Optimization of the parameters in SVM-based spectrum sensing model

摘要：

　　The process of spectrum sensing plays a key part in Cognitive Radio (CR).In this paper,Support Vector Machine(SVM)is applied to spectrum sensing which can judge whether the Primary User (PU) exists or not for its good performance in many fields.Data set including training set and testing set with different characteristics is generated by laboratory instruments to verify the SVM-based spectrum sensing model of good performance in actual communication environment.To solve the problem of huge training set,the paper focuses on researching the optimal parameters of the number and dimension of training set.The simulation results show that the optimal parameter of Radial Basis Function (RBF) and the size of training set and testing set are different with different characteristics of training set and testing set.

关键词： Spectrum Sensing Support Vector Machine Primary User Radial Basis Function

作者: Xuping Zhai Tian Meng Bingbing Yang

作者单位: Key Laboratory of Specialty Fiber Optics and Optical Access Networks Shanghai University, 200072 Shanghai, China

会议类型: 国际会议

会议名称: 2016 2nd International Conference on Mechanical, Electronic and Information Technology Engineering(2016机械、电子和信息技术国际会议)

会议地点: 重庆

会议语种:英文

页码: 329-332

在线出版日期: 2016-03-21（万方平台首次上网日期，不代表论文的发表时间）