Performance Analysis Of Joint Parameter Estimation For MIMO Radar

Multiple-input multiple-output(MIMO)radar employs more than one transmit stations and receive stations to detect and estimate targets which can greatly improve radar performance.This thesis considers the joint estimation for target position and velocity using MIMO radar,and the best estimation performance bound Cramer-Rao bound(CRB)is derived.This thesis investigates the joint estimation performance in the nonideal situations.Then considering if phase synchronization can be realized,the noncoherent estimation and coherent estimation are analyzed.Although the main focus is on passive radar,the results can be applied to both active and passive radar.For skywave MIMO over-the-horizon(OTH)radar,the joint estimation of ionospheric and target parameters is analyzed.Estimation under the mismatched signal model is also presented.Radar designers can compare the mean squared erros(MSE)of their developed approaches with the CRBs provided in this paper.If MSE is close enough to the CRB,these developed approaches can be deemed ?good enough?.First,this thesis develops the MIMO radar received signal model of noncoherent estimation and coherent estimation.Specifically,the signal model allows nonorthogonal signal,dependent reflection coefficients,and dependent noise.Two cases for random reflection coefficients and deterministic unknown reflection coefficients are considered,and the maximum likelihood(ML)estimation for both cases are given.The effects of phase information and reflection coefficients on the estimation performance are analyzed based on a simplified ML.The CRB is derived.The effects of signal orthogonality,dependency of reflection coefficients,and dependency of noise on the estimation performance are analyzed numerically.Then,for skywave OTH MIMO radar,the effects of ionospheric parameter estimation error on the target parameter estimation are studied.To fully use the ionospheric parameter information in received signal,this thesis proposes a way of joint estimating ionospheric and target parameters.The ML of joint estimation and traditional estimation are given,and the CRB is also presented.Finally,considering that under the MIMO radar signal model mismatch,the traditional CRB cannot bound the mismatched MSE.Through redefining score function and Fisher information matrix,the mismatched CRB for a specific mismatched case is derived for the noncoherent estimation with random reflection coefficients.