A fast off-grid stap algorithm based on marginal likelihood maximization
Xinying Zhang, Tong Wang · IET conference proceedings. · 2024
Space-time adaptive processing (STAP) based on sparse recovery (SR) have become a promising algorithm for the benefit of reducing the training samples requirements. However, most SR-STAP approaches suffer the off-grid effect due to the discretization of the angle-Doppler plane. To mitigate the off-grid effect, an efficient clutter subspace estimation algorithm is proposed based on marginal likelihood maximization in this paper. The algorithm selects atoms from a dense dictionary incrementally to estimate clutter subspace, which avoids the effect of the strong coherence between the atoms. Moreover, the proposed algorithm only uses a part of the dictionary, leading to high computational efficiency. Numerical experiments are conducted to prove that the proposed algorithm has superior clutter suppression performance with low computational complexity.