Study On Target Detection And Parameter Estimation Approaches Of Fully Polarimetric Radar

The acquisition and utilization of polarimetric information plays an important role in improving the performance of radars' target detection and parameter estimation.Polarization has been widely used in the field of weather observation,remote sensing,air defense,and etc.At present,advances in information technology promote the rapid development of polarimetric radar.On one hand,the development of radar antenna technology and digital computing technology gave birth to the fully polarimetric radar with variable transmit and receive(Tx/Rx)polarizations.On the other hand,the improvement of the means of information measurement and the diversity of the information sources make it possible to grasp the prior information of the target and environment to develop the knowledge-aided adaptive radar system.If the priori information on the target and environment could be effectively used,also the radar waveform and Tx/Rx polarizations are optimized to construct adaptive knowledge-aided radar system,“matching illumination”is expected to become true to gain a better performance on target detection and parameter estimation than conventional polarization radars.To this end,it is urgent to develop novel signal processing technology under the condition of new information acquiring method and measurement system.With facing the fully polarimetric radar and by the means of adaptive waveform design and Tx/Rx polarization optimization,the problem of target detection and parameter estimation is deeply studied in this dissertation,and the research results are verified by using measured target data in microwave anechoic chamber and the reported clutter parameters.Main research results include:In the aspect of target detection,aiming at the cases of the prior knowledge of target and the clutter partially known and unknown,based on Tx/Rx polarization optimization,the method of improving the output Signal to Clutter plus Noise Ratio(SCNR)and the target detection algorithm are studied,respectively.Precisely,as to the case of the prior knowledge of target and clutter partially known,based on the Generalized Fractional Programming(GFP),an iterative algorithm to optimize transmit waveform and receive filter bank,is designed.Considering the minimum output SCNR of the filter bank(each filter corresponding to a particular Target Aspect Angle(TAA))as the figure of merit,and forcing the transmit waveform to match a power constraint and also a similarity one,the goal of optimal design is to maximize such minimum SCNR.At the stage of performance analysis,measured data is used to verify the validity of the algorithm,and then the influence of the factors on the output SCNR,including the uncertainty range of TAA,the bank size and etc,is systematically analysed.After that,the performance differences between the proposed algorithm and the existing typical algorithms,are compared.The analytical results show that,the fully polarimetric radar with waveform optimization has a significant SCNR gain with respect to conventional polarimetric radar and the single polarization radar with waveform optimization.Within the whole TAA uncertainty interval,compared to the existing transmit waveform and receive filter optimization methods,the proposed algorithm has a taller and more stable SCNR behavior.As to the case of the prior knowledge of target and clutter unknown,the measurement model of vector measurement system with variable transmit polarization is built,and a polarimetric detector under the heavy non-Gaussian clutter is proposed based on Generalized Likelihood Ratio Test(GLRT).The analytical performance expression of the detector is derived,and it is shown that it is a Constant False Alarm Rate(CFAR)test.Then also the relationship between variable transmit polarization vector and the performance expression is pointed out.In the performance analysis stage,by employing the simulated data and also the real clutter data,the performance of the proposed detector is verified.The experimental results show that,in the slowly-moving target detection under non Gaussian clutter background scenario,the proposed algorithm has a better performance against non-Gaussian clutter than two other polarization detectors;With a given false alarm rate,the probability of detection of our method is higher than the aforementioned detectors and its performance is close to Optimal Polarization Detector(OPD)with a known target and clutter information.It is also demonstrated that the polarization waveform optimization has a performance gain on target detection than the conventional fullpolarization approach.Furthermore,the proposed polarization detector is generalized to the scalar measurement system with joint Tx/Rx polarization optimization.As a consequence,jointly optimizing Tx/Rx polarizations to form a scalar measurement is confirmed to achieve a comparable detection performance to the vector measurement system with only transmit polarization optimization.In the aspect of target parameter estimation,after the measurement model of the scalar measurement system with variable transmit and receive polarizations is built,by considering the mean square error(MSE)of the target scattering vector's estimation as the figure of merit,the problem formulation of optimal target scattering vector estimation based on jointly Tx/Rx polarization optimization,is provided.With regard to how to tackle the optimization problem at hand,the error of the approach proposed by J.Xiao,who is a researcher of Washington university in St.Louis,is shown firstly.Then a sequential waveform polarization method for the estimation of target scattering matrix in the presence of clutter is proposed.The proposed sequential algorithm,based on the concept of sequential minimum MSE estimation,to determine the coefficients of the scattering matrix,guarantees the convergence and the resulting computational complexity is linear with the number of iterations.The effectiveness of the proposed method is validated through numerical results,underlining the performance improvement given by joint Tx/Rx polarization optimization for the scalar system.In order to obtain the same estimation error,the required SCNR value of the scalar system with joint Tx/Rx polarization optimization<the scalar system with only transmit polarization optimization< the scalar system with fixed Tx/Rx polarizations.Also,the results show that the vector system with transmit polarization optimization provides a comparative performance as the scalar measurement system employing joint Tx/Rx polarization optimization.By closely considering the practical demand of current development of radar technology,the novel system optimization methods and the relative signal processing algorithms are proposed in this dissertation.Also the simulation experiments and real data experiments are employed to verify the results.The research results are of important reference value for improving the capability of polarimetric radars on target detection and characteristics acquisition.