Waveform Design And Signal Processing For A Novel Automotive Radar

As a measure to enhance the safety of automotive, Automotive Radar has been under development since the mid 1960's. Automotive Radar has not been popular as other automotive electronics apart from the consideration of cost issues, what the principle reason is the low capability of target detection. In this paper, a radar network with four Coded Stepped-FMCW radars is considered.The Coded Stepped-FMCW, which provides high range resolution and good anti- interference capability, is easy to be applied in a single target situation. Considered the real applications, an error approach algorithm for multiple vehicle target detection is proposed.In automotive applications, situations with multiple targets almost always occur. Therefore, a high-performance data association is very important to separate and distinguish between these targets. To avoid errors in the data-association step and resulting ghost targets, this paper describes the bottom-up processing technique. It shows very good results in multi-target situations and reduces the number of ghost targets drastically. Target tracking is most important in multiple-target situations. Therefore, two different tracking algorithms are described and compared in this paper. The Unscented Kalman Filter (UKF) proves to be better than the Extended Kalman Filter (EKF) in the same condition.