FOC-Based Gridless Harmonic Retrieval Joint MM-Estimation: DOA Estimation for FMCW Radar Against Unknown Colored Clutter-Noise

Yuqian Mao,Gong Zhang,Changyun Qi,Xueling Yang,Wei Xiong
DOI: https://doi.org/10.1109/JSEN.2022.3148984
IF: 4.3
2022-01-01
IEEE Sensors Journal
Abstract:In this paper, we present a direction-of-arrival (DOA) estimation algorithm named FOC (fourth-order-cumulates)-based gridless harmonic retrieval joint MM-estimation (FGMM), for FMCW radar against unknown coloured (i.e. spatially and temporally coloured) clutter-noise. First, the target's intermediate frequency (IF) signal is resolved and reconstructed by the FOC-based gridless harmonic retrieval method. Then, we regard the signal expression of each array sensor as an independent form of multiple regression and further use MM-estimation to achieve the DOA by averaging the regression coefficient of the reconstructed target's IF signal from each array sensor one at a time. This work focuses on suppressing the damaging effect of the temporal and spatial correlation of the coloured clutter-noise. By introducing the FOC-based gridless harmonic retrieval method and MM-estimation, FGMM algorithm suppresses the negative influence of the temporal correlation of the coloured clutter-noise. Moreover, by exploiting the existence of the same clutter-noise distribution on each sensor whether the clutter-noise is spatially coloured or not, FGMM algorithm effectively eliminate the damage of the spatial correlation of the coloured clutter-noise. The experimental results validate the above effectiveness of FGMM algorithm.
What problem does this paper attempt to address?