Synthetic aperture radar (SAR) exploits very high spatial resolution via temporal integration and ownship motion to reduce the background clutter power in a given resolution cell to allow detection of non-moving targets. Ground moving target indicator (GMTI) radar, on the other hand, employs much lower resolution processing, but exploits the physical aperture and relative differences in the space-time response between moving targets and clutter for detection. Therefore, SAR and GMTI represent two different temporal processing resolution scales which have typically been optimized and demonstrated independently to work well for detecting either stationary (in the case of SAR) or exo-clutter (in the case of GMTI) targets. Based on this multi-resolution interpretation of airborne radar data processing there appears to be an opportunity to develop detection techniques that attempt to optimize the signal processing resolution scale (e.g., length of temporal integration) to match the dynamics of a target of interest. This paper investigates signal processing techniques that exploit long CPIs to improve the detection performance of GMTI radar.

The Knowledge-Aided Sensor Signal Processing and Expert Reasoning (KASSPER) program aims to improve airborne Ground Moving Target Indicator (GMTI) radar performance by taking into account all available prior knowledge. One powerful piece of information is that the radar return signal is a superposition of near-ideal plane-waves. A plane-wave signal and clutter model of sampled GMTI radar data can be used to calibrate the receive array, suppress clutter, and detect moving targets. Each range gate is processed independently. Sample covariance matrices are unnecessary. The synthetic KASSPER Challenge Datacube is processed to demonstrate performance.

This paper presents a framework for incorporating knowledge sources directly in the space-time beamformer of airborne adaptive radars. The algorithm derivation follows the usual linearly-constrained minimum-variance (LCMV) space-time beamformer with additional constraints based on a model of the clutter covariance matrix that is computed using available knowledge about the operating environment. This technique has the desirable property of reducing sample support requirements by ?blending? the information contained in the observed radar data and the a priori knowledge sources. Applications of the technique to both full degree-of-freedom (DoF) and reduced DoF beamformer algorithms are considered. The performance of the knowledge-aided beamforming techniques are demonstrated using high-fidelity X-band radar simulation data.

A technique for incorporating past coherent processing interval (CPI) radar data into knowledge-aided space-time adaptive processing (KASTAP) is described. The technique forms earth-based clutter reflectivity maps to provide improved knowledge of clutter statistics in nonhomogeneous terrain environments. The maps are utilized to calculate predicted clutter covariance matrices as a function of range. Using a data set provided under the DARPA Knowledge-Aided Sensor Signal Processing and Expert Reasoning (KASSPER) program, predicted clutter statistics are compared to measured statistics to verify the accuracy of the approach. Robust STAP weight vectors are calculated using a technique that combines covariance tapering, adaptive estimation of gain and phase corrections, knowledge-aided pre-whitening, and eigenvalue rescaling. Several performance metrics are calculated, including signal-to-interference plus noise (SINR) loss, target detections and false alarms, receiver operating characteristic (ROC) curves, and tracking performance. The results show a significant benefit to using knowledge-aided processing based on multiple CPI clutter reflectivity maps.

Abstract?Space-time adaptive processing (STAP) plays an important role in ground moving target indication (GMTI). Heterogeneous clutter environments prevent STAP from achieving its theoretical performance bounds. The incorporation of a priori knowledge into the signal processing architecture holds the potential to greatly enhance detection performance by mitigating heterogeneous clutter effects. In this paper we propose one possible knowledge-aided STAP approach comprised of the following elements: a knowledge-aided prediction/estimation filter, a discrete matched filter, and a partially adaptive STAP applied to the clutter residual, assisted by knowledge-aided training. We focus our discussion on justifying the aforementioned elements and independently characterizing their performance potential. Using both measured and simulated data, we find the potential for substantial performance improvement.