Performance evaluation of non-Gaussian channel estimation techniques on ultrawideband impulse radio channels

摘要

We study the problem of adaptive channel estimation for a multipath channel on an ultrawideband (UWB) impulse radio (IR) system. The purpose of the work is to demonstrate that the statistical characteristics of the UWB-IR environment are strongly non-Gaussian and that as a result, non-Gaussian signal processing techniques are both efficient and powerful in a UWB-IR environment. The multipath channel estimation problem has been chosen as an illustrative example because UWB-IR systems often operate in complex multipath environments. We study both blind and training-based estimation techniques and demonstrate that even simple non-Gaussian strategies can achieve significant performance improvement when compared to more common second-order estimation techniques.