Novel energy detection scheme in cognitive radio

Cognitive radio, as a prominent technology to solve frequency scarcity by dynamical spectrum access, attract many research interests. In order to fill voids in the wireless spectrum while cleverly avoid interference to the existing communication system, real-time spectrum sensing is quite necessary for a cognitive radio system. Through study about performance of energy detection, it finds that the larger the noise fluctuations, the sharper decline exhibit in energy detection performance, especially at low SNR. In this paper, the detection threshold was calibrated to reduce detection error due to noise uncertainty. The sparse nature of energy change is exploited to amend the judgment result. Simulation results show that under guarantee the advantage of the traditional energy detection, the proposed differential energy detection can effectively improve accurate detection performance of the idle spectrum for the cognitive users in real-time.