频谱感知是认知无线电系统的关键技术之一,针对基于支持向量机的频谱感知方法中核函数选取的单一性和核函数参数的不确定性,提出一种基于混合核函数支持向量机的频谱感知算法,将两种核函数混合构造新的核函数,采用量子粒子群算法对其中的参数进行优化,并引入主成分分析方法对样本进行降维并提取其全局特征。实验结果表明,该模型较传统方法在低信噪比下无线环境中的分类精度上有了明显提高,在信噪比为-10 dB的无线环境中能完全识别出主用户,为频谱感知提供了一种可靠性高的设计方案。
Spectrum sensing is one of the key technologies in cognitive radio systems .For the oneness selection of kernel function and the uncertainty of kernel function parameter in support vector machine (SVM ) based spectrum sensing ,a modified spectrum sensing algorithm based on SVM with hybrid kernel function was proposed .It mixes two kernel function to construct a new kernel function ,and uses the quantum particle swarm optimization (QPSO ) to optimize the parameters in this kernel function .principal component analysis (PCA ) algorithm is used to reduce the dimension of samples and the global features are extracted .Results show that the new model has great accuracy than SVM‐based spectrum sensing model under low signal‐to‐noise ratio (SNR) and could identify the primary user under-10 dB ,providing a compact scheme for spectrum sensing .