Robust Channel Estimation in WPM Systems

Wavelet packet modulation (WPM) is a potential technique to support high bit rates required wireless multimedia service. Effective channel estimation can improve WPM system performance significantly through channel state information (CSI) acquisition. However the conventional least squares channel estimation is not always appropriate especially in the presence of non-Gaussian channel. In this paper, we drive a Huber channel estimation, which reduces the effect of gross errors sufficiently by choosing proper weight function so as to obtain the best estimate in the non-Gaussian channel. The simulation results demonstrate that the Huber channel estimation method can significantly improve the performance of WPM systems in a non-Gaussian channel.