Soft-decision based position estimation for wireless localization in indoor scenarios

Wireless localization in non-line of sight scenarios suffers from ambiguous range measurements. To solve the resulting problems, a novel probabilistic localization approach is investigated in this work. Following up the soft-ranging algorithm from a former work, we define an algorithm for soft-decision based position estimation. In this paper, we focus on the calculation and evaluation of direct-path probability density functions. The functions serve as a ranging metric instead of the common single-value range estimates. Supporting multiple range hypotheses with corresponding probabilities used as weights, we expect a major improvement of the localization accuracy in critical non-line of sight conditions. The proposed algorithm is evaluated using the IEEE 802.15.4a channel model.