Waveform Optimization For MIMO Radar

In recent years, multiple-input multiple-output (MIMO) radar has been drawn more and more attention from researchers and engineers. Unlike that the traditional phased-array radar only can transmit coherent waveforms, MIMO radar can employ multiple antennas to simultaneously transmit arbitrary signal, which is so-called waveform diversity. MIMO radar can take many potential advantages over phased-array radar by exploiting waveform diversity, such as target fading mitigation, resolution improvement, and interference suppression etc.. Fully exploiting these potentials can lead to substantially enhanced target detection, parameter estimation, as well as target tracking and recognition performance. This thesis considers some issues in colocated MIMO radar signal processing, including target parameter identifiability under general waveform assumptions, waveform optimization for improving parameter estimation accuracy and detection performance for space-time adaptive processing (STAP). The main contributions of this thesis are summarized as follows:1. Under general waveform assumption, the problem of parameter identifiability is investigated. Firstly, we propose a set of rules to determine DOFs of the MIMO radar with more flexible waveform diversity, which is based on the time-bandwidth product. Based on these rules, the sufficient and necessary conditions for parameter identifiability of MIMO radar with more flexible waveform diversity can be obtained, which can be stated as follows:the maximum number of targets that can be uniquely localized, depends on not only the geometry of MIMO radar, but also the rank and structure of the WCM. It is noticed that the proposed conditions can also be adopted in the cases of both uncorrelated and coherent waveforms. Therefore, they have rather more application.2. Under certain constraints associated with the biased estimator, we consider the problem of joint optimization of MIMO radar waveform and biased estimator with prior information on targets of interest for improving parameter estimation accuracy in the presence of clutter. The joint WCM and biased estimator design is formulated in terms of a rather complicated nonlinear optimization problem. Consequently, the optimization problem can not be easily solved. Under an approximating rational assumption, this problem is solved resorting to a convex relaxation that belongs to the SDP class, and hence can be handled very efficiently. An optimal solution of the initial joint optimization problem is then constructed through a suitable approximation to an optimal solution of the relaxed one (in a least squares (LS) sense). Because some prior information on targets of interest and the joint optimization of MIMO radar waveform and biased estimator are exploited, the proposed method can significantly improve parameter estimation accuracy compared to uncorrelated waveforms.3. The problem of robust optimization is investigated in the case that the parameters exploited by the waveform optimization to improve the parameter estimation performance are estimated with errors. The proposed method explicitly incorporates certain parameter uncertainty model into the optimization problem, and then the worst-case performance of parameter estimation over the convex set about the parameter estimate can be improved through optimizing the transmitted waveforms, so does the overall systematic performance. Because the problem of the robust waveform optimization is a rather complicated nonlinear function with respect to the optimization variables, and hence it is difficult to be solved. By exploiting some matrix inequalities, the original problem can be relaxed as an SDP. Due to the fact that the proposed method considers the joint optimization of the transmitted waveforms and the initial parameter estimation error, and then the optimized waveforms is more robust against the parameter estimation error compared to the uncorrelated waveforms.4. The problem of improving the detection performance of MIMO-STAP by exploiting waveform optimization is considered. Because it can be proved that maximization of the output signal-interference-plus-noise-ratio (SINR) is tantamount to maximization of the detection performance in the case of Gaussian noise, the output SINR can be regarded as the object function to optimize the transmitted waveforms. However, it can be seen that the output SINR is a nonlinear function with respect to the waveform covariance matrix (WCM). Therefore, this waveform optimization problem can not be easily solved. By employing the diagonal loading (DL) approach, the clutter covariance matrix can be reshaped as a positive definite one, and hence the original problem can be cast as an SDP. The robust waveform optimization is also considered to improve the worst-case detection performance of MIMO-STAP. Because the transmitted waveform can maximize the output SINR, the proposed method can significantly improve the detection performance of MIMO radar compared to uncorrelated waveforms.