Robust Iterative Adaptive Approach for Radar Short CPI Processing
Sandun Kodituwakku, Văn Đức Nguyễn, Mike D. E. Turley · 2022 IEEE Radar Conference (RadarConf22) · 2022
Heavy tapers are used in radar Doppler processing for controlling the sidelobes of strong clutter returns. However, when short coherent processing intervals (CPIs) are used, such processing could mask a significant portion of the Doppler spectrum due to clutter mainlobe broadening. In this paper, we propose a spectral estimation technique based on the Iterative Adaptive Approach (IAA) for radar Doppler processing to overcome the above limitation. Given the IAA technique does not require an estimation of training data based covariance matrix, it simultaneously addresses the problem of lack of homogeneous training data in short CPIs. We further improve the standard IAA technique by adding robustness to the covariance estimation using a concept borrowed from the robust adaptive beamforming literature. We demonstrate the superiority of robust IAA (IAA-R) compared to window based method and standard IAA on simulated data as well as data collected from the Jindalee Operational Radar Network (JORN). Probability of detection analysis shows a significant performance improvement especially for slow moving targets.