基于高阶累积量的调制方式识别研究  

Study on Modulation Recognition Based on Higher-order Cumulants

作　　者：李跃[1] 郭兴吉 赵欣[1,3] LI Yue;GUO Xingji;ZHAO Xin(City College of Southwest University of Science arid Technology,Mianyang 621000,Sichuan,China;Mianyang Normal University,Mianyang 621000,Sichuan,China;Mianyang Focational and Technical College,Mianyang 621000,Sichuan,China)

机构地区：[1]西南科技大学城市学院,四川绵阳621000 [2]绵阳师范学院,四川绵阳621000 [3]绵阳职业技术学院,四川绵阳621000

出　　处：《西南科技大学学报》2018年第3期64-68,共5页Journal of Southwest University of Science and Technology

基　　金：国家自然科学基金项目(61072138)

摘　　要：通信信号调制方式自动识别技术在非协作通信的信号识别中具有重要作用,基于高阶累积量的方法,对常见的8种数字信号的调制方式识别进行了研究。调制方式包括2ASK,BPSK,QPSK,8PSK,16QAM,32QAM,16APSK,32APSK,给出了各调制信号的二、四、六、八阶累积量理论值,提取5个特征参数,根据不同调制信号的特征参数差异情况,设计合理的分类决策树和阈值对信号进行有效识别。仿真结果显示,在信噪比大于-3 d B时,信号识别率可达90%以上。Modulation automatic recognition technology of communication signals plays an important role in non-cooperative communication signal recognition. In this paper,modulation recognition of eight common digital signals was studied based on the high-order cumulant method. The modulation methods included 2 ASK,BPSK,QPSK,8 PSK,16 QAM,32 QAM,16 APSK and 32 APSK. The theoretical values of the second,fourth,sixth and eighth cumulants of each modulated signal were calculated and five characteristic parameters were extracted. A reasonable classification of decision trees and thresholds were then designed for effective signal identification according to the differences of characteristic parameters of different modulation signals. The simulation results show that the signal recognition rate can reach more than90% when the SNR is more than-3 d B.

关 键 词：高阶累积量 调制识别 APSK 特征参数 

分 类 号：TP311[自动化与计算机技术—计算机软件与理论]