Characterization of Smart Phone Received Signal Strength Indication for WLAN Indoor Positioning Accuracy Improvement

Abstract

Considering that indoor positioning applications based on wireless local area network location fingerprinting would be mainly used on the mobile devices. This paper investigates the differences of received signal strength indication (RSSI) between different smart phones and the distributions of RSSI are also analyzed. The statistical analysis of experimental results shows the differences of RSSI between different smart phones are not trivial. Nearly 65% of the RSSI histograms are significantly peaked relative to Gaussian distribution and 65% of them are left-skewed distribution. Therefore, taking skewness and kurtosis coefficients into account, Gaussian distribution is not sufficient to ensure an accurate modeling of the RSSI. The impacts of human behavior on RSSI distribution are explored and two types of human behavior are revealed to be the cause of bi-modal distribution. The statistical data analysis could enable smart phone indoor positioning systems designers to improve positioning performance and to model location fingerprinting based indoor positioning systems.