International Symposium on Telecommunication (IST)

Home / 11th International Symposium on Telecommunication (IST'2024)

DOA Estimation in FMCW Automotive Radars with Interference: ML and VMD Approaches

Authors :

Alireza Zaherfekr¹ Mohammad Javad Ghoreishian² Ataollah Ebrahimzadeh³

1- Department of Electrical & Computer Engineering , Babol Noshirvani university of technology 2- Department of Electrical & Computer Engineering , Babol Noshirvani university of technology 3- Department of Electrical & Computer Engineering , Babol Noshirvani university of technology

Keywords :

Automotive Radar،DoA،VMD Method،Interference،Interference،Machine learning

Abstract :

Today, with the progress and development of road transportation systems, the advanced driver assistance systems (ADAS) have improved driving quality and reduced road accidents. One of the most important processes to implement the ADAS system is the accurate location of vehicles. On the other hand, with smart vehicles equipped with radar systems, problems such as interference will be inevitable. In this paper, we deal with the problem of estimating the Direction of Arrival (DOA) of vehicles under the interference assumption in frequency division multiplexing-multiple input-multiple output (FDM-MIMO) car radar systems. To estimate the DOA, we propose two approaches. In the first approach, with the help of a designed machine learning network, waveforms that are associated with interference are identified and excluded from the MUSIC algorithm for the DOA estimation. In the second approach, instead of removing the waveforms contaminated with interference, we use the variational mode decomposition (VMD) technique to reduce the interference of the waveforms. Finally, the DOA is estimated with the help of a novel weighted-MUSIC algorithm whose weights are determined based on the signal to the interference ratio (SIR) output from the VMD algorithm. In the end, to evaluate the proposed schemes and compare them with conventional schemes, we will perform extensive numerical simulation.