With the progress and upgrade of wireless telecommunication technologies, the lack of spectrum resources is pretty serious. As a method to improve spectrum utilization, cognitiveradio has received widespread attentio...
详细信息
With the progress and upgrade of wireless telecommunication technologies, the lack of spectrum resources is pretty serious. As a method to improve spectrum utilization, cognitiveradio has received widespread attention. The technology of spectrumsensing, one of the most significant parts of cognitiveradio, has important research value. The existing spectrumsensing technology is difficult to detect the main user signal at a lower signal-to-noise-ratio(SNR).To solve this problem, wavelet transform and artificial neural network based spectrum sensing in cognitive radio is put forward. The spectrum detection perceptual classifier is designed through the wavelettransform, cyclostationary feature and the artificialneuralnetwork. Firstly, the wavelet threshold method is used to process the received signal. The cyclic characteristic value of the processed signal is calculated. Secondly, the eigenvalues are learned and stored through the artificialneuralnetwork. Finally, the test data is brought into the trained network for classification. Verifying by simulation experiments, the proposed algorithm has better accuracy and effectiveness than the original method in the low SNR environment.
暂无评论