Research On Beam Alignment Algorithm Of Millimeter Wave System

Millimeter wave(mm Wave)wireless communication is regarded as ones of the most vital realm of wireless communication technology in the future.Traditional sub-6G radio frequency and wireless communication technology can barely meet the fast-growing demand for data of human.However,millimeter wireless communication,with its abundant spectrum resource and huge channel capacity,can lead to more quick and more reliable communication.So,it is getting more attention,and has tremendous potential for explicitness.Millimeter wave(mm Wave)usually refers to radio spans form 30 GHz to 100 GHz.Despite its abundant available spectrum resource,communication at mm Wave suffers from severe path loss,which limit its propagation distance.The beamforming gains of large-scale arrays are required to increase communication distance.By using beamforming at Rx and Tx,energy of the signal can be concentrated on the specific direction.Thus,the severe pass loss can be overcome,and link quality can be improved.This requires beam training to determine the optimal beam direction and achieve beam alignment.Binary beam search based on hierarchical codebook is a time-efficient method of beam training.By increasingly narrowing the searching interval,the optimal beam direction can be estimated with small training overhead.Unfortunately,due to hardware restrictions,the transition area between two adjacent training beams often leads to error propagation,which limits the accuracy of beam alignment.In this paper,we propose three algorithms to tackle the error propagation.The first is referred to as shifting beams with overlapped interval.The training beams of every stage of binary search are shifted,then a threshold is set to detect whether the optimal beam direction is in the center of the search interval.Compared to traditional binary search,this algorithm lowers the beam misalignment error rate when the optimal beam direction is in the center and edge of the searching interval.The second algorithm is also designed to tackle the high misalignment rate of the center and edge area by using three additional beams which aligned the high-misalignment-rate areas.The third one can be regarded as a combination of the first and second algorithm.In this algorithm,a additional beam aligning the center area is transmitted after shifting the two original beams.Besides,a hierarchical codebook for multi-user beam alignment and a beam alignment algorithm based on this codebook is proposed in this paper,which can simultaneously achieve beam alignment for all users in a cell.Simulation results demonstrate that this algorithm can achieve satisfying trade-off between training overhead and misalignment error rate.