상세정보
학술지
Wireless Distance Estimation Based on Error Correction of Bluetooth RSSI
- 발행일
- 201506
- 출처
- IEICE Transactions on Communications, v.E98.B no.6, pp.1018-1031
- ISSN
- 1745-1345
- 출판사
- 일본, 전자정보통신학회 (IEICE)
- 협약과제
- 14MS9200, (통합)인간친화형 디바이스(스킨패치, 멀티모달 서피스) 및 디바이스 소셜 프레임워크 기술 개발, 배창석
- 초록
- In this paper, we propose an error-correction low-pass filter (EC-LPF) algorithm for estimating the wireless distance between devices. To measure this distance, the received signal strength indication (RSSI) is a popularly used method because the RSSI of a wireless signal, such as Wi-Fi and Bluetooth, can be measured easily without the need for additional hardware. However, estimating the wireless distance using an RSSI is known to be difficult owing to the occurrence of inaccuracies. To examine the inaccuracy characteristics of Bluetooth RSSI, we conduct a preliminary test to discover the relationship between the actual distance and Bluetooth RSSI under several different environments. The test results verify that the main reason for inaccuracy is the existence of measurement errors in the raw Bluetooth RSSI data. In this paper, the EC-LPF algorithm is proposed to reduce measurement errors by alleviating fluctuations in a Bluetooth signal with responsiveness for real-time applications. To evaluate the effectiveness of the EC-LPF algorithm, distance accuracies of different filtering algorithms are compared, namely, a low-pass filer (LPF), a Kalman filter, a particle filter, and the EC-LPF algorithm under two different environments: An electromagnetic compatibility (EMC) chamber and an indoor hall. The EC-LPF algorithm achieves the best performance in both environments in terms of the coefficient of determination, standard deviation, measurement range, and response time. In addition, we also implemented a meeting room application to verify the feasibility of the ECLPF algorithm. The results prove that the EC-LPF algorithm distinguishes the inside and outside areas of a meeting room without error.
- KSP 제안 키워드
- Actual distance, Best performance, Bluetooth RSSI, Coefficient of determination, Error Correction, Filtering algorithms, Measurement errors, Measurement range, Meeting room, Particle filters, Preliminary test