UWB based pre-localization algorithm for aiding target location in a multipath environment

Abstract

When a UWB signal is transmitted, we expect there to be a number of reflected multipath signals because of reflections due to objects in the environment. As a result of these reflected signals, it is impossible to define the ellipses that would be used for target detection as described by existing elliptical based localization schemes when we consider a highly multipath rich environment such as a typical UWB propagation channel unless we are able to manipulate the multipath propagation scenario and reduce it to a two-path propagation one. To tackle this problem, a pre-localization algorithm is proposed. In our proposed algorithm, we use the reflection properties of UWB signals to extract information from the reflected signals in the multipath environment and ultimately reduce the multipath propagation scenario into a two-path one based on these extracted information. Our process of extraction involves the sampling of the received signals at three receivers during regular intervals, correlating the sampled signals with a predefined database of template reflected signals; and finally using a decision engine to determine the signals that would be used for generating the ellipse for target localization. Consequently, we were able to differentiate multipath signals from one another using the relationship between individual multipath signals and their respective angle of incidences; and ultimately select a set of reflected signals that would be suitable for the localization process described by existing elliptical based localization schemes.