信道状态信息的有效获取是实现大规模MIMO通信的前提。在实际环境中，大规模MIMO信道呈现时、空非平稳和快变特性，不符合传统的信道假设，从而导致现有的信道状态信息获取技术性能劣化。本项目拟提出非平稳快变条件下的信道状态信息获取理论与方法，具体包括：1.探索统计信道状态信息获取原理，提出基于隐马尔科夫模型的估计方法，实现存储和计算能力约束下的最优估计；2.探索瞬时信道状态信息获取原理，提出基于动态因子分析和最优遗忘因子的估计方法，实现复杂度低、鲁棒性高的信道估计与追踪；3.在瞬时信道状态信息获取基础上提出信道预测方法，优化导频和发射序列，实现能谱高效的联合信道预测与数据检测；4.揭示信道状态信息获取中的误差传递与累积误差对系统性能的作用机理，分析制约系统极限性能的关键因素。通过此研究，本项目拟突破现有信道状态信息获取技术受限于实际大规模MIMO信道的非平稳快变特性而难以应用的瓶颈。

Accurate channel-state-information (CSI) acquisition is the premise for the realization of massive MIMO. As the number of antennas increases, massive MIMO channels become non-stationary in both time domain and spatial domain. In addition, the channel is fast-varying in ultra-high bands, which are expected to be widely adopted by 5G and future wireless communication systems. Since most of the existing methods for channel estimation are proposed under stationary and semi-static assumptions, the application of them in real massive MIMO systems becomes difficult. In this project, we focus on massive MIMO CSI acquisition methods in the non-stationary and fast-varying environment. Specifically, the following problems will be solved. 1. Based on hidden Markov modelling, the statistical CSI will be estimated. The optimal estimator will be obtained by solving the optimization problem under both storing and computing constrains. 2. Based on dynamic factor analyzing and optimal forgetting factor designing, robust and low-complexity instantaneous CSI acquisition with good tracking ability will be realized. 3. Based on the estimated CSI, a joint channel prediction and data detection will be proposed. By optimizing the pilot and transmitting sequence, the optimal spectrum-energy efficiency of the proposed scheme will be obtained. 4. Taking the error propagation into account, the cumulative channel estimation error and its effect on the performance of massive MIMO systems will be analyzed. The expected results of our project will provide solutions to the problems that cannot be solved by traditional methods and will facilitate the application of massive MIMO in future wireless communication systems.