基于小波变换的通信信号符号速率估计算法研究  

Research on Symbol Rate Estimation Algorithm Based on Wavelet Transform

作　　者：石昊翔 洪玮 张明社[1] 吴怡[1] Shi Haoxiang;Hong Wei;Zhang Mingshe;Wu Yi(The 39th Research Institute of China Electronics Technology Group Corporation,Xi'an)

机构地区：[1]中国电子科技集团公司第39研究所,西安710065

出　　处：《测控与通信》2023年第4期35-40,共6页

摘　　要：非协作条件下的通信信号符号速率估计是后续信号解调与译码的重要基础,对无线电频谱监测与通信信号侦察等任务有着重要的意义。常见的符号速率估计算法可大致分为3类:基于信号时域瞬时特征的估计算法、基于包络检测或循环检测的非线性变换算法、基于小波变换的算法。其中前2类算法在低信噪比下性能下降明显。为解决这一问题,本文对基于小波变换的符号速率估计算法进行了深入的研究,讨论了母小波函数的选择与小波变换后处理算法对符号速率估计的性能影响。通过理论分析与仿真实验验证了本文提出的算法在较低信噪比、较低数据量的情况下仍能保持较高的估计准确率和鲁棒性。The symbol rate estimation of communication signals under non-cooperative conditions is an important basis for signal demodulation and decode.It has significant influence on the tasks such as radio spectrum surveillance and communication reconnaissance.The algorithms on symbol rate estimation could be divided into three types roughly.One is based on transient features of the signal in time filed,and one is based on cyclic correlation or envelope spectrum,and another one is based on wavelet transform.The first two types of algorithms suffer obvious performance decline when the SNR is low.To solve the problem,the algorithm based on wavelet transform is studied.The optimal wavelet function choice is deduced,and two different post-processing algorithms are studied.Theoretic analysis and simulation show that the proposed algorithm and parameter configuration has many advantages such as high estimation accuracy,low dataset requirement and good robustness in low SNR environment.

关 键 词：小波变换 符号速率 信号参数估计 

分 类 号：TN9[电子电信—信息与通信工程]