Towards super-resolution computational imaging with GeoSTAR configured mm-band sensor array radar: An aggregated adaptive beamforming compressed sensing approach

Abstract

We address a new adaptive beamforming (ABF) approach for attaining virtual super-resolution performances of radar imaging with differently configured mm-band compressed sensing (CS) array radars. Our new aggregated ABF-CS approach is an adaptive beamforming-oriented generalization of the conventional matched spatial filtering (MSF) method for radar image formation based on the advanced descriptive experiment design regularization (DEDR) framework for radar imagery enhancement. First, we optimize the sensor array configuration employing the celebrated GeoSTAR geometry to attain the desired shape of the MSF system point spread function (PSF). At the second (reconstructive) stage, the low resolution MSF image is next enhanced via performing the aggregated ABF-CS post-processing aimed at attaining the overall super-high resolution remote sensing (RS) performances. The effectiveness of the new aggregated ABF-CS radar imaging method is corroborated via extended simulations of different DEDR-related imaging techniques using the specialized elaborated software that we refer to as “Virtual Remote Sensing Laboratory” (VRSL). The unified ABF-CS computational imaging method has been adapted to the DEDR-optimized GeoSTAR sensor array configuration and exemplified in the reported simulations of super-high resolution localization of the multiple closely spaces targets performed with the elaborated VRSL software. The latter are indicative of the superior operational efficiency of the imaging radar system that employs the new ABF-CS method adapted to the DEDR-optimized GeoSTAR configuration over other tested competing techniques.