联合循环平稳特征PCA与RVM的频谱感知

针对无线信道环境中低信噪比情况下主用户信号检测率较低的问题，提出了一种基于循环平稳特征主成分分析(PCA)与相关向量机(RVM)的认知网络频谱感知算法。该算法结合了主成分分析算法与相关向量机分类方法，应用于解决认知网络频谱感知问题。首先对信号循环平稳特征参数进行特征提取，通过主成分分析进行降维提取信号主成分，生成训练样本和待测样本，并完成对相关向量机的训练，再采用训练完成的相关向量机算法分别对有无主用户情况下的信号进行分类检测，最后获得主用户信号存在性的感知判断。仿真实验表明，与人工神经网络、支持向量机和最大最小特征值算法相比较，所提算法在低信噪比情况下具有较高的分类检测性能，检测率最大可提高61.6%，有效地实现了对主用户信号的感知。

Spectrum sensing by combining cyclostationary PCA with RVM.

For the low accuracy rate of the primary user detection in the wireless channel environment,this paper proposes a method based on cyclostationary principal component analysis(PCA) and relevant vector machine(RVM) for spectrum sensing under the low signal to noise ratio(SNR) environment in cognitive radio.This method combines PCA with the RVM classification algorithm to solve spectrum sensing problem in cognitive network.A set of cyclic spectrum features are first calculated,and the PCA is applied to extract the most discriminate feature vector as training samples and testing samples for classification.The RVM is trained by training samples.Finally,the trained RVM is utilized to detect and decide the existence of the primary user.It is observed that the maximum increase of the detection probability of the proposed algorithm can be increased about 61.6% in comparison with artificial neural network(ANN),support vector machine(SVM) and maximum-minimum eigenvalue(MME).Simulation results show the proposed algorithm can effectively detect primary user signals.