A Higher-Order Propagator Method for 2D-DOA Estimation in Massive MIMO Systems

Massive multiple-input multiple-output (MIMO) systems under time division duplexing (TDD), assume reciprocity between the uplink and downlink channels making it essential for the base station (BS) to estimate two-dimensional direction-of-arrival (2D-DOA) for downlink operations. Owning to the large dimension, computational complexity becomes a major concern for these systems in addition to the estimation accuracy. Matrix and tensor-based subspace schemes provide high accuracy estimates but offer tremendous computational load to perform singular value decomposition (SVD), root computations and/or peak search. Keeping in view, we propose a novel higher-order propagator method (HOPM) for DOA estimation which benefits from the inherent multidimensional structure and enhances the estimation accuracy with an added advantage of inexpensive computability over SVD. Moreover, we also developed a higher-order unitary PM (HOUPM) which provided further performance improvement at a much lower computational requirement. Numerical simulations are included to verify the effectiveness of proposed methods.