频谱细化分数阶Fourier变换的LFM信号滤波算法  

Filtering Algorithm for Noisy LFM Signal Based on Spectrum Zooming Fractional Order Fourier Transform

作　　者：伍龙 黄凯峰 肖理庆 徐小军 WU Long;HUANG Kai-feng;XIAO Li-qing;XU Xiao-jun(School of Mechanical and Electrical Engineering/Huainan Normal University,Huainan 232038,China)

机构地区：[1]淮南师范学院机械与电气工程学院,安徽淮南232038

出　　处：《山东农业大学学报（自然科学版）》2020年第1期163-167,共5页Journal of Shandong Agricultural University：Natural Science Edition

基　　金：安徽高校自然科学研究重点项目(KJ2017A458);淮南师范学院科研重点项目(2016xj02zd)。

摘　　要：针对传统分数阶Fourier变换域含噪线性调频(LFM)信号滤波算法计算量大的问题,提出基于频谱细化与分数阶Fourier变换的改进滤波算法。将含噪的线性调频信号与延迟信号在时域内点对点相乘,相乘后其结果再做FFT变换,估计出LFM信号的调频率范围,然后利用频谱细化提高调频率的估计精度,根据精确调频率计算出分数阶傅里叶变换最佳变换阶次,在最优分数阶Fourier域对含噪LFM信号做窄带通滤波处理。建立MATLAB含噪LFM信号滤波仿真模型,仿真结果表明:改进算法不但计算量明显减少,滤波后的信噪比比传统分数阶Fourier算法提高了近1.41 dB,验证了所提改进算法的有效性。Aiming at a large amount of computation in traditional fractional Fourier transform(FRFT) domain noisy linear frequency modulation(LFM) signal filtering algorithm, an improved filtering algorithm based on spectral refinement and FRFT is proposed. First, the spectrum of the array multiplication of the noisy LFM signal and its delayed version is analyzed to obtain the rough estimate of the chirp rate, and then chirp-Z transform is applied to estimate the chirp rate accurately.Second, the optimum transformation order of FRFT is calculated according to the chirp rate, and then the LFM signal is filtered in the optimal fractional Fourier domain. The simulation model of LFM signal filtering with noise in MATLAB is established. The simulation results show that : The improved algorithm not only reduces the computational complexity obviously, The signal-to-noise ratio(SNR) of the filter is improved by nearly 1.41 dB compared with the traditional fractional Fourier algorithm, The validity of the improved algorithm is verified.

关 键 词：LFM信号 分数阶FOURIER 频谱细化 信号滤波 

分 类 号：TN911[电子电信—通信与信息系统]