrogation signal received by the first portable terminal.

3. The learning device for the authentication system according to claim 1 or 2, wherein, The learning data includes first environmental information, which represents the setting environment of the reading device at the time the first query signal was sent. The input to the learned model includes second environmental information, which represents the setting environment of the reading device at the time when the second query signal was sent.

4. The learning device of the authentication system according to claim 3, wherein, The information about the environment in which the reading device is set includes information about the positional relationship between the reading device and the structures located around the reading device.

5. The learning device for the authentication system according to claim 3, wherein, When the reading device transmits wireless signals via electromagnetic waves, the information about the environment in which the reading device is set up includes information about the electromagnetic waves or electromagnetic fields at the authentication location.

6. The learning device for the authentication system according to claim 1 or 2, wherein, The first terminal information includes information about the posture of the first portable terminal relative to the reading device. The second terminal information includes information about the orientation of the second portable terminal relative to the reading device.

7. The learning device for the authentication system according to claim 1 or 2, wherein, The first terminal information includes information about the holding location of the first portable terminal by its owner. The second terminal information includes information about the holding location of the second portable terminal by its owner.

8. The learning device for the authentication system according to claim 1 or 2, wherein, The first terminal information includes at least any one of the following: the model information of the first portable terminal and the model information of the wireless communication module mounted on the first portable terminal. The second terminal information includes at least any one of the following: the model information of the second portable terminal and the model information of the wireless communication module mounted on the second portable terminal.

9. The learning device for the authentication system according to claim 1 or 2, wherein, When the reading device transmits wireless signals via electromagnetic waves, the first terminal information includes information about the electromagnetic waves or electromagnetic fields at the location of the first portable terminal, and the second terminal information includes information about the electromagnetic waves or electromagnetic fields at the location of the second portable terminal.

10. An inference device for an authentication system, wherein, The inference device of the authentication system includes: The acquisition department acquires operational data; and The estimation unit uses the learned model to obtain an estimation result based on the operational data acquired by the acquisition unit. The learned model is pre-generated through the following learning process: When the first portable terminal receives a first interrogation signal (a wireless signal) from a reading device located at an authentication location, the received strength is correlated with the output; and the transmission strength of the first interrogation signal from the reading device and the first terminal information are correlated with the input. The first portable terminal is located at a predetermined authentication distance from the reading device, and the first terminal information includes at least any information related to the state and category of the first portable terminal. The operational data includes: The transmission strength of the second interrogation signal, which is a wireless signal, sent by the reading device; and The second terminal information includes at least any of the following information: status and category, associated with a second portable terminal that receives the second interrogation signal from the reading device and is held by a user. The estimation unit estimates the reception strength of the second interrogation signal received by the second portable terminal when it is located at a position at a distance from the reading device equal to the authentication distance, and uses this as the estimation result.

11. The inference apparatus of the authentication system according to claim 10, wherein, The learned model includes a learned sub-model. The learned sub-model is a pre-generated model through the following learning process, in which the following response information is associated with the output, and the first terminal information is associated with the input, wherein the response information represents the relationship between the transmission strength of the first interrogation signal sent by the reading device and the reception strength of the first interrogation signal received by the first portable terminal. The estimation unit obtains the estimation result using the following response information and the transmission strength of the second interrogation signal sent by the reading device, wherein the response information is the information output by the learned sub-model with the second terminal information as input, and the response information indicates the relationship between the transmission strength of the second interrogation signal sent by the reading device and the reception strength of the second interrogation signal received by the second portable terminal when it is located at a distance of the authentication distance from the reading device.

12. The reasoning apparatus of the authentication system according to claim 10 or 11, wherein, The learned model is a model generated during the learning process by mapping the first environmental information to the input. This first environmental information represents the setting environment information of the reading device at the time when the first query signal was sent. The operational data includes second environmental information, which indicates the setting environment of the reading device at the time the second interrogation signal was sent.

13. The inference device of the authentication system according to claim 12, wherein, The information about the environment in which the reading device is set includes information about the positional relationship between the reading device and the structures located around the reading device.

14. The inference device of the authentication system according to claim 12, wherein, When the reading device transmits wireless signals via electromagnetic waves, the information about the environment in which the reading device is set up includes information about the electromagnetic waves or electromagnetic fields at the authentication location.

15. The inference apparatus of the authentication system according to claim 10 or 11, wherein, The first terminal information includes information about the posture of the first portable terminal relative to the reading device. The second terminal information includes information about the orientation of the second portable terminal relative to the reading device.

16. The inference apparatus of the authentication system according to claim 10 or 11, wherein, The first terminal information includes information about the holding location of the first portable terminal by its owner. The second terminal information includes information about the holding location of the second portable terminal by its owner.

17. The inference apparatus of the authentication system according to claim 10 or 11, wherein, The first terminal information includes at least any one of the following: the model information of the first portable terminal and the model information of the wireless communication module mounted on the first portable terminal. The second terminal information includes at least any one of the following: the model information of the second portable terminal and the model information of the wireless communication module mounted on the second portable terminal.

18. The reasoning apparatus of the authentication system according to claim 10 or 11, wherein, When the reading device transmits wireless signals via electromagnetic waves, the first terminal information includes information about the electromagnetic waves or electromagnetic fields at the location of the first portable terminal, and the second terminal information includes information about the electromagnetic waves or electromagnetic fields at the location of the second portable terminal.

19. The reasoning apparatus of the authentication system according to claim 10 or 11, wherein, When the second portable terminal receives the second interrogation signal with a reception strength exceeding a preset reception threshold, and the second portable terminal sends a response signal as a wireless signal to the reading device, the inference device of the authentication system includes a reception correction unit. This reception correction unit uses the estimation result of the estimation unit to correct the reception threshold of the second portable terminal, so that the second portable terminal sends the response signal at a position at a distance from the reading device equal to the authentication distance.

20. The inference apparatus of the authentication system according to claim 19, wherein, When the authentication distance is set according to each user holding the second portable terminal, the reception correction unit corrects the reception threshold accordingly for each user holding the second portable terminal.

21. The inference apparatus of the authentication system according to claim 10 or 11, wherein, When the second portable terminal receives the second interrogation signal with a reception strength exceeding a preset reception threshold, and the second portable terminal sends a response signal as a wireless signal to the reading device, the inference device of the authentication system includes a transmission correction unit. This transmission correction unit uses the estimation result of the estimation unit to correct the transmission strength of the second interrogation signal sent by the reading device, so that the second portable terminal sends the response signal at a position at a distance from the reading device equal to the authentication distance.

22. The inference device of the authentication system according to claim 12, wherein, When the second portable terminal receives the second interrogation signal with a reception strength exceeding a preset reception threshold, and the second portable terminal sends a response signal as a wireless signal to the reading device, the inference device of the authentication system includes a transmission correction unit. This transmission correction unit uses the estimation result of the estimation unit to correct the transmission strength of the second interrogation signal sent by the reading device, so that the second portable terminal sends the response signal at a position at a distance from the reading device equal to the authentication distance. If the information about the setting environment of the reading device represented by the second environmental information contained in the operating data has not changed relative to the information about the setting environment of the reading device before the acquisition unit acquired the operating data, the transmission correction unit retains the correction of the transmission strength of the second interrogation signal sent by the reading device.

23. The inference apparatus of the authentication system according to claim 21, wherein, The inference device of the authentication system includes a reception correction unit, which uses the estimation result of the estimation unit to correct the reception threshold of the second portable terminal, so that the second portable terminal sends the response signal at a position at a distance from the reading device equal to the authentication distance. While the transmission correction unit corrects the transmission strength of the second interrogation signal sent by the reading device, the reception correction unit corrects the reception threshold of the third portable terminal other than the second portable terminal.

24. A learning method for an authentication system, wherein, The learning method of the authentication system includes: The steps involve the computer acquiring learning data; and In the generation step, the computer uses the learning data obtained in the acquisition step to generate a learned model. The learning data includes: The transmission strength of the first interrogation signal, which is a wireless signal, sent by the reading device located at the authentication location; The reception strength of the first interrogation signal received by the first portable terminal, which is positioned at a location at a preset authentication distance from the reading device; as well as The first terminal information includes at least any of the information regarding the status and category associated with the first portable terminal. The learned model takes the transmission strength of the second interrogation signal, which is a wireless signal sent by the reading device, and the second terminal information as input, and takes the reception strength of the second interrogation signal received by the second portable terminal when it is located at a distance of the authentication distance from the reading device as output. The second terminal information includes at least any information from the state and category associated with the second portable terminal held by the user.

25. A reasoning method for an authentication system, wherein, The reasoning method of the authentication system has the following features: The acquisition process involves the computer acquiring runtime data; and In the estimation step, the computer uses the learned model to obtain an estimation result based on the runtime data obtained in the acquisition step. The learned model is pre-generated through the following learning process: When the first portable terminal receives a first interrogation signal (a wireless signal) from a reading device located at an authentication location, the received strength is correlated with the output; and the transmission strength of the first interrogation signal from the reading device and the first terminal information are correlated with the input. The first portable terminal is located at a predetermined authentication distance from the reading device, and the first terminal information includes at least any information related to the state and category of the first portable terminal. The operational data includes: The transmission strength of the second interrogation signal, which is a wireless signal, sent by the reading device; and The second terminal information includes at least any of the following information: status and category, associated with a second portable terminal that receives the second interrogation signal from the reading device and is held by a user. In the estimation step, the reception strength of the second interrogation signal received by the second portable terminal when it is located at a distance from the reading device equal to the authentication distance is estimated as the estimation result.

26. A storage medium that records the learning process of an authentication system, wherein, The learning process of the authentication system is executed by the computer: The steps to obtain learning data; and The generation step uses the learning data obtained in the acquisition step to generate the learned model. The learning data includes: The transmission strength of the first interrogation signal, which is a wireless signal, sent by the reading device located at the authentication location; The reception strength of the first interrogation signal received by the first portable terminal, which is positioned at a location at a preset authentication distance from the reading device; as well as The first terminal information includes at least any of the information regarding the status and category associated with the first portable terminal. The learned model takes the transmission strength of the second interrogation signal, which is a wireless signal sent by the reading device, and the second terminal information as input, and takes the reception strength of the second interrogation signal received by the second portable terminal when it is located at a distance of the authentication distance from the reading device as output. The second terminal information includes at least any information from the state and category associated with the second portable terminal held by the user.

27. A storage medium that records the reasoning program of an authentication system, wherein, The reasoning program of the authentication system causes the computer to execute: Obtaining steps, obtaining runtime data; and The estimation step involves using the learned model to obtain an estimation result based on the operational data acquired in the acquisition step. The learned model is pre-generated through the following learning process: When the first portable terminal receives a first interrogation signal (a wireless signal) from a reading device located at an authentication location, the received strength is correlated with the output; and the transmission strength of the first interrogation signal from the reading device and the first terminal information are correlated with the input. The first portable terminal is located at a predetermined authentication distance from the reading device, and the first terminal information includes at least any information related to the state and category of the first portable terminal. The operational data includes: The transmission strength of the second interrogation signal, which is a wireless signal, sent by the reading device; and The second terminal information includes at least any of the following information: status and category, associated with a second portable terminal that receives the second interrogation signal from the reading device and is held by a user. In the estimation step, the reception strength of the second interrogation signal received by the second portable terminal when it is located at a distance from the reading device equal to the authentication distance is estimated as the estimation result.

28. An authentication system, wherein, The authentication system has the following features: The acquisition department acquires learning data; The generation unit uses the learning data acquired by the acquisition unit to generate a learned model; and The estimation unit uses the learned model to obtain an estimation result based on the operational data acquired by the acquisition unit. The learning data includes: The transmission strength of the first interrogation signal, which is a wireless signal, sent by the reading device located at the authentication location; The reception strength of the first interrogation signal received by the first portable terminal, which is positioned at a location at a preset authentication distance from the reading device; as well as The first terminal information includes at least any of the information regarding the status and category associated with the first portable terminal. The generation unit corresponds the received strength of the first interrogation signal received by the first portable terminal with the output, and corresponds the transmitted strength of the first interrogation signal sent by the reading device and the first terminal information with the input to generate the learned model. The operational data includes: The transmission strength of the second interrogation signal, which is a wireless signal, sent by the reading device; and The second terminal information includes at least any of the following information: status and category, associated with a second portable terminal that receives the second interrogation signal from the reading device and is held by a user. The estimation unit estimates the reception strength of the second interrogation signal received when the second portable terminal is located at a position at a distance from the reading device equal to the authentication distance, and uses this as the estimation result.

Images