Base station, communication method, and communication program

Abstract

Provided is a base station that relays communication between terminals and transmits information related to AI processing to an AI server, the base station being capable of suppressing an amount of communication to the AI server. This base station relays communication between terminals and communicates with an AI server that executes AI processing. The base station comprises: a reception unit that receives a signal from a terminal; a determination unit that determines, on the basis of the signal from the terminal, whether the terminal is moving or not; a transmission control unit that, in a case where the terminal is moving, performs control so as to transmit the information of the terminal to the AI server at a relatively high frequency, and that, in a case where the terminal is not moving, performs control so as to transmit the information of the terminal to the AI server at a relatively low frequency; and a transmission unit that transmits the information of the terminal to the AI server at the frequency controlled by the transmission control unit.

Description

Base station, communication method and communication program

[0001] The present invention relates to a base station that relays communications between terminals and transmits information to an AI server, as well as a communication method and a communication program.

[0002] In recent years, various AI applications have been implemented using RIC (RAN Intelligent Controller). In the field of communications, this AI application often utilizes information from devices such as smartphones, tablet devices, and mobile phones. However, when transmitting device information to an AI server that performs AI processing for AI application, a large amount of information is sent, which can cause traffic congestion in some cases. Patent Document 1 discloses a technology for increasing the frequency of information transmission from a device when the device is moving.

[0003] JP 2014-174827 A

[0004] Incidentally, while the above-mentioned patent document can reduce the amount of communication between a base station and a terminal, it cannot reduce the amount of communication between a base station and an AI server.

[0005] Therefore, the present invention has been made in consideration of the above problems, and aims to provide a base station that can reduce the amount of communication between the base station and an AI server that performs AI processing using information from terminals with which the base station communicates.

[0006] In order to solve the above problem, a base station according to one embodiment of the present invention is a base station that relays communications between terminals and communicates with an AI server that performs AI processing, and is equipped with a receiving unit that receives signals from the terminal, a determining unit that determines whether the terminal is moving based on the signal from the terminal, a transmission control unit that controls the terminal to send information about the terminal to the AI ​​server at a relatively high frequency when the terminal is moving, and to send information about the terminal to the AI ​​server at a relatively low frequency when the terminal is not moving, and a transmitting unit that transmits information about the terminal to the AI ​​server at a frequency controlled by the transmission control unit.

[0007] The base station may include a memory unit that stores a learning model that has learned the correspondence between a first signal received by the base station from a terminal at a first time, a second signal received by the base station from the terminal at a second time after the first time, and whether the terminal is moving between the first time and the second time, and the determination unit may determine whether the terminal is moving based on the third signal received by the base station from the terminal at a third time, a fourth signal received by the base station at a fourth time after the third time, and the learning model.

[0008] In the above base station, the learning model is a model that has learned the correspondence between feature points based on the difference between the first signal and the second signal and whether or not the terminal is moving, and the judgment unit may judge whether or not the terminal is moving based on the feature points based on the difference between the third signal and the fourth signal and the learning model.

[0009] In the base station, the feature point based on the difference between the first signal and the second signal may be a feature point based on the difference between the signal waveform of the first signal and the signal waveform of the second signal, and the feature point based on the difference between the third signal and the fourth signal may be a feature point based on the difference between the signal waveform of the third signal and the signal waveform of the fourth signal.

[0010] In addition, in order to solve the above problem, a communication method according to one aspect of the present invention includes a receiving step in which a computer of a base station that relays communication between terminals and communicates with an AI server that performs AI processing receives a signal from the terminal; a determination step in which the computer determines whether the terminal is moving based on the signal from the terminal; a transmission control step in which, if the terminal is moving, the computer controls the information of the terminal to be sent to the AI ​​server at a relatively high frequency, and if the terminal is not moving, the computer controls the information of the terminal to be sent to the AI ​​server at a relatively low frequency; and a transmission step in which the information of the terminal is sent to the AI ​​server at a frequency controlled by the transmission control step.

[0011] In addition, in order to solve the above problem, a communication program according to one embodiment of the present invention provides a base station computer that relays communications between terminals and communicates with an AI server that performs AI processing with the following functions: a receiving function that receives signals from the terminal; a determination function that determines whether the terminal is moving based on the signal from the terminal; a transmission control function that controls the terminal so that, if the terminal is moving, the terminal sends information about the terminal to the AI ​​server at a relatively high frequency, and if the terminal is not moving, the terminal sends information about the terminal to the AI ​​server at a relatively low frequency; and a transmission function that sends the terminal information to the AI ​​server at a frequency controlled by the transmission control function.

[0012] According to the base station of the present invention, the AI ​​server transmits information about the terminals used for AI processing at high frequency for mobile terminals, where the situation of the terminal may change from moment to moment, while transmitting information at low frequency for stationary terminals, where the situation of the terminal is likely to remain stable. This reduces communication volume compared to transmitting information for all terminals uniformly. As a result, the AI ​​server can obtain result information showing the results of highly accurate AI processing, despite reducing the amount of information transmitted from the base station.

[0013] 7 is a system diagram showing an overview of a communication system. FIG. 8 is a block diagram showing an example of the configuration of a base station. FIG. 9 is a block diagram showing an example of the configuration of an AI server. FIG. 10 is a system diagram showing an example of an AI server acquiring location information of a terminal. FIG. 11 is a sequence diagram showing an example of interaction between devices during learning. FIG. 12 is a flowchart showing an example of the operation of the AI ​​server in the interaction of FIG. 5. FIG. 13 is a sequence diagram showing an example of interaction between devices when AI processing is executed. FIG. 14 is a flowchart showing an example of the operation of a base station in the interaction of FIG.

[0014] A base station according to the present invention will be described below with reference to the drawings.

[0015] <Overview> Fig. 1 is a system diagram showing an example of the configuration of a communication system 1 including a base station 100 according to the present invention, and is a diagram showing an overview of the present invention. As shown in Fig. 1, the communication system 1 includes base stations 100 (100a, 100b), an AI server 200, and terminals 300 (300a, 300b, 300c). Note that although the number of base stations 100 shown in Fig. 1 is two, this number is not limited. Also, although the number of terminals 300 shown in Fig. 1 is three, this number is not limited.

[0016] The base station 100 is a communication device that relays communications between terminals 300 within its coverage area and is an information processing device (computer system) with so-called base station functionality. The base station 100 relays communications between terminals 300 and can also communicate with the AI ​​server 200. The AI ​​server 200 is an information processing device (computer system) with the function of executing AI processing using information from the terminals 300. The AI ​​server 200 may be referred to as a Radio Access Network Intelligent Controller (RIC). Therefore, each base station 100 uploads (transmits) information about the terminals to which it is connected to the AI ​​server 200. In the example of FIG. 1, the base station 100a communicates with terminals 300a and 300b, and the base station 100b communicates with terminal 300c. The terminal 300 may be, for example, a mobile communication module such as a smartphone, a tablet terminal, or a mobile phone, or may be an IoT (Internet of Things) device or the like.

[0017] In FIG. 1, terminal 300a is shown moving, while terminals 300b and 300c are shown not moving. The base station 100 according to this embodiment transmits information about the moving terminal 300 to the AI ​​server 200 at a high frequency, and transmits information about the stationary terminal 300 to the AI ​​server 200 at a low frequency. Therefore, as shown in the figure, the base station 100a transmits information about terminal 300a to the AI ​​server 200 at a high frequency, while transmitting information about terminal 300b to the AI ​​server 200 at a low frequency. Furthermore, the base station 100b transmits information about terminal 300c to the AI ​​server 200 at a low frequency. By configuring the base station 100 in this manner, the overall communication volume can be reduced compared to when information about all terminals 300 is uniformly transmitted to the AI ​​server 200. As a result, delays in information transmission from the base station 100 to the AI ​​server 200 can be reduced, and the processing load on the base station 100 and the AI ​​server 200 can be reduced. This will be described in detail below.

[0018] <Configuration> <Configuration of base station 100> Fig. 2 is a block diagram showing an example configuration of the base station 100. The base station 100 functions as a base station that relays communications between terrestrial terminals 300, and is a computer system that operates according to a predetermined program.

[0019] 2, the base station 100 includes a communication unit 110, a control unit 130, and a storage unit 140. The base station 100 may also include an input unit 120 and an output unit 150.

[0020] The communication unit 110 is a communication interface that has the function of communicating with devices external to the base station 100. The communication unit 110 has the function of communicating with the AI ​​server 200 and the terminal 300 as external devices.

[0021] The communication unit 110 transmits terminal information to the AI ​​server 200 at a set frequency in accordance with instructions from the control unit 130. The communication unit 110 also communicates with the terminal 300. Note that the communication unit 110 may be realized as separate communication devices, one for communicating with the AI ​​server 200 and one for communicating with the terminal 300, or may be realized as a single communication device capable of communicating with both the AI ​​server 200 and the terminal 300.

[0022] The input unit 120 has a function of receiving input from an operator of the base station 100 or the like and transmitting the input content to the control unit 130. The input unit 120 may be realized by an input device such as a mouse, a keyboard, or a touch panel, or in the case of voice input, by a microphone.

[0023] The control unit 130 is a processor having a function of controlling each unit of the base station 100. The control unit 130 may be realized by a single core or a multi-core. The control unit 130 executes various programs stored in the storage unit 140 and uses various data to realize the functions of the base station 100. The control unit 130 is necessary for generating a learning model for determining whether the terminal 300 is moving, and has the following functions: a function of transmitting a first signal and a second signal transmitted from the terminal 300 at different times to the AI ​​server 200 as explanatory variables for learning; and a function of determining whether the terminal 300 is moving, determining a frequency at which information about the terminal 300 is to be transmitted to the AI ​​server 200, and transmitting the information to the AI ​​server 200 at that frequency.

[0024] The control unit 130 transmits to the AI ​​server 200 a first signal transmitted from and received by the terminal 300 and a second signal transmitted from the same terminal 300 but transmitted and received at a different time than the first signal. The first signal and the second signal are transmitted to the AI ​​server 200 as data that can serve as explanatory variables for generating a mobility learning model 141 for determining whether the terminal 300 is moving. The first signal and the second signal are signals of the same type and, as described above, are signals received at different times. Here, the time difference between the first signal and the second signal may be, but is not limited to, a predetermined arbitrary time difference. For example, the time difference may be 1 second, 5 seconds, or 10 seconds, but is not limited to these time differences. The first signal and the second signal may be signals that can be used as information specifying a signal deviation related to movement detection based on their differential features (feature points). As an example, the first signal and the second signal may be an IQ (In-phase / Quadrature-phase) signal when receiving 5QI (5G QoS (Quality of Service) Identifier) ​​and receiving DMRS (DeModulation Reference Signal) or SRS (Sounding Reference Signal). However, this is not limited to this. Note that 5QI is a quality identifier used in 5G communication networks. DMRS is a reference signal that serves as a demodulation standard in 5G communication networks. Furthermore, the SRS is a reference signal in the uplink used by the base station 100 to measure the uplink channel quality, reception timing, and the like. The differential feature may be the difference between the waveforms of the signals. The waveform of the IQ signal changes depending on the reception environment around the base station 100 and the transmission environment of the terminal 300, and the waveform is likely to be different especially when the terminal 300 is located in a different location. Therefore, by using the received waveforms of signals transmitted from the terminal 300 at different times, it is possible to easily use the waveforms as information for estimating whether or not the terminal 300 is moving, which has the advantage of making it easier to identify the movement of the terminal 300.Furthermore, the differential feature may be the difference in the amount of Doppler shift that can be detected by an RU (Radio Unit) in a 5G communication network, or may be the difference between the amount of Doppler shift at a first time and the amount of Doppler shift at a second time.

[0025] The control unit 130 may transmit analog signals when the first signal and the second signal are received to the AI ​​server 200, or may transmit digitally converted signals with a resolution sufficient to detect differential characteristics between when the vehicle is moving and when it is not moving to the AI ​​server 200.

[0026] The base station 100 stores the mobile learning model 141 generated and transmitted by the AI ​​server 200 based on the first signal and the second signal in the memory unit 140.

[0027] The control unit 130 includes a determination unit 131, a transmission control unit 132, and a transmission unit 133 as functions realized by the control unit 130.

[0028] The determination unit 131 determines whether each terminal 300 communicating with the base station 100 is moving. The determination unit 131 determines whether each terminal 300 is moving based on a third signal and a fourth signal transmitted from the terminal 300. The third signal and the fourth signal are the same type of signal and are transmitted by the terminal 300 at different times. Here, the time difference between the third signal and the fourth signal may be any predetermined time difference, but is not limited to this. For example, the time difference may be 1 second, 5 seconds, or 10 seconds, but is not limited to these time differences. The time difference between the third signal and the fourth signal is preferably the same as the time difference between the first signal and the second signal, but does not have to be the same. The third signal, the fourth signal, and the first signal are the same type of signal. Furthermore, the first signal to the fourth signal are preferably signals that are determined to be transmitted periodically from the terminal 300 to the base station 100. The determination unit 131 may identify a difference feature between the third signal and the fourth signal, and input the identified difference feature into the mobility learning model 141 to determine (estimate) whether the terminal 300 is moving. The determination result by the mobility learning model 141 may be output as information of 1 or 0 indicating whether the terminal 300 is moving, or may be output as information on the percentage of the probability of moving or not moving. The determination unit 131 transmits the determination result to the transmission control unit 132.

[0029] The transmission control unit 132 determines and sets the transmission frequency for transmitting information of each terminal to the AI ​​server 200 according to the determination result of the determination unit 131. The transmission control unit 132 may determine and set the transmission frequency of the information of the corresponding terminal each time a determination result is transmitted from the determination unit 131. That is, when the determination unit 131 determines that the terminal 300 is not moving, the transmission control unit 132 sets the information transmission frequency of the terminal 300 to "low." On the other hand, when the determination unit 131 determines that the terminal 300 is moving, the transmission control unit 132 sets the information transmission frequency of the terminal 300 to "high." Here, a "low" transmission frequency means that the transmission frequency per unit time is relatively low, and may be lower than a "high" transmission frequency. Furthermore, a "high" transmission frequency means that the transmission frequency per unit time is relatively high, and may be higher than a "low" transmission frequency. The "low" transmission frequency may be, for example, 2 times per minute, but is not limited to this. The "high" transmission frequency may be, for example, 12 times per minute, but is not limited to this. The "low" and "high" transmission frequencies may be predetermined fixed values ​​or may be variable values ​​that vary depending on the communication situation, but the magnitude relationship between the "low" and "high" transmission frequencies is maintained. The transmission control unit 132 stores information indicating the determined transmission frequency in the storage unit 140, in association with the ID of each terminal.

[0030] The transmission unit 133 transmits information about each terminal to the AI ​​server 200 via the communication unit 110 at a frequency set by the transmission control unit 132 for each terminal.

[0031] The control unit 130 may receive information about the AI ​​processing executed by the AI ​​server 200 by transmitting information about the terminal 300 to the AI ​​server 200, and may perform control based on the received information. As an example, the AI ​​processing executed by the AI ​​server 200 may be communication settings for each terminal 300, such as a frequency to be assigned to the terminal 300, communication timing, or specification of a precoding matrix for beamforming.

[0032] The memory unit 140 has a function of storing various programs and data required for the operation of the base station 100. The memory unit 140 can be realized, for example, by a hard disk drive (HDD), a solid state drive (SSD), a flash memory, etc., but is not limited to these. The memory unit 140 may store, for example, a program for determining whether a terminal is moving, a program for determining the frequency of transmitting terminal information to the AI ​​server 200, and a program for transmitting terminal information to the AI ​​server 200. The memory unit 140 may also store a mobility learning model 141. The mobility learning model 141 is a trained model generated by the AI ​​server 200 and is a model that can determine whether a terminal is moving from two signals transmitted from the terminal at different times. Details of the mobility learning model 141 will be described later.

[0033] The output unit 150 has a function of outputting designated information in accordance with an instruction from the control unit 130. The output unit 150 may, for example, output text information or image information, in which case the output unit 150 is realized by a monitor provided in or connected to the base station 100. The output unit 150 may, for example, output audio information, in which case the output unit 150 is realized by a speaker provided in or connected to the information processing device.

[0034] The above is an example of the configuration of the base station 100.

[0035] <Configuration example of AI server 200> Fig. 3 is a block diagram showing a configuration example of the AI ​​server 200. The AI ​​server 200 is a base station that functions as a base station that relays communications between ground terminals 300, and is a computer system that operates according to a predetermined program.

[0036] 3, the AI ​​server 200 includes a communication unit 210, a control unit 230, and a storage unit 240. The AI ​​server 200 may also include an input unit 220 and an output unit 250.

[0037] The communication unit 210 is a communication interface having a function of communicating with devices external to the AI ​​server 200. The communication unit 210 communicates with each base station 100 and the management device 400 as external devices. The communication unit 210 receives a first signal or a second signal (or a signal indicating a differential feature between the first signal and the second signal) from the base station 100 and transmits it to the control unit 230. The communication unit 210 may also transmit the mobile learning model 141 to the base station 100 or the management device 400 in accordance with an instruction from the control unit 230. The communication unit 210 may also transmit information indicating the processing results of the AI ​​processing to an external device, for example, the base station 100, in accordance with an instruction from the control unit 230.

[0038] The input unit 220 has a function of receiving input from an operator of the AI ​​server 200 or the like and transmitting the input content to the control unit 230. The input unit 220 may be realized by an input device such as a mouse, keyboard, or touch panel, or in the case of voice input, by a microphone.

[0039] The control unit 230 is a processor having the function of controlling each unit of the AI ​​server 200. The control unit 230 may be realized by a single core or by multiple cores. The control unit 230 realizes the functions of the AI ​​server 200 by executing various programs stored in the storage unit 240 and using various data.

[0040] The control unit 230 includes a learning unit 231 and an AI processing unit 232 as functions realized by the control unit 230.

[0041] The learning unit 231 generates a movement learning model 141 for determining whether the terminal 300 is moving or not, based on the first signal and the second signal transmitted from the communication unit 210. The learning unit 231 acquires the first position of the terminal 300 when the first signal is transmitted and the second position of the terminal 300 when the second signal is transmitted.

[0042] An example of obtaining location information of the terminal 300 will be described below with reference to FIG. 4. FIG. 4 illustrates an example of a communication system 1 including a base station 100 and an AI server 200, and further including a management device 400 and a service providing device 500. The management device 400 is a higher-level device of the base station 100 and may be an information processing device also referred to as a center device. The management device 400 accumulates information from each base station 100 regarding communications at each base station 100. The information transmitted from the base station 100 to the management device 400 is a signal transmitted from the terminal 300 with which the base station 100 is communicating, and may include a TMSI (Temporary Mobile Subscriber Identity) or an IMSI (International Mobile Subscriber Identity) capable of identifying the terminal 300, and may also include time information at the time of transmission (or reception by the base station 100). The service providing device 500 may be a device that provides applications for various purposes to the terminal 300, such as a game application, a schedule application, or a navigation application. Some of these applications require the user of the terminal 300 to consent to the transmission of location information. That is, the service providing device 500 may retain location information of the terminal 300. Therefore, as an example, the management device 400 may cooperate with such a service providing device 500, acquire location information of the terminal 300 from the service providing device 500, and transmit the location information to the AI ​​server 200. That is, the management device 400 transmits to the service providing device 500 the identification information (TMSI or IMSI) of the terminal 300 in the first signal transmitted from the base station 100 and information on the transmission time of the first signal. The service providing device 500 then returns to the management device 400 the location information corresponding to the transmission time of the terminal 300 corresponding to the identification information. In this way, the AI ​​server 200 can acquire the location information of the terminal 300.

[0043] In the example shown in FIG. 4 , both terminals are the same terminal 300, but at time t1 it is communicating with base station 100a and at time t2 it is communicating with base station 100b. In FIG. 4 , the terminal at time t1 is shown as terminal 300t1, and the terminal at time t2 is shown as terminal 300t2. At time t1, terminal 300 transmits a first signal to base station 100a. Then, base station 100a transmits the first signal to AI server 200 as a signal related to movement detection. Meanwhile, base station 100a simultaneously transmits information related to the first signal to management device 400. The information related to the first signal includes identification information of terminal 300 that transmitted the first signal and information on the transmission time. Management device 400 transmits this information to service providing device 500, and service providing device 500 transmits the identification information of terminal 300 and location information of terminal 300 corresponding to the transmission time to management device 400. The management device 400 transmits the received terminal location information together with the identification information and transmission time of the terminal 300 to the AI ​​server 200. This allows the AI ​​server 200 to obtain the location information of the terminal 300 at time t1, i.e., when the first signal was transmitted.

[0044] Similarly, at time t2, terminal 300 transmits a second signal to base station 100b. Then, base station 100b transmits the second signal to AI server 200 as a signal related to movement detection. Meanwhile, base station 100b simultaneously transmits information about the second signal to management device 400. The information about the second signal includes identification information of terminal 300 that transmitted the second signal and information about the transmission time. Management device 400 transmits this information to service providing device 500, and service providing device 500 transmits to management device 400 the identification information of terminal 300 and location information of terminal 300 corresponding to the transmission time. Management device 400 transmits the received terminal location information to AI server 200 together with the identification information and transmission time of terminal 300. This allows AI server 200 to obtain location information of terminal 300 at time t2, i.e., at the time the second signal was transmitted.

[0045] In this way, the learning unit 231 can acquire the first position of the terminal 300 when the first signal is transmitted and the second position of the terminal 300 when the second signal is transmitted. The learning unit 231 determines whether the terminal 300 is moving based on the acquired first and second positions. That is, the learning unit 231 determines whether the terminal 300 is moving based on whether the distance between the first and second positions is within a predetermined distance. The learning unit 231 then calculates the difference between the first and second signals and uses this as a differential feature. The learning unit 231 performs learning using this differential feature as an explanatory variable and training data with binary information indicating whether the terminal 300 is moving as a target variable. The learning unit 231 trains multiple (large numbers) of such training data to generate the mobility learning model 141.

[0046] Note that even after generating the mobile learning model 141, the learning unit 231 may appropriately receive the first signal and the second signal from the base station 100 and perform re-learning. Re-learning may be performed at a timing when the AI ​​processing unit 232 is not performing AI processing, or at a timing when a predetermined number of first signals and second signals have been accumulated.

[0047] The AI ​​processing unit 232 executes predetermined AI processing based on information about each terminal 300 transmitted from the base station 100. The predetermined AI processing may be processing specific to the terminal 300, or comprehensive processing based on information about multiple terminals 300. For example, the predetermined AI processing may be determining communication settings for the terminal 300 (determining the frequency to be used, the communication timing, the beamforming direction, etc.). For example, when applying AI processing (AI (Artificial Intelligence) / ML (Machine Learning)) to beamforming for a terminal, it is possible to realize processing such as sending inference data used in AI processing for beamforming for a moving terminal at a high frequency and sending inference data used in AI processing for beamforming for a stationary terminal at a low frequency. Alternatively, the predetermined AI processing may be people flow analysis, for example. The AI ​​processing unit 232 may output information indicating the results of the AI ​​processing via the output unit 250, or may transmit the information to an external device (e.g., the base station 100) via the communication unit 210.

[0048] The storage unit 240 has a function of storing various programs and data required for the operation of the AI ​​server 200. The storage unit 240 can be realized by, for example, a hard disk drive (HDD), a solid state drive (SSD), a flash memory, etc., but is not limited to these. The storage unit 240 may store, for example, a program for learning whether or not the terminal is moving based on the first signal and the second signal and generating the movement learning model 141, a program for executing AI processing, etc.

[0049] The output unit 250 has a function of outputting designated information in accordance with an instruction from the control unit 230. The output unit 250 may, for example, output text information or image information, in which case the output unit 250 is realized by a monitor provided in or connected to the base station 100. The output unit 250 may, for example, output audio information, in which case the output unit 250 is realized by a speaker provided in or connected to the information processing device.

[0050] The above is an example configuration of the AI ​​server 200.

[0051] The terminal 300 is similar to a typical information processing device having a configuration similar to that of a communication terminal such as a smartphone or tablet terminal, and therefore a detailed description using a block diagram will be omitted.

[0052] <Operation> Using Figures 5 to 8, the operation of the base station 100 and the AI ​​server 200 in the communication system 1 according to this embodiment will be described.

[0053] First, the process of generating the mobile learning model 141 will be described with reference to FIGS.

[0054] FIG. 5 is a sequence diagram showing an example of interactions between devices in the communication system 1 when generating the mobile learning model 141.

[0055] As shown in FIG. 5, the terminal 300 transmits a first signal to the base station 100 at a certain time (step S501).

[0056] When the base station 100 receives the first signal, it transmits it to the AI ​​server 200 (step S502). When the AI ​​server 200 receives the first signal, it identifies the first position of the terminal 300 at the time the first signal was transmitted (step S503).

[0057] On the other hand, after transmitting the first signal, the terminal 300 transmits the second signal (step S504).

[0058] When the base station 100 receives the second signal, it transmits it to the AI ​​server 200 (step S505). When the AI ​​server 200 receives the second signal, it identifies the second position of the terminal 300 at the time the second signal was transmitted (step S506).

[0059] The AI ​​server 200 performs learning using the differential features between the received first signal and the received second signal as explanatory variables and information indicating whether the terminal is moving based on the first position identified in step S503 and the second position identified in step S506 as a target variable, and generates a mobility learning model 141. The AI ​​server 200 transmits the generated mobility learning model 141 to each base station 100 (step S508).

[0060] The base station 100 receives the mobile learning model 141, stores it in the storage unit 140 (step S509), and ends the process.

[0061] By performing learning based on a large number of first signals and second signals, the AI ​​server 200 can generate a movement learning model 141 that can determine (estimate) whether the terminal 300 is moving or not based solely on the signals transmitted from the terminal 300. The process shown in Fig. 5 can be automated by a program, so that a large amount of training data required for learning can be automatically generated, and learning can be performed without the effort of manually preparing training data.

[0062] FIG. 6 is a flowchart showing an example of the operation of the AI ​​server 200 that realizes the exchange shown in FIG.

[0063] 6, the communication unit 210 of the AI ​​server 200 receives the first signal transmitted from the base station 100 (step S601). The communication unit 210 transmits the received first signal to the control unit 230.

[0064] The learning unit 231 of the control unit 230 identifies the first location of the terminal 300 corresponding to the first signal at the time of transmitting the first signal (step S602). The method of identifying the first location is as described above.

[0065] The communication unit 210 further receives the second signal transmitted from the base station 100 (step SS603). The communication unit 210 transfers the received second signal to the control unit 230.

[0066] The control unit 230 identifies the second location of the terminal 300 corresponding to the second signal at the time of transmitting the second signal (step S604). The method of identifying the second location is as described above.

[0067] The learning unit 231 determines whether the terminal 300 is moving based on the first location identified in step S602 and the second location identified in step S604 (step S605). Specifically, the learning unit 231 may determine whether the terminal 300 is moving based on whether the distance between the first location and the second location is within a predetermined distance. Here, the predetermined distance may be any distance that can be determined to be moving (the environment around the terminal 300 has changed).

[0068] The learning unit 231 generates differential features between the first signal and the second signal, generates training data using the differential features as explanatory variables and the information on whether the vehicle is moving, which is the determination result of step S605, as a target variable, and performs training (step S606). Any algorithm may be used for training, and examples include, but are not limited to, a decision tree, a neural network, and a support vector machine. The learning unit 231 performs training based on multiple combinations of the first signal and the second signal to generate a mobility training model 141 (step S607).

[0069] The learning unit 231 transmits the generated mobile learning model 141 to each base station 100 via the communication unit 210 (step S608), and ends the processing. Note that the learning unit 231 may store the generated mobile learning model 141 in the storage unit 240.

[0070] Note that the terminal 300 simply transmits signals as a normal terminal would, and therefore a description thereof will be omitted using a flowchart. Similarly, the base station 100 simply relays signals from the terminal 300 to the AI ​​server 200, and therefore a description thereof will be omitted using a flowchart. In this way, the AI ​​server 200 can automatically learn and generate a mobility learning model 141 for determining whether the terminal 300 is moving.

[0071] Next, the operation during operation will be described with reference to FIGS.

[0072] FIG. 7 is a sequence diagram showing an example of interactions between devices in the communication system 1 during operation of AI processing.

[0073] 7, the terminal 300 transmits a third signal to the base station 100 (step S701). After that, the terminal 300 transmits a fourth signal to the base station 100 (step S702).

[0074] The base station 100 determines whether the terminal is moving or not based on the third signal, the fourth signal, and the mobile learning model 141 (step S703). Then, the base station 100 transmits the terminal information to be used for AI processing to the AI ​​server 200 with high frequency if the terminal is moving and with low frequency if the terminal is not moving (step S704).

[0075] In the AI ​​server 200, the AI ​​processing unit 232 executes AI processing based on the terminal information received from the base station 100 (step S705).

[0076] FIG. 8 is a flowchart showing an example of the operation of the base station 100 when the exchange shown in FIG. 7 is performed.

[0077] 8, the communication unit 110 of the base station 100 receives the third signal transmitted from the terminal 300 (step S801). The communication unit 110 transmits the third signal to the control unit .

[0078] Furthermore, the communication unit 110 receives a fourth signal transmitted from the same terminal 300 (step S802). The communication unit 110 transmits the fourth signal to the control unit 130. Whether the fourth signal is transmitted from the same terminal 300 can be determined from, for example, an international mobile subscriber identity (IMSI) or a temporary mobile subscriber identity (TMSI).

[0079] The determination unit 131 determines whether the terminal 300 is moving by using the differential feature between the third signal and the fourth signal and the mobility learning model 141 (step S802). That is, the determination unit 131 generates a differential signal from the third signal and the fourth signal as a differential feature, and determines whether the terminal 300 is moving by inputting the generated differential feature to the mobility learning model 141. The determination unit 131 transmits the determination result to the transmission control unit 132.

[0080] If it is determined that the terminal 300 is moving (YES in step S804), the transmission control unit 132 sets the transmission frequency of the corresponding terminal 300 (terminal 300 that transmitted the third signal and the fourth signal) to "high" and registers it in the storage unit 140 (step S805). On the other hand, if it is determined that the terminal 300 is not moving (NO in step S804), the transmission control unit 132 sets the transmission frequency of the corresponding terminal 300 (terminal 300 that transmitted the third signal and the fourth signal) to "low" and registers it in the storage unit 140 (step S806).

[0081] The transmitting unit 133 transmits the information of the terminal 300 to be used for the pre-specified AI processing to the AI ​​server 200 via the communication unit 110 based on the setting of the transmission frequency of the terminal 300 registered in the storage unit 140 (step S807), and ends the processing. The processing shown in FIG. 8 is executed appropriately at any timing. In addition, the terminal information transmitted to the AI ​​server 200 here may be the same type of signal as the first to fourth signals, or may be information with different content, and is determined by the AI ​​processing executed by the AI ​​server 200.

[0082] The terminal information transmitted by the base station 100 to the AI ​​server 200 is processed by the AI ​​processing unit 232.

[0083] Note that, as for terminal 300, it simply transmits signals in the same way as a normal terminal, and therefore explanation using a flowchart will be omitted. Similarly, explanation using a flowchart will be omitted for AI server 200. In this way, base station 100 can determine whether to send information about terminal 300 to AI server 200 at a high frequency or a low frequency depending on whether terminal 300 is moving or not, and can transmit information at the determined frequency.

[0084] <Summary> As described above, the base station 100 can determine whether the terminal 300 is moving without directly acquiring location information from the terminal 300. Based on the determination result, the base station 100 can determine the transmission frequency for transmitting information about the terminal 300 to the AI ​​server 200, and transmit the information about the terminal 300 at the determined transmission frequency. When the terminal 300 is moving, the base station 100 transmits the information about the terminal 300 to the AI ​​server 200 at a high frequency. When the terminal 300 is not moving, the base station 100 transmits the information about the terminal 300 at a low frequency, which is lower than the high frequency. This reduces the total amount of information transmitted to the AI ​​server 200 compared to when information is transmitted uniformly to all terminals 300 connected to the base station 100. As a result, delays in transmitting information from the base station 100 to the AI ​​server 200 can be minimized, and the processing loads on the base station 100 and the AI ​​server 200 can be reduced.

[0085] <Modifications> It goes without saying that the base station 100 and the AI ​​server 200 according to the above embodiment are not limited to the above embodiment, and may be realized by other methods. Various modifications will be described below.

[0086] (1) In the above embodiment, the AI ​​server 200 has been described as a server having two functions: a function to generate a mobile learning model 141 for determining whether the terminal 300 is moving, and a function to perform AI processing. However, these functions may each be realized by a separate server, and the function of generating the mobile learning model 141 may be realized by an external device.

[0087] (2) In the above embodiment, the mobile learning model 141 is transmitted from the AI ​​server 200 to the base station 100, and during operation, the base station 100 determines whether the terminal 300 is moving. However, this is not limited to this. The base station 100 may be configured to request the AI ​​server 200 to determine whether the terminal 300 is moving by transmitting the third signal and the fourth signal to the AI ​​server 200, and to receive the determination result from the AI ​​server 200.

[0088] (3) In the above embodiment, the method by which the AI ​​server 200 identifies the location of the terminal 300 corresponding to the first signal and the second signal was described using FIG. 4 , but this is not limited thereto. For example, if there is a channel over which location information can be transmitted from the terminal 300 to the base station 100, the terminal 300 may transmit its own location information when transmitting the first signal and the second signal to the base station 100, and the base station 100 may transmit the received location information to the AI ​​server 200 along with the first signal or the second signal. The AI ​​server 200 may use the location information from the base station 100. Alternatively, when the base station 100 is communicating with the terminal 300 by performing beamforming on the terminal 300, the base station 100 may estimate the location of the terminal 300 based on the direction of the beamforming and the distance to the terminal 300 determined based on the reception strength of the signal transmitted from the terminal 300 at that time, and transmit this estimated location to the AI ​​server 200.

[0089] (4) In the above embodiment, an example was shown in which the base station 100 transmitted the first signal and the second signal to the AI ​​server 200. However, the control unit 130 of the base station 100 may convert the first signal and the second signal based on the first signal and the second signal into information indicating the difference characteristics between the first signal and the second signal, and transmit the converted information to the AI ​​server 200. Converting into information indicating the difference characteristics between the first signal and the second signal may, for example, involve generating a signal by subtracting the analog signal of the second signal from the analog signal of the first signal, but is not limited to this. In this case, the base station 100 may transmit to the AI ​​server 200 information indicating whether the terminal 300 is moving, along with the information. The method by which the base station 100 acquires the location information of the terminal 300 may be the same as the method by which the AI ​​server 200 acquires the location information described in the above embodiment, or the method described in the above modification (3) may be used.

[0090] (5) In the above embodiment, the mobile learning model 141 generated by the AI ​​server 200 may be a model common to all base stations 100, or may be a model specific to each base station. If the model is specific to each base station, the AI ​​server 200 can generate a mobile learning model 141 specific to the base station 100 by performing learning using only the first signal and the second signal transmitted from a single base station 100.

[0091] (6) The learning unit of the AI ​​server 200 in the above embodiment may be provided in the base station 100, and the base station 100 may perform learning on its own to generate the mobile learning model 141.

[0092] (7) In the above embodiment, an example was shown in which one base station server 100 receives the first signal and the second signal or the third signal and the fourth signal, transmits them to the AI ​​server 200, and determines whether the terminal 300 is moving. However, between the transmission of the first signal and the second signal, or between the transmission of the third signal and the fourth signal, the terminal 300 may perform a handover to the base station 100. In this case, it may be possible to determine whether the signals from the terminal 300 are from the same terminal by the SMSI, TMSI, or the like included in the signals, and the setting by the transmission control unit 132 may be performed by the base station 100 after the handover.

[0093] (8) The program for the base station 100 of the present disclosure to determine whether the terminal of each embodiment is moving, set the frequency of transmitting information to the AI ​​server 200, and communicate with the AI ​​server 200, as well as the program for the AI ​​server 200 to generate the mobility learning model 141, may be provided in a state stored in a computer-readable storage medium. The storage medium can store the program in a "non-transitory tangible medium." The storage medium can include any suitable storage medium, such as an HDD or SSD, or an appropriate combination of two or more thereof. The storage medium may be volatile, non-volatile, or a combination of volatile and non-volatile. Note that the storage medium is not limited to these examples and may be any device or medium capable of storing a program.

[0094] The base station 100 and the AI ​​server 200 can realize the functions of the multiple functional units shown in each embodiment by, for example, reading a program stored in a storage medium and executing the read program. The program may also be provided to the base station 100, the AI ​​server 200, etc. via any transmission medium (such as a communication network or broadcast waves). The base station 100 and the AI ​​server 200 realize the functions of the multiple functional units shown in each embodiment by executing a program downloaded via the Internet, for example. This program may be executed by the base station 100, etc.

[0095] The program can be implemented using, for example, a scripting language such as ActionScript or JavaScript (registered trademark), an object-oriented programming language such as Objective-C or Java (registered trademark), or a markup language such as HTML5, but is not limited to these.

[0096] At least a part of the processing in the base station 100 and the AI ​​server 200 may be realized by cloud computing consisting of one or more computers. Furthermore, each functional unit of the base station, the AI ​​server 200, and the terminal 300 may be realized by one or more circuits that realize the functions described in the above embodiments, and the functions of multiple functional units may be realized by one circuit.

[0097] (9) The various techniques and processes shown in the above embodiments and variant examples may be combined as appropriate within the scope of achieving the purpose of generating the mobile learning model 141 or transmitting information about the terminal 300 from the base station 100 to the AI ​​server 200.

[0098] REFERENCE SIGNS LIST 100 Base station 110 Communication unit 120 Input unit 130 Control unit 131 Determination unit 132 Transmission control unit 133 Transmission unit 140 Storage unit 150 Output unit 200 AI server 210 Communication unit 220 Input unit 230 Control unit 231 Learning unit 232 AI processing unit 240 Storage unit 250 Output unit 300, 300a, 300b, 300c Terminal 400 Management device 500 Service providing device

Claims

1. A base station that relays communications between terminals and communicates with an AI server that performs AI processing, comprising: a receiving unit that receives signals from the terminal; a determining unit that determines whether the terminal is moving based on the signal from the terminal; a transmission control unit that controls the transmission of information about the terminal to the AI ​​server at a relatively high frequency when the terminal is moving, and the transmission of information about the terminal to the AI ​​server at a relatively low frequency when the terminal is not moving; and a transmitting unit that transmits information about the terminal to the AI ​​server at the frequency controlled by the transmission control unit.

2. The base station according to claim 1, further comprising a memory unit that stores a learning model that has learned the correspondence between a first signal received by the base station from the terminal at a first time, a second signal received by the base station from the terminal at a second time after the first time, and whether the terminal is moving between the first time and the second time, and wherein the judgment unit judges whether the terminal is moving based on a third signal received by the base station from the terminal at a third time, a fourth signal received by the base station at a fourth time after the third time, and the learning model.

3. The base station described in claim 2, characterized in that the learning model is a model that has learned the correspondence between feature points based on the difference between the first signal and the second signal and whether the terminal is moving or not, and the judgment unit judges whether the terminal is moving or not based on feature points based on the difference between the third signal and the fourth signal and the learning model.

4. The base station according to claim 3, characterized in that the feature point based on the difference between the first signal and the second signal is a feature point based on the difference between the signal waveform of the first signal and the signal waveform of the second signal, and the feature point based on the difference between the third signal and the fourth signal is a feature point based on the difference between the signal waveform of the third signal and the signal waveform of the fourth signal.

5. A communication method in which a computer at a base station that relays communications between terminals and communicates with an AI server that performs AI processing executes the following steps: a receiving step in which the computer receives a signal from the terminal; a determining step in which the computer determines whether the terminal is moving based on the signal from the terminal; a transmission control step in which, if the terminal is moving, the computer controls the computer to send information about the terminal to the AI ​​server at a relatively high frequency, and, if the terminal is not moving, the computer controls the computer to send information about the terminal to the AI ​​server at a relatively low frequency; and a transmission step in which the computer transmits information about the terminal to the AI ​​server at the frequency controlled by the transmission control step.

6. A communications program that implements, in a base station computer that relays communications between terminals and communicates with an AI server that performs AI processing, the following functions: a receiving function that receives signals from the terminal; a determination function that determines whether the terminal is moving based on the signal from the terminal; a transmission control function that controls the terminal to send information about the terminal to the AI ​​server at a relatively high frequency if the terminal is moving, and to send information about the terminal to the AI ​​server at a relatively low frequency if the terminal is not moving; and a transmission function that transmits information about the terminal to the AI ​​server at a frequency controlled by the transmission control function.

Images