Research On Channel Estimation And Data Transmission Based On Sparsity And Reciprocity Of Angular Domain In FDD Massive MIMO Systems

Massive MIMO,as one of the key technologies of the fifth generation mobile communication system,can effectively improve the spectrum efficiency and system capacity.In this thesis,channel estimation and data transmission in FDD massive MIMO systems are studied.Based on the sparsity and reciprocity of the angular domain in FDD massive MIMO systems,the angular sparsity based pilot reuse estimation method and BTBC scheme are applied to CSI acquisition and data transmission respectively.Firstly,the conventional channel estimation,precoding and diversity techniques are summarized.The problem of the conventional channel estimation method in FDD massive MIMO systems is analyzed.The linear and nonlinear precoding methods are compared.The encoding and decoding processes of the OSTBC are introduced.The technologies mentioned-above are simulated and analyzed for the following research and application.Secondly,the angular sparsity based pilot reuse channel estimation method is studied.The channel sparsity of angular domain in massive MIMO systems is analyzed.By exploiting the sparsity,the channel information can be extracted effectively,which reduces the processing complexity.Also based on the sparsity of angular domain,user grouping and pilot reusing methods are discussed.In terms of the angle reciprocity,the DL training and feedback overhead are compressed.Numerical results show that the angular sparsity based pilot reuse channel estimation method has good MSE performance and do reduce training and feedback cost.Thirdly,this thesis proposes an adaptive user grouping method that can be applied in angular sparsity based pilot reuse channel estimation.The channel characteristics under different angular spread are analyzed.According to the relationship between channel energy point distribution and angular spread,the dynamic acquisition algorithm for energy point index set and the corresponding adaptive user grouping method are provided in this thesis.The simulation results show that the channel estimation adopting the proposed adaptive user grouping method can maintain good estimation performance under different angular spread.Compared with the original method,the system overhead can be reduced obviously.Finally,the BTBC for FDD massive MIMO systems is studied.Based on angle reciprocity,the energy point index set and the beamforming target direction can be calculated and obtained through UL CSI.User grouping and beamforming are jointly processed and optimized.Based on OSTBC,a BTBC transmission scheme is given.According to the grouping criteria,a less complex user grouping method for data transmission is proposed.The simulation results show that the proposed grouping method can reduce the calculation complexity while maintaining the similar grouping result.In addition,they also show that the BTBC can provide good data transmission performance and can also enhance spectral efficiency for FDD massive MIMO systems.