该文针对线性调频信号,提出一种基于分数阶波包变换分析方法——高斯短时分数阶傅里叶变换。通过旋转角度的搜索及高斯窗口宽度的调整,能够在低信噪比条件下对多分量LFM信号进行检测,避免交叉项的出现,并能得到参数的估计值。通过推导分析给出变换结果的解析表达式,计算机仿真结果也表明了该变换的有效性。
Based on fractional wave packet transform, a new hybrid time-frequency transform: Gauss Short-time Fractional Fourier Transform (GSFrFT) is proposed for the detection and estimation of Linear Frequency-Modulated (LFM) signals. Multiple LFM signals can be detected in low SNR by choosing the rotated angle and cross-items can be avoided. Analytical expression of the transform result is given. The simulation results also show that the GSFrFT is valid.