Pilot symbols distribution for compressive sensing based NBI mitigation in UWB systems

Abstract

Ultra wideband (UWB) technology is promising a cutting edge in delivering high data rate for short range wireless communication systems. Because of their large bandwidth, UWB signals may encounter some problems especially with high sampling rate requirements. Moreover, coherence existence with other narrowband systems is a major concern which needs to be addressed through proper mechanisms. Since narrowband interference (NBI) signals have sparse representation in the discrete cosine transform (DCT) domain, they can be estimated and suppressed using Compressive Sensing (CS). CS also has the ability to reduce the high sampling rate requirements. For training based NBI mitigation with CS, three pilot groups symbol are used to estimate the NBI signal subspace, UWB signal subspace, and provide information about the channel. In this paper, the distribution of pilot symbols among the three groups is investigated in the presence of strong NBI. The investigation is based on the bit error rate (BER). The influence of each pilot group symbols is also studied. Simulation results show that the third pilot group symbols is the most dominant one; hence more symbols should be assigned to estimate the channel information.