Sparsity Adaptive Algorithm for Ultra-Wideband Channel Estimation

Ultra-wideband (UWB) channel estimation based on the theory of compressive sensing needs to predict sparsity of the channel. Considering the sparseness of the UWB channel in time domain, the problem of channel estimation can be transformed into the reconstruction of the sparse vector in compressive sensing theory. Sparsity adaptive regularization compressive sampling matching pursuit (SARCoSaMP) algorithm is proposed in this paper. The ideas of adaptive and regularization are introduced based on compressive sampling matching pursuit (CoSaMP) algorithm. The number of the selected atoms is controlled automatically in order to approach channel sparsity K gradually. The UWB channel is estimated accurately although the sparsity of the channel is not available. Results show that the proposed algorithm can be effectively used in ultra-wideband channel estimation and it is significantly superior to CoSaMP and sparsity adaptive matching pursuit (SAMP) algorithm.