A New Way of Ultra-Wideband Channel Estimation Based on Bayesian Compressive Sensing

In this paper, in order to solve the problem that the sampling rate in ultra-wideband (UWB) channel estimation is too high, we discuss the applicability of Bayesian Compressive Sensing (BCS) used in UWB channel estimation. We solve the problem by using the time domain sparse of the impulse response of the UWB channel and establishing the probability model of the Compressive Sensing (CS) measurement. We accomplish the channel estimation by optimizing maximum a posteriori (MAP) of the channel. The simulation results show that the proposed scheme needs a very low sampling rate to recover the channel accurately. And the BCS algorithm has a better performance than the basis pursuit (BP) algorithm and the traditional least square (LS) algorithm in bit error rate (BER).