Clustered ML Channel Estimation for Ultra-Wideband Signals

The multipath capture of ultrawideband (UWB) communications systems can be dependent on the accuracy of channel estimation. Furthermore, inefficient modeling of the channel often leads to over-parametrization and increased channel estimation error. Relying on a channel model based on clusters, this letter proposes a maximum likelihood estimation strategy which exploits the properties of the UWB channel and offers performance improvement of about 2 dB over less parametric schemes. The robustness of the proposed algorithm in the presence of pulse distortion is investigated and the corresponding BER degradation is observed to be small even when experimental data are employed.