Weighted Energy Detection Receivers for NC-UWB Green Communications Using Parameter Optimized Improved Normalized LMS Algorithm

Abstract

Energy detection (ED) receiver is one of two popular receivers for non-coherent (NC-) ultra-wideband (UWB) Green Communications. In this paper, weighted ED (WED) receivers are further investigated for NC-UWB systems using different adaptive filtering algorithms. In the conventional normalized least mean square (NLMS) algorithm, a fixed step-size is used to update the weighting coefficients of WED receivers, and therefore there must be a trade-off between the steady state error and the convergence speed. Correspondingly, an improved NLMS (INLMS) algorithm was recently proposed to address this issue. In this paper, parameter optimization is discussed for the INLMS algorithm and accordingly the parameter optimized (PO-) INLMS algorithm is presented. Moreover, convergence behavior and bit error rate (BER) performance are evaluated for adaptive WED receivers. Simulation results show that compared to the traditional NLMS algorithm, PO-INLMS possesses significant benefits in convergence rate together with about 0.3-dB-gain in terms of the BER performance, and therefore the PO-INLMS based WED receiver can be considered as a promising candidate for energy-aware NC-UWB Green Communications.