Radio quality estimation method and program

A hybrid method combining time-series and spatial estimation enhances wireless quality prediction accuracy by using machine learning to adaptively select estimates based on AP switching times, addressing accuracy issues in existing Wi-Fi prediction systems.

WO2026239612A1PCT designated stage Publication Date: 2026-11-19NT T INC
View PDF 0 Cites 0 Cited by

Patent Information

Application Number
PCT/JP2025/017917
Authority / Receiving Office
WO · WO
Patent Type
Applications
Current Assignee / Owner
Filing Date
2025-05-16
Publication Date
2026-11-19

AI Technical Summary

Technical Problem

Existing wireless quality prediction methods, particularly for Wi-Fi, suffer from degraded accuracy during AP switching due to reliance on time-series data from previously connected access points.

Method used

Implement a hybrid approach combining time-series prediction and spatial estimation, using machine learning models to calculate future wireless quality based on current throughput and terminal location, and selectively outputting estimates from either method based on the time since AP switching.

Benefits of technology

Improves wireless quality estimation accuracy by leveraging spatial estimation during AP switching periods, ensuring more reliable predictions.

✦ Generated by Eureka AI based on patent content.

Smart Images

  • Figure JP2025017917_19112026_PF_FP_ABST
    Figure JP2025017917_19112026_PF_FP_ABST
Patent Text Reader

Abstract

The present invention improves estimation accuracy of the radio quality by causing a computer to execute: a time series prediction procedure for calculating an estimation value of future radio quality on the basis of time series data about radio quality of a terminal; a spatial estimation procedure for calculating an estimation value of the radio quality on the basis of a future location of the terminal at a certain point in time; and a selection procedure for selecting, as an output target, the estimation value calculated by the spatial estimation procedure when the certain point in time is within a prescribed period after the terminal switches an access point, and selecting, as an output time, the estimation value calculated by the time series prediction procedure when the certain point in time is not within the prescribed period.
Need to check novelty before this filing date? Find Prior Art

Description

Wireless quality estimation method and program

[0001] The present invention relates to a wireless quality estimation method and a program.

[0002] Conventionally, there is a technology for predicting the future wireless quality of a terminal through time-series prediction. For example, a method is being studied in which time-series data such as received power observed by a terminal is learned using RNN, which is one of machine learning methods, to predict the received power several seconds later.

[0003] In time-series prediction, future prediction is performed based on the latest trend of wireless quality, so real-time prediction is possible regardless of the wireless scheme.

[0004] Wakao et al., "Quality Prediction Technology for Optimal Use of Multiple Wireless Accesses", NTT Technical Journal, April 2020

[0005] However, in the case of Wi-Fi (registered trademark), for example, at the moment when the access point (AP) is switched by roaming (switching of the connection destination AP), the wireless quality of the currently connected AP is predicted from the time-series data of the previously connected AP, which poses a problem of degraded accuracy.

[0006] The present invention has been made in view of the above points, and aims to improve the estimation accuracy of wireless quality.

[0007] Therefore, in order to solve the above problem, a computer executes: a time-series prediction procedure for calculating an estimated value of future wireless quality based on time-series data of the wireless quality of a terminal at a certain point in time; a spatial estimation procedure for calculating an estimated value of wireless quality based on the future position of the terminal at the certain point in time; and a selection procedure of selecting the estimated value calculated by the spatial estimation procedure as an output target when the certain point in time is within a predetermined period after the terminal switches access points, and selecting the estimated value calculated by the time-series prediction procedure as an output target when the certain point in time is not within the predetermined period.

[0008] The estimation accuracy of wireless quality can be improved.

[0009] This figure shows an example configuration of the wireless quality estimation system in an embodiment of the present invention. This figure shows an example hardware configuration of the terminal 10 in an embodiment of the present invention. This figure shows an example functional configuration of the wireless quality estimation system in an embodiment of the present invention. This figure is for explaining the spatial estimation unit 21. This figure is for explaining the time series prediction unit 11. This is a flowchart for explaining an example of a processing procedure executed by the terminal 10. This figure shows a second example functional configuration of the wireless quality estimation system in an embodiment of the present invention.

[0010] Embodiments of the present invention will be described below with reference to the drawings. Figure 1 is a diagram showing an example of the configuration of a wireless quality estimation system in an embodiment of the present invention.

[0011] In Figure 1, terminal 10 is a communication terminal capable of wireless communication using a wireless LAN such as Wi-Fi (registered trademark). For example, a smartphone, tablet device, PC (Personal Computer), or a communication terminal mounted on a mobile device such as a vehicle or robot may be used as terminal 10.

[0012] The spatial estimation device 20 is one or more computers capable of predicting the wireless quality of terminal 10 based on the terminal's location (future location). In this embodiment, the case in which throughput is used as the wireless quality is described, but other indicators may be used as the wireless quality. Terminal 10 can connect to the spatial estimation device 20 via wireless communication or the like. Although only one terminal 10 is shown in Figure 1 for convenience, the spatial estimation device 20 can predict the wireless quality for multiple terminals.

[0013] Figure 2 shows an example of the hardware configuration of terminal 10 in an embodiment of the present invention. Terminal 10 includes a drive device 100, an auxiliary storage device 102, a memory device 103, a processor 104, and an interface device 105, etc., which are all interconnected by bus B.

[0014] The program that enables processing on terminal 10 is provided on a recording medium 101 such as a CD-ROM. When the recording medium 101 containing the program is set in the drive device 100, the program is installed from the recording medium 101 to the auxiliary storage device 102 via the drive device 100. However, the program does not necessarily have to be installed from the recording medium 101; it may also be downloaded from another computer via a network. The auxiliary storage device 102 stores the installed program as well as necessary files and data.

[0015] The memory device 103 reads and stores a program from the auxiliary storage device 102 when a program startup command is received. The processor 104 is either a CPU or a GPU (Graphics Processing Unit), or both a CPU and a GPU, and executes functions related to the terminal 10 according to the program stored in the memory device 103. The interface device 105 is used as an interface for connecting to a wireless network.

[0016] The spatial estimation device 20 also has the same hardware configuration as shown in Figure 2.

[0017] Figure 3 shows an example of the functional configuration of a wireless quality estimation system in an embodiment of the present invention. In Figure 3, the spatial estimation device 20 has a spatial estimation unit 21. The spatial estimation unit 21 is realized by processing that one or more programs installed in the spatial estimation device 20 cause the processor of the spatial estimation device 20 to execute.

[0018] The spatial estimation unit 21 calculates an estimated throughput of terminal 10 based on the future position of terminal 10 at a given time. The throughput of terminal 10 refers to the throughput of wireless communication that terminal 10 is able to perform.

[0019] Figure 4 is a diagram illustrating the spatial estimation unit 21. The spatial estimation unit 21 can be implemented using a machine learning model (e.g., a neural network) that has learned the correspondence between location information and throughput.

[0020] Alternatively, the spatial estimation device 20 may store a history of multiple locations and the throughput at each location for the past few days. In this case, the spatial estimation unit 21 may output the average throughput for the past few days at the location related to the input location information as the estimated throughput for that location.

[0021] Furthermore, if the correspondence between position and throughput differs depending on the wireless communication method, the spatial estimation device 20 may be provided for each wireless communication method.

[0022] Terminal 10 includes a time-series prediction unit 11, a communication unit 12, a selection unit 13, and an output unit 14. Each of these units is realized by processing that one or more programs installed on terminal 10 cause the processor 104 to execute.

[0023] The time series prediction unit 11 calculates an estimated throughput for the future (x seconds after a certain point in time) based on the time series data of the terminal's wireless quality at a given point in time.

[0024] Figure 5 is a diagram illustrating the time series forecasting unit 11. The time series forecasting unit 11 can be implemented using a machine learning model (for example, a neural network such as an RNN (Recurrent Neural Network)) that has already learned the correspondence between past throughput time series data and future throughput (x seconds later). In other words, training data for the machine learning model of the time series forecasting unit 11 is used, which consists of a set of throughput time series data and the throughput x seconds later of that time series data.

[0025] The communication unit 12 controls communication via a wireless LAN (Local Area Network) such as Wi-Fi (registered trademark). When the terminal 10 performs roaming (switching to a different access point), the communication unit 12 detects the roaming and notifies the selection unit 13 that roaming has occurred.

[0026] The selection unit 13 selects the estimated throughput to be output from among the estimated throughput by the time series forecasting unit 11 and the estimated throughput by the spatial estimation unit 21 at a given point in time. If the point in time is within a predetermined period since the terminal 10 started roaming, the selection unit 13 selects the estimated throughput calculated by the spatial estimation unit 21 as the output target. If the point in time is not within the predetermined period, the selection unit 13 selects the estimated throughput calculated by the time series forecasting unit 11 as the output target.

[0027] The output unit 14 outputs the estimated throughput selected by the selection unit 13.

[0028] The estimated throughput output from the output unit 14 can be used for controlling a mobile device equipped with the terminal 10. For example, if the mobile device is a robot, and it is predicted that the estimated throughput after x seconds (for example, 5 seconds) will deteriorate to a level where communication is impossible, this can be used as a trigger for instructions to change the route or control the speed, thereby preventing an uncontrollable state.

[0029] The following describes the processing steps performed by terminal 10. Figure 6 is a flowchart illustrating an example of the processing steps performed by terminal 10. The processing steps in Figure 6 are performed periodically (for example, at 1-second intervals).

[0030] In step S101, the time series prediction unit 11 calculates the estimated throughput x seconds later based on the time series data of throughput measured at terminal 10 over a recent period (for example, the past few seconds). The calculated estimated throughput is input to the selection unit 13.

[0031] Next, the selection unit 13 estimates the position of the terminal 10 after x seconds and obtains the estimated throughput for the estimated position information by querying the spatial estimation unit 21 (S102). The value of x is assumed to be known to the selection unit 13. The position of the terminal 10 after x seconds may also be estimated based on control information including the movement path of the mobile body on which the terminal 10 is mounted, for example. In this case, the selection unit 13 transmits the position information of the terminal 10 after x seconds to the spatial estimation unit 21. The spatial estimation unit 21 receives this position information and calculates the estimated throughput. If the spatial estimation device 20 has a spatial estimation unit 21 for each wireless system, the selection unit 13 queries the spatial estimation unit 21 corresponding to the wireless system used by the terminal 10 for the estimated throughput for the estimated position information.

[0032] Next, the selection unit 13 determines whether the elapsed time from the last notification of roaming implementation to the present time is less than or equal to a predetermined time (S103). If the elapsed time is less than or equal to the predetermined time (Yes in S103), the selection unit 13 selects the estimated throughput obtained from the spatial estimation unit 21 as the output target (S104). On the other hand, if the elapsed time exceeds the predetermined time (No in S103), the selection unit 13 selects the estimated throughput obtained from the time series prediction unit 11 as the output target (S105).

[0033] Next, the output unit 14 outputs the estimated throughput selected by the selection unit 13 (S106).

[0034] In the above example, the spatial estimation device 20 is shown to have a spatial estimation unit 21, but the terminal 10 may also have a spatial estimation unit 21.

[0035] Figure 7 shows a second example of the functional configuration of a wireless quality estimation system according to an embodiment of the present invention. In Figure 7, the same parts as in Figure 3 are denoted by the same reference numerals, and their descriptions are omitted as appropriate. In Figure 7, the terminal 10 has a spatial estimation unit 21.

[0036] As described above, according to this embodiment, during a predetermined period from roaming when the accuracy of estimating wireless quality based on time-series data deteriorates, the estimated value of wireless quality based on location information is adopted (selected). Therefore, the accuracy of estimating wireless quality can be improved.

[0037] The functions of the elements disclosed herein may be implemented using circuits or processing circuitry that include general-purpose processors, special-purpose processors, integrated circuits, ASICs (Application Specific Integrated Circuits), FPGAs (Field Programmable Gate Arrays), conventional circuits, and / or combinations thereof that are programmed using one or more programs stored in one or more memories, or otherwise configured to perform the disclosed functions. A processor is considered processing circuitry or circuitry because it includes transistors and other circuits. A processor may be a programmed processor that executes programs stored in memory. In this disclosure, a circuit, unit, or means is hardware that performs the enumerated functions, or hardware programmed to perform the enumerated functions. Hardware may be any hardware disclosed herein that is programmed or configured to perform the enumerated functions.

[0038] There is a memory for storing a computer program that includes computer instructions. These computer instructions provide logic and routines that enable hardware (e.g., processing circuitry or circuitry) to perform the methods disclosed herein. This computer program can be implemented in commonly known forms, such as computer-readable storage media, computer program products, memory devices, recording media such as CD-ROMs and DVDs, and / or memory in FPGAs and ASICs.

[0039] Although embodiments of the present invention have been described in detail above, the present invention is not limited to these specific embodiments, and various modifications and changes are possible within the scope of the gist of the present invention as described in the claims.

[0040] 10 Terminal 11 Time series prediction unit 12 Communication unit 13 Selection unit 14 Output unit 20 Spatial estimation device 21 Spatial estimation unit 100 Drive device 101 Recording medium 102 Auxiliary storage device 103 Memory device 104 Processor 105 Interface device B Bus

Claims

1. A wireless quality estimation method characterized in that a computer executes: a time-series prediction procedure for calculating an estimated value of future wireless quality based on time-series data of the wireless quality of a terminal at a certain point in time; a spatial estimation procedure for calculating an estimated value of wireless quality based on the future location of the terminal at the aforementioned point in time; and a selection procedure for selecting the estimated value calculated by the spatial estimation procedure as the output target if the aforementioned point in time is within a predetermined period since the terminal switched access points, and selecting the estimated value calculated by the time-series prediction procedure as the output target if the aforementioned point in time is not within the predetermined period.

2. The wireless quality estimation method according to claim 1, characterized in that the wireless quality is throughput.

3. A program for causing a computer to execute: a time-series prediction procedure for calculating an estimated value of future wireless quality based on time-series data of the wireless quality of a terminal at a certain point in time; a spatial estimation procedure for calculating an estimated value of wireless quality based on the future location of the terminal at the aforementioned point in time; and a selection procedure for selecting the estimated value calculated by the spatial estimation procedure as the output target if the aforementioned point in time is within a predetermined period since the terminal switched access points, and selecting the estimated value calculated by the time-series prediction procedure as the output target if the aforementioned point in time is not within the predetermined period.