基于混合移相器的毫米波大规模MIMO混合预编码算法研究OA

To address the problem of low energy efficiency in millimeter wave massive MIMO systems,a hybrid precoding iter-ative minimization algorithm based on discrete mixed precision phase shifter is proposed.Firstly,the algorithm follows the basic idea of alternating minimization algorithm and applies orthogonal constraints to the digital precoder by utilizing its orthogonal charac-teristics.Then it extracts the phase of the analog precoder from the equivalent precoder determined by the digital precoder and the unconstrained optimal digital precoder and perform mixed quantization processing on the extracted phase.Finally,the definition of dual norm and Hölder inequality are used to approximate optimized objective function.The simulation results show that the pro-posed algorithm significantly improves the energy efficiency of the system compared to other single precision algorithms while ensur-ing spectral efficiency close to the all digital precoding scheme.Moreover,the proposed algorithm has the properties of rather fast convergence speed and low computational complexity.