Maximum likelihood signal estimation for polarization sensitive arrays

The authors consider the problem of separating and estimating the waveforms of superimposed signals received by a polarization-sensitive array. Signal estimation is accomplished by a two-step maximum likelihood procedure: (i) The directions of arrival and polarization parameters of all the signals are estimated. (ii) The estimated signal is obtained as a linear combination of the array outputs, with weights which are computed from the estimated direction/polarization parameters. The objective of this study is to analyze the quality of the estimated signal in terms of the output signal-to-interference ratio (SIR) and output signal-to-noise ratio (SNR). Closed-form expressions are derived for the output SIR and SNR of a general diversely polarized array. By evaluating these expressions for selected test cases it is shown that polarization-sensitive arrays can provide significantly higher output SIR and SNR than uniformly polarized arrays. The performance improvement is especially significant for closed spaced sources with sufficiently different polarization characteristics