Reciprocal Two Stages Spectrum Sensor to overcome the Noise Uncertainty

Abstract

Noise uncertainty is the most crucial issue that affects the cognitive radio spectrum sensing process. The majority of robust wideband spectrum sensing approaches employ a fixed threshold determined by the amount of noise. These sensing approaches are not accurate since the noise is unpredictable. In this work, a novel spectrum sensing framework was proposed that enhances the detection results in the noise uncertainty presence, and decreases its effect. The proposed framework consists of two different stages; each has a specific rule, the first one is wavelet denoising, the second stage is an adaptive threshold energy detector. The noise information we get in the first sensing process will be used to make the energy detector adapt its threshold in order to overcome the noise uncertainty. The proposed model's detection performance beats a variety of conventional and hybrid techniques in the presence of noise uncertainty. That of ED on its own, and also the previous two-stages spectrum sensing techniques. The suggested model improves detection performance in the noise uncertainty presence, according to simulation findings, and the main goal of this accomplished and assessed effort is to not raise the computational cost.