Region of Interest Based Adaptive High Resolution Parameter Estimation with Applications in Automotive Radar

Abstract

High resolution processing for automotive radar is required to enhance the probability of target detection and classification in proceeding stages of driver assistance systems. As a result of the large computational demand associated with high resolution processing of the entire data set, real-time application of such methods remains a challenge. This paper presents a computationally efficient approach to achieve this goal by only processing Regions of Interest with the capability of utilizing a multi-dimensional variable resolution to meet necessary requirements for different applications.