A Study On MIMO Radar Target Estimation Theory In Complex Environment

Multiple Input Multiple Output(MIMO)radar has spatial diversity and waveform diversity gain,which can improve the target parameter estimation performance.Classical MIMO radar estimation theory usually relies on ideal assumptions,they are orthogonal transmitted signals,spatially uncorrelated target reflection coefficients,and white noise.However,recently academia and industry find that it is difficult to achieve perfect orthogonal signals for transmission,which may lead to spatially correlated noise.Similar problems also appear in the application of passive MIMO radar,joint radar-communication and other systems.In some extreme cases,there is even the influence of spatially correlated target reflection coefficients.These situations no longer satisfy the ideal assumptions,and the classical MIMO radar estimation theory is no longer applicable.Considering that the above cases do not usually act independently on the system,it may be difficult to analyze these cases alone to get a full picture of the problem,and it is necessary to analyze their impact on MIMO radar systems from a higher dimension.Therefore,in this dissertation,factors such as orthogonal/non-orthogonal transmitted signals,spatial-independent/spacedependent noise,clutter,and spatial-independent/space-dependent target reflection coefficients are unifiedly modeled as complex environments,so that the complex environment model can be extended and cover common non-ideal complex situations,and then study the theory of MIMO radar parameter estimation in complex environments.The main contributions and innovations of this dissertation are summarized as follows:(1)For the target parameter estimation using active and passive MIMO radar in complex environments,the Maximum Likelihood estimate(ML)of joint estimation of target position and velocity is given under coherent and non-coherent processing,respectively.The Cramér-Rao Bound(CRB)is derived to analyze the complex environment’s effect on estimation performance based on CRB,and then some methods are given to improve the estimation performance of the system in the complex environment.(2)The optimization for target parameter estimation performance using passive MIMO radar is studied in complex environment,and the Weighted CRB(WCRB)is given as the system estimation performance evaluation metric.Under system complexity constraints,an optimization design is given to minimize the WCRB by joint selection of Signal Of Opportunity(SOO)and receiving station placement.Different from the optimal design of traditional PMIMO radar system,the joint design of this dissertation can select noncooperative SOO through the matched filter structure,while allowing the receiving station to be placed in a continuous area.(3)In the cooperative coexistence radar communication integrated system,the problem of target parameter estimation performance degradation caused by the complex environment with the overlapping of radar and communication non-orthogonal signals is addressed.Based on the performance evaluation metric WCRB,a joint optimization design for Joint Transmitter Selection and Receiver Placement optimization(JTSRP)is proposed,and genetic algorithm based method is given to efficiently solve the mixed integer nonlinear optimization problem.Numerical examples show that JTSRP is more capable of improving the target parameter estimation performance of the cooperative coexistence radar communication integrated system than other non-joint optimization method.At the same time,JTSRP can maintain the performance gains brought by the cooperation between radar and communications in the coexisting integrated systems.(4)To estimate the target direction of arrival(DOA)using co-located antenna MIMO radar in a complex clutter environment,a DOA estimation method based on supervised learning is proposed,where the neural network featuring with certain physical meaning is proposed to spontaneous find the optimal receiving sparse array structure and the corresponding optimal beamforming weights during training.Under the system design of optimal beamforming weights combined with broadband transmission signals to suppress clutter,and optimal sparse array structure to fully improve the resolution of target DOA,high-precision estimation of target DOA can be efficiently achieved.