Study On Airborne Radar STAP And Wideband Radar MTD Technologies In Nonhomogeneous Environments

Compared with ground-based radar,the maximum detection range of airborne radar is extended through the elevated position of the platform,especially for the low altitude targets.The clutter returns received by airborne radar operating in down-looking mode are very strong,and have the characteristics of space-time coupling.In that case,the space-time processing is needed to suppress clutter,i.e.space-time adaptive processing(STAP).In modern warfare,radar is expected to have the capability to provide more information about the target,not just the conventional capabilities of detecting and tracking targets.Consequently,wideband radar has drawn attention and been widely used for its high parameter identifiability and the ability of acquiring more target information.As for traditional STAP algorithms,it is required to have a certain amount of independent and identically distributed(IID)training samples to estimate the clutter-and-noise covariance matrix.However,the homogeneous assumption on the secondary data may be invalid in practical application.And the heterogeneity of secondary data may be caused by many effects such as variations in terrain,manmade structures,which can lead to severe degradation in the performance of STAP algorithms.When designing wideband radar moving target detector(MTD),similar problem will be encountered.And if such heterogeneous training samples are used directly to estimate the clutter covariance matrix,the detection performance will degrade dramatically.Therefore,it is necessary to cope with heterogeneous samples for practical use whether for airborne radar STAP or wideband radar moving target detection.This dissertation mainly focuses on airborne radar clutter suppression and wideband radar moving target detection in heterogeneous environment.The main contents of this dissertation can be summarized as the following four aspects:1.The knowledge-aided STAP(KA-STAP)methods available show significantly performance degradation under the circumstance of inaccurate environmental knowledge.To solve this problem,two KA-STAP approaches based on dynamic environment sensing are proposed.(1)With transmitted signal being orthogonal waveform,the clutter information is achieved.Then the clutter information and platform parameters are used and a clutter covariance matrix at future time is obtained incorporating system parameters.Finally,the space-time processor can be built based on the combination of the predicted clutter covariance matrix and the sample covariance matrix.(2)This method is based on method 1 and makes full use of the spatial characteristics of the clutter radar cross section(RCS),thus the robustness of the reconstruction algorithm is improved by jointly sensing of multiple measurements.2.Pre-whitened STAP and color loading algorithms can obtain fast convergence rate and achieve better clutter suppression performance in heterogeneous clutter environment.However,these algorithms show significantly performance degradation in the presence of array errors.To address this issue,firstly we introduce three common error models of array and focus on the equivalent model of the array errors in the presence of three kinds of errors.Then we study the effects of the array errors on KA-STAP algorithms.At last,we propose gain-phase errors online calibration based on clutter data and direction-dependent errors online calibration based on clutter data according to the form of the array errors,respectively.3.The problem of wideband radar moving target detection in heterogeneous clutter is studied.When designing traditional wideband radar adaptive detector,it is assumed that range migration is negligible and effective training samples for estimating the clutter covariance matrix is sufficient.However,taking into account high range resolution and the heterogeneity of the real environment,the above assumptions are not always hold.To solve this problem,we propose a knowledge-aided wideband radar moving target detection method.This method eliminates the effects of range migration by transforming the original detection problem from range domain to range frequency domain,and reduces the influences of heterogeneous sample by incorporating prior information into moving target detector.4.Range migration can be compensated through Keystone transform without the exact target velocity.However,it is unfortunate that the methods based on Keystone transform can not be used in the case of multiple targets with velocity ambiguity.To solve this problem,we propose a detection algorithm of moving targets for wideband radar based on joint-sparse recovery.Firstly,a prewhitening operation is performed to filter the clutter.Then,the existing non-ambiguous representation of the wideband signals is used,and a jointly row sparse representation of the wideband signals is derived in frequency/slow-time domain.Finally the detection problem is solved via joint-sparse recovery.