基于短时分数阶傅里叶变换域滤波的多项式相位信号时频检测

The use of short time fractional fourier transform domain filtering method for multi-polynomial phase signals detection in low SNR condition

针对低信噪比下多个多项式相位信号提出了一种基于短时分数阶滤波的时频检测方法。分数阶傅里叶变换作为一种线性变换,能够实现线性调频信号检测与分离。而多项式相位信号在短时间内可以由线性调频信号提供良好的近似,故可以采用短时分数阶傅里叶变换实现多线性调频分量的检测与分离。对每个短时信号的时频分析进行叠加组合,即得到多个多项式相位信号的时频分析检测。计算机模拟仿真证明了此方法的有效性。

A method for multi-polynomial phase signals (PPS) detection in low signal to noise condition is proposed. As a linear transform, fractional Fourier transform (FrFT) can detect and separate LFM signals in multi-LFM signals condition. On the other hand, PPS can be view as LFM signal in a short time. So, short time fractional Fourier transform (STFrFT) filtering method is used to separate the PPS from noisy environment. After that, we fit the time-frequency representation together, and get the results. The validity of this method has been show by computer simulation.