Wireless quality prediction system, wireless quality prediction device, wireless quality prediction method, and program

Abstract

According to the present invention, in order to improve the accuracy of predicting the wireless quality of wireless terminals in various communication environments, a wireless quality prediction system comprises: an information acquisition unit configured to acquire information about the wireless terminal, including the terminal position of the wireless terminal that performs wireless communication; a plurality of prediction units configured to predict, on the basis of the information about the wireless terminal, predicted values of received power of the wireless terminal after a predetermined time has elapsed by using mutually different prediction methods; a calculation unit configured to calculate predictive data regarding the wireless quality of the wireless terminal after the predetermined time has elapsed, by weighting the predicted values of the received power predicted by the plurality of prediction units in accordance with the terminal position of the wireless terminal; and a notification unit configured to notify the wireless terminal of the prediction data.

Description

[Technical Field]

[0001] The present invention relates to a wireless quality prediction system, a wireless quality prediction device, a wireless quality prediction method, and a program. [Background technology]

[0002] There is a wireless quality prediction technology that predicts the wireless quality of a wireless terminal that communicates with a wireless base station for stable use of wireless communication, etc. For example, a technology is known in which past quality information is stored in a wireless quality prediction server located on a network, and a predicted value of wireless quality at the terminal's location is calculated by machine learning (see, for example, Non-Patent Document 1). [Prior art documents] [Non-patent literature]

[0003] [Non-Patent Document 1] Naoki Shibuya et al., "A Study on Wireless Communication Quality Prediction Using Machine Learning," IEICE General University, B-6-74, Mar, 2022 Summary of the Invention [Problem to be solved by the invention]

[0004] The method for predicting wireless quality is not limited to the above method, and various other methods can be applied, such as predicting wireless quality using recurrent neural networks (RNNs) that perform time-series prediction from time-series information.

[0005] However, these methods use different approaches to predict wireless quality, and each method may not be suitable for the actual communication environment, resulting in the problem that the prediction accuracy of each method varies depending on the communication environment.

[0006] The embodiments of the present invention have been made in consideration of the above-mentioned problems, and provide a wireless quality prediction system that improves the prediction accuracy of the wireless quality of a wireless terminal in various communication environments. [Means for solving the problem]

[0007] In order to solve the above problem, a wireless quality prediction system according to an embodiment of the present invention includes: After a specified time has passed Terminal position of wireless terminal and time series data of the received power of the wireless terminal. an information acquisition unit configured to acquire information of the wireless terminal, the information including a first prediction unit configured to predict the received power of the wireless terminal after a predetermined time has elapsed, using a terminal position of the wireless terminal after the predetermined time has elapsed; and A predicted value of the received power of the wireless terminal after a predetermined time has elapsed. a second prediction unit configured to predict; a plurality of prediction units; of, The wireless communication device includes a calculation unit configured to calculate predicted data regarding the wireless quality of the wireless terminal after the predetermined time has elapsed by weighting the data according to the terminal location of the wireless terminal, and a notification unit configured to notify the wireless terminal of the predicted data. [Effects of the Invention]

[0008] According to an embodiment of the present invention, it is possible to provide a wireless quality prediction system that improves the accuracy of predicting the wireless quality of a wireless terminal in a variety of communication environments. [Brief explanation of the drawings]

[0009] [Figure 1] 1 is a diagram illustrating an example of a system configuration of a wireless quality prediction system according to an embodiment of the present invention. [Figure 2] FIG. 1 is a diagram for explaining an overview of the processing of a wireless quality prediction system according to an embodiment of the present invention. [Figure 3] 10 is a flowchart illustrating an example of a process for predicting wireless quality according to the first embodiment. [Figure 4] FIG. 4 is a diagram for explaining a degradation risk value of wireless quality according to the first embodiment. [Figure 5] FIG. 10 is a diagram for explaining a method for calculating a risk value according to Modification 3. [Figure 6] 10 is a flowchart illustrating an example of a process of predicting wireless quality according to the second embodiment. [Figure 7] FIG. 10 is a diagram illustrating an expected value of received power according to the second embodiment. [Figure 8] FIG. 1 is a diagram illustrating an example of a hardware configuration of a wireless quality prediction device according to an embodiment of the present invention. [Figure 9] FIG. 2 is a diagram illustrating an example of a hardware configuration of a wireless terminal according to the present embodiment. [Figure 10] FIG. 1 is a diagram illustrating an example of a conventional wireless quality prediction process. [Figure 11] FIG. 1 is a diagram for explaining a problem with the conventional technology. DETAILED DESCRIPTION OF THE INVENTION

[0010] Hereinafter, an embodiment of the present invention will be described with reference to the drawings. The embodiment described below is merely an example, and the embodiment to which the present invention is applied is not limited to the following embodiment.

[0011] <About wireless quality prediction technology> Before describing the system configuration of the wireless quality prediction system according to this embodiment, an example of a conventional wireless quality prediction technique will be described. The wireless quality prediction technique is a technique for predicting the wireless quality of a wireless terminal that performs wireless communication with a wireless base station, for example, in order to ensure stable use of wireless communication.

[0012] Fig. 10 is a diagram showing an example of a conventional wireless quality prediction process. This diagram shows an overview of the wireless quality prediction process disclosed in Non-Patent Document 1. In the example of Fig. 10, a wireless quality prediction server 1 acquires performance data such as location information, radio wave intensity, and throughput from multiple wireless terminals 2 (step S1). Furthermore, the wireless quality prediction server 1 uses the acquired performance data to execute a learning process for machine learning a neural network 10 so that the radio wave intensity, throughput, etc. of the wireless terminals 2 and 3 can be predicted based on the location information of the wireless terminals 2 and 3 (step S2).

[0013] After the learning process, the wireless quality prediction server 1 receives an inquiry about wireless quality from a wireless terminal 3 having a mobile function such as a vehicle (step S3). In response to this, the wireless quality prediction server 1 inputs the location information of the wireless terminal 3 included in the inquiry about wireless quality into the trained neural network 10, thereby executing an inference process to predict the radio wave strength, throughput, etc. of the wireless terminal 3 (step S4). The wireless quality prediction server 1 also notifies the wireless terminal 3 of the predicted values ​​of wireless quality such as the radio wave strength, throughput, etc. of the wireless terminal 3 (step S5).

[0014] By the above process, for example, when the wireless terminal 3 wants to know the wireless quality after a predetermined time has elapsed (after n seconds), it estimates the position of the wireless terminal 3 after the predetermined time has elapsed and transmits a wireless quality inquiry including the estimated position to the wireless quality prediction server 1. This allows the wireless terminal 3 to acquire the wireless quality such as the radio wave intensity and throughput of the wireless terminal 3 after the predetermined time has elapsed.

[0015] The above-described wireless quality prediction process is an example. The wireless quality prediction server 1 may predict the received power (an example of wireless quality) corresponding to the location information of the wireless terminal 3 using, for example, three-dimensional data such as a building DB (Database) or CAD (Computer Aided Design) data, and a radio wave propagation estimation technique such as ray tracing.

[0016] In addition, the wireless quality prediction server 1 may predict the received power of the wireless terminal 3 after a predetermined time has elapsed using an RNN (Recurrent Neural Networks) or the like that performs time series prediction from time series data of the received power of the wireless terminal 3 (for example, the received power for the last t seconds, etc.).

[0017] (About the assignment) 11 is a diagram for explaining the problems of the conventional technology. As described above, the wireless quality prediction server 1 can predict the wireless quality (e.g., received power) of the wireless terminal 2 after a predetermined time has elapsed using various methods. However, each method has a problem in that the prediction accuracy of the wireless quality varies depending on the surrounding wireless environment.

[0018] 11, for example, for a wireless terminal 3a that is indoors and has line-of-sight with the wireless base station 20a, method A may have higher accuracy in predicting wireless quality. However, for a wireless terminal 3b that is indoors and does not have line-of-sight with the wireless base stations 20a and 20b, method B may have higher accuracy in predicting wireless quality than method A. Furthermore, for a wireless terminal 3c that is outdoors and has line-of-sight with the wireless base station 20c, method C may have higher accuracy in predicting wireless quality than methods A and B.

[0019] As described above, conventional techniques have the problem that each method may not be suitable depending on the actual communication environment, and the prediction accuracy of each method varies depending on the communication environment.

[0020] Therefore, a wireless quality prediction system 100 according to this embodiment has a system configuration as shown in FIG. 1, for example, in order to improve the prediction accuracy of the wireless quality of a wireless terminal in various communication environments.

[0021] <System configuration> 1 shows an example of the system configuration of a wireless quality prediction system according to this embodiment. The wireless quality prediction system 100 includes a wireless terminal 120 that can communicate with a wireless base station 101 via a predetermined wireless communication 102, and a wireless quality prediction device 110 that predicts the wireless quality of the wireless terminal 120.

[0022] The wireless quality predicting device 110 is, for example, an information processing device having a computer configuration or a system including multiple computers. The wireless quality predicting device 110 executes a wireless quality prediction process to predict the wireless quality of the wireless terminal 120 after a predetermined time has elapsed (after n seconds) based on, for example, wireless terminal information transmitted by the wireless terminal 120.

[0023] The wireless terminal 120 is a wireless station that communicates with the wireless base station 101 using a predetermined wireless communication such as 5G (5th Generation) or LTE (Long Term Evolution). The wireless terminal 120 can communicate with the wireless quality prediction device 110 via the wireless base station 101, for example. The wireless terminal 120 transmits information about the wireless terminal to the wireless quality prediction device 110, requests a predicted value of wireless quality after a predetermined time has elapsed, and performs predetermined control such as handover based on the predicted value of wireless quality notified by the wireless quality prediction device 110.

[0024] The wireless terminal 120 according to this embodiment may be, for example, a wireless station such as a smartphone carried by a user, or may be, for example, a wireless station with a mobile function such as a robot or a vehicle.

[0025] <Functional configuration> (Functional configuration of wireless terminal) The wireless terminal 120 realizes, for example, a position acquisition unit 121, a position estimation unit 122, a state acquisition unit 123, a received power acquisition unit 124, an inquiry unit 125, a control determination unit 126, a terminal control unit 127, and a wireless communication unit 128 by causing a computer included in the wireless terminal 120 to execute a predetermined program. Note that at least a part of the above functional configurations may be realized by hardware.

[0026] The position acquisition unit 121 acquires the terminal position (current position) of the wireless terminal 120 by using a positioning device such as a GPS (Global Positioning System) device provided in the wireless terminal 120, for example.

[0027] The position estimation unit 122 calculates the terminal position (estimated position) of the wireless terminal 120 after a predetermined time has elapsed (n seconds later) from, for example, the terminal position of the wireless terminal 120 acquired by the position acquisition unit 121 and the movement of the wireless terminal 120 measured by a sensor such as an acceleration sensor or an angle sensor. Alternatively, the position estimation unit 122 may store a history of the current position of the wireless terminal 120 acquired by the position acquisition unit 121, and estimate the terminal position (estimated position) of the wireless terminal 120 after a predetermined time has elapsed based on this history. Furthermore, if the wireless terminal 120 is a robot, a vehicle, or the like that moves along a predetermined movement path, the position estimation unit 122 may estimate the terminal position (estimated position) of the wireless terminal 120 after a predetermined time has elapsed based on, for example, movement path information of the wireless terminal 120, etc.

[0028] The state acquisition unit 123 acquires the terminal state (orientation, speed, etc.) of the wireless terminal 120 based on, for example, sensor data from an acceleration sensor, an angle sensor, etc. provided in the wireless terminal 120. The state acquisition unit 123 also estimates the terminal state of the wireless terminal 120 after a predetermined time has elapsed (after n seconds) based on sensor data from the acceleration sensor, angle sensor, etc. For example, the state acquisition unit 123 may estimate the terminal state of the wireless terminal 120 after n seconds using a machine learning model or the like that has been trained in advance using the sensor data of the wireless terminal 120 as feature quantities and the terminal state of the wireless terminal 120 after n seconds as training data. Alternatively, the state acquisition unit 123 may store a history of the acquired terminal states of the wireless terminal 120 and estimate the terminal state of the wireless terminal after a predetermined time has elapsed based on this history.

[0029] The received power acquisition unit 124 acquires information about the received power of the wireless terminal 120 that performs wireless communication with the wireless base station 101. For example, the received power information may be information such as a received signal strength indicator (RSSI) output by the wireless communication unit 128. The received power acquisition unit 124 also acquires data (time-series data) received by the wireless terminal 120 for the most recent t seconds.

[0030] The inquiry unit 125 transmits to the wireless quality predicting device 110 wireless terminal information including, for example, the terminal position (current position and estimated position) of the wireless terminal 120, an estimate of the terminal state (orientation, speed, etc.) of the wireless terminal 120 after a predetermined time has elapsed, and information such as received power. Preferably, the terminal position of the wireless terminal includes, for example, the terminal position (current position) of the wireless terminal 120 acquired by the position acquisition unit 121 and the terminal position (estimated position) of the wireless terminal 120 after a predetermined time has elapsed (n seconds later) estimated by the position estimation unit 122. Furthermore, the received power includes, for example, the latest received power acquired by the received power acquisition unit 124 and the received power (time series data) for the most recent t seconds.

[0031] The control determination unit 126 receives a predicted value of wireless quality transmitted from the wireless quality prediction device 110 in response to the wireless terminal information transmitted by the inquiry unit 125, and determines the control content of the wireless terminal 120 based on the received predicted value of wireless quality. For example, the control determination unit 126 determines whether or not the wireless terminal 120 will initiate handover based on the predicted value of wireless quality after a predetermined time has elapsed (after n seconds), which is received from the wireless quality prediction device 110.

[0032] Furthermore, if the wireless terminal 120 is, for example, a wireless terminal with a movement function such as a robot or vehicle that holds movement route information and moves along a predetermined route, the control determination unit 126 may determine whether to change the movement route, etc., based on a predicted value of wireless quality after a predetermined time has elapsed.

[0033] The terminal control unit 127 controls the entire wireless terminal 120. For example, based on the determination result by the control determination unit 126, the terminal control unit 127 starts handover of the wireless terminal 120, changes the moving route, etc.

[0034] The wireless communication unit 128 performs predetermined wireless communication with the wireless base station 101, for example, under control of the terminal control unit 127. Also, under control of the terminal control unit 127, the wireless communication unit 128 executes, for example, handover processing to change the wireless base station 101 to which it is connected.

[0035] (Functional configuration of wireless quality prediction device) The wireless quality predicting device 110 implements, for example, an information acquiring unit 111, a plurality of predicting units 112a, 112b, 112c, ..., a calculating unit 113, a notifying unit 114, and a storage unit 115 by executing a predetermined program on one or more computers included in the wireless quality predicting device 110. In the following description, when referring to any of the plurality of predicting units 112a, 112b, 112c, ..., the term "predicting unit 112" is used.

[0036] The information acquisition unit 111 executes information acquisition processing to acquire information about a wireless terminal transmitted by an inquiry unit 125 of the wireless terminal 120 that performs wireless communication with the wireless base station 101. The information acquisition unit 111 also notifies a plurality of prediction units 112 of the acquired information about the wireless terminal, and notifies a calculation unit 113 of information about the received power.

[0037] The plurality of prediction units 112a, 112b, 112c, etc. execute prediction processes using different prediction methods to predict the predicted value of the received power of the wireless terminal 120 after a predetermined time has elapsed (after n seconds) based on the information of the wireless terminal notified by the information acquisition unit 111. Specific examples of each prediction unit 112 will be described later.

[0038] The calculation unit 113 performs a calculation process to weight the predicted values ​​of received power predicted by the multiple prediction units 112a, 112b, 112c, ... according to the terminal position of the wireless terminal 120, and calculates predicted data regarding the wireless quality of the wireless terminal 120 after a predetermined time has elapsed. For example, the calculation unit 113 calculates, as the predicted data regarding the wireless quality of the wireless terminal 120, a degradation risk value (probability that the communication quality will become equal to or lower than a threshold) indicating the risk that the communication quality of the wireless terminal 120 will deteriorate after a predetermined time has elapsed. Alternatively, the calculation unit 113 may calculate, as the predicted data regarding the wireless quality of the wireless terminal 120, a value indicating the communication quality of the wireless terminal 120 after a predetermined time has elapsed (for example, a predicted value of received power or a predicted value of throughput, etc.).

[0039] The notification unit 114 notifies (transmits) the predicted data on the wireless quality of the wireless terminal 120 calculated by the calculation unit 113 to the wireless terminal 120 that transmitted the information about the wireless terminal.

[0040] The storage unit 115 is realized by, for example, a program executed by a computer included in the wireless quality prediction device 110, a storage device, a memory, etc. The storage unit 115 stores, for example, threshold values ​​used by the calculation unit 113 to determine deterioration of wireless quality. Preferably, the storage unit 115 stores a plurality of threshold values ​​that are set in advance according to, for example, the purpose of use, the model of the wireless terminal 120, the location, etc.

[0041] 1 is an example. For example, at least some of the multiple prediction units 112a, 112b, 112c, etc. may be realized by a computer or the like external to the wireless quality prediction device 110. Furthermore, the storage unit 115 may use a computer or a cloud service external to the wireless quality prediction device 110. Furthermore, at least some of the multiple prediction units 112 and calculation unit 113 may be included in the wireless terminal 120.

[0042] (Processing Overview) 2 is a diagram illustrating an overview of the processing of the wireless quality prediction system according to this embodiment. In the example of FIG. 2, the prediction unit 112a predicts the received power of the wireless terminal 120 by, for example, radio wave propagation estimation such as ray tracing based on, for example, three-dimensional data such as a building DB or CAD data, and information such as the terminal position and terminal state of the wireless terminal 120. For example, the prediction unit 112a can calculate a predicted value of the received power of the wireless terminal 120 after a predetermined time has elapsed (n seconds later) based on the terminal position and terminal information of the wireless terminal 120. Note that the prediction unit 112a may predict the received power of the wireless terminal 120 based on the three-dimensional data and the terminal position of the wireless terminal 120, regardless of the terminal state of the wireless terminal 120.

[0043] The prediction unit 112b has a received power learner 201 that has undergone machine learning to output a predicted value of the received power of the wireless terminal 120 based on, for example, the terminal position and terminal state of the wireless terminal 120 from past performance data or the like. For example, the prediction unit 112b can predict a predicted value of the received power of the wireless terminal 120 after a predetermined time has elapsed (n seconds later) based on the terminal position and terminal information of the wireless terminal 120. Note that the prediction unit 112b may predict the received power of the wireless terminal 120 based on the terminal position of the wireless terminal 120, regardless of the terminal state of the wireless terminal 120.

[0044] The prediction unit 112a and the prediction unit 112b are examples of a first prediction unit configured to predict the received power of the wireless terminal 120 after a predetermined time has elapsed using the terminal position of the wireless terminal 120 after a predetermined time has elapsed.

[0045] The prediction unit 112c has a received power learner 202 that has undergone machine learning to output a predicted value of the received power of the wireless terminal 120 after a predetermined time has elapsed, based on, for example, the terminal position of the wireless terminal 120 and the received power for the most recent t seconds from past performance data, etc. For example, the prediction unit 112c can predict the predicted value of the received power of the wireless terminal 120 after a predetermined time has elapsed, based on the terminal position of the wireless terminal 120 and the received power of the wireless terminal 120 for the most recent t seconds (time series data).

[0046] The prediction unit 112c is an example of a second prediction unit configured to predict the received power of the wireless terminal 120 after a predetermined time has elapsed using the terminal position of the wireless terminal 120 and time series data of the received power of the wireless terminal 120.

[0047] The calculation unit 113 has a wireless quality learner 210 that learns the weights of the multiple prediction units 112 at each terminal position using the received power of the wireless terminal 120 predicted by the multiple prediction units 112 at multiple terminal positions and the correct values ​​of the received power at the multiple terminal positions. For example, the calculation unit 113 can calculate a predicted value of the wireless quality after a predetermined time has elapsed based on the terminal position of the wireless terminal 120 after a predetermined time has elapsed (n seconds later) and the predicted values ​​of the received power of the wireless terminal 120 after the predetermined time has elapsed predicted by the multiple prediction units 112.

[0048] <Processing flow> Next, the processing flow of the wireless quality prediction method according to this embodiment will be described by taking a specific example as an example.

[0049] [Example 1] Fig. 3 is a flowchart illustrating an example of a process for predicting wireless quality according to the first embodiment. This process illustrates a specific example of the process described in Fig. 2. At the start of the process illustrated in Fig. 3, the received power learner 201, the received power learner 202, and the wireless quality learner 210 described in Fig. 2 are assumed to have completed learning.

[0050] In step S301, the information acquisition unit 111 of the wireless quality predicting device 110 acquires wireless terminal information transmitted by the wireless terminal 120. Here, the wireless terminal information includes the terminal position (current position and estimated position) of the wireless terminal 120, an estimated value of the terminal state (orientation, speed, etc.) of the wireless terminal 120 after a predetermined time has elapsed (n seconds), and information such as the received power of the wireless terminal 120. Furthermore, the received power information of the wireless terminal 120 includes information (time series data) of the received power of the wireless terminal 120 for the most recent t seconds. Note that the values ​​of "n seconds", "t seconds", etc. are assumed to be appropriate values ​​that have been set in advance by, for example, an administrator who manages the wireless quality prediction system 100 or a designer who designed the wireless quality prediction system 100.

[0051] In step S302, the prediction unit 112a of the wireless quality prediction device 110 uses estimated values ​​of the terminal position and terminal state of the wireless terminal 120 after a predetermined time has elapsed (n seconds) to predict the received power of the wireless terminal 120 after the predetermined time has elapsed, using the radio wave propagation estimation described above.

[0052] In step S303, the prediction unit 112b of the wireless quality prediction device 110 uses estimated values ​​of the terminal position and terminal state of the wireless terminal 120 after a predetermined time has elapsed (n seconds) to predict the received power of the wireless terminal 120 after the predetermined time has elapsed using the received power learning device 201.

[0053] In step S304, the prediction unit 112c of the wireless quality prediction device 110 predicts the received power of the wireless terminal 120 after a predetermined time has elapsed using the terminal position (current position) of the wireless terminal 120 and the received power of the wireless terminal 120 for the last t seconds, using the received power learner 202. Note that the received power of the wireless terminal 120 for the last t seconds includes, for example, time-series data of the received power measured by the wireless terminal 120 at predetermined time intervals. Furthermore, the processes of steps S302 to S304 are executed, for example, in parallel by the prediction units 112a to 112c.

[0054] In step S305, the calculation unit 113 of the wireless quality prediction device 110 weights the predicted values ​​of received power predicted by the multiple prediction units 112a to 112c with weights according to the terminal position of the wireless terminal 120, and calculates a deterioration risk value of wireless quality after a predetermined time has elapsed.

[0055] Fig. 4 is a diagram for explaining a deterioration risk value of wireless quality according to the first embodiment. In Fig. 4, it is assumed that a wireless terminal 120 that performs wireless communication with a wireless base station 101 moves from its current location to an estimated location after a predetermined time has elapsed (n seconds).

[0056] Furthermore, it is assumed that the prediction unit 112a predicts that the received power of the wireless terminal 120 at the estimated position after a predetermined time has elapsed will be −82 dBm. Similarly, it is assumed that the prediction unit 112b predicts that the received power of the wireless terminal 120 at the estimated position will be −77 dBm, and the prediction unit 112c predicts that the received power of the wireless terminal 120 at the estimated position will be −80 dBm.

[0057] The calculation unit 113 calculates the risk (probability) that the received power of the wireless terminal 120 will be equal to or lower than the threshold value after a predetermined time has elapsed, using the predicted values ​​of the received power predicted by the multiple prediction units 112a to 112c and the threshold value obtained from the storage unit 115. For example, it is assumed that the threshold value is −80 dBm and the weights of the prediction units 112a, 112b, and 112c are the initial value of 1:1:1. In this case, the calculation unit 113 calculates that the wireless quality degradation risk value is approximately 66% (= 2 / 3) because the prediction units 112a and 112c predict that the received power of the wireless terminal 120 will be equal to or lower than the threshold value after a predetermined time has elapsed.

[0058] Furthermore, when the wireless terminal 120 actually arrives at the estimated position, the calculation unit 113 uses the actual measurement value "-80 dBm" of the received power of the wireless terminal 120 and the predicted values ​​of the plurality of prediction units 112a to 112c to learn the weight of each prediction unit. For example, the calculation unit 113 trains the wireless quality learner 210 so that the weight of the prediction unit 112 whose predicted value at the estimated position is closer to the actual measurement value is increased.

[0059] The calculation unit 113 repeatedly executes the same process to train the wireless quality learner 210. The terminal position and estimated position of the wireless terminal 120 may be pinpoint or may be, for example, a 5 m square cell. The feature used for training may also be data other than position (for example, environmental information acquired by LiDAR, or the positions of other terminals, etc.).

[0060] Furthermore, when wireless terminal 120 passes through the same location, calculation unit 113 calculates a deterioration risk value of wireless quality using the learned weights. For example, it is assumed that the weights after learning of prediction units 112a, 112b, and 112c are 2:1:3. In this case, the deterioration risk value of wireless quality is approximately 83% (=5 / 6).

[0061] Here, the description of the flowchart will continue, returning to Fig. 3. In step S306, the notification unit 114 of the wireless quality predicting device 110 notifies (transmits) to the wireless terminal 120 the deterioration risk value of the wireless quality predicted by the calculation unit 113.

[0062] In step S307, the wireless terminal 120 executes predetermined control such as handover or route change based on the wireless quality deterioration risk value notified by the wireless quality predicting device 110. For example, if the wireless quality deterioration risk value is equal to or greater than a predetermined value, the control determination unit 126 of the wireless terminal 120 determines to start handover to another wireless base station and instructs the terminal control unit 127 to start the handover. Furthermore, the terminal control unit 127 controls the wireless communication unit 128 in accordance with the instruction from the control determination unit 126 to start the handover.

[0063] As described above, according to the first embodiment, the wireless quality prediction system 100 can improve the prediction accuracy of the wireless quality degradation risk value in various communication environments by weighting the multiple prediction units 112 by the terminal position of the wireless terminal 120.

[0064] (Variation 1) The calculation unit 113 may calculate the wireless quality degradation risk value using a method other than the weighting method described above. For example, the calculation unit 113 may perform risk prediction by using a method used in machine learning classification (such as "Naive Bayes") to calculate the probability that the predicted value from each prediction unit 112 and the location information will be equal to or less than a threshold value, using these predicted values ​​as feature quantities. Alternatively, the calculation unit 113 may calculate the probability that the predicted values ​​from each prediction unit 112 will be correct, and perform a total risk prediction.

[0065] (Variation 2) The calculation unit 113 may adopt the lowest predicted value among the predicted values ​​from each prediction unit 112 for each terminal position (or estimated position), and determine whether the adopted predicted value is equal to or less than a threshold value. Alternatively, the calculation unit 113 may determine the probability that the lowest predicted value itself is correct using machine learning or the like, and use this probability as the risk value.

[0066] (Variation 3) 1 to 4, each prediction unit 112 predicts a uniquely determined predicted value, but each prediction unit 112 may calculate a probability distribution 501 of received power, for example, as shown in Fig. 5. In this case, the calculation unit 113 weights the probabilities for the received power values ​​from each prediction unit 112 so that the probability distributions calculated by each prediction unit 112 become a joint probability distribution, and creates a weighted probability distribution 502 of received power so that the sum is 1.

[0067] Furthermore, the calculation unit 113 calculates a deterioration risk value of wireless quality using the weighted received power probability distribution 502. For example, if the required quality (threshold) is -70 dBm or higher, the sum of the probability distributions below -70 dBm, "0.6", becomes the risk value.

[0068] In this way, the processes of the multiple prediction units 112 and the calculation unit 113 can be modified or applied in various ways.

[0069] [Example 2] In the first embodiment, the wireless quality predicting device 110 calculates, as a predicted value of wireless quality, a degradation risk value (probability that communication quality will become equal to or lower than a threshold) indicating a risk that communication quality of the wireless terminal 120 will deteriorate after a predetermined time has elapsed. However, without being limited to this, the wireless quality predicting device 110 may calculate an expected value (for example, an expected value of received power or an expected value of throughput) indicating wireless quality of the wireless terminal 120 after a predetermined time has elapsed.

[0070] Fig. 6 is a flowchart illustrating an example of a process of predicting wireless quality according to the embodiment 2. Among the processes illustrated in Fig. 6, the processes of steps S301 to S304 are the same as the process of predicting wireless quality according to the embodiment 1 described in Fig. 3, and therefore, description thereof will be omitted here.

[0071] In step S601, the calculation unit 113 of the wireless quality prediction device 110 weights the predicted values ​​of received power predicted by the multiple prediction units 112 with weights according to the terminal position of the wireless terminal 120, and calculates the expected value of the received power of the wireless terminal 120 after a predetermined time has elapsed (n seconds).

[0072] 7 is a diagram illustrating the expected value of received power according to the second embodiment. For example, it is assumed that the predicted value of received power predicted by the prediction unit 112a is −82 dBm and the weight of the prediction unit 112a is 2. It is also assumed that the predicted value of received power predicted by the prediction unit 112b is −77 dBm and the weight of the prediction unit 112b is 1. It is also assumed that the predicted value of received power predicted by the prediction unit 112c is −80 dBm and the weight of the prediction unit 112c is 3.

[0073] In this case, calculation unit 113 calculates expected value E of received power of wireless terminal 120 after a predetermined time has elapsed (n seconds) using (Equation 1) in Fig. 7. In the example of Fig. 7, expected value E of received power of wireless terminal 120 after a predetermined time has elapsed (n seconds) is -80.2 dBm.

[0074] 6, the description of the flowchart will be continued. In step S602, the notification unit 114 of the wireless quality predicting device 110 notifies the wireless terminal 120 of the expected value of the received power of the wireless terminal 120 after a predetermined time has elapsed, as calculated by the calculation unit 113.

[0075] In step S603, the wireless terminal 120 executes predetermined control such as handover or route change based on the expected value of the received power notified by the wireless quality predicting device 110.

[0076] As described above, in the second embodiment, the wireless quality prediction system 100 can improve the prediction accuracy of the wireless quality degradation risk value in a variety of communication environments by weighting the multiple prediction units 112 based on the terminal position of the wireless terminal 120.

[0077] <Hardware configuration example> (Hardware configuration of wireless quality prediction device) Fig. 8 is a diagram showing an example of the hardware configuration of the wireless quality predicting device 110 according to this embodiment. The wireless quality predicting device 110 has, for example, the configuration of a computer 800 as shown in Fig. 8. In the example of Fig. 8, the computer 800 has a processor 801, a memory 802, a storage device 803, a communication device 804, an input device 805, an output device 806, a bus B, etc.

[0078] The processor 801 is, for example, an arithmetic unit such as a CPU (Central Processing Unit) that executes predetermined programs to realize various functions. The memory 802 is a storage medium readable by the computer 800, and includes, for example, a RAM (Random Access Memory) and a ROM (Read Only Memory). The storage device 803 is a computer-readable storage medium, and may include, for example, a HDD (Hard Disk Drive), an SSD (Solid State Drive), various optical disks, and magneto-optical disks.

[0079] The communication device 804 includes one or more pieces of hardware (communication devices) for communicating with other devices via a wireless or wired network. The input device 805 is an input device (e.g., a keyboard, a mouse, a microphone, a switch, a button, a sensor, etc.) that receives input from the outside. The output device 806 is an output device (e.g., a display, a speaker, an LED lamp, etc.) that outputs to the outside. Note that the input device 805 and the output device 806 may be integrated into one device (e.g., an input / output device such as a touch panel display).

[0080] The bus B is commonly connected to the above components and transmits, for example, address signals, data signals, and various control signals. The processor 801 is not limited to a CPU, and may be, for example, a DSP (Digital Signal Processor), a PLD (Programmable Logic Device), or an FPGA (Field Programmable Gate Array).

[0081] (Wireless terminal hardware configuration) 9 is a diagram showing an example of the hardware configuration of a wireless terminal according to this embodiment. The wireless terminal 120 includes, for example, a GPS device 901, a sensor 902, and the like in addition to the hardware configuration of the computer 800 described in FIG.

[0082] The GPS device 901 is a positioning device that receives positioning signals transmitted by GPS satellites and outputs position information indicating the position of the wireless terminal 120. The sensor 902 is a device that detects the movement of the wireless terminal 120, such as an acceleration sensor, an angle sensor, or an IMU (Inertial Measurement Unit).

[0083] (supplement) The wireless quality prediction device 110 and the wireless terminal 120 in this embodiment are not limited to being realized by dedicated devices, but may be realized by a general-purpose computer. In this case, a program for realizing this function may be recorded on a computer-readable recording medium, and the program recorded on this recording medium may be read into a computer system and executed to realize the function. Note that the term "computer system" here includes hardware such as an OS and peripheral devices.

[0084] Additionally, "computer-readable recording media" includes various storage devices such as portable media such as flexible disks, optical magnetic disks, ROMs, and CD-ROMs, as well as hard disks built into computer systems. Furthermore, "computer-readable recording media" may also include devices that dynamically store programs for a short period of time, such as communication lines when transmitting programs via networks such as the Internet or communication lines such as telephone lines, and devices that store programs for a certain period of time, such as volatile memory within computer systems that serve as servers or clients in such cases.

[0085] Furthermore, the above program may be one that realizes part of the above-mentioned functions, or may be one that can realize the above-mentioned functions in combination with a program already recorded in a computer system, or may be one that is realized using hardware such as a PLD (Programmable Logic Device) or FPGA (Field Programmable Gate Array).

[0086] <Effects of the embodiment> According to this embodiment, it is possible to provide a wireless quality prediction system 100 that improves the prediction accuracy of the wireless quality of a wireless terminal in various communication environments. For example, the wireless quality prediction system 100 according to this embodiment can improve the prediction accuracy of the wireless quality by using multiple prediction methods compared to prediction by using only a single prediction method.

[0087] Furthermore, according to the present embodiment, it is possible to predict the risk value of wireless quality degradation with higher accuracy and provide the predicted risk value of wireless quality degradation to the wireless terminal 120. As a result, the wireless terminal 120 can perform control such as switching the wireless base station 101 to which it is connected, controlling communication parameters, or changing the travel route, thereby enabling stable use of wireless communication and applications.

[0088] <Summary of the embodiment> This specification discloses at least the wireless quality prediction system, wireless quality prediction device, wireless quality prediction method, and program described in the following sections. (Section 1) an information acquisition unit configured to acquire information about a wireless terminal performing wireless communication, the information including a terminal location of the wireless terminal; a plurality of prediction units configured to predict, based on information of the wireless terminal, a predicted value of received power of the wireless terminal after a predetermined time has elapsed using different prediction methods; a calculation unit configured to weight the predicted values ​​of the received power predicted by the plurality of prediction units in accordance with a terminal position of the wireless terminal, and calculate predicted data regarding a wireless quality of the wireless terminal after the predetermined time has elapsed; a notification unit configured to notify the wireless terminal of the prediction data; A wireless quality prediction system comprising: (Section 2) 2. The wireless quality prediction system according to claim 1, wherein the prediction data includes a degradation risk value indicating a risk that the wireless quality of the wireless terminal will deteriorate after the predetermined time has elapsed. (Section 3) The calculation unit a learning device that learns weights according to each terminal position using predicted values ​​of the received power predicted by the plurality of prediction units at a plurality of terminal positions and correct values ​​of the received power at the plurality of terminal positions; a terminal position of the wireless terminal and the predicted values ​​of the received power predicted by the plurality of prediction units are input to the learning device to calculate the predicted data; 3. The wireless quality prediction system according to claim 1 or 2. (Section 4) the information about the wireless terminal includes a terminal position of the wireless terminal after the predetermined time has elapsed; the plurality of prediction units include a first prediction unit configured to predict a received power of the wireless terminal after the predetermined time has elapsed, using a terminal position of the wireless terminal after the predetermined time has elapsed; 3. The wireless quality prediction system according to claim 1 or 2. (Section 5) the information about the wireless terminal includes time-series data of the received power of the wireless terminal; the plurality of prediction units include a second prediction unit configured to predict the received power of the wireless terminal after the predetermined time has elapsed, using time-series data of the terminal position of the wireless terminal and the received power; 5. The wireless quality prediction system according to claim 4. (Section 6) an information acquisition unit configured to acquire information about a wireless terminal performing wireless communication, the information including a terminal location of the wireless terminal; a plurality of prediction units configured to predict, based on information of the wireless terminal, a predicted value of received power of the wireless terminal after a predetermined time has elapsed using different prediction methods; a calculation unit configured to weight the predicted values ​​of the received power predicted by the plurality of prediction units in accordance with a terminal position of the wireless terminal, and calculate predicted data regarding a wireless quality of the wireless terminal after the predetermined time has elapsed; a notification unit configured to notify the wireless terminal of the prediction data; A wireless quality prediction device comprising: (Section 7) Wireless quality prediction system A process of acquiring information about a wireless terminal performing wireless communication, including a terminal location of the wireless terminal; a process of predicting a predicted value of received power of the wireless terminal after a predetermined time has elapsed using a plurality of different prediction methods based on information of the wireless terminal; a process of weighting the predicted values ​​of the received power predicted by the plurality of prediction methods according to a terminal position of the wireless terminal, and calculating predicted data regarding the wireless quality of the wireless terminal after the predetermined time has elapsed; a process of notifying the wireless terminal of the prediction data; A wireless quality prediction method for performing the above. (Section 8) 8. A program that causes a computer to execute the wireless quality prediction method according to claim 7.

[0089] Although the present embodiment has been described above, the present invention is not limited to such a specific embodiment, and various modifications and changes are possible within the scope of the gist of the present invention described in the claims. [Explanation of symbols]

[0090] 100 Wireless Quality Prediction System 110 Wireless Quality Prediction Device 111 Information Acquisition Department 112 Prediction Department 112a prediction unit (an example of a first prediction unit) 112b prediction unit (an example of a first prediction unit) 112c prediction unit (second prediction unit) 113 Calculation Unit 114 Notification Department 120 Wireless Terminal 210 Wireless Quality Learning Machine (Learner)

Claims

1. An information acquisition unit configured to acquire information about a wireless terminal, the information including a terminal position of the wireless terminal after a predetermined time has elapsed and time series data of received power of the wireless terminal; a plurality of prediction units including a first prediction unit configured to predict a received power of the wireless terminal after the predetermined time has elapsed, using a terminal position of the wireless terminal after the predetermined time has elapsed, and a second prediction unit configured to predict a predicted value of the received power of the wireless terminal after the predetermined time has elapsed, using time-series data of the position of the wireless terminal and the received power; a calculation unit configured to weight the plurality of prediction units according to a terminal position of the wireless terminal and calculate predicted data regarding a wireless quality of the wireless terminal after the predetermined time has elapsed; a notification unit configured to notify the wireless terminal of the prediction data; A wireless quality prediction system comprising:

2. The wireless quality prediction system according to claim 1 , wherein the prediction data includes a degradation risk value indicating a risk that the wireless quality of the wireless terminal will deteriorate after the predetermined time has elapsed.

3. The calculation unit a learning device that learns weights according to each terminal position using predicted values ​​of the received power predicted by the plurality of prediction units at a plurality of terminal positions and correct values ​​of the received power at the plurality of terminal positions; a terminal position of the wireless terminal and the predicted values ​​of the received power predicted by the plurality of prediction units are input to the learning device to calculate the predicted data; The wireless quality prediction system according to claim 1 or 2.

4. An information acquisition unit configured to acquire information about a wireless terminal, the information including a terminal position of the wireless terminal after a predetermined time has elapsed and time series data of received power of the wireless terminal; a plurality of prediction units including a first prediction unit configured to predict a received power of the wireless terminal after the predetermined time has elapsed, using a terminal position of the wireless terminal after the predetermined time has elapsed, and a second prediction unit configured to predict a predicted value of the received power of the wireless terminal after the predetermined time has elapsed, using time-series data of the position of the wireless terminal and the received power; a calculation unit configured to weight the plurality of prediction units according to a terminal position of the wireless terminal and calculate predicted data regarding a wireless quality of the wireless terminal after the predetermined time has elapsed; a notification unit configured to notify the wireless terminal of the prediction data; A wireless quality prediction device comprising:

5. Wireless quality prediction system A process of acquiring information about the wireless terminal including a terminal position of the wireless terminal after a predetermined time has elapsed and time-series data of the received power of the wireless terminal; a process of predicting a predicted value of the received power of the wireless terminal after the predetermined time has elapsed using a plurality of prediction units, the prediction unit including: a first prediction unit configured to predict a received power of the wireless terminal after the predetermined time has elapsed using a terminal position of the wireless terminal after the predetermined time has elapsed; and a second prediction unit configured to predict a predicted value of the received power of the wireless terminal after the predetermined time has elapsed using time series data of the position of the wireless terminal and the received power; a process of weighting the plurality of prediction units according to a terminal location of the wireless terminal and calculating prediction data regarding wireless quality of the wireless terminal after the predetermined time has elapsed; a process of notifying the wireless terminal of the prediction data; A wireless quality prediction method for performing the above.

6. A program that causes a computer to execute the wireless quality prediction method according to claim 5.

Images