Low complexity large scale MIMO processing

大规模多输入多输出(MIMO)系统，或称大规模MIMO，是多用户多天线技术，构成了最近文献中提出的解决即将到来的无线网络紧缩的领跑者之一，这将导致物联网和5G系统的订户数量及其数据饥渴应用的不断增加。在该方案中，我们将解决在大规模MIMO发送和接收两端设计低复杂度的信号处理算法的问题。我们将利用我们在为中小型MIMO系统设计高效译码算法方面的经验，并利用海量MIMO的性质来提出新的技术，以利用新的信道特性，即信道矩阵稀疏性和围绕其平均值的信道硬化，同时保持高效的信号处理算法。系统复杂度将分布在能够承受更复杂算法的基站和具有低复杂度处理约束的终端之间。在基站，我们将考虑设计有限反馈信道状态信息的模拟和数字混合波束形成算法，以减少昂贵的射频链数量，提高系统效率。然后，我们将考虑多用户调度算法的设计，该算法利用稀疏信道矩阵来通过可管理大小的块有效地减少稀疏信道矩阵。还将对波束形成和调度所需的训练序列的设计进行深入的调查和分析。最后，我们将解决分布式大规模MIMO系统的设计问题，包括实时处理、估计误差、同步和干扰管理等实际考虑因素。针对下一代无线通信标准的高效收发机设计的研究和开发将对加拿大IT行业产生重要的商业利益，这项提议将导致两名HQP的形成，他们将在加拿大经济中最重要和最快速增长的部门之一获得理论和实践技能。

Massive multiple-input multiple-output (MIMO) systems, or large scale MIMO, are multiuser multiple antenna technologies that constitute one of the front runners proposed in the recent literature for solving the impending wireless network crunch that will result from the ever increasing number of subscribers and their data-hungry applications envisioned for the Internet of Things and 5G systems. In this proposal, we will tackle the problem of designing low complexity signal processing algorithms at both ends of large-scale MIMO transmission and reception. We will use our experience in designing efficient decoding algorithms for MIMO systems at a small-to-medium scale and exploit the nature of the massive MIMO in order to propose novel techniques taking advantages of the new channel characteristics, i.e., channel matrix sparsity and channel hardening around its mean value, while maintaining efficient signal processing algorithms. The system complexity will be distributed between the base stations, which can afford more complex algorithms, and the end-user terminals with low complexity processing constraints. At the base stations, we will consider the design of hybrid analog and digital beamforming algorithms with limited feedback on channel state information in order to decrease the number of costly radio frequency chains and improve the system efficiency. Then, we will consider the design of multiuser scheduling algorithms that take advantage of the sparse channel matrix for efficiently reducing it by blocks of manageable sizes. The design of training sequences required for beamforming and scheduling will also be investigated and analyzed in depth. Finally, we will tackle the problem of designing distributed massive MIMO systems, with practical considerations such as real-time processing, estimation errors, synchronization and interference management. Research and development on efficient transceivers designs targeting next generation wireless communications standards will have important commercial benefits to the Canadian IT sector, and this proposal will lead to the formation of two HQP who will gain both theoretical and practical skills in one of the most important and fast growing sectors of the Canadian economy.********************