Research And Design Of Large-scale 3D MIMO System Signal Detection Scheme For 5G

3D MIMO technology is one of the key technologies in 5G wireless communication systems which can largely reduce inter-cell interferences,increase system throughput and spectral efficiency.Existing signal detection technologies are mostly designed for 2D MIMO system.However,with the acceleration of propulsion speed,a signal detection method for 3D MIMO systems which balances signal detection performance and computational complexity has become an important research direction.In this paper,the signal detection scheme for 5G is researched and combined with matrix approximate inversion method,spatial modulation technique and compressed sensing theory.The main work of this paper is as follows:First of all,we researched and analyzed signal detection schemes suitable for 2D MIMO systems including the maximum likelihood detection algorithm,the linear detection algorithm and the non-linear detection algorithm.During these schemes,the ZF algorithm,the MMSE algorithm,the SIC algorithm and the QR decomposition algorithm are mainly studied.We analyzed advantages and disadvantages of these schemes and pointed out the inadaptability of traditional signal detection method in 3D MIMO system.Secondly,based on the Richardson matrix decomposition method,the Jacobi iteration algorithm,the GS iteration algorithm and the SOR iteration algorithm are mainly analyzed in order to reduce the computational complexity of linear signal detection methods.We designed a signal detection scheme based on iteration residual correction method for large-scale 3D MIMO systems and analyzed its feasibility and effectiveness by means of Monte Carlo simulation.Thirdly,an improved structured subspace pursuit algorithm is proposed and a large-scale 3D MIMO system signal detection scheme based on compressed sensing theory is designed.Spatial modulation technology with high information transmission speed and high energy efficiency is introduced into 3D MIMO system.Due to the block sparseness characteristics of the uplink received signals,the structured compressed sensing theory is applied to detect the signals.At the same time,through the normalization process,the range of signal detection problems is extended from Gaussian white noise environment to non-Gaussian white noise environment,which means the signal detection problems can be well solved in a variety of environments.