COMET-Based Direct Position Determination of Incoherently Distributed Sources: Algorithms and Performance Analysis

This correspondence proposes two direct position determination (DPD) algorithms for incoherently distributed sources based on covariance matching estimation technique (COMET). First, a low-complexity estimator DPD-COMET is obtained via least squares fitting between the sample covariance matrices from all arrays and theoretical covariance matrices. Then, by minimizing the mean square error (MSE) to attain Cramér-Rao bound-level performance, the optimal weighted estimator DPD-OWCOMET is developed. Additionally, the asymptotic MSE distributions for both algorithms are rigorously derived. Numerical simulations demonstrate that both algorithms outperform the two-step localization methods and validate their theoretical results.