RSS-based indoor positioning with biased estimator and local geographical factor

Abstract

Considering the instability of received signal strength (RSS) in the indoor environment, this paper presents a feasible RSS based indoor positioning method by introducing a biased but optimized distance estimator, which is transformed from the Log-Normal (LN) fading model. On the other hand, the existed positioning methods, like maximum likelihood estimation (MLE) and least square estimation (LSE), always require at least three reference distances, which sometimes cannot be met in practice due to the situation that some nodes may be too far from one or multiple reference nodes. This paper proposes a positioning method considering the situation that one of three reference nodes receiving abnormal or no RSS value from a mobile node. The motivation of using RSS as the reference information is that compared with methods with other metrics like time of flight (TOF) or angle of arrival (AOA), RSS based method has the advantages of low complexity, low device requirement and low cost. The experiment results show that the proposed method has better performance than the MLE algorithm for a common LN model.