An N-Selective MRC Rake Receiver with LMS Adaptive Equalizer for UWB Systems

This paper proposes a receiver to reduce 181 for high speed UWB transmission. The structure consists of an N-selective maximum ratio combiner (MRC) rake receiver and a least-mean-square (LMS) equalizer. This sub-optimal rake receiver processes only a subset of the available resolved multi-path components. A sliding correlation algorithm is used to obtain parameters for the N strongest paths. The LMS equalizer is designed to assist the receiver to reduce ISI. Benefit of this adaptive scheme is the equalizer does not need to know parameters of the entire multi-path component. A short training period with known information sequence is required to initially adjust the tap weights. The equalizer acts as a linear filter to suppress ISI. Simulation results for N=8, 16, and 32 at different values of SNR show significant improvement in BER. There is only a small trade off in system complexity to obtain such gain