Multi-feature collaborative spectrum sensing method based on support vector machine

To solve the shortage of spectrum resources and improve the utilization of spectrum is a new challenge for current wireless communication services. The traditional spectrum sensing method is limited by uncertain noise, user prior information and other factors, and the detection performance is poor under low signal-to-noise ratio. The decision threshold of the blind spectrum detection method based on eigenvalues is approximate, and the detection performance is affected. To this end, a multi-feature collaborative spectrum sensing method based on support vector machine is proposed. The new feature is constructed by weighted fusion of the energy values of multiple cognitive users. We use the new feature and the detection statistics in the blind detection method based on eigenvalues as the feature vector. The support vector machine model is obtained by training, and then used for spectrum sensing. The simulation results show that the detection accuracy of the proposed method in this paper has great improvement under low signal-to-noise ratio. When the signal-to-noise ratio is -10dB, the detection probability reaches 95%. Improvement of new feature in the proposed method can reduce the amount of calculation and shorten the sensing time.