Design of Orthogonal Matching Pursuit (OMP) for Sub-Nyquist Wideband Spectrum Sensing in Cognitive Radio on FPGA

Abstract

Cognitive radio (CR) is one of the best option for enhancing spectrum usage in the future generation of cellular-based networks. An essential stage of cognitive radio is the spectrum sensing process, which detects unused spectrum holes in order to use all available frequencies of the radio spectrum. The sampling rate over a multi-GHz bandwidth is the main challenge in the spectrum sensing architecture design. More recently, to be able to use a low sampling rate, Compressed Sensing (CS) technology has been used. In this paper, an enhanced spectrum sensing architecture for reconstructing the original signal is presented. The architecture is based on a modified Orthogonal Matching Pursuit (OMP) algorithm. The proposed architecture is implemented with 1024 samples, a measurement vector with a size of 256 and a sparsity of 36. The proposed architecture can be executed on 132 MHz clock frequency, using Xilinx Virtex 6 FPGA, with a speed up of 32% in comparison with previously published work.