Electromagnetic vector-sensor array pro-cessing for distributed source localization

We consider the problem of direction-of-arrival (DOA) estimation for distributed signals with electromagnetic vector sensors, of which each provides measurements of the complete electric and magnetic fields induced by electromagnetic (EM) signals. In this paper, we consider situations where the sources are distributed not only in space with a deterministic angular signal density, but also in polarization with partially polarized components. A distributed signals general model with electromagnetic vector-sensor array (EMVS-DIS) is established with some reasonable assumptions. Based on the EMVS-DIS model, the minimum-variance distortionless response (MVDR) estimators for distributed source DOA are derived. MVDR estimators do not require the knowledge of the effective dimension of the pseudosignal subspace. We compare our method with the distributed signal MUSIC-like estimator in electromagnetic vector-sensor arrays. The simulation studies show significant advantages in using the proposed EMVS-DIS model with electromagnetic vector sensors. Simulation results show that the new MVDR method outperforms the MUSIC-like algorithm by reducing the estimation RMSE and improving resolution performance for scenario with distributed sources. A robustness study of MVDR localizer was also conducted via simulations.