基于混合导频辅助的大规模MIMO上行链路信道估计

针对多小区多用户大规模MIMO上行链路传输，本文提出了一种基于混合导频的信道估计方案.通过在一个数据帧内设计时分导频及隐含导频两种导频格式，进行对信道参数的联合估计.相较传统的单一导频方案，混合导频可实现两种导频的优势互补.在天线数较大的前提下，分析推导出利用混合导频信道估计的上行链路吞吐率近似封闭解，从而明确了混合导频中时分导频与隐含导频的优化准则.仿真结果表明，时分导频适用于数据帧长较长的传输环境，隐含导频在数据帧长较短的情况下更具优势，而两者的结合可极大提高系统的性能.

Hybrid Pilot-aided Channel Estimation in Massive MIMO Uplink

This paper concerns with the uplink of multiuser multi-cell massive MIMO systems, and presents a hybrid pilot-aided channel estimation technique, where both time-multiplexed (TM) pilots and time-superimposed (TS) pilots are jointly employed for channel estimation in a frame. We show that the design of hybrid pilots combines the complementary advantages between TM pilots and TS pilots, and thereby, allows for a superior solution compared to conventional pilot schemes. In large-number-of-antennas regime, we asymptotically analyze the closed-form expression of the uplink throughput, and obtain the pilot-design-criteria for using both TM pilots and TS pilots. Numerical analysis and system simulation results show that TM pilot is more effective for the uplink transmission of a relatively large frame-size, while TS pilot has the substantial advantage in frame-based transmission of a limited size. It is obvious that such hybrid scheme can significantly improve the performance of massive MIMO systems.