Communication quality estimation device, machine learning method, communication quality estimation method, and program

Abstract

The purpose of the present disclosure is to estimate, with as high accuracy as possible, whether it is possible to continuously satisfy desired communication quality, even in the case of wireless communication by a mobile entity of which the position changes.　To achieve this purpose, the present disclosure relates to a communication quality estimation device that learns the communication quality of wireless communication corresponding to the position of a mobile entity through machine learning in a learning phase, wherein the communication quality estimation device has an acquisition unit that acquires prescribed position information indicating the position of the mobile entity and information pertaining to communication quality as a measured value of the wireless communication at the aforementioned position, a section estimation processing unit that performs section estimation processing on the communication quality to thereby calculate an upper-limit value that represents the communication quality on the higher-order side of the section estimation and a lower-limit value that represents the communication quality on the lower-order side of the section estimation, and a machine learning unit that trains a first machine learning model through machine learning using the prescribed position information and the upper-limit value and trains a second machine learning model through machine learning using the prescribed position information and the lower-limit value.

Description

[Technical Field]

[0001] The present disclosure relates to a technology for performing machine learning on communication quality of wireless communication according to past positions of a mobile body and estimating communication quality of wireless communication according to future destinations of the mobile body. [Background technology]

[0002] In applications such as remote control or remote monitoring of vehicles, robots, and other mobile objects, it is common to perform remote communication via wireless communication between a server at a center and a communication terminal installed on the mobile object at the site. Furthermore, since mobile objects are on the move, high availability of wireless communication quality is required for remote communication, taking safety into consideration. It is considered effective to estimate (predict) whether stable wireless communication quality will be available at the mobile object's destination, and if there is a risk, to take measures such as slowing down the moving speed in advance or reducing the video transmission rate to prevent video interruptions. Therefore, it is important to estimate the communication quality at the destination.

[0003] As an example of estimating wireless communication quality at a destination, there is a technology that estimates communication quality at a destination in advance based on a history that indicates the relationship between the past positions of a mobile body (communication equipment mounted on a mobile body) and communication quality (see Non-Patent Document 1). Non-Patent Document 1 proposes a technology that uses past measured values ​​(actual values) of communication quality to estimate communication quality of throughput as a median or average value by machine learning or the like. [Prior art documents] [Non-patent literature]

[0004] [Non-Patent Document 1] Kawamura, et al., "Optimal Selection of Multi-Wireless Networks for Autonomous Agricultural Machinery Driving Based on Prediction of Wireless Communication Quality," IEICE General Conference, B-6-3, 2021. Summary of the Invention [Problem to be solved by the invention]

[0005] However, in wireless communication between mobile bodies whose locations change, there may be large changes in the radio wave environment and fluctuations in communication quality, so there is a problem that the conventional method of estimating communication quality as a median or average value cannot accurately estimate whether the desired communication quality can be continuously achieved.

[0006] The present invention has been made in consideration of the above points, and aims to estimate as accurately as possible whether a desired communication quality can be continuously achieved, even in the case of wireless communication by a mobile body whose position changes. [Means for solving the problem]

[0007] In order to solve the above problem, the invention of claim 1 is a communication quality estimation device that, in a learning phase, machine-learns the communication quality of wireless communication according to the position of a mobile body, and includes: an acquisition unit that acquires predetermined location information indicating the position of the mobile body and information on communication quality as a measured value of the wireless communication at that position; a section estimation processing unit that performs section estimation processing on the communication quality to calculate an upper limit value that is a communication quality value on the upper side of the section estimation and a lower limit value that is a communication quality value on the lower side of the section estimation; and a machine learning unit that trains a first machine learning model using the predetermined location information and the upper limit value, and trains a second machine learning model using the predetermined location information and the lower limit value. [Effects of the Invention]

[0008] As described above, according to the present invention, it is possible to estimate as accurately as possible whether the desired communication quality can be continuously achieved, even in the case of wireless communication by a mobile body whose position changes. [Brief explanation of the drawings]

[0009] [Figure 1] 1 is a diagram illustrating the overall configuration of a communication system according to an embodiment of the present invention. [Figure 2]1 is a diagram illustrating an electrical hardware configuration of a communication quality estimation device according to an embodiment of the present invention. [Figure 3] FIG. 2 is a diagram illustrating an electrical hardware configuration of a communication terminal according to the present embodiment. [Figure 4] FIG. 2 is a functional configuration diagram of a communication quality estimation device in a learning phase. [Figure 5] FIG. 2 is a functional configuration diagram of a communication quality estimation device in an estimation phase. [Figure 6] 10 is a flowchart showing a machine learning method executed by the communication quality estimation device in a learning phase. [Figure 7] 10 is a flowchart showing a machine learning method executed by the communication quality estimation device in a learning phase. [Figure 8] 10 is a flowchart showing a communication quality estimation method executed by a communication quality estimation device in an estimation phase. DETAILED DESCRIPTION OF THE INVENTION

[0010] Hereinafter, an embodiment of the present invention will be described with reference to the drawings.

[0011] [System configuration of the embodiment] First, the overall configuration of the communication system of this embodiment will be explained with reference to Fig. 1. Fig. 1 is a diagram showing the overall configuration of the communication system according to this embodiment.

[0012] As shown in Fig. 1, a communication system 1 of this embodiment is constructed by a communication quality estimation device 3 and a communication terminal 5. The communication terminal 5 is managed and used by a user. The user refers to the output result of the communication quality estimation device and determines how to respond thereafter.

[0013] Furthermore, the communication quality estimation device 3 and the communication terminal 5 can communicate with each other via a communication network 100 such as the Internet. The communication network 100 may be connected wirelessly or by wire.

[0014] The communication quality estimation device 3 is configured by one or more computers. When the communication quality estimation device 3 is configured by multiple computers, it may be referred to as a "communication quality estimation device" or a "communication quality estimation system."

[0015] The communication quality estimation device 3 estimates communication quality according to the location (geographical coordinates) of a communication device of a mobile body by performing interval estimation based on past location information and measurement values ​​of communication quality at the location indicated by this location information, thereby learning the variance (upper and lower limits) of communication quality by machine learning, and estimating communication quality in advance based on the location of the mobile body's destination. The mobile body is, for example, a vehicle, an aircraft, a ship, or the like.

[0016] The communication terminal 5 is a computer, and a notebook computer is shown as an example in Fig. 1. In Fig. 1, a user operates the communication terminal 5. Note that the communication quality estimation device 3 may perform processing independently without using the communication terminal 5.

[0017] [Hardware configuration] <Hardware configuration of the communication quality estimation device> Next, the electrical hardware configuration of the communication quality estimation device 3 will be described with reference to Fig. 2. Fig. 2 is a diagram showing the electrical hardware configuration of the communication quality estimation device.

[0018] As shown in FIG. 2, the communication quality estimation device 3 is a computer that includes a CPU (Central Processing Unit) 301, a ROM (Read Only Memory) 302, a RAM (Random Access Memory) 303, an SSD (Solid State Drive) 304, an external device connection I / F (Interface) 305, a network I / F 306, a media I / F 309, and a bus line 310.

[0019] Of these, the CPU 301 controls the overall operation of the communication quality estimation device 3. The ROM 302 stores programs such as an IPL (Initial Program Loader) used to drive the CPU 301. The RAM 303 is used as a work area for the CPU 301.

[0020] The SSD 304 reads or writes various data under the control of the CPU 301. Note that instead of the SSD 304, a hard disk drive (HDD) may be used.

[0021] The external device connection I / F 305 is an interface for connecting various external devices, such as a display, a speaker, a keyboard, a mouse, a USB (Universal Serial Bus) memory, and a printer.

[0022] The network I / F 306 is an interface for performing data communication via the communication network 100 .

[0023] The media I / F 309 controls reading and writing (storing) of data from and to a recording medium 309m such as a flash memory, etc. The recording medium 309m includes a DVD (Digital Versatile Disc) and a Blu-ray Disc (registered trademark).

[0024] The bus line 310 is an address bus, a data bus, etc. for electrically connecting the components such as the CPU 301 shown in FIG.

[0025] <Hardware configuration of communication terminal> Next, the electrical hardware configuration of the communication terminal 5 will be described with reference to Fig. 3. Fig. 3 is a diagram showing the electrical hardware configuration of the communication terminal.

[0026] As shown in FIG. 3, the communication terminal 5 is a computer and includes a CPU 501, a ROM 502, a RAM 503, an SSD 504, an external device connection I / F (Interface) 505, a network I / F 506, a display 507, a pointing device 508, a media I / F 509, and a bus line 510.

[0027] Of these, the CPU 501 controls the overall operation of the communication terminal 5. The ROM 502 stores programs such as IPL used to drive the CPU 501. The RAM 503 is used as a work area for the CPU 501.

[0028] The SSD 504 reads or writes various data under the control of the CPU 501. Note that instead of the SSD 504, an HDD (Hard Disk Drive) may be used.

[0029] The external device connection I / F 505 is an interface for connecting various external devices, such as a display, a speaker, a keyboard, a mouse, a USB memory, and a printer.

[0030] The network I / F 506 is an interface for performing data communication via the communication network 100 .

[0031] The display 507 is a type of display means such as a liquid crystal display or organic EL (Electro Luminescence) display that displays various images.

[0032] The pointing device 508 is a type of input means for selecting and executing various instructions, selecting a processing target, moving a cursor, etc. If the user uses a keyboard, the function of the pointing device 508 may be turned off.

[0033] The media I / F 509 controls reading and writing (storing) of data from and to a recording medium 509m such as a flash memory, etc. The recording medium 509m includes DVDs, Blu-ray Discs (registered trademarks), etc.

[0034] The bus line 510 is an address bus, a data bus, or the like for electrically connecting the components such as the CPU 501 shown in FIG.

[0035] [Functional configuration of the communication quality estimation device] The functional configuration of the communication quality estimation device 3 according to this embodiment in the learning phase and estimation (prediction) phase will be described.

[0036] <Learning phase functional configuration> First, each function of the communication quality estimation device 3 in the learning phase will be described with reference to Fig. 4. Fig. 4 is a functional configuration diagram of the communication quality estimation device in the learning phase.

[0037] As shown in Fig. 4, the communication quality estimation device 3 includes an acquisition unit 31, an interval estimation processing unit 33, and a machine learning unit 35. These units are functions realized by instructions from the CPU 301 in Fig. 2 based on a program. In addition, a communication quality DB 40 and machine learning models 41a and 42a are built in the RAM 303 or the SSD 304.

[0038] The acquisition unit 31 acquires learning data (location information, measurement values ​​of communication quality) from the communication terminal 7 and stores the data in the communication quality DB 40. The acquisition unit 31 also serves as an input unit for inputting learning data for machine learning.

[0039] "Location information" is information indicating the past geographical location (longitude, latitude) of a mobile object (communication equipment mounted on the mobile object).

[0040] "Measured communication quality" refers to the measured value (actual value) of each element of the communication quality of wireless communication of a mobile unit paired with the location information of the mobile unit in one training data. Examples of each element include uplink throughput, downlink throughput, delay time, jitter, packet loss, received power, etc.

[0041] The interval estimation process 33 reads learning data from the communication quality DB 40 periodically or at any timing, and performs interval estimation processing on the measurement values ​​of communication quality included in this learning data. For example, the interval estimation process 33 calculates an upper limit value (a value of communication quality on the upper side) p1 and a lower limit value (a value of communication quality on the lower side) p2 of the interval estimation in a 95% confidence interval based on the measurement values ​​of communication quality associated with other location information. This processing will be described in detail later. Note that the 95% parameter is changeable.

[0042] The machine learning unit 35 trains the machine learning model 41a based on the location information and the upper limit value p1 calculated by the interval estimation processing unit 33. The machine learning unit 35 also trains the machine learning model 42a based on the same location information and the lower limit value p2 calculated by the interval estimation processing unit 33. The machine learning models 41a and 42a perform machine learning using a known machine learning algorithm such as a neural network.

[0043] This completes the description of the functional configuration of the learning phase.

[0044] <Functional configuration of the estimation phase> Next, a description will be given of each function of the communication quality estimation device 3 in the estimation phase. Fig. 5 is a functional configuration diagram of the communication quality estimation device in the estimation phase.

[0045] As shown in Fig. 5, the communication quality estimation device 3 includes an acquisition unit 31, an estimation unit 37, and an output unit 39. These units each have a function realized by an instruction from the CPU 301 in Fig. 2 based on a program. A trained machine learning model 30b is stored in the RAM 303 or the SSD 304. Note that functional components similar to those in the learning phase are denoted by the same reference numerals, and description thereof will be omitted.

[0046] The estimation unit 37 uses the trained machine learning model 41b to output an estimated value e1, which is the upper limit value of each element of communication quality, based on the specific location information acquired by the acquisition unit 31. Furthermore, the estimation unit 37 uses the trained machine learning model 42b to output an estimated value e2, which is the lower limit value of each element of communication quality, based on the specific location information acquired by the acquisition unit 31.

[0047] The output unit 39 outputs the estimation result from the communication quality estimation device 3 based on the estimated values ​​e1 and e2 output by the estimation unit 37. Examples of output include displaying the result on a display connected to the external device connection I / F 305 in Fig. 2, transmitting the result to the communication terminal 5 or the like via the network I / F 306, and the like.

[0048] [Processing or Operation of the Communication Quality Estimation Device] Next, the processing or operation of the communication quality estimation device 3 in the learning phase and estimation phase will be described with reference to FIGS.

[0049] <Processing or operation in the learning phase> 6 and 7 are flowcharts showing the machine learning method executed by the communication quality estimation device in the learning phase.

[0050] S11: The acquisition unit 31 acquires location information and measurement values ​​of communication quality as learning data from the communication terminal 7, and stores them in the communication quality DB 40.

[0051] S12: The section estimation processing unit 33 reads out, from the communication quality DB 40, predetermined location information and a measurement value of communication quality associated with the predetermined location information.

[0052] S13: The section estimation processing unit 33 reads out each measurement value of communication quality (referred to as a "measurement value (actual value) group Dp") stored in association with location information relating to a position within a predetermined distance range (for example, a radius L [m]) from a position (point) relating to the predetermined location information. Note that this measurement value group Dp is a group of measurement values ​​indicated for each element of communication quality, such as throughput.

[0053] S14: The section estimation processing unit 33 performs section estimation processing for each element of communication quality based on the measurement value of communication quality at the position indicated by the predetermined position information and the measurement value group Dp.

[0054] Here, the interval estimation processing unit 33 calculates a 100(1-α)% confidence interval as shown below.

[0055]

number

[0056]

number

[0057] S16: The section estimation processing unit 33 determines whether the processes (S12 to S15) have been completed for all of the location information and related communication quality measurement values ​​stored in the communication quality DB 40. If the processes (S12 to S15) have not been completed for all of the location information and related communication quality measurement values ​​(S16; NO), the process returns to the above-mentioned process S12. On the other hand, if the processes (S12 to S15) have been completed for all of the location information and related communication quality measurement values ​​(S16; YES), the process proceeds to process S17.

[0058] S17: The machine learning unit 35 reads the upper limit p1 and the location information associated with this upper limit p1 from the communication quality DB 40, and trains the machine learning model 41a based on this upper limit p1 and the location information.

[0059] S18: The machine learning unit 35 reads the lower limit p2 and the location information associated with this lower limit p2 (the same location information as in S17) from the communication quality DB 40, and trains the machine learning model 42a based on this lower limit p2 and the location information.

[0060] This completes the description of the processing or operation in the learning phase.

[0061] <Processing or operation in the estimation phase> FIG. 8 is a flowchart showing a communication quality estimation method executed by the communication quality estimation device in the estimation phase.

[0062] S21: The acquisition unit 31 acquires specific location information as input data based on a direct input from the communication terminal 5 or to the own device (communication quality estimation device 3).

[0063] S22: The estimation unit 37 uses the trained machine learning model 41b to output an estimated value e1 as an upper limit value of each element of communication quality based on specific location information.

[0064] S23: The estimation unit 37 uses the trained machine learning model 42b to output an estimated value e2 as the lower limit of each element of communication quality based on the specific location information.

[0065] S24: The output unit 39 outputs information about the estimation result.

[0066] This completes the description of the processing or operation in the estimation phase.

[0067] [Effects of the embodiment] As described above, according to this embodiment, it is possible to estimate the range of change in communication quality of wireless communication, and to determine the use of wireless access taking into account the lower and upper limits of quality estimation based on location information, etc., and the probability of that estimation. This has the effect of making it possible to estimate as accurately as possible whether a desired communication quality can be continuously achieved, even in the case of wireless communication using communication equipment of a mobile body whose location changes.

[0068] 〔supplement〕 The present invention is not limited to the above-described embodiment, and may have the following configurations or processes (operations). (1) The communication quality estimation device 3 can be realized by a computer and a program, but this program can also be recorded on a (non-transitory) recording medium or provided via the communication network 100. (2) In the above embodiment, a notebook computer is shown as an example of a communication terminal 5, but this is not limited to this and may be, for example, a desktop computer, a tablet terminal, a smartphone, a smartwatch, a car navigation device, a refrigerator, a microwave oven, etc. (3) Each CPU 301, 501 may be a single CPU or may be multiple CPUs. [Explanation of symbols]

[0069] 1. Communication Systems 3. Communication quality estimation device 5. Communication terminals 31 Acquisition section (input section) 33 Section estimation processing unit 35 Machine Learning Department 37 Estimation part 39 Output section 40 Communication Quality DB 41a Machine learning model (an example of the first machine learning model) 41b Trained machine learning model (an example of the first trained machine learning model) 42a Machine learning model (an example of a second machine learning model) 42b Trained machine learning model (an example of a second trained machine learning model)

Claims

1. A communication quality estimation device that performs machine learning to estimate communication quality of wireless communication according to a position of a mobile object in a learning phase, an acquisition unit that acquires predetermined location information indicating a location of the mobile object and information on communication quality as a measurement value of the wireless communication at the location; a section estimation processing unit that performs section estimation processing on the communication quality to calculate an upper limit value that is a value of communication quality on an upper side of the section estimation and a lower limit value that is a value of communication quality on a lower side of the section estimation; a machine learning unit that trains a first machine learning model using the predetermined location information and the upper limit value, and trains a second machine learning model using the predetermined location information and the lower limit value; A communication quality estimation device having the above configuration.

2. 2. The communication quality estimation device according to claim 1, wherein the section estimation processing unit performs the section estimation process for communication qualities associated with the predetermined location information and location information associated with locations within a predetermined distance from the location indicated by the predetermined location information.

3. the section estimation processing unit performs the section estimation process for each element of the communication quality; the machine learning unit trains a first machine learning model using the predetermined location information and the upper limit value for each element, and trains a second machine learning model using the predetermined location information and the lower limit value for each element; The communication quality estimation device according to claim 1 or 2.

4. The communication quality estimation device according to claim 3 , wherein the element of the communication quality is an upstream throughput, a downstream throughput, a delay time, a jitter, a packet loss, or a received power.

5. A machine learning method executed by a communication quality estimation device that performs machine learning to estimate communication quality of wireless communication according to a position of a mobile object in a learning phase, comprising: an acquisition process for acquiring predetermined location information indicating the location of the mobile object and information on communication quality as a measurement value of the wireless communication at the location; a section estimation process for calculating an upper limit value, which is a value of communication quality on an upper side of the section estimation, and a lower limit value, which is a value of communication quality on a lower side of the section estimation, by performing a section estimation process on the communication quality; a machine learning process for training a first machine learning model using the predetermined location information and the upper limit value, and training a second machine learning model using the predetermined location information and the lower limit value; Machine learning methods that perform

6. A communication quality estimation device that estimates communication quality of wireless communication at a specific position using a trained machine learning model in an estimation phase, an acquisition unit that acquires specific location information indicating the specific location; an estimation unit that uses a first trained machine learning model to output a first estimated value as an upper limit value that is a communication quality value on the upper side of section estimation performed on the communication quality based on the specific location information, and that uses a second trained machine learning model to output a second estimated value as a lower limit value that is a communication quality value on the lower side of section estimation performed on the communication quality based on the specific location information; an output unit that outputs information of the estimation result indicating the first estimated value and the second estimated value; A communication quality estimation device having the above configuration.

7. A communication quality estimation method executed by a communication quality estimation device that estimates communication quality of wireless communication at a specific position using a trained machine learning model in an estimation phase, comprising: The communication quality estimation device an acquisition process for acquiring specific location information indicating the specific location; an estimation process using a first trained machine learning model to output a first estimated value as an upper limit value that is a communication quality value on the upper side of the section estimation performed on the communication quality based on the specific location information, and using a second trained machine learning model to output a second estimated value as a lower limit value that is a communication quality value on the lower side of the section estimation performed on the communication quality based on the specific location information; an output process for outputting information on the estimation result indicating the first estimated value and the second estimated value; A communication quality estimation method that performs the above.

8. A program that causes a computer to execute the method according to claim 5 or 7.

Images