基于参数预估计和滑动FFT的MFSK信号类内识别方法  

A within Class Recognition Algorithm based on the Parameter Pre-estimation and Sliding FFT

作　　者：张华娣[1] 楼华勋[1,2] ZHANG Huadi;LOU Huaxun(No.36 Institute of CETC,Jiaxing Zhejiang 314033,China;Science and Technology on Communication Information Security Control Laboratory,Jiaxing Zhejiang 314033,China)

机构地区：[1]中国电子科技集团公司第三十六研究所,浙江嘉兴314033 [2]通信信息控制和安全技术重点实验室,浙江嘉兴314033

出　　处：《通信技术》2021年第12期2602-2605,共4页Communications Technology

摘　　要：针对在调制指数h较小的情况下,对多进制频移键控(M-ary Frequency Shift Keying,MFSK)信号的识别正确率不高的问题,提出了一种基于参数预估计和滑动快速傅里叶变换(Fast Fourier Transform,FFT)的MFSK信号类内识别方法。该方法先估计出MFSK信号的码元速率,再计算码元宽度内信号的频率,并对各码元的频率种类进行归并,最后按照频率种类数来实现MFSK信号的类内识别。对该方法的适应性的理论分析和计算机仿真结果表明,该方法大大提高了MFSK信号的识别正确率,并有效降低了识别的信噪比门限。Aiming at the shortcomings of low recognition accuracy on the MFSK(M-ary Frequency Shift Keying)signals under the condition of the relatively small modulation index,this paper proposes a within class recognition algorithm based on parameter pre-estimation and sliding FFT(Fast Fourier Transform).This algorithm first estimates the symbol rate of the MFSK signals,and then calculates the spectrum peak of the signals within every symbol.All the frequency of every symbol is calculated and merged.Finally,the number of frequency type is used to carry out the MFSK signals within classification.The results of the theory analysis and simulation indicate that the algorithm can greatly improve the accuracy of identification in MFSK signals and decrease the SNR threshold of identification effectively.

关 键 词：滑动快速傅里叶变换 多进制频移键控 码速率 识别 

分 类 号：TN911[电子电信—通信与信息系统]