ETRI-Knowledge Sharing Plaform



논문 검색
구분 SCI
연도 ~ 키워드


학술대회 Learning based Wi-Fi RTT Range Estimation
Cited 0 time in scopus Download 5 time Share share facebook twitter linkedin kakaostory
정부금, 정병창, 임진혁, 유윤식, 박혜숙
International Conference on Information and Communication Technology Convergence (ICTC) 2021, pp.1030-1032
21HR4600, 공공 업무·임무용 정보통신자원의 노출을 최소화하는 지능적 스텔스화 기술개발, 박혜숙
In Wi-Fi RTT range estimation, accuracy is the most critical issue. But current estimated values using Wi-Fi RTT with 802.11mc FTM protocol are often randomly far away from the true range. These inaccuracies and fluctuations make it difficult to estimate the distance of mobile devices and Wi-Fi access points needed for indoor location-based services. In this paper, we present learning-based system model to get generalized probabilistic distribution. We made a deep learning model using existing measured range values on each certain range as training data. To improve accuracy, we used multiple correlated parameters detected with 802.11mc FTM. We verified the performance of our model using real test data. It is shown that it can guarantee the stability with high accuracy for true range estimation. Our system can be used as a base framework for other various situations or more learning algorithms to enhance development efficiency.
KSP 제안 키워드
Access point, High accuracy, Indoor Location-based services, Learning model, Learning-based, Mobile devices, Range estimation, Test data, Wi-fi access, based system, deep learning(DL)