New algorithm for joint estimation of DOA and polarisation based on electrically large and complete electromagnetic vector sensor

摘要

With the development of estimation algorithms for polarisation sensitive array (PSA) parameters, research has been focused on complete/incomplete electromagnetic vector sensor (EMVS) that is mainly based on electrically 'short' dipoles and/or electrically 'small' circular loops. Recently, electrically 'long' dipoles and electrically 'large' circular loops are further preferred and have been proposed to utilise more efficiently radiation characteristics. However, almost all available such EMVSs are only composed of either long dipoles or large circular loops. To sufficiently utilise relevant electromagnetic information, in this study an electrically 'large' EMVS is proposed, which is composed of a triad of long dipoles and a triad of large circular loops, co-located, and orthogonally oriented. With formulated array manifold of the proposed EMVS, a close-form algorithm is developed to estimate an incident source's azimuth-elevation direction-of-arrival (DOA) and polarisation simultaneously without prior knowledge, and verified that its estimation accuracy is closely approaching corresponding to Cramer-Rao Bound. It is demonstrated that when with inaccurate prior knowledge, the proposed algorithm has better performance than existing algorithms that were designed for the incomplete EMVS with either long dipoles or large loops. Finally, simulation results also show that with increased electrical size of the EMVS, the estimation accuracy is significantly improved.