WO2024100881A1 - 通信状態推定装置、通信状態推定システム、通信状態推定方法、及び通信状態推定プログラム - Google Patents

通信状態推定装置、通信状態推定システム、通信状態推定方法、及び通信状態推定プログラム Download PDF

Info

Publication number
WO2024100881A1
WO2024100881A1 PCT/JP2022/042068 JP2022042068W WO2024100881A1 WO 2024100881 A1 WO2024100881 A1 WO 2024100881A1 JP 2022042068 W JP2022042068 W JP 2022042068W WO 2024100881 A1 WO2024100881 A1 WO 2024100881A1
Authority
WO
WIPO (PCT)
Prior art keywords
communication
user
motion information
communication quality
communication state
Prior art date
Legal status (The legal status is an assumption and is not a legal conclusion. Google has not performed a legal analysis and makes no representation as to the accuracy of the status listed.)
Ceased
Application number
PCT/JP2022/042068
Other languages
English (en)
French (fr)
Japanese (ja)
Inventor
尚志 永田
理一 工藤
馨子 高橋
智明 小川
Current Assignee (The listed assignees may be inaccurate. Google has not performed a legal analysis and makes no representation or warranty as to the accuracy of the list.)
NTT Inc
Original Assignee
Nippon Telegraph and Telephone Corp
Priority date (The priority date is an assumption and is not a legal conclusion. Google has not performed a legal analysis and makes no representation as to the accuracy of the date listed.)
Filing date
Publication date
Application filed by Nippon Telegraph and Telephone Corp filed Critical Nippon Telegraph and Telephone Corp
Priority to PCT/JP2022/042068 priority Critical patent/WO2024100881A1/ja
Priority to JP2024556986A priority patent/JPWO2024100881A1/ja
Publication of WO2024100881A1 publication Critical patent/WO2024100881A1/ja
Anticipated expiration legal-status Critical
Ceased legal-status Critical Current

Links

Images

Classifications

    • H—ELECTRICITY
    • H04—ELECTRIC COMMUNICATION TECHNIQUE
    • H04W—WIRELESS COMMUNICATION NETWORKS
    • H04W24/00—Supervisory, monitoring or testing arrangements
    • H04W24/06—Testing, supervising or monitoring using simulated traffic

Definitions

  • This disclosure relates to a communication state estimation device, a communication state estimation system, a communication state estimation method, and a communication state estimation program.
  • the communication quality changes according to changes in the environment, such as the movement of objects in the vicinity. Due to changes in the environment, it may become impossible to meet the service provided by the communication device or the communication quality required by the system.
  • fifth-generation communications such as "IEEE802.11ad” and cellular communications use high frequencies in the millimeter range, so blocking caused by obstacles between the transmitter and receiver in wireless communications has a significant impact on communication quality.
  • Non-patent literature 1 and 2 describe a method for estimating the receiving power of millimeter wave radio waves based on physical information obtained from a depth camera
  • non-patent literature 2 describes a method for estimating millimeter wave throughput based on physical information obtained from a depth camera.
  • non-patent documents 1 and 2 do not take into account the movement of the user carrying the communication device when predicting communication quality. That is, non-patent documents 1 and 2 acquire physical space information from a depth camera, and predict communication quality when the wireless communication channel of millimeter wave communication is interrupted due to the passage of an object, based on this information. Non-patent documents 1 and 2 make no mention of predicting communication quality based on physical space information using motion including the positional relationship between the user and the communication device, and the user's moving speed and acceleration. This poses the problem that communication conditions such as communication quality cannot be predicted with high accuracy.
  • This disclosure has been made in consideration of the above circumstances, and its purpose is to provide a communication state estimation device, a communication state estimation system, a communication state estimation method, and a communication state estimation program that are capable of predicting with high accuracy the communication state of a communication terminal carried by a user (person, robot).
  • a communication status estimation device includes an information acquisition unit that acquires user motion information from detection data from the motion detection unit, a communication quality acquisition unit that acquires the communication quality of the communication terminal carried by the user, and an estimation unit that compares the motion information with the communication quality, extracts motion information whose relationship with the communication quality is higher than a predetermined value, and, when the motion information of the user to be estimated is acquired, estimates the communication status of the communication terminal carried by the user to be estimated based on the motion information whose relationship is higher than the predetermined value.
  • a communication state estimation system includes a motion detection unit installed around a user, and a communication state estimation device that estimates the communication state of a communication terminal carried by the user.
  • the communication state estimation device includes an information acquisition unit that acquires motion information of the user from data detected by the motion detection unit, a communication quality acquisition unit that acquires the communication quality of the communication terminal carried by the user, and an estimation unit that compares the motion information with the communication quality, extracts motion information whose relationship with the communication quality is higher than a predetermined value, and, when motion information of a user to be estimated is acquired, estimates the communication state of the communication terminal carried by the user to be estimated based on the motion information whose relationship is higher than a predetermined value.
  • a communication status estimation method acquires user motion information from detection data from a motion detection unit, acquires communication quality of a communication terminal carried by the user, compares the motion information with the communication quality, extracts motion information whose relationship to the communication quality is higher than a predetermined value, and, when the motion information of a user to be estimated is acquired, estimates the communication status of the communication terminal carried by the user to be estimated based on the motion information whose relationship is higher than a predetermined value.
  • One aspect of the present disclosure is a communication state estimation program for causing a computer to function as the above-mentioned communication state estimation device.
  • This disclosure makes it possible to predict with high accuracy the communication status of a communication terminal owned by a user.
  • FIG. 1 is a block diagram showing the configuration of a communication state estimating system and its peripheral devices according to an embodiment.
  • FIG. 2 is a flowchart showing a processing procedure of the communication state estimating system according to the embodiment.
  • FIG. 3 is an explanatory diagram that illustrates a flow of calculating a correlation value between motion information and communication quality using machine learning based on the motion information and communication quality collected during data collection.
  • FIG. 4A is a graph showing which parts contributed significantly to predictions when learning a prediction model, and shows data when the communication terminal was inserted in a backpack carried by the user.
  • FIG. 4B is a graph showing which parts contributed significantly to predictions when learning the prediction model, and shows data when the user held the communication terminal in his/her left hand.
  • FIG. 4A is a graph showing which parts contributed significantly to predictions when learning a prediction model, and shows data when the communication terminal was inserted in a backpack carried by the user.
  • FIG. 4B is a graph showing which parts contributed significantly to predictions when learning the prediction model, and shows
  • FIG. 5A is a graph showing estimated values of communication quality predicted based on location coordinates and actual measured values of communication quality.
  • FIG. 5B is a graph showing estimated values when communication quality is predicted based on motion information and actual measured values of communication quality.
  • FIG. 6 is a block diagram showing the hardware configuration of this embodiment.
  • FIG. 1 is a block diagram showing the configuration of a communication state estimation system according to an embodiment.
  • the communication state estimation system 100 includes a motion detection unit 1 and a communication state estimation device 2.
  • the motion detection unit 1 is installed around a person or a robot.
  • the robot is, for example, a robot that imitates a human, and is equipped with, for example, a right hand, a left hand, a torso, and a head, each of which can operate independently.
  • the person and the robot are collectively referred to as "user P1".
  • User P1 possesses a communication terminal 51.
  • the communication terminal 51 is, for example, a smartphone, a tablet terminal, or a PC (personal computer), and is capable of voice calls, data communications, etc.
  • the motion detection unit 1 includes at least one of a sensor 11 worn by the user P1 and a camera 12 installed around the user P1 to capture images of the surroundings.
  • Motion information is information about the position, speed, acceleration, and combinations of these of the part. Specifically, motion information is information such as “the right hand is raised upwards” and “the left hand is moving at a speed of XX [m/sec]."
  • Camera 12 is installed at an appropriate location around user P1, and captures an image of user P1 to obtain image data. Because the installation position of camera 12 is fixed, it is possible to detect motion information of user P1 by analyzing the image data.
  • the motion detection unit 1 which includes the sensor 11 and the camera 12, is connected to the communication state estimation device 2 wirelessly or via a wire.
  • the motion detection unit 1 transmits motion information detected by the sensor 11 and image data captured by the camera 12 to the communication state estimation device 2.
  • the motion detection unit 1 includes the sensor 11 and the camera 12 will be described, but a configuration in which only one of the sensor 11 and the camera 12 is included may also be used.
  • devices capable of detecting the above-mentioned motion information may also be installed.
  • the communication state estimation device 2 collects the communication quality when communication terminals 51 held by multiple users P1 are communicating, and the motion information of each user P1 obtained from the motion detection unit 1, and collects associated data indicating the relationship between the motion information and the communication quality. Furthermore, when the motion information of the user P1 to be estimated is obtained, the communication state of the communication terminal 51 held by this user P1 is estimated based on the associated data.
  • the communication state includes the communication quality of the communication terminal 51, and the manner in which the user P1 holds the communication terminal 51.
  • Communication quality includes, for example, bandwidth throughput, RSSI (Received Signal Strength Indicator) which indicates the received signal strength of radio waves, RSRP (Reference Signal Received Power) which indicates the radio wave strength of the base station, RSRQ (Reference Signal Received Quality) which indicates the received strength of radio waves, and SINR (Signal to Interference and Noise) which indicates the received quality of the signal.
  • RSSI Receiveived Signal Strength Indicator
  • RSRP Reference Signal Received Power
  • RSRQ Reference Signal Received Quality
  • SINR Signal to Interference and Noise
  • the manner in which user P1 holds communication terminal 51 refers to, for example, user P1 holding communication terminal 51 in his right hand, or communication terminal 51 being inserted in a backpack carried by user P1.
  • the configuration of communication state estimation device 2 will be specifically described below.
  • the communication state estimation device 2 includes a communication quality acquisition unit 21, an information acquisition unit 22, and an estimation unit 23.
  • the communication quality acquisition unit 21 acquires the communication quality when the communication terminal 51 owned by the user P1 is communicating.
  • the information acquisition unit 22 acquires at least one of the motion information detected by the sensor 11 and image data captured by the camera 12. Based on the image data transmitted from the camera 12, the information acquisition unit 22 calculates the motion information of the user P1 contained in the image data. That is, the information acquisition unit 22 acquires the motion information of the user from the detection data of the motion detection unit 1 installed around the user.
  • Well-known image processing techniques can be used as a method for calculating the motion information from the image data.
  • the estimation unit 23 acquires the communication quality from the communication quality acquisition unit 21, and acquires the motion information from the information acquisition unit 22. Specifically, the estimation unit 23 calculates a correlation value between the motion information and the communication quality using machine learning based on the communication quality and the motion information. Based on the calculated correlation value, the estimation unit 23 extracts motion information whose correlation value with the communication quality is higher than a predetermined value. Note that the calculation of the correlation value may employ a statistical method other than machine learning.
  • feature importance and singular value decomposition may also be analyzed, and motion information whose analysis results are higher than a predetermined value may be extracted. That is, the estimation unit 23 analyzes at least one of the factors of the correlation value, feature importance, and singular value decomposition based on the time-series data of the motion information and communication quality, and outputs motion information whose factor is higher than a predetermined threshold value.
  • the estimation unit 23 also acquires motion information of the user P1 to be estimated, and estimates the communication state of the communication terminal 51 held by the user P1 to be estimated based on the motion information in which the relationship is higher than a predetermined value.
  • the communication state includes the communication quality and the manner in which the user P1 holds the communication terminal 51.
  • the estimation unit 23 generates a communication quality prediction model that indicates the relationship between the extracted motion information and the communication quality during communication, and when the motion information of the user P1 to be estimated is input, this model is used to estimate the communication quality (communication state) of the communication terminal 51 held by the user P1 to be estimated.
  • the estimation unit 23 generates a prediction model based on the motion information of the user P1 to be estimated and the motion information whose correlation value with the communication quality is higher than a predetermined value, and estimates the manner in which the user P1 is holding the communication terminal 51. For example, it estimates that the user P1 is holding the communication terminal 51 in his right hand, that the communication terminal 51 is inserted in a backpack that the user P1 is carrying, and the like.
  • the estimation unit 23 compares the motion information with the communication quality, extracts the motion information whose relationship with the communication quality is higher than a predetermined value, and when the motion information of the user P1 to be estimated is obtained, it estimates the communication state of the communication terminal 51 held by the user P1 to be estimated based on the motion information whose relationship with the communication quality is higher than a predetermined value.
  • Fig. 2 is a flowchart showing the processing procedure of the communication state estimation system 100
  • Fig. 3 is an explanatory diagram showing a schematic flow of calculating a correlation value between motion information and communication quality using machine learning based on the motion information and communication quality collected during data collection.
  • the communication quality acquisition unit 21 acquires the communication quality when the communication terminal 51 is communicating.
  • the communication quality includes bandwidth throughput, RSSI, RSRP, RSRQ, SINR, and the time required for downloading.
  • the information acquisition unit 22 acquires motion information from the motion detection unit 1.
  • the motion information is, for example, information on the skeletal coordinates of the user P1, as shown by "C1" in FIG. 3.
  • the estimation unit 23 acquires the communication quality from the communication quality acquisition unit 21, and acquires the motion information from the information acquisition unit 22.
  • the estimation unit 23 calculates the correlation value between the motion information acquired in the processing of step S11 and the communication quality by using techniques such as machine learning and statistics.
  • the correlation value is calculated by machine learning using a neural network that takes motion information as input and communication quality as output.
  • motion information that has a high correlation with wireless communication quality is determined from the weight values of a neural network consisting of an input layer W1, at least one intermediate layer W2, and an output layer W3, and the importance of the input of the random forest.
  • the communication quality of the communication terminal 51 owned by the user P1 and the manner in which the communication terminal 51 is owned can be estimated.
  • the predicted results of communication quality are output, and in “C4" the output results are evaluated.
  • feature importance and singular value decomposition may be analyzed to automatically generate motion information whose correlation value with respect to communication quality is higher than a predetermined value.
  • the communication quality of communication terminal 51 owned by user P1 can be estimated with high accuracy.
  • feature importance and singular value decomposition may be analyzed to determine motion information whose correlation value with respect to communication quality is higher than a predetermined value.
  • step S13 the estimation unit 23 estimates the communication quality (communication state) of the communication terminal 51 based on the motion information output in the processing of step S12. Specifically, the estimation unit 23 uses the motion information obtained in the processing of step S12 to generate a communication quality model indicating the relationship between the motion information and the communication quality. When the motion information of the user P1 to be estimated is obtained, the estimation unit 23 predicts the communication quality of the communication terminal 51 owned by the user P1 based on this motion information using the communication quality model.
  • the estimation unit 23 estimates the manner in which the user P1 is holding the communication terminal 51 (communication state) based on the motion information of the user P1 to be estimated obtained in the processing of step S11. For example, it estimates that the user P1 is holding the communication terminal 51 in his/her right hand, or that the communication terminal 51 is inserted in a backpack that the user is carrying. In this way, it is possible to estimate the communication state (communication quality, the manner in which the user P1 is holding the communication terminal 51) of the communication terminal 51 held by the user P1 based on the motion information of the user P1.
  • Figures 4A and 4B are graphs showing the importance during learning, i.e., which parts of the body contributed greatly to predictions when learning the prediction model.
  • the head and torso of user P1 contributed greatly to the learning of the prediction model, and it is estimated that there is a high possibility that communication terminal 51 is inserted into the backpack. According to experiments conducted by the inventors, it was confirmed that communication terminal 51 is actually inserted into the backpack carried by user P1.
  • the left hand, torso, and left shoulder of user P1 contribute significantly to the learning of the prediction model, and it is estimated that there is a high possibility that communication terminal 51 is held in the left hand. According to experiments conducted by the inventors, it was confirmed that communication terminal 51 is actually held in the left hand.
  • FIG. 5A is a graph showing the estimated and measured values when estimating communication quality based on the location information (internal information) of user P1 acquired by a GPS receiver, where the graph shown with a solid line shows the measured values and the graph shown with a dashed dotted line shows the estimated values. It can be seen from the graph in FIG. 5A that when internal information is used, the measured values and the estimated values are almost the same.
  • FIG. 5B is a graph showing estimated and measured values when estimating the communication quality of communication terminal 51 owned by user P1 based on the motion information used in this embodiment, with the graph shown with a solid line showing the measured values and the graph shown with a dashed dotted line showing the estimated values. From the graph in FIG. 5B, it can be seen that when motion information is used, the measured and estimated values are nearly identical, and results are nearly equivalent to those obtained when the internal information shown in FIG. 5A is used.
  • the communication state estimation device 2 includes an information acquisition unit 22 that acquires motion information of the user P1 from the detection data of the motion detection unit 1, a communication quality acquisition unit 21 that acquires the communication quality of the communication terminal 51 carried by the user P1, and an estimation unit 23 that compares the motion information with the communication quality, extracts motion information whose relationship to the communication quality is higher than a predetermined value, and, when the motion information of the user P1 to be estimated is acquired, estimates the communication state (communication quality, carrying manner) of the communication terminal 51 carried by the user P1 to be estimated based on the motion information whose relationship is higher than the predetermined value.
  • the communication state estimation device 2 is capable of estimating communication quality with high accuracy, and if an estimation result indicates that communication quality will decrease, for example, it becomes possible to carry out processing in advance, such as "reducing the frame rate when transmitting video.”
  • the manner in which the communication terminal 51 is held by the user P1 who is the subject of estimation (communication state). For example, it is possible to estimate whether the user P1 is holding the communication terminal 51 in his/her right hand, or whether the communication terminal 51 is inserted in a backpack that the user P1 is carrying.
  • the estimation unit 23 analyzes at least one of the factors, correlation value, feature importance, and singular value decomposition, based on the time-series data of the motion information and communication quality, and extracts motion information whose factor has a numerical value higher than a predetermined threshold. This makes it possible to extract motion information that affects communication quality with high accuracy.
  • the motion detection unit 1 includes at least one of a sensor 11 that detects the position, speed, and acceleration of parts of the user P1, and a camera that captures an image of the user P1, making it possible to detect the motion information of the user P1 with high accuracy.
  • the communication state estimation device 2 of the present embodiment described above may be, for example, a general-purpose computer system including a CPU (Central Processing Unit, processor) 901, memory 902, storage 903 (HDD: Hard Disk Drive, SSD: Solid State Drive), communication device 904, input device 905, and output device 906, as shown in FIG. 6.
  • the memory 902 and storage 903 are storage devices.
  • the CPU 901 executes a predetermined program loaded onto the memory 902, thereby realizing each function of the communication state estimation device 2.
  • the communication state estimation device 2 may be implemented in one computer, or in multiple computers.
  • the communication state estimation device 2 may also be a virtual machine implemented in a computer.
  • the program for the communication state estimation device 2 can be stored on a computer-readable recording medium such as a HDD, SSD, USB (Universal Serial Bus) memory, CD (Compact Disc), or DVD (Digital Versatile Disc), or can be distributed via a network.
  • a computer-readable recording medium such as a HDD, SSD, USB (Universal Serial Bus) memory, CD (Compact Disc), or DVD (Digital Versatile Disc), or can be distributed via a network.
  • the computer-readable recording medium is, for example, a non-transitory recording medium.
  • the present invention is not limited to the above embodiment, and many variations are possible within the scope of the gist.

Landscapes

  • Engineering & Computer Science (AREA)
  • Computer Networks & Wireless Communication (AREA)
  • Signal Processing (AREA)
  • Mobile Radio Communication Systems (AREA)
PCT/JP2022/042068 2022-11-11 2022-11-11 通信状態推定装置、通信状態推定システム、通信状態推定方法、及び通信状態推定プログラム Ceased WO2024100881A1 (ja)

Priority Applications (2)

Application Number Priority Date Filing Date Title
PCT/JP2022/042068 WO2024100881A1 (ja) 2022-11-11 2022-11-11 通信状態推定装置、通信状態推定システム、通信状態推定方法、及び通信状態推定プログラム
JP2024556986A JPWO2024100881A1 (https=) 2022-11-11 2022-11-11

Applications Claiming Priority (1)

Application Number Priority Date Filing Date Title
PCT/JP2022/042068 WO2024100881A1 (ja) 2022-11-11 2022-11-11 通信状態推定装置、通信状態推定システム、通信状態推定方法、及び通信状態推定プログラム

Publications (1)

Publication Number Publication Date
WO2024100881A1 true WO2024100881A1 (ja) 2024-05-16

Family

ID=91032190

Family Applications (1)

Application Number Title Priority Date Filing Date
PCT/JP2022/042068 Ceased WO2024100881A1 (ja) 2022-11-11 2022-11-11 通信状態推定装置、通信状態推定システム、通信状態推定方法、及び通信状態推定プログラム

Country Status (2)

Country Link
JP (1) JPWO2024100881A1 (https=)
WO (1) WO2024100881A1 (https=)

Citations (2)

* Cited by examiner, † Cited by third party
Publication number Priority date Publication date Assignee Title
JP2019195136A (ja) * 2018-05-02 2019-11-07 日本電気株式会社 管理装置、データ抽出方法及びプログラム
WO2021214997A1 (ja) * 2020-04-24 2021-10-28 三菱電機株式会社 情報処理装置、情報処理方法及び情報処理プログラム

Patent Citations (2)

* Cited by examiner, † Cited by third party
Publication number Priority date Publication date Assignee Title
JP2019195136A (ja) * 2018-05-02 2019-11-07 日本電気株式会社 管理装置、データ抽出方法及びプログラム
WO2021214997A1 (ja) * 2020-04-24 2021-10-28 三菱電機株式会社 情報処理装置、情報処理方法及び情報処理プログラム

Also Published As

Publication number Publication date
JPWO2024100881A1 (https=) 2024-05-16

Similar Documents

Publication Publication Date Title
JP7082113B2 (ja) 散歩を自動的に検出および開始するためのシステムおよび方法
EP3163543B1 (en) Alarming method and device
KR101666009B1 (ko) 다중 셀 환경에서 피간섭 단말을 검출하고 간섭 제어를 수행하는 통신 시스템
CN116708086B (zh) 感知方法、装置及通信设备
Yang et al. Fine-grained adaptive location-independent activity recognition using commodity WiFi
US10667107B2 (en) Method and system for entry and exit monitoring using bluetooth low energy (BLE) beacons
WO2021084519A1 (en) System for multi-path 5g and wi-fi motion detection
JP7226575B2 (ja) 通信端末及び通信品質予測方法
KR102705914B1 (ko) 향상된 채널 추정을 위한 전자 환경의 격리
EP3276370A1 (en) Method and apparatus for detecting motion activity
US12279190B1 (en) Respiration monitoring based on noisy channel state information (CSI) data
US10901076B2 (en) Temporal analysis for user speed estimation in wireless networks
CN116707673A (zh) 感知方法、装置及通信设备
Lv et al. Accurate bandwidth and delay prediction for 5G cellular networks
CN112739984A (zh) 手机多模态位置感测
WO2024100881A1 (ja) 通信状態推定装置、通信状態推定システム、通信状態推定方法、及び通信状態推定プログラム
CN117998398A (zh) 确定模型适用性的方法、装置及通信设备
KR20240021181A (ko) 동적 시간 도메인 채널 표현을 위한 시스템 및 방법
CN118302649A (zh) 隔离电子环境以改进信道估计
KR101347885B1 (ko) 무선 인지 장치 및 무선 인지 방법
Manjappa et al. Estimating crowd distribution using smart bulbs
WO2026028391A1 (ja) 無線通信装置及び通信品質予測方法
WO2024100893A1 (ja) モデル生成装置及びモデル生成方法並びにモデル生成プログラム
Oliveira et al. A multi-scale statistical control process for mobility and interference identification in ieee 802.11
KR20160001084A (ko) 사용자에 대한 상황 정보 리포팅 방법 및 상황 정보를 이용하는 방법

Legal Events

Date Code Title Description
121 Ep: the epo has been informed by wipo that ep was designated in this application

Ref document number: 22965205

Country of ref document: EP

Kind code of ref document: A1

WWE Wipo information: entry into national phase

Ref document number: 2024556986

Country of ref document: JP

NENP Non-entry into the national phase

Ref country code: DE

122 Ep: pct application non-entry in european phase

Ref document number: 22965205

Country of ref document: EP

Kind code of ref document: A1