Segmented CS with measurement allocation for UWB channel estimation

Abstract

Compressed sensing (CS) is an effective way to lower the sampling rate of UWB signals. Segmented compressed sensing (SCS) can reduce the storage space and computation complexity greatly. This paper presents a new SCS scheme with measurement allocation for UWB channel estimation. The received pilot signal is divided into segments of equal size as in SCS. To improve the channel estimation accuracy, a method is proposed to allocate measurements among segments according to the average power delay profile (APDP) of UWB channel. Bernoulli random matrices are generated to perform sensing and orthogonal matching pursuit algorithm is employed for signal recovery. Simulation results show that better channel estimation accuracy can be achieved by using our proposed scheme.