HSBCS framework based performance detection for DS-UWB Systems

Abstract

Multi-users interference in direct sequence ultra wideband (DS-UWB) communication is the hotspot research for wireless communication. In this paper, using the wavelet coefficient transform signal in DS-UWB communication, we establish the tree-based hierarchy shrinkage Bayesian compressive sensing (HSBCS) framework for multi-users interference models and noise model generalization. The Markov chain Monte Carlo (MCMC) multi-user interference algorithm by use of HSBCS framework is proposed to suppress noise in the presence of multi-user interference in the DS-UWB communications. The target of this paper is to detect the noise suppression performance of multi-user interference in the DS-UWB communications including normal mean square error (NMSE) performance and peak signal-to-noise-ratio (PSNR) performance using HSBCS framework. Simulation results show that NMSE and PSNR performance of tree-based HSBCS algorithm is better than that of the other algorithms with no tree structure. Otherwise, with more users increasing, the error probability performance of MCMC multiuser interference algorithm in the presence of multiuser interference by use of HSBCS framework is lowest and gradually approach zero.