Multipath Assisted Positioning with Band-Limited Signals in an Urban Environment

While location-awareness in cellular radio networks has arisen more and more interest in recent years, positioning using cellular radio signals might show very weak performance due to shadowing and multipath propagation, especially in urban and indoor scenarios. Beyond, the number of base stations within communication range is often too low for positioning. The standard methods for positioning try to mitigate the effect of multipath propagation on the line-of-sight path. Though, various recently published results show that multipath components arising from reflections and scattering can actually be exploited, and hence the number of base stations required for positioning can be decreased. However, these results are based on the assumption of a high system bandwidth, e.g. 100 MHz or more. This is much higher than the bandwidth of cellular signals currently deployed. By performing measurements where we emulate a multipath scenario, we show that the tracking of a receiver is possible using a 3GPP-LTE system of a bandwidth of only 20 MHz in a simple urban scenario. We apply an advanced signal processing algorithm to track multipath components impinging at a moving receiver, and feed those results into a particle filter for position estimation. Our results indicate that tracking the receiver is possible with only two base stations, even if one of them is not in line-of-sight. The positioning error is always below 7 meters.