基于短时傅里叶变换与相关解调的参数估计算法

Parameters Estimation Based on Short Time Fourier Transform and Correlation Demodulation

针对傅里叶变换无法估计对称三角线性调频连续波信号的调制周期,短时傅里叶变换参数估计精度低的问题,提出了基于最佳频率分辨率窗宽的高斯窗短时傅里叶变换与改进的相关解调相结合的参数估计算法。该算法首先使用基于瞬时频率的窗宽倍增搜索算法获取最佳频率分辨率窗宽,其次使用高斯窗短时傅里叶变换求得对称三角线性调频连续波信号的中心频率和调制带宽,再使用改进的相关解调求出信号的调频斜率,进而求出信号的调制周期。仿真结果表明:本文算法不但提高了短时傅里叶变换对于对称三角线性调频连续波信号中心频率,调制带宽和调制周期的估计精度,而且具备良好的实时性。

Aiming at the problem that Fourier transform couldn＇t estimate the modulation period of symmetrical triangular linear frequency modulation continuous wave（STLFMCW）signal,and short time Fourier transform（STFT）couldn＇t estimate parameters accurately,an estimation method based on optimal frequency resolution ratio gauss window width STFT and improved correlation demodulation was proposed.Firstly,the method acquired the optimal frequency resolution ratio window width through the measure of window width doubly searching,and then made use of the gauss window STFT to obtain the STLFMCW signal＇s center frequency and the modulation bandwidth.Then for the purpose of attaining the signal＇s modulation period,the method got the signal＇s frequency modulation gradient by means of the improved correlation demodulation.MATLAB simulation indicated that the method could improve the STFT parameters estimation precision in a short time.