Super-resolution algorithm for joint range-azimuth-Doppler estimation in automotive radars
Yalin Gurcan, Alexander G. Yarovoy · 2017
The problem of localization of targets and determining their velocities by means of radar is formulated as a joint estimation problem over single, three-dimensional domain. Joint estimation algorithms remove problems with data association and allow to exploit relations between individual estimation problems. To exploit wide bandwidth of the sensing signals, a new signal model is formulated based on time-dependant steering vectors for wideband automotive LFMCW radar applications. Joint three dimensional range-azimuth-Doppler estimation is performed using MUSIC approach based on the proposed signal model. The signal model is further improved based on investigation of inter-pulse Doppler effects on the radar array. Results of proposed model are compared to narrow-band MUSIC based joint estimation approach.