基于凸优化的MIMO雷达的优化滤波器组设计

Optimal filter bank design for MIMO radar based on covex optimization

在正交波形MIMO雷达信号处理中,发射信号分离是其中的一个重要环节,一般可以通过匹配滤波器组来实现,而匹配滤波器组对信号分离是基于发射信号满足理想正交条件的,实际中由于很难得到理想的正交波形集,因此将会造成信号分离能力下降或分离不彻底,为此,提出一种优化滤波器组设计方法,以最小化滤波输出峰值旁瓣电平为目标函数,采用凸优化方法进行全局优化设计,计算机仿真表明,该方法所设计的优化滤波器组可以显著提高非理想正交波形MIMO雷达的信号分离能力。

The extraction of transmitting signals is an important step in signal processing for MIMO radar.Usually,it can be realized by matched filter bank which principle is based on ideal orthogonal condition for transmitting waveforms.But it is hard to get ideal orthogonal waves.So with the matched filter bank,the signal extracting capability descends acutely.Therefore,a new method to design optimal filter bank is proposed.The object function is formulated through minimizing peak sidelobe level of filter output.Covex optimization is used to get the global optimal filters.The computer simulations verify that the optimal filter bank can improve the extracting capability for MIMO radar with nonideal orthogonal transmitting waves.