A Spectrum Sensing Algorithm to Improve Detection Rates in Cognitive Radio Systems

Abstract

To solve the problem of low detection rate of primary user signal under low signal-to-noise ratio in wireless channel environment, a spectrum sensing method of cognitive radio network based on quantum particle swarm optimization extreme learning machine (ELM) algorithm is proposed. According to the characteristics of extreme learning machine algorithm, quantum particle swarm optimization (QPSO) is employed to optimize parameters of extreme learning machine, and QPSO-ELM model with structural risk idea is constructed. The model which reduces the empirical risk of the algorithm improves the generalization ability of the model and improves the spectrum sensing performance of the algorithm. Simulation experiments show that compared with the three machine learning algorithms of artificial neural network (ANN), support vector machine (SVM) and extreme learning machine (ELM), the spectrum sensing performance of the algorithm is improved by 16%, 28% and 9% respectively when the signal-to-noise ratio is -15dB. And the simulation experiments proves that the algorithm proposed in this paper has higher performance under the condition of low signal-to-noise ratio and can effectively realize the spectrum sensing of the main user signal.