Hierarchical Codebook Design and Search Algorithm in Millimeter-Wave MIMO

Abstract

In a millimeter-wave communication system, using a large-scale antenna array to improve system performance also brings about the problem that how to efficiently search the best beam direction in the angle domain at both the transmitter and the receiver. Because of the problem that the use of the exhaustive search method may cause excessively high time complexity, the hierarchical search method is widely used, and the performance of the search results depends on the design of the hierarchical codebook. In this thesis, by combining two phase shifters in parallel, the problem of the phase shifter amplitude is effectively solved, and a hierarchical codebook is designed for the proposed millimeter wave MIMO system model. In addition, this thesis proposes two hierarchical search algorithms based on the tree structure features of hierarchical codebooks. In this thesis, by combining the properties of discrete-time Fourier transform with the traditional method of designing hierarchical codebooks, an optimization objective function is proposed, and the gradient descent algorithm is used to solve the optimization objective function to obtain the hierarchical codebook. According to the hierarchical search algorithm of the tree, taking the maximum SNR at the receiver as the goal, hierarchical codebook is searched. Finally, the Matlab simulation results show that the performance of the hierarchical codebook proposed in this thesis is better than the performance of the traditional hierarchical codebook.