DoA-Aided Channel Estimation for One-Bit Quantized Systems

This paper introduces a novel channel estimation technique for one-bit quantized systems, which uses structural side information given by a direction-of-arrival (DoA) estimate. Following previous work on DoA-aided minimum mean square error (MMSE) channel estimation, the principle of decomposing the estimation problem into two orthogonal subspaces is extended for one-bit quantized signals. The MMSE estimate is formulated within each subspace by utilizing the well-known Bussgang decomposition. To this end, we derive an approximation of the arcsine law, which facilitates decomposing the receive covariance matrix into the two distinct subspaces. Numerical simulations demonstrate the superiority of our proposed two-stage estimation approach compared to state-of-the-art methods.