基于PSO-LSSVM的非重构宽带压缩频谱感知方法  

Non-reconfigurable Broadband Compressed SpectrumSensing Method Based on PSO-LSSVM

作　　者：殷晓虎[1] 李加美 谢豪 YIN Xiaohu;LI Jiamei;XIE Hao(School of Communication and Information Engineering,Xi’an University of Science and Technology,Xi’an 710600,China)

机构地区：[1]西安科技大学通信与信息工程学院,陕西西安710600

出　　处：《无线电工程》2023年第5期1207-1213,共7页Radio Engineering

基　　金：陕西省科技计划项目(2020GY-029)。

摘　　要：频谱感知是认知无线电的关键技术之一,而宽带环境使采样设备压力大,宽带压缩频谱感知有效地解决了此问题,但压缩重构造成巨大计算复杂度,并且传统检测方法受门限限制,在低信噪比下检测率低。提出将机器学习与非重构压缩感知技术应用到频谱感知当中的方法,原始信号经压缩感知得到测量值,对其进行采样协方差预处理后,通过改进PSO-LSSVM算法对数据进行训练分类,解决门限困扰,并适用于宽带频谱感知。仿真结果表明,此方法在低信噪比下,与传统检测相比有较好的检测结果。Spectrum sensing is one of the key technologies of cognitive radio,and the broadband environment puts great pressure on the sampling equipment.Broadband compressed spectrum sensing effectively solves this problem,but compressed reconfiguration causes huge computational complexity.The traditional detection method is limited by the threshold,so the detection rate is low in low signal-to-noise ratio.A method of applying machine learning and non-reconfigurable compressed sensing technology to spectrum sensing is proposed.The original signal is compressed to get the measured value,and after preprocessing the sampling covariance,the improved PSO-LSSVM algorithm is used to train and classify the data,which solves the threshold puzzle and is suitable for broadband spectrum sensing.The simulation results show that this method has better detection results compared with traditional detection in low signal to noise ratio.

关 键 词：宽带频谱感知 PSO-LSSVM 压缩感知 非重构 

分 类 号：TN92[电子电信—通信与信息系统]