Estimation and Modelling of NLoS Time-Variant Multipath for Localization Channel Model in Mobile Radios

Time Based (TB) localization in terrestrial communications mobile radio as a complementation to global navigation satellite systems has gained recently plenty of interests. As an essential tool to develop suitable algorithms for joint communications and localizations in mobile radio networks, the wireless channel model has a growing significance. For both communications and localizations, the time evolution of multipath components is essential especially for tracking applications. Moreover, the Non Line-of-Sight (NLoS) error due to an undetectable line-of-sight path introduces an additional range error for TB localizations, which needs to be taken into account in the channel model. In this paper we present a space-alternating generalized expectation-maximization based Kalman filter method to efficiently estimate and track the time-variant multipath components based on a channel measurement campaign. The modeling of time-variant multipath characteristics, in terms of NLoS error, variance of path number, and the life span, are presented in this paper as well.