Direction finding by sparse-based approach for dual-polarised conformal array in the presence of gain and phase uncertainties

This article investigates the problem of robust direction-of-arrival (DOA) estimation with unknown sensor gain and phase uncertainties. A Dual-Polarised Regularised Multiple FOCal Underdetermined System Solver is proposed to estimate the 2-D DOA for the dual-polarised conformal array. Each polarisation signal is decomposed into two orthogonal components, which are regarded as a pair of coherent signals, and the polarisation parameters are merged into the signal waveform. A sparse-based method based on the coherent signals block is presented, implementing weight sharing and grid pruning within the block to alleviate the negative influence of the errors. Meanwhile, gain and phase errors are modelled as the equivalent noise, which can be suppressed with the adaptive adjustment for the regularisation parameter. As the insensitivity of the sparse-based algorithm to coherent signals, the proposed algorithm can realise the blind polarisation 2-D DOA estimation. The Cramer-Rao bound is derived, and simulations exhibit the effectiveness and robustness of the proposed method with the unknown sensor gain and phase uncertainties.