An Off-Grid Massive MIMO Channel Estimator with Unknown Angular Cluster Prior

By exploiting both the inherent sparsity in the continuous angular domain and the unknown angular cluster prior, we propose a super-resolution channel estimator for massive multiple-input multiple-output (MIMO) systems. The off-grid channel estimation is formulated by a weighted atomic norm minimization problem promoting the angular cluster-sparse prior, rather than known a priori, which is acquired by iteratively updating the angular cluster boundaries within the on-grid angular domain. Simulation results show the superior performance of the proposed algorithm over the existing off-grid estimator.