论文部分内容阅读
Compared with some other array errors self-calibration algorithms,the rank reduction estimator(RARE) can provide the “decoupling” direction-of-arrival(DOA) estimation and thus can avoid both iterative computation and local convergence.In this paper,we focus on the angle resolution performance of the RARE in the finite sample case.First,the spatial spectrum of the RARE is approximately represented as the quadratic function of the perturbations of the signal eigenvector.Then,the statistical expectation of the spatial spectrum of the RARE is presented,based on which the explicit formula for the mean signal-to-noise ratio(SNR) resolving threshold is derived.The theoretical analysis is verified through the numerical experiments which are in the context of mutual coupling self-calibration for some uniform arrays.Furthermore,the impacts of the power ratio and correlation coefficient of the sources on the resolution capability of the RARE are illustrated via numerical experiments.
Compared with some other array errors self-calibration algorithms, the rank reduction estimator (RARE) can provide the “decoupling ” direction-of-arrival (DOA) estimation and thus can avoid both iterative computation and local convergence.In this paper, we focus on the angle resolution performance of the RARE in the finite sample case. First, the spatial spectrum of the RARE is approximately represented as the quadratic function of the perturbations of the signal eigenvector. Then, the statistical expectation of the spatial spectrum of the RARE is presented, based on which the explicit formula for the mean signal-to-noise ratio (SNR) resolving threshold is derived. The theoretical analysis is verified through the numerical experiments which are in the context of mutual coupling self-calibration for some uniform arrays. Futuremore, the impacts of the power ratio and correlation coefficient of the sources on the resolution capability of the RARE are illustrated via numerical experiments.