“…Many DOA estimation methods usable for conventional scalar arrays (CSAs) have been extended to PSA, such as estimation of signal parameters via rotational invariance techniques (ESPRIT) [4][5][6], multiple signal classification (MUSIC) [7][8][9][10], maximum likelihood (ML) [11][12][13] and so on. Besides the conventional super-resolution algorithms, there are some specific DOA estimation algorithms for PSA, such as (i) vector cross-product DOA estimator for PSA composed of electromagnetic vector-sensor (EMVS) [1,14], which has smaller computational complexity and easier implementation than those of super-resolution algorithms; (ii) a decorrelation technique, that is, polarisation smoothing algorithm for coherent source's DOA estimation, which is applicable to arbitrary array structure and does not decrease the effective array aperture [13,[15][16][17]; (iii) multi-nions algebra-based super-resolution algorithms [8,9,[17][18][19], which are more robust to modellisation errors and coherent noise compared with conventional super-resolution algorithms; (iv) moreover, [20] shows how a dipole or a loop triad by itself suffices for multisource azimuth/elevation direction finding and polarisation estimation. Based on the framework of complex-valued non-orthogonal joint diagonalisation algorithms, [21] presents a novel strategy to exploit the multi-dimensional structure present in the second-order statistics of EMVS outputs for simultaneous DOA and polarisation estimation.…”