ENHANCED AND EFFICIENT ISAR IMAGE FOCUSING USING THE DISCRETE GABOR REPRESENTATION IN AN OVERSAMPLING SCHEME

Abstract

Inverse synthetic aperture radar (TSAR) imaging is one of the most well-known techniques of radar target recognition. One of the most important issues in TSAR imaging is the improvement of the image smeared by a moving target. In this paper, we propose the discrete Gabor representation (DGR) in an oversampling scheme as an effective means of obtaining a well-focused TSAR image with a short calculation time. In contrast to other linear time-frequency transforms (TFTs), the DGR obtains Gabor coefficients using the analysis window frames derived from the clearly defined Gaussian-type synthesis window. The oversampling scheme of the DGR leads to accurate calculations of the Gabor coefficients, which denote signal time-frequency amplitude. Since each Gabor coefficient is compartmentally assigned to the associated unit cell of the time-frequency grid, the DGR can show an excellent time-frequency concentration and can effectively discriminate the Doppler components of prominent point-scatterers. The application results demonstrate that the DGR not only has enhanced focusing performance but also retains computational efficiency. The DGR in the oversampling scheme is expected to facilitate high-quality TSAR imaging in radar target recognition.