Indoor localization for Bluetooth low energy using wavelet and smoothing filter

Abstract

This paper presents an investigation on the impact of the Bluetooth Low Energy (BLE) received signal strength indicator (RSSI) value for indoor localization in real line-of-sight (LOS) and dynamic non-LOS environments. Experimentation demonstrates that the RSSI value of BLE signals is unstable in indoor environments. The principle underlying this behavior is discussed in this paper. Two self-adaptive filters (smoothing filter and wavelet filter) are applied to stabilize and de-noise the RSSI sequence. We demonstrate that the stability of localization performance is enhanced by employing these filters. The standard deviation of the RSSI sequence is reduced from 4.6 meters to 0.8 meters, which means that a steady increase in the accuracy and stabilization of the localization is achieved.