Research On The Uplink Receiving Technology In Massive MIMO Distributed Antenna Systems

Massive MIMO(Multiple-Input Multiple-Output)technology ensures higher data transfer rate,stronger system reliability,lower energy consumption,so it's being the focus of next-generation mobile communi-cations technology.The distributed antenna technology improves the spectral efficiency,reduces the blind spots,especially in the application of indoor or outdoor hot coverage,so it's a very promising technology.When a large number of RAUs(Remote Access Unit)are deployed in a distributed system,massive MIMO technology can be a perfect application.In the massive MIMO distributed antenna systems,the base station is configured with hundreds or even more antennas and serves multiple users at the same time,resulting in a sharp increase in the complexity of the base station receiver.The design of low complexity receiver becomes one of the key issues.Based on the massive MIMO distributed antenna systems,three aspects of contents are studied in this paper:the detection method based on Neumann series approximation,channel sparsity technology in uplink receivers,and the uplink detection technology under 5G experimental system.Firstly,the model of massive MIMO distributed antenna system and fading characteristics of wireless channel are stuied.And then the channel model of massive MIMO distributed antenna system channel is introduced.In addition,the uplink sum data rate of massive MIMO distributed antenna system is derived,and the typical linear detection algorithm is introduced.Secondly,the detection algorithm based on Neumann series approximation is analyzed.The MMSE-NSA(Minimum Mean Square Error-Neumann Series Approximation)algorithm is mainly discussed.The convergence,complexity and error of MMSE-NSA algorithm are analyzed.The simulation results show that the MMSE-NSA algorithm can approximate the performance of the MMSE algorithm with lower compu-tational complexity when the number of base station is large than that of the user relatively.Then,MMSE-TNSA(MMSE-Triangle NS A)algorithm is proposed for the problem of the slow convergence of MMSE-NSA algorithm in the relevant channel.The MMSE-TNSA algorithm improves the convergence speed and stability of the MMSE-NSA algorithm by introducing more non-diagonal elements.Then,the channel sparsity technology in the uplink receiver is analyzed.The sparsity of the channel matrix in the uplink receiver is considered to obtain the massive sparse matrix.Then the PARDISO(Parallel Direct Sparse Solver)is used to invert the massive sparse matrix.The simulation results show that the channel matrix can be greatly thinned with a little performance loss in the massive MIMO distributed antenna systems.In the multi-core case,the optimized PARDISO solver in the Intel MKL(Math Kernel Library)library is used to solve the inverse of massive sparse matrix in the time of(?)(K).Finally,the uplink detection technology under the 5G experimental system is analyzed.The architec-ture of 5G experimental system is described,and the multi-user interference suppression receiver under joint MMSE scheme is studied.In order to reduce the computational complexity of MMSE algorithm,MMSE-DA(MMSE-Diagonal Approximation)algorithm,which uses a diagonal matrix to approximate the inter-user interference,is analyzed.Thus the dimension of matrix is reduced from the number of transmit antennas to the number of streams sent by the user.In order to further improve the system performance,the iterative soft decision interference cancellation algorithm is studied.The simulation results show that the performance of system can be improved by 1 time of iterations.