Iterative Channel Estimation for Pulse-Based UWB Wireless Communications

Three iterative channel estimation schemes for pulse-based ultra-wideband (UWB) systems are proposed based on multiple iterations of least mean square (LMS) algorithm. The regular sliding window (SW) channel estimator is utilized for computing the initial values of iterative process. Compared to the regular SW method, these schemes can significantly improve estimation accuracy since the effects of the intersymbol interference and multipath interference can be reduced gradually through iterations. Moreover, unlike the successive cancellation (SC) estimation method, the proposed methods save the processing time of channel estimation due to their parallel-type processing manners. In addition, the performance of the proposed hybrid scheme, one of three schemes, can approach that of the SC method w ith iterations and gain a lower complexity. As a result, the proposed methods are more flexible and attractive since they can provide the tradeoff between the complexity and accuracy. Finally, simulations demonstrate the effectiveness of the proposed methods.