Array Position Calibration for Robust DOA Estimation of Incoherently Distributed Sources

Existing calibration methods designed for point-like source models degrade in multipath scattering environments due to unmodeled angular spread and coupling between position errors and directions of arrival (DOAs). To address this issue, we propose a joint estimation framework for partially calibrated arrays that simultaneously recovers DOAs, angular spreads, and sensor perturbations. Specifically, we leverage a generalized in coherently distributed (ID) source model that captures stochastic angular spread and position–DOA coupling, and then formulate a weighted least-squares (WLS) objective aligned with this model. We also develop an alternating optimization (AO) solver with closed-form updates for covariance and noise terms and iterative refinements for DOAs and positions, together with identifiability guarantees and a proof of local convergence. Moreover, the Cram´ er–Rao lower bound is derived as a theoretical benchmark. Extensive simulations validate the effectiveness and robustness of the proposed method.