Adaptive stochastic resonance aided energy detection with modified periodogram

Spectrum sensing is a fundamental technology to detect the presence of primary user (PU) in cognitive radio. Usually, it requires that the spectrum sensing scheme has a good performance even in extremely low signal-to-noise ratio (SNR) environments. In this paper, we proposed a novel stochastic resonance (SR) aided spectrum sensing scheme to meet this requirement. After analyzing the optimality of SR effect in signal detection, we proposed an adaptive SR (ASR) aided spectrum sensing framework, in which the parameters of SR system is adaptively adjusted according to the power of noise. Furthermore, by filtering out the energy due to direct current (DC) component of ASR system output signal in frequency domain, we propose a modified energy detection based on periodogram (P-ED) so that much better performance can be achieved under very low SNR region. Compared with the existing energy detection (ED) based on SR, simulation results show that the proposed novel spectrum sensing method can significantly improve the detection performance under very low SNR region.