Mimo Radar Parameter Estimation And Waveform Design Method

In the field of modern wireless communication, Multiple Input Multiple Output(MIMO)multiple-antenna technology has been explored and put into practical use for years. It adopts multiple antennas at both ends of the wireless links, making good use of spatial resources, which improves system capacity,frequency utilization ratio and dependability without increasing the transmitting power or system bandwidth. This is a huge breakthrough for intelligent antennas in the wireless mobile communication field and the key technology that is indispensably used in the new generation mobile communication systems. Enlightened by the development and the advantages of MIMO technology, the concept of MIMO radar has been proposed by radar designers recently. More and more attention has been paid to the development and the future of MIMO radar. However the study is still in the primary stage home and abroad.Unlike beamforming which assumes that signals at transmitters and receivers are highly correlated, MIMO radar concept explores independence between signals at the both the transmitting and receiving arrays which consist of multiple omni-directional antennas with different phase centers. Independence between signals leads to some unique advantages for MIMO radar, such as the constant Radar Cross Section (RCS) of targets, significantly decreasing the signal fading. Radar detection ability is improved by the obtained diversity gains.Nowadays, studies on MIMO radar mainly focus on DOA estimation, target detection performance analysis, waveform design and multi-channel signal modeling of active radars. Based on co-located MIMO radars, popular models of MIMO radar and the existing parameter estimation algorithms, including DOA estimation and complex amplitude estimation, are discussed. Target identifiability and waveform design problems are investigated. For MIMO radar with widely separated antennas, the waveform design method in the extended target model is discussed by simulations. Then, A waveform design method is proposed, which applies the technique of Kronecker Structured Matrix Estimation in the solution of the design above. This method can be viewed as the supplement to the existing design methods. Firstly, the popular models of MIMO radar are introduced, and the differences of signal processing methods in different models are discussed. Considering the angular spread, a virtual representation of the channel model is discussed.Secondly, based on the co-located MIMO radar, nonparametric spatial spectrum estimation algorithms including Capon,Amplitude and Phase Estimation(APES),Generalized Likelihood Ratio Test(GLRT) and iterative GLRT(iGLRT) based on noncoherent subarray architecture are discussed, and their performances are analyzed through computer simulations.Thirdly, parameter identifiability of MIMO radar is discussed, and its advantages in comparison with the traditional radars are verified through simulations. Based on the co-located MIMO radar, the parameter estimation and target number detection with ML and BIC are analyzed. Further, a minor generalization is proposed, combining the GLRT and BIC for parameter estimation and target number detection, respectively. This method is validated by MATLAB simulation results.In the end, waveform design of MIMO radar is investigated. Here, the waveform design problems based on co-located MIMO radar and widely-separated MIMO radar are discussed, respectively. Moreover, optimum radar waveform design based on widely-separated MIMO radar viewed as a FIR system is considered, and the technique of Kronecker Structured Matrix Estimation is applied to the solution of the design presented above. Both the maximum likelihood estimation and separable least squares framework estimation are verified through simulation results. It is shown that the two methods have almost the same performance which improves as the signal duration or total transmitting power increases.