Researchand Implemention Of Massive MIMO Precoding Technology

In massive MIMO system, the base station was equipped with a large number of antennas.It could achieve higher energy and spectral efficiency, and thus become one of key technicalfor 5G system. However the large number of antennas could significantlyincreasethe complexity of the precoding algorithm.This paper will focus on the research of low complexity precoding algorithms in massive MIMO.First of all, we give the massive MIMO system architecture and capacity analysis, research on the LTE-Advanced downlink multi-user precoding technology and analyzethe key points of massive MIMO precoding technology.Secondly, weresearch theMaximalRatioTransmit(MRT)and Zero forcing(ZF)precodingalgorithmfor Massive MIMO. The complexity and system spectral efficiency of the two algorithms are analyzed.Additionally, we study the Truncated Polynomial Expansion(TPE) procoding algorithm with less complexity, and a Schmidt orthogonalinterference procoding algorithm is introduced into massive MIMO scene. TPE algorithmusing polynomial cut off approximation to calculate the inverse matrix for positive definite hermite matrix. The inverse process is used in the iteration for procoding algorithm.Schmidt orthogonal interference procoding algorithm eliminates interference with Schmidt orthogonal iteration. A simulation platform is built to verify the feasibility of Schmidt orthogonal interference algorithm. Simulation result shows that the proposed algorithm has less complexity but higher system performance compared with ZF algorithm.Finally, the thesis designs a Schmidt orthogonalinterference procoding hardware module using Xilinx VC707 FPGA. The test of functional simulation and the under link test can verify the reasonableness of this design.