Low Complexity Adaptive Spectrum Sensing using Modified FRM Filter Bank

Abstract

ABSTRACT Spectrum Sensing (SS) has been profoundly used in the Cognitive Radio (CR) framework due to its ability to identify spectrum holes and avoid intrusion for Primary Users (PUs). Several SS approaches based on filter banks (FBs) have been proposed in the literature. These techniques have the benefit of reusing the sensing filter bank for transmission as well, thereby providing an additional way to communicate at no extra cost. But the implementation complexity issues involved with the filter banks for CR were not taken into account. In this paper, we propose a Wideband Spectrum Sensing (WSS) based on a low complexity reconfigurable Modified FRM filter bank (ModFRM-FB) for detecting multiple active channels simultaneously. The proposed method uniformly divides the wideband signals into multiple-narrowband signals and each sub-band is sensed to evaluate the occupancy status of the sub-band by energy detection. Unlike traditional sensing algorithms, an adaptive threshold paradigm for energy detection is used to modify the threshold according to the noise power. This improves the detection performance even at low Signal-to-Noise Ratio (SNR) levels. The hardware complexity of our proposal is found to be comparatively lower than previous techniques in the literature.