KR20210079882A - Method for Estimating Distance Using Channel Amplitude Infomration Estimated from LTE signal Based on Neural Network - Google Patents

Method for Estimating Distance Using Channel Amplitude Infomration Estimated from LTE signal Based on Neural Network Download PDF

Info

Publication number
KR20210079882A
KR20210079882A KR1020190172101A KR20190172101A KR20210079882A KR 20210079882 A KR20210079882 A KR 20210079882A KR 1020190172101 A KR1020190172101 A KR 1020190172101A KR 20190172101 A KR20190172101 A KR 20190172101A KR 20210079882 A KR20210079882 A KR 20210079882A
Authority
KR
South Korea
Prior art keywords
neural network
lte signal
artificial neural
estimated
amplitude information
Prior art date
Application number
KR1020190172101A
Other languages
Korean (ko)
Inventor
서지원
이하림
Original Assignee
연세대학교 산학협력단
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 연세대학교 산학협력단 filed Critical 연세대학교 산학협력단
Priority to KR1020190172101A priority Critical patent/KR20210079882A/en
Publication of KR20210079882A publication Critical patent/KR20210079882A/en

Links

Images

Classifications

    • G—PHYSICS
    • G01—MEASURING; TESTING
    • G01S—RADIO DIRECTION-FINDING; RADIO NAVIGATION; DETERMINING DISTANCE OR VELOCITY BY USE OF RADIO WAVES; LOCATING OR PRESENCE-DETECTING BY USE OF THE REFLECTION OR RERADIATION OF RADIO WAVES; ANALOGOUS ARRANGEMENTS USING OTHER WAVES
    • G01S11/00—Systems for determining distance or velocity not using reflection or reradiation
    • G01S11/02—Systems for determining distance or velocity not using reflection or reradiation using radio waves
    • G—PHYSICS
    • G06—COMPUTING; CALCULATING OR COUNTING
    • G06N—COMPUTING ARRANGEMENTS BASED ON SPECIFIC COMPUTATIONAL MODELS
    • G06N3/00—Computing arrangements based on biological models
    • G06N3/02—Neural networks
    • G06N3/08—Learning methods
    • G06N3/082—Learning methods modifying the architecture, e.g. adding, deleting or silencing nodes or connections
    • H—ELECTRICITY
    • H04—ELECTRIC COMMUNICATION TECHNIQUE
    • H04W—WIRELESS COMMUNICATION NETWORKS
    • H04W64/00—Locating users or terminals or network equipment for network management purposes, e.g. mobility management

Landscapes

  • Engineering & Computer Science (AREA)
  • Physics & Mathematics (AREA)
  • Theoretical Computer Science (AREA)
  • General Physics & Mathematics (AREA)
  • General Health & Medical Sciences (AREA)
  • General Engineering & Computer Science (AREA)
  • Biophysics (AREA)
  • Computational Linguistics (AREA)
  • Data Mining & Analysis (AREA)
  • Evolutionary Computation (AREA)
  • Artificial Intelligence (AREA)
  • Molecular Biology (AREA)
  • Computing Systems (AREA)
  • Biomedical Technology (AREA)
  • Life Sciences & Earth Sciences (AREA)
  • Mathematical Physics (AREA)
  • Software Systems (AREA)
  • Health & Medical Sciences (AREA)
  • Radar, Positioning & Navigation (AREA)
  • Remote Sensing (AREA)
  • Computer Networks & Wireless Communication (AREA)
  • Signal Processing (AREA)
  • Position Fixing By Use Of Radio Waves (AREA)

Abstract

Disclosed is a method for estimating distance using channel amplitude information estimated from LTE signal based on artificial neural network. The disclosed method includes the steps of: obtaining amplitude information of an estimated channel from an LTE signal collected through an LTE signal receiver in a plurality of environments; and obtaining a distance between a base station and a user terminal by inputting the obtained amplitude information into an artificial neural network model, wherein the artificial neural network model has a CNN structure, and the CNN structure includes a convolutional layer, a pooling layer, and an FC layer. According to the disclosed method, multiple signals exist in an urban area and the accuracy of positioning can be guaranteed in an NLOS environment.

Description

LTE 신호에서 추정된 채널 진폭 정보를 이용한 인공 신경망 기반 거리 추정 방법{Method for Estimating Distance Using Channel Amplitude Infomration Estimated from LTE signal Based on Neural Network}Method for Estimating Distance Using Channel Amplitude Infomration Estimated from LTE signal Based on Neural Network

본 발명은 거리 추정 방법에 관한 것으로서, 더욱 상세하게는 LTE 신호에서 추정된 채널 진폭 정보를 이용한 인공 신경망 기반 거리 추정 방법에 관한 것이다. The present invention relates to a distance estimation method, and more particularly, to an artificial neural network-based distance estimation method using channel amplitude information estimated from an LTE signal.

통신 신호로부터 측위 또는 거리 추정을 위해 다양한 방법들이 연구되고 있다. 무선 신호를 이용하여 측위하는 방법에는 time-of-arrival (TOA), time-difference-of-arrival (TDOA) 등 시간 지연을 이용하는 방법과, received-signal-strength indicator (RSSI), channel state information (CSI) 등 신호 세기를 이용하는 방법이 있다.Various methods are being studied for positioning or distance estimation from a communication signal. A method of positioning using a radio signal includes a method using a time delay such as time-of-arrival (TOA), time-difference-of-arrival (TDOA), received-signal-strength indicator (RSSI), channel state information ( There is a method using signal strength such as CSI).

통신 신호를 이용하여 측위를 수행하는 선행문헌 중 하나로 10-2019-0053470호가 있다. 해당 선행문헌은 무선 신호 정보를 수집하는 데이터 수집 툴, 인공 신경망 구조의 측위 알고리즘을 학습시키고 학습된 결과를 검증하는 트레이닝 서버, 입력된 무선 신호 정보에 대해 인공 신경망 구조의 측위 알고리즘을 적용하여 위치를 추정하는 측위 서버를 개시하고 있다. 해당 선행문헌은 핑거 프린팅 방식을 사용하고 핑거프린팅 데이터베이스를 필요로 하기 때문에 별도의 저장소가 필요하다. As one of the prior documents for performing positioning using a communication signal, there is No. 10-2019-0053470. The prior literature is a data collection tool that collects wireless signal information, a training server that learns a positioning algorithm of an artificial neural network structure and verifies the learned result, and applies a positioning algorithm of an artificial neural network structure to input wireless signal information to determine the location. A positioning server for estimating is disclosed. Since the prior literature uses a fingerprinting method and requires a fingerprinting database, a separate storage is required.

통신 신호를 이용하여 측위를 수행하는 또 다른 선행문헌으로 10-2019-0072900호가 있다. 해당 선행문헌은 서로 다른 지점에 위치하는 n개의 수신부로부터 무선 신호를 수신한 뒤 시간, 주파수, 안테나에 따른 신호의 전력, 크기 및 위상을 추출하고 딥러닝을 이용하여 측위를 수행하는 시스템을 개시한다. 해당 선행문헌은 다수의 안테나를 필요로 하는 문제가 있다. As another prior document for performing positioning using a communication signal, there is No. 10-2019-0072900. The prior literature discloses a system for receiving radio signals from n receivers located at different points, extracting the power, magnitude and phase of the signals according to time, frequency, and antenna, and performing positioning using deep learning. . The prior literature has a problem that requires a plurality of antennas.

한편, 전통적인 측위 방식 중 하나인 GPS(Global Positioning System) 등의 GNSS(Global Navigation Satellite System) 방식은 반사로 인해 발생하는 다중 신호와 NLOS(Non Line of Sight)의 영향으로 측위 정확도가 떨어지는 문제점이 있다. On the other hand, GNSS (Global Navigation Satellite System) such as GPS (Global Positioning System), which is one of the traditional positioning methods, has a problem in that positioning accuracy is lowered due to the influence of multiple signals and NLOS (Non Line of Sight) generated due to reflection .

본 발명은 다중 신호가 존재하고 NLOS로 인해 위성항법 신호를 신뢰할 수 없는 도심 지역에서 측위의 정확도를 보장할 수 있는 인공 신경망 거리 추정 방법을 제안한다. The present invention proposes an artificial neural network distance estimation method that can ensure the accuracy of positioning in an urban area where multiple signals exist and satellite navigation signals are not reliable due to NLOS.

상기와 같은 목적을 달성하기 위해, 본 발명의 일 측면에 따르면, 다수의 환경에서 LTE 신호 수신기를 통해 수집된 LTE 신호에서 추정된 채널의 진폭 정보를 획득하는 단계; 상기 획득한 진폭 정보를 인공 신경망 모델에 입력하여 기지국과 사용자 단말기간 거리를 획득하는 단계를 포함하되, 상기 인공 신경망 모델은 CNN 구조를 가지고, 상기 CNN 구조는 콘볼루션 레이어, 풀링 레이어 및 FC 레이어를 포함하는 LTE 신호를 이용한 인공 신경망 기반 거리 추정 방법이 제공된다. In order to achieve the above object, according to an aspect of the present invention, the method comprising: obtaining amplitude information of a channel estimated from an LTE signal collected through an LTE signal receiver in a plurality of environments; and inputting the obtained amplitude information into an artificial neural network model to obtain a distance between a base station and a user terminal, wherein the artificial neural network model has a CNN structure, and the CNN structure includes a convolutional layer, a pooling layer and an FC layer. An artificial neural network-based distance estimation method using an LTE signal including

본 발명에 의하면, 다중 신호가 존재하고 NLOS로 인해 위성항법 신호를 신뢰할 수 없는 도심 지역에서 측위의 정확도를 보장할 수 있는 장점이 있다. According to the present invention, there is an advantage in that the accuracy of positioning can be ensured in an urban area where multiple signals exist and a satellite navigation signal is not reliable due to NLOS.

도 1은 본 발명의 일 실시에에 따른 CSI(Channel State Information) 추정 방법의 전체적인 흐름을 나타낸 도면.
도 2는 본 발명의 거리 추정 방법의 전체적인 흐름을 나타낸 도면.
도 3은 본 발명에서 제안하는 인공 신경망 모델을 나타낸 도면.1 is a view showing the overall flow of a method for estimating CSI (Channel State Information) according to an embodiment of the present invention.
2 is a view showing the overall flow of the distance estimation method of the present invention.
3 is a view showing an artificial neural network model proposed by the present invention.

이하에서는 첨부한 도면을 참조하여 본 발명을 설명하기로 한다. 그러나 본 발명은 여러 가지 상이한 형태로 구현될 수 있으며, 따라서 여기에서 설명하는 실시예로 한정되는 것은 아니다.Hereinafter, the present invention will be described with reference to the accompanying drawings. However, the present invention may be embodied in several different forms, and thus is not limited to the embodiments described herein.

그리고 도면에서 본 발명을 명확하게 설명하기 위해서 설명과 관계없는 부분은 생략하였으며, 명세서 전체를 통하여 유사한 부분에 대해서는 유사한 도면 부호를 붙였다.And in order to clearly explain the present invention in the drawings, parts irrelevant to the description are omitted, and similar reference numerals are attached to similar parts throughout the specification.

명세서 전체에서, 어떤 부분이 다른 부분과 "연결"되어 있다고 할 때, 이는 "직접적으로 연결"되어 있는 경우뿐 아니라, 그 중간에 다른 부재를 사이에 두고 "간접적으로 연결"되어 있는 경우도 포함한다.Throughout the specification, when a part is said to be "connected" with another part, it includes not only the case where it is "directly connected" but also the case where it is "indirectly connected" with another member interposed therebetween. .

또한 어떤 부분이 어떤 구성 요소를 "포함"한다고 할 때, 이는 특별히 반대되는 기재가 없는 한 다른 구성 요소를 제외하는 것이 아니라 다른 구성 요소를 더 구비할 수 있다는 것을 의미한다.In addition, when a part "includes" a certain component, this means that other components may be further provided without excluding other components unless otherwise stated.

이하 첨부된 도면을 참고하여 본 발명의 실시예를 상세히 설명하기로 한다.Hereinafter, embodiments of the present invention will be described in detail with reference to the accompanying drawings.

본 발명에서 활용한 채널 진폭 정보는 LTE 다운링크 신호의 물리 계층에 위치한 cell specific reference signal (CRS)로부터 추정된다.The channel amplitude information utilized in the present invention is estimated from a cell specific reference signal (CRS) located in the physical layer of the LTE downlink signal.

본 발명에서 개발된 기술은 여러 야외 환경에서 학습된 인공 신경망 모델을 기반으로 추정된 채널 진폭 정보를 입력 받으면 LTE 기지국과 사용자 단말기 간의 거리를 출력하는 시스템에 관한 내용이다.The technology developed in the present invention relates to a system for outputting a distance between an LTE base station and a user terminal when receiving channel amplitude information estimated based on an artificial neural network model learned in various outdoor environments.

도 1은 본 발명의 일 실시에에 따른 CSI 추정 방법의 전체적인 흐름을 나타낸 도면이다. 1 is a diagram illustrating an overall flow of a CSI estimation method according to an embodiment of the present invention.

CSI는 시간과 주파수 두 축을 가지는 2차원 정보로, 시간축만을 가지는 1차원 정보인 RSSI에 비하여 정보량이 많아 더 정확한 거리 계산이 가능하다.CSI is two-dimensional information having two axes of time and frequency, and more accurate distance calculation is possible than RSSI, which is one-dimensional information having only a time axis, because of its large amount of information.

CSI는 LTE 다운링크 신호의 물리 계층에 위치한 pilot symbol로부터 추정되며, 이는 상용 LTE 신호 수신기에서 사용되는 방법이다.CSI is estimated from a pilot symbol located in the physical layer of the LTE downlink signal, which is a method used in commercial LTE signal receivers.

LTE 다운링크 신호의 pilot symbol의 명칭은 cell specific reference signal (CRS)이며, 도 1과 같은 과정으로 추정될 수 있다.The name of the pilot symbol of the LTE downlink signal is a cell specific reference signal (CRS), and can be estimated through the same process as in FIG. 1 .

도 2는 본 발명의 거리 추정 방법의 전체적인 흐름을 나타낸 도면이다. 2 is a diagram showing the overall flow of the distance estimation method of the present invention.

본 발명은 신호 수신기로부터 LTE 신호를 수신하고, 수신된 LTE 신호에서 채널 진폭 정보를 추정하며, 추정된 채널 진폭 정보를 인공 신경망 모델에 입력한 뒤, 예상되는 기지국과 사용자 단말기 간 거리를 출력하는 과정을 포함한다.The present invention is a process of receiving an LTE signal from a signal receiver, estimating channel amplitude information from the received LTE signal, inputting the estimated channel amplitude information into an artificial neural network model, and outputting an expected distance between a base station and a user terminal includes

본 발명은 1) 기지국에서 송신된 LTE 신호 데이터를 단말기로 수집, 2) 수집된 신호 물리계층의 CRS로부터 채널 진폭 정보 추정, 3) 추정된 채널 진폭 정보를 인공 신경망 모델에 입력, 4) 인공 신경망 모델에 통과되면 최종적으로 기지국과 단말기간 거리를 출력하는 과정을 포함한다.The present invention is 1) collecting LTE signal data transmitted from a base station to a terminal, 2) estimating channel amplitude information from the CRS of the collected signal physical layer, 3) inputting the estimated channel amplitude information into an artificial neural network model, 4) artificial neural network If the model is passed, it includes the process of finally outputting the distance between the base station and the terminal.

본 발명에서 제안하는 방법으로 단말과 다수의 기지국 간의 거리 추정치를 여러 개 구하면, 다변측량 (multilateration)을 이용하여 단말의 위치를 구할 수 있다.When several distance estimates between the terminal and a plurality of base stations are obtained by the method proposed in the present invention, the location of the terminal can be obtained using multilateration.

유사하게, 본 발명에서 제안하는 방법으로 다수의 단말과 기지국 간의 거리 추정치를 여러 개 구하면, 다변측량 (multilateration)을 이용하여 기지국의 위치를 구할 수 있다.Similarly, if a plurality of distance estimates between a plurality of terminals and a base station are obtained by the method proposed in the present invention, the location of the base station can be obtained using multilateration.

도 3은 본 발명에서 제안하는 인공 신경망 모델을 나타낸 도면이다. 3 is a diagram illustrating an artificial neural network model proposed by the present invention.

도 3을 참조하면, LTE 신호 특성상 채널 진폭 정보는 2차원으로 표현된다. 채널 진폭 정보를 convolutional neural network (CNN) 구조에 입력한다. Convolutional layer에서는 입력 데이터에서 특징 (feature)를 추출한다. 데이터의 양이 매우 크기 때문에 큰 데이터를 적은 양의 데이터로 줄이는 과정을 진행한다. 학습 과정에서 각 레이어들의 변수 (커널 사이즈, 필터 개수 등)가 조정된다. 마지막 fully connected layer에서는 회귀 (regression)을 진행한다. 최종적으로는 기지국과 단말기 간 거리를 추출할 수 있다.Referring to FIG. 3 , channel amplitude information is expressed in two dimensions due to the characteristics of the LTE signal. The channel amplitude information is input into a convolutional neural network (CNN) structure. In the convolutional layer, features are extracted from the input data. Since the amount of data is very large, the process of reducing large data to small data is performed. Variables (kernel size, number of filters, etc.) of each layer are adjusted during the learning process. In the last fully connected layer, regression is performed. Finally, the distance between the base station and the terminal can be extracted.

이상에서 설명한 바와 같은 본 발명에서 제안하는 인공 신경망 기반 거리 추정 방법은 기지국과 사용자 단말 간의 거리를 구하는 기술으로, 사용자 단말의 위치를 추정하는 데에 활용될 수 있다.As described above, the artificial neural network-based distance estimation method proposed by the present invention is a technique for obtaining a distance between a base station and a user terminal, and can be utilized to estimate the location of the user terminal.

또한, 본 발명에서 제안하는 LTE 기반 측위 방식은 GPS등 위성항법신호의 수신이 어려운 도심 환경에 사용될 수 있다.In addition, the LTE-based positioning method proposed in the present invention can be used in an urban environment in which it is difficult to receive satellite navigation signals such as GPS.

아울러, 본 발명에서 제안하는 기술은 기존의 핑거프린팅 방식과는 달리 데이터베이스를 따로 저장할 필요가 없어 간편한 방식이다.In addition, the technique proposed in the present invention is a convenient method because there is no need to separately store a database unlike the existing fingerprinting method.

이상에서 기술한 실시예들은 모든 면에서 예시적인 것이며 한정적이 아닌 것으로 이해해야만 한다.It should be understood that the embodiments described above are illustrative in all respects and not restrictive.

예를 들어, 단일형으로 설명되어 있는 각 구성 요소는 분산되어 실시될 수도 있으며, 마찬가지로 분산된 것으로 설명되어 있는 구성 요소들도 결합된 형태로 실시될 수 있다.For example, each component described as a single type may be implemented in a dispersed form, and likewise components described as distributed may be implemented in a combined form.

본 발명의 범위는 후술하는 특허청구범위에 의하여 나타내어지며, 특허청구범위의 의미 및 범위 그리고 그 균등 개념으로부터 도출되는 모든 변경 또는 변형된 형태가 본 발명의 범위에 포함되는 것으로 해석되어야 한다.The scope of the present invention is indicated by the following claims, and all changes or modifications derived from the meaning and scope of the claims and their equivalent concepts should be construed as being included in the scope of the present invention.

Claims (1)

다수의 환경에서 LTE 신호 수신기를 통해 수집된 LTE 신호에서 추정된 채널의 진폭 정보를 획득하는 단계;
상기 획득한 진폭 정보를 인공 신경망 모델에 입력하여 기지국과 사용자 단말기간 거리를 획득하는 단계를 포함하되,
상기 인공 신경망 모델은 CNN 구조를 가지고, 상기 CNN 구조는 콘볼루션 레이어, 풀링 레이어 및 FC 레이어를 포함하는 것을 특징으로 하는 LTE 신호를 이용한 인공 신경망 기반 거리 추정 방법.




obtaining amplitude information of an estimated channel from an LTE signal collected through an LTE signal receiver in a plurality of environments;
Comprising the step of obtaining the distance between the base station and the user terminal by inputting the obtained amplitude information into an artificial neural network model,
The artificial neural network model has a CNN structure, and the CNN structure is an artificial neural network-based distance estimation method using an LTE signal, characterized in that it includes a convolutional layer, a pooling layer, and an FC layer.
KR1020190172101A 2019-12-20 2019-12-20 Method for Estimating Distance Using Channel Amplitude Infomration Estimated from LTE signal Based on Neural Network KR20210079882A (en)

Priority Applications (1)

Application Number Priority Date Filing Date Title
KR1020190172101A KR20210079882A (en) 2019-12-20 2019-12-20 Method for Estimating Distance Using Channel Amplitude Infomration Estimated from LTE signal Based on Neural Network

Applications Claiming Priority (1)

Application Number Priority Date Filing Date Title
KR1020190172101A KR20210079882A (en) 2019-12-20 2019-12-20 Method for Estimating Distance Using Channel Amplitude Infomration Estimated from LTE signal Based on Neural Network

Publications (1)

Publication Number Publication Date
KR20210079882A true KR20210079882A (en) 2021-06-30

Family

ID=76602151

Family Applications (1)

Application Number Title Priority Date Filing Date
KR1020190172101A KR20210079882A (en) 2019-12-20 2019-12-20 Method for Estimating Distance Using Channel Amplitude Infomration Estimated from LTE signal Based on Neural Network

Country Status (1)

Country Link
KR (1) KR20210079882A (en)

Cited By (1)

* Cited by examiner, † Cited by third party
Publication number Priority date Publication date Assignee Title
WO2024106947A1 (en) * 2022-11-15 2024-05-23 경희대학교 산학협력단 Mimo communication system and communication method, and computing device for performing same

Cited By (1)

* Cited by examiner, † Cited by third party
Publication number Priority date Publication date Assignee Title
WO2024106947A1 (en) * 2022-11-15 2024-05-23 경희대학교 산학협력단 Mimo communication system and communication method, and computing device for performing same

Similar Documents

Publication Publication Date Title
US11391806B2 (en) Estimating characteristics of objects in environment
CN108702585B (en) Hybrid fingerprinting/OTDOA location techniques and systems
US9121922B2 (en) Access point location identification methods and apparatus based on absolute and relative harvesting
US8816908B2 (en) Optimal storage scheme for access point coverage data
US9432811B2 (en) Estimating the lateral location of a wireless terminal based on temperature and atmospheric pressure
US8433250B2 (en) Estimating whether a wireless terminal is indoors using pattern classification
CN107211249A (en) The frequency offset compensation that time difference is measured in being determined for position
CN112327340B (en) Terminal positioning accuracy evaluation method, device, equipment and medium
CN104735620A (en) Accurate positioning method based on multiple base stations
EP2404386A1 (en) Screening terminal positions at a terminal
CN105163382A (en) Indoor region location optimization method and system
Lee et al. Neural network-based ranging with LTE channel impulse response for localization in indoor environments
CN109996178B (en) Positioning method, apparatus, device and medium
CN105408761B (en) The method and apparatus for effectively reducing number of access point in radio map
KR102052519B1 (en) Indoor Positioning Method and Apparatus Based on Bluetooth Low Energy
KR20210079882A (en) Method for Estimating Distance Using Channel Amplitude Infomration Estimated from LTE signal Based on Neural Network
US10123298B1 (en) Estimating the location of a wireless terminal based on detection of whether it is moving or stationary
KR20220150604A (en) Apparatus and Method for Machine Learning-based Positioning Database Creation and Positioning of Uncollected Points using Matching Feature with Wireless Communication Infrastructure
CN111868548A (en) Method and system for geographical positioning of group evolution terminals
CN115474268B (en) Multi-carrier signal tracking and high-precision positioning method and system based on deep learning
Raspopoulos et al. Location-dependent information extraction for positioning
CN105074736A (en) Supporting coverage area modeling
Choi et al. RSS bias compensation in BLE beacon based positioning system
JP2015064232A (en) Determination device, network node, determination method, and program
Raja Critical analysis and modelling of location finding services