The traditional beam selection algorithms determine the optimal beam direction by feeding back the perfect channel state information (CSI) in a millimeter wave (mmWave) massive Multiple-Input Multiple-Output (MIMO) system. Popular beam selection algorithms mostly focus on the methods of feedback and exhaustive search. In order to reduce the extra computational complexity coming from the redundant feedback and exhaustive search, a position fingerprint (PFP)-based mmWave multi-cell beam selection scheme is proposed in this paper. In the proposed scheme, the best beam identity (ID) and the strongest interference beam IDs from adjacent cells of each fingerprint spot are stored in a fingerprint database (FPDB), then the optimal beam and the strongest interference beams can be determined by matching the current PFP of the user equipment (UE) with the PFP in the FPDB instead of exhaustive search, and the orthogonal codes are also allocated to the optimal beam and the strongest interference beams. Simulation results show that the proposed PFP-based beam selection scheme can reduce the computational complexity and inter-cell interference and produce less feedback, and the system sum-rate for the mmWave heterogeneous networks is also improved.
Keywords: beam selection; heterogeneous networks; millimeter wave; multi-cell; position fingerprint matching.