Gridless Target Parameter Estimation For MIMO Radar

Multiple input multiple output(MIMO)radar exploits the degrees of freedom at both the transmitter and receiver,which endows it with superior performance in resolution,anti-interference and target detection.The existing literatures predominantly revolve around the target parameter estimation of MIMO radar with linear arrays,wherein the parameter estimation is beset by challenges such as low estimation accuracy and high computational complexity.Moreover,since the beamforming of MIMO radar is only a function of angle,it fails to resolve targets at distinct ranges along the same direction.On the contrary,the beamforming of frequency diverse array(FDA)radar is a function of both angle and range,which facilitates the incorporation of FDA technique with MIMO radar to construct FDA-MIMO radar.Most of the existing studies concentrate on target parameter estimation of FDA-MIMO radar based on uniform linear array.Owing to the coupling of angle and range in FDA-MIMO radar,it will impair the estimation accuracy of target parameters.Most of the existing MIMO radar and FDA-MIMO radar parameter estimation algorithms necessitate grid partitioning,thus encountering problems such as high computational complexity and insufficient estimation accuracy.To tackle these problems,this thesis mainly investigates the gridless target parameter estimation method based on MIMO radar and FDA-MIMO radar.In view of the problems of low estimation accuracy and high computational complexity in parameter estimation based on planar array bistatic MIMO radar,a gridless target parameter estimation method is proposed.Firstly,based on the single snapshot scenario where both the transmitter and receiver are uniform planar arrays(UPAs),a bistatic MIMO radar model is established.Secondly,it is proved that the atomic norm minimization(ANM)problem of this model is equivalent to a semi-definite programming(SDP)problem,and by solving the SDP problem,angle estimation is obtained without discretizing the four-dimensional angle space.Moreover,an algorithm implementation based on alternating direction method of multipliers(ADMM)is derived to enhance computational efficiency.Finally,simulation experiments verify that the proposed algorithm improves estimation accuracy while reducing computational complexity.In regard to the problem of angle and range coupling in FDA-MIMO radar,a gridless angle and range estimation method suitable for sparse linear array as receiving array for monostatic FDAMIMO radar is proposed.Firstly,a monostatic FDA-MIMO radar model is established,and based on this,an angle and range decoupled model is derived.Secondly,utilizing the semi-positive definite,low-rank and other properties of the received signal covariance matrix,combined with the covariance matching criterion,an optimization problem based on low-rank matrix reconstruction is proposed,and solved by using the idea of alternating projection algorithm.Furthermore,a Cramér-Rao lower bound for monostatic FDA-MIMO radar based on virtual array idea is derived.Finally,simulation experiments verify that the proposed algorithm can achieve high-resolution target parameter estimation while reducing computational complexity.Lastly,in respect to the problem of angle and range coupling in monostatic planar array FDAMIMO radar,a gridless parameter estimation method for angle and range decoupling in monostatic planar array FDA-MIMO radar is proposed.Firstly,a monostatic planar array FDA-MIMO radar signal model is established,and based on this,an angle and range decoupled model is derived.Secondly,an optimization problem based on low-rank matrix reconstruction suitable for this model is proposed,and an algorithm implementation based on alternating projection is derived to accelerate computation.Besides,a Cramér-Rao lower bound for monostatic planar array FDA-MIMO radar is derived.Lastly,simulation experiments verify that the proposed algorithm has higher resolution while having lower computational complexity.