Research On Target Detection Algorithm In Duplicating Electromagnetic Environment

Target detecting radar in sea clutter such as airborne radar,shore based radar,ship borne radar,and so on,plays an important role in both the civil and the military field.These radar echoes will inevitably be disturbed by sea clutter.Therefore,the research on the statistical properties of sea clutter and target detection in sea clutter has received an extensive attention.In this thesis,we first discuss the target detection in sea clutter,and summarize the statistical characteristics of sea clutter.Aiming at the limited secondary samples in spatially heterogeneous sea clutter,an adaptive diagonal-loading covariance matrix estimator is proposed,and its performance is compared with the counterparts by using IPIX radar data in the simulation experiments.Due to the non-stationary of sea clutter in a long observation time,a distributed feature-based detector based on speckle coherence factor is proposed,and its performance is compared by using IPIX radar data in a simulation experiment.Three parts are mainly included in the thesis:1)The summary of sea clutter statistical characteristics and the basic model.At first,the statistical characteristics of sea clutter are summarized,and then the common models of sea clutter are summarized and contrasted.2)Research on adaptive detector in sea clutter.Firstly,we give three classical adaptive detectors,e.g.generalized likelihood ratio test(GLRT)detector,adaptive matched filter(AMF)detector and adaptive normalized matched filter(ANMF)detector,and three estimators,e.g.sample covariance matrix(SCM)estimator,normalized sample covariance matrix(NSCM)estimator and approximate maximum likelihood(AML)estimator,and compare the detectors' performance.Then,due to the less secondary samples in heterogeneous clutter,we propose a new covariance estimator based on the adaptive diagonal loading.Finally,experimental results using IPIX radar data sets show that,to detect the distributed targets in sea clutter by GLRT detector,the proposed estimator is robust to jamming and the performance improves in different range resolution sea clutter,especially as high as 15 d B in 3m range resolution.3)Research on the distributed feature-based detector for floating small targets in sea clutter.Several feature-based detectors are briefly introduced.Due to the non-stationary clutter at long observation time,a distributed feature-based detector relying on a speckle coherence factor is proposed.By using the secondary samples,the NSCM estimator is obtained firstly,and then is normalized.We obtain the speckle's consistency factor by the normalized NSCM.Finally,experimental results using IPIX radar data sets show that,the proposed detector removes the non-stationary texture.