Research on The Multi-target Recognition of Millimeter Wave LFMCW Automotive Anti-Collision Radar

For linear frequency modulated continuous wave automotive collision avoidance radar system, it has advantages of not existing the blind area, high detection sensitivity, not easy to be interfered and low operating voltage. It becomes the focus of research in recent years. In practice, we often meet the case that there a multi-targets in front of radar. Generally, the radar can distinguish the targets by using stretching and FFT processing. When there are two similar distance and different speeds targets in front of the radar, their one-dimensional FFT spectrum will overlap. In this paper, due to the different speeds of the targets, the speeds can be calculated by the application of the two-dimensional FFT conversion. We use CZT transform to extract the amplitude information of each target. Then the difference between the amplitude of the spectrum is normalized, and the distance of the target is corrected. So the radar can recongize the targets with similar distance but different speed. It greatly enhances the distance measurement accuracy of each target.