Cooperative spectrum sensing algorithm based on improved energy detection and Sevcik fractal dimension

In cognitive radio networks, the single-node spectrum sensing is susceptible to noise uncertainty, and the traditional energy detection method has poor detection performance in high noise power scenarios. As the Sevcik fractal dimension(SFD) is insensitive to noise and can distinguish signal from noise waveform, a cooperative spectrum sensing (CSS) method based on energy detection and SFD is proposed by combining adaptive threshold energy detection method SFD method. Firstly, the received signal is detected and judged by energy detection method, and then the signal that primary user (PU) does not exist is rechecked by SFD. Finally, all the detection results are fused by K-out-of-N rule, and final decision is obtained according to fusion result. Simulation results show that the proposed spectrum sensing method is insensitive to noise, and the detection performance is significantly improved at low signal-noise-rate (SNR).