One-Bit Wideband DOA Estimation via a Time-Domain Weighted SPICE Framework

Wideband direction-of-arrival (DOA) estimation with one-bit analog-to-digital converters (ADCs) is of great interest in designing antenna array systems on resource-limited platforms. The one-bit wideband signal model is formulated as a sparse signal reconstruction problem based on the concept of convolution sparse coding. A unified weighted sparse iterative covariance estimation framework is proposed for one-bit wideband DOA estimation, where a one-bit wideband penalty term based on the maximum likelihood function is introduced. Then, a fitting alternating minimization strategy formed based on the group-sparsity concept is proposed as an effective solution. This framework stands out from traditional methods by eliminating the necessity for hyperparameter tuning and prior knowledge on the number of sources, thus broadening its applicability. In addition, within this one-bit wideband signal model, the corresponding Cramer-Rao bound and its asymptotic lower bound are derived. Simulation results demonstrate that our proposed framework including four variants offers superior estimation performance with the wideband signal sampled by one-bit ADCs, especially in the case where a small number of snapshots are involved. Experimental results further validate the effectiveness of the proposed methods.