Research On Waveform Optimization And Parameter Estimation For MIMO Radar

Multiple-input multiple-output (MIMO) radar which is proposed in recent years has bean drawn much attention from researchers and engineers in radar research area. MIMO radar can take many potential advantages over phased-array radar by combining waveform diversity and space diversity. By dexploiting these potentials, MIMO radar has several advantages, such as much improved angle resolution, enhanced target parameter identifiability, improved performance on weak target detection and low speed target dection, better target parameter estimation, as well as the essential beakthrough of the traditional radar techniques performance.Aiming at waveform diversity, target identifiability and therir computational complexity, in this dissertation, many methods on waveform optimization and parameter estimation for MIMO radar are proposed. The main works and contributions of this dissertation are as follows:(1) Transmit beampattern synthesis and waveform design for narrowband MIMO radarAn omnidirectional beampattern design algorithm is firstly proposed to synthesize the covariance matrix for omnidirectional beampattern design in MIMO radars. Some existing criteria can be seen as the special cases of this proposed framework.In some applications, we hope the distortionless response to match the desired beampattern in the mainlobe regions but maintain the sidelobes to be lower than the given threshold values. To overcome this disadvantage, the following algorithm is proposed. The proposed algorithm, referred to as the beampattern design with sidelobe control (SCBD), includes three different design aspects to synthesize the covariance matrix of the beampattern design with sidelobe control, which can guarantee the distortionless response in the mainlobes domain, and maintain the sidelobes to be lower than the given threshold values.(2) Waveform design based on transmit beampattern synthesis regarding unidirectional and wide-beam signals for wideband MIMO radarFirstly, a new method, which satisfies the unit-modulus or low peak-to-average power ratio (PAPR) constraints, is proposed to synthesize transmit waveforms that match a given unidirectional transmi beampattern. Secondly, another approach is proposed to generate power spectral density matrices using mixed norms that match a given wider main-beam beampattern. By comparison with the existing WBFIT method, the designed transmit beampattern not only matches the desired wideband transmit beampattern in the mainlobe region but also maintains the sidelobes below a specified threshold value-40dB.(3) The direction of arrival (DOA) estimation and complex amplitude estimation of the stationary targets for MIMO radarCapon can provide accurate estimates of the target locations. However, the Capon amplitude estimates are significantly biased downward. The APES provides accurate amplitude estimates of the target locations, but at the cost of lower resolution. To overcome these disadvantages, adapitive parameter estimation is proposed for colocated MIMO radar via sparse modeling. The Capon and APES fail to resolve the two closely spaced targets at θ2=-25°and θ3=-21°, but proposed method provides quite accurate target parameter estimates.(4) Accurate range-angle-Doppler estimation of moving target for MIMO RadarA regularized iterative reweighted minimization approach (RIRMA) is proposed to accurately estimate properties (angle, range, and Doppler) of targets for MIMO radar. In RIRMA, a nonlinear model of is reformulated as a linear model and a sequence of weighed lq (0<q≤1) minimization optimization problems are solved. By comparison with the existing iterative algorithm for reweighted l1 minimization, a closed-form solution of the optimization problem is obtained for next iteration in the optimization procedure. By comparison with existing DAS, IAA, and IAA-R, RIRMA provides accurate angle-range estimates at the target locations, and even in the range-Doppler dimension, RIRMA can also give excellent Doppler estimate.The above proposed methods aiming at waveform optimization and parameter estimation for MIMO radar are verified using simulated data of MIMO radar experimental system. Several numerical examples are provided to prove the effectiveness of the proposed methods.