Research On Key Technologys Of Hybrid Precoding For Millimeter Wave MIMO System

With the rapid development of mobile Internet and mobile terminal, the updating of communication technology is motivated by the demand of mobile data service. Multi antenna technology, as the key technology to improve the spectrum efficiency and system throughput rate, has become the most active part in the field of wireless communication. Currently, the study of MIMO technology is developed from traditional MIMO to millimeter wave MIMO, from two dimensional sapce to three dimensional space, from small scale MIMO to a large scale MIMO. In this paper, we will study key technologys of hybrid beamforming for millimeter wave MIMO System. The main work of this paper are as follows:Firstly, digital precoding technology, analog beamforming and the hybrid precoding technology based on the sparse characteristics of millimeter wave is studied. Taking IEEE 802.11ad as an example, the beam training mechanism of antenna array is analyzed. Precoding matrix design criteria in shared hybrid systems as well as design algorithm is elaborated. MATLAB simulation shows that the performance of the hybrid precoding system can be close to digital prcoding.Secondly, this paper focus on the design of codebook for large scale 3D MIMO system based on limited feedback. When uniform planar array is configured at the base station and path loss is taken into consideration, this paper proposes a new codebook design method based on Kronecker product, which requires the amplitude of the beam coverage area is not less than ? attenuation of that of the beam center. The proposed algorithm starts from the edge of the coverage of the base station, calculates each code word successively until the whole area is covered. Compared with the traditional DFT codebook, the characteristics of the proposed codebook are analyzed. MATLAB simulation shows that the proposed vertical code can match the channel better and improve the spectrum efficiency of the system.Thirdly, this paper focuses on analog beamforming technology of hybrid system. Taking the advantage of the sparsity of the millimeter wave channel, a low complexity beam search algorithm is proposed based on the fact that he channel gain between the matched antenna arrays is much higher than that of the unmatched antenna arrays. Compared with exhaustive search algorithm and pair-wise search algorithm, the complexity of the proposed algorithm is much lower. Matlab simulation shows that the performance of the proposed algorithm is only slightly lower than that of the optimal exhaustive search algorithm, and is better than that of pair-wise algorithm. Considering the huge computation overhead for exhaustive search, the proposed algorithm is more easily achieved.Finally, this paper focuses on the feedback of Equivalent channel. Based on the high speed systolic array and Givens rotation, a matrix compression scheme is proposed. Pipeline structure is implemented with strong expansibility. MATLAB fixed-point simulation shows that the algorithm has good stability. The link level simulation is carried out on the simulation platform of IEEE 802.11aj (45 GHz), and the reasonable quantization method is given.