Study On New Systems And Techniques For Improving Radar Imaging Performances

Radar imaging is a powerful tool in retrieving target information within the microwave regime. The long detection range, wide field of view (FOV), and environment adaptability of radar imaging systems are what optical imaging systems cannot rival, thus radar imaging is playing an increasingly crucial role in many military and civil applications such as battlefield investigation, terrain mapping, homeland defense. With the rapid development of electronic and information technology, and the upgrade of applications and demands, the field of radar imaging is experiencing a series of revolutionary changes. Traditional Synthetic Aperture Radar (SAR) and Inverse Synthetic Aperture Radar (ISAR) are incapable of meeting the growingly strict requirements on radar imaging, which is forced to gradually develop towards new areas inclucing bistatic/multistatic cooperated imaging, ultra wideband (UWB) and wide angle imaging, and super-resolution imaging. The quality and resolution of the target image can be dramatically improved by combining data from multiple antennas and multiple viewing angles. Besides, it is of great importance to seek for new imaging systems to break the resolution limitation of real aperture radar imaging (RARI), which could also be implemented on moving platform to perform forward looking imaging.This dissertation demonstrates efforts on enhancing signal to noise ratio (SNR), resolution and target feature of radar images, by establishing and researching new imaging systems of bistatic/multistatic ISAR, wide angle radar imaging (WARI) and radar correlated imaging, as well as by exploiting new techniques including image fusion and sparse reconstruction. The main content of this dissertation is as follows:In chapter I, a brief overview of the traditional SAR and ISAR is given, followed by an introduction to new trends on radar imaging system and technique. The main purpose of this dissertation is subsequently shown.In chapter II, a novel method of fusing bistatic ISAR images in data domain is introduced. Bistatic ISAR system can receive echoes from different observation angles simultaneously, thus a finer image with more abundant information of a certain target can be obtained by fusing the two images attained by bistatic radar system. However, the dis-match of dimensions and shapes of the two target images caused by bistatic angle highly complicates the image fusion process. This chapter, based on the bistatic ISAR turn table model, proposes an ISAR image fusion method combined with parameter estimation. The Doppler differences between two adjacent sub-apertures of prominent scattering points are firstly extracted to estimate the rotation velocity (RV), then the half bistatic angle is estimated by exploiting the geometrical relations of two ISAR images, and finaly the two data sets from different observation angles are mapped onto a global coordinate system to perform the image fusion process, resulting in an ISAR image of higher quality.In chapter â…¢, distributed ISAR with its resolution enhancing possibility is introduced, by taking the multistatic ISAR with straight baseline formation as an example. Firstly the geometrical model for distributed ISAR is established, and the principle of resolution enhancement is analyzed. For the scenario where overlapping of viewing angle exists, a novel signal combining method for distributed ISAR based on 2D correlation function is presented and verified via simulation results. For the scenario where angular gap between adjacent sensors exists, the matrix model for data sampling and image formation is introduced, and a Quasi-Newton solver is presented to focus high resolution images and to overcome the error induced by data gap.In chapter IV, a novel method of full aperture image composition for wide angle radar imaging (WARI) is presented. WARI possesses higher capabilities of target resolution and information extraction by collecting echoes through larger angular extent. However, in WARI scenario, regular approximations do not hold, so Fourier based methods cannot be applied directly. To deal with this problem, the signal model is firstly analyzed, then the full aperture data is divided into several overlapping sub-apertures and Polar Format Algorithm (PFA) is utilized to obtain sub-images at different aspects. Finally the rotationally aligned subimages are used to perform full aperture image composition in the NMF (Non-negative Matrix Factorization) subspace via solving a Sparsity Enhanced NMF (SENMF) problem, resulting a synthesized image with target features enhanced and SNR improved.In chapter V, quantum imaging (or correlated imaging) theory is introduced and the concept is extended into microwave regime. By analyzing the signal model, the necessity of achieving radar correlated imaging is deduced, which is to generate randomly fluctuating microwave radiation fields that are controllable and measurable. Then the technique to realize this necessity based on modulating a phased array is given, together with the analysis of effect of the array distribution on the self correlation function of the field. The resolution of the imaging result breaks the Rayleigh criterion if compressive sensing (CS) based approach is utilized. By transmitting wideband signals, this novel imaging mechanism can achieve 3D super-resolution imaging on a forward looking radar platform in motion. Simulation results are given to verify the validity of the proposed system, as well as the analysis on model error and the results of real imaging experiment. This innovative system does not rely on aperture synthesis, and can retrieve the target image with fewer data samples than Nyquist sampling requirement, making it highly promising in many applications.In chapter VI, conclusions of this dissertation are drawn.