摘要
信噪比低于-5 dB时,信号恢复难度大.提出了一种新的信号估计算法,基于加窗Wigner-Ville分布(pseudo Wigner-Ville distribution,PWVD)的时频峰值滤波(time-frequency peak filtering,TFPF)PWVD-TFPF,即利用PWVD与信号时频的关系设计一种计算窗函数长度的方法,同时将此窗与魏格纳分布结合用于TFPF.TFPF是通过时频分布重建信号.首先将信号经过FM调制为解析信号,采用PWVD集中信号的能量,而噪声能量均匀分布于整个时频平面,提高了信号的信噪比,可准确估计出信号的频率.将时频分析方法用于低信噪比实信号估计中,使频率估计在信噪比低于0 dB时依然可行.此设计能够在信噪比达到-8 dB时恢复信号,并且信号的频率估计结果基本达到Cramer-Rao lower Bound(CRLB).
In the signal to noise ratio is less than-5 dB,the signal recovery is very difficult.A new signal estimation algorithm is proposed that time-frequency peak filtering(TFPF)based on the pseudo Wigner-Ville distribution(PWVD)ie PWVD-TFPF.That is,a method of calculating the length of the window function is designed when using the relation of PWVD and signal frequency which is combined with the Wegener distribution for TFPF.TFPF can reconstruct the signal by time-frequency distribution.Firstly,the noisy signal is modulated by FM as the analytic signal,and then the energy of the signal is concentrated by PWVD,while the noise energy is evenly distributed over the whole time-frequency plane,which improves the signal-to-noise ratio of the signal and accurately estimate the frequency of the signal.The above time-frequency analysis method is used in the real signal estimation so that the frequency estimation is still feasible when the signal-to-noise ratio(SNR)is less than 0 dB.This method can recovery the signal in noise level down to a SNR of-8 dB,and the frequency estimation result of the signal basically reaches Cramer-Rao lower Bound(CRLB).
作者
刘亚南
赵加祥
王增科
许薇
Liu Yanan;Zhao Jiaxiang;Wang Zengke;Xu Wei(Signal Processing and Sensor Network Laboratory,College of Electronic Information and Optical Engineering,Nankai University,Tianjin 30071,China;School of Electronics and Information Engineering,Tianjin Polytechnic University,Tianjin 300387,China)
出处
《南开大学学报（自然科学版）》
CAS
CSCD
北大核心
2020年第3期1-4,共4页
Acta Scientiarum Naturalium Universitatis Nankaiensis
基金
天津市应用基础与前沿技术研究计划(14JCYBJC16100)
国家自然科学基金(61501324)。
