Adaptive Two Thresholds Based Energy Detection For Cooperative Spectrum Sensing

Abstract

In cognitive radio networks, secondary users need to conduct spectrum sensing to properly detect the presence of primary signals that may have much lower power than noise plus interference power. In such a case, time-varying noise plus interference, which is briefly called noise uncertainty in this paper, can substantially degrade the sensing reliability of hard information combining (HIC) and soft information combining (SIC). To improve the sensing reliability in the circumstance with heavy noise uncertainty, we propose an adaptive two thresholds based energy detection and a two stage HIC based cooperative decision for cooperative spectrum sensing. The proposed sensing technique has shown better performance than conventional HIC and comparable performance with SIC when a small number of sensing nodes are used in spectrum sensing.