Signal processing for automotive radar

With rising accident rates, researchers are looking for solutions to reduce fatalities. Some enhance car designs to protect drivers more adequately. Some propose improvements to the traffic and road systems to reduce the chances of accidents. Still others propose the installation of special gadgets to improve the situational awareness of drivers and to alert them to dangerous circumstances. A motor vehicle may be equipped with a radar sensor that checks the spatial environment around the vehicle. The radar is the most commonly adopted sensor for this purpose due to its all-weather capability. In this paper, a 77-GHz FMCW automotive radar signal processor is developed and implemented using the Renesas (previously Hitachi) SH7615 solutions engine development board. The project includes building additional required hardware, the DSP algorithm and a data simulator program to generate test data for testing. The main purpose of the automotive radar signal processor is to detect legitimate targets from unwanted clutter (e.g. road surfaces), and to extract target information from the radar returns. In this paper, the application of the new AND-OR CFAR is also introduced. The simple mathematical fusion models of the AND-OR CFAR are provided and explained here as well. This manuscript is divided into sections. The first section will delve into the basics and the building blocks of the automotive radar. The signal processing portion of this work will be presented in the second section. The data simulator built to simulate testing data is explained in the third section. Results from testing are put forward in section four. Conclusions drawn from the experience are presented in the last section.