This study provides an investigation of radar penetration imaging using ultra wideband (UWB) random noise waveforms from both theoretical and signal processing points of view. The principles governing the operation of a UWB random noise radar are discussed, including the analytical description of a UWB Gaussian process, the mathematical model and the implementation of a correlation receiver, and the range-Doppler resolution in random noise radar imaging. To address the problem of foliage penetration (FOPEN) synthetic aperture radar (SAR) imaging, a statistical-physical model for foliage transmission is developed and validated. Foliage obscuration impact on SAR images is analyzed by means of the paired-echo method. A comparative study between step-frequency and random noise SAR imaging is made to demonstrate the suitability of a random noise radar to be deployed as a FOPEN SAR. An experimental UWB random noise radar system is developed and field tested. In the signal processing aspect; a range sidelobe suppression technique called apodization filtering is developed to adopt the uncharacteristic point spread function (PSF) of a UWB random noise radar. The unique features for UWB radar penetration imaging are discussed and a 2-D FOPEN random noise SAR imaging model is established. Based on this model, FOPEN random noise SAR image formation and processing techniques are developed. The impact of frequency and aspect dependent target signatures on the SAR images are studied. More dedicated image processing techniques in terms of better target detection, discrimination and identification, are also investigated. These include three-dimensional (3-D) interferometric inverse SAR (ISAR) imaging technique and image resolution enhancement deploying spatially variant apodization (SVA) with various applications.
展开▼
