Adaptive Doppler Compensation for Mitigating Range Dependence in Forward-Looking Airborne Radar

Abstract

In this paper, we present ground moving target indication (GMTI) signal processing algorithm encompassing clutter suppression, target detection and parameter estimation. One of the most significant yet least publicized is the need of the GMTI mode for a forward-looking airborne radar. The integration of GMTI mode in a forward-looking airborne radar allows reconnaissance and surveillance operations in all weather conditions. In this context, space time adaptive processing (STAP) offers a unique prospect of enabling the GMTI mode in forward looking airborne radar. STAP is a two-dimensional filter designed to suppress platform motion-induced clutter Doppler spread. Interference is characterized by a covariance matrix. In the case of a forward-looking airborne radar, the clutter Doppler is dependent on range. Clutter Doppler dependency on the range renders the training cells heterogeneous. The heterogeneity effects are particularly prominent in the near range bins. Non-homogeneous training cells have a deleterious effect on STAP performance. In this study, we propose an adaptive Doppler compensation to mitigate the degraded STAP performance in the near range bins. The adaptivity feature circumvents the need for the availability of radar parameters in real-time. The real time implementation of STAP is impeded by requirements of a large number of training samples and covariance matrix inversion. Therefore, there is a dire need to devise a framework to detect and estimate target parameters within the STAP. In this regard, we propose an efficient STAP algorithm to detect and estimate target parameters. STAP weights are applied to the input data to obtain a 3D array. The range projection of the 3D array is utilized to detect and estimate the range of the target, while the angle–Doppler projection is used to estimate spatial and temporal parameters of the target. Most of the literature on STAP is geared towards a known covariance matrix. The assumption of a known covariance matrix may degrade STAP performance because of the inherent mismatches between the actual and assumed target steering vectors. In this study, we estimate the covariance matrix based on the synthetic data generated from a model of an airborne phased array radar. The developed STAP algorithms closely mimic a real-time implementation scheme in an airborne radar platform. The results of the proposed algorithm are validated through target parameter estimation and STAP metrics on synthetic data.