Fig. 3

The curve of the relationship between SNR and detection probability. Considering the change of SNR between – 15 db and 5 dB, we can compare the detection probability of the paper algorithm and the classical algorithm in the case of different number of cooperative users. After the cognitive user obtains the local detection results through compressed sensing, the detection probability is obtained according to the actual spectrum occupation of the channel and the simulated spectrum occupation. With the increase of the number of cooperative users, our method has higher detection probability than the traditional method. However, no matter it is single-user detection or multi-user detection, it is not difficult to find that when the SNR is very low, the detection probability is very low between – 15 db and 10 dB, and the performance of cooperative spectrum detection does not have many advantages. This low SNR state is equivalent to a severe shadow-fading or other fading in real environments. Cognitive users in this environment will affect the performance of cooperative spectrum detection and even reduce the detection performance of cognitive users with high SNR. Therefore, synergetic spectrum detection not only improves the detection performance, but also reduces it