基于先验信息的自适应核函数设计

Design of Adaptive Kernel Based on Priori Knowledge

针对多分量线性调频信号 ,提出一种新的自适应核函数 ,并给出了有效的核参数估计方法。相对于自适应高斯核函数而言 ,该核函数既能获得更好的交叉项抑制效果 ,又避免了较大的信号能量损失 ,因而是一种更好的核函数。计算机仿真结果验证了其有效性。

In this paper, based on the priori knowledge, a new adaptive kernel for reducing the cross terms of WVD and an efficient kernel estimation method are presented. Compared to adaptive Gaussian kernel, the new adaptive kernel has better performance in reducing cross terms of WVD and avoiding large loss of signal energy than the adaptive Gaussian kernel. Simulation results confirm its better performance.