报告题目：CSI acquisition inmassive MIMO systems
报 告 人：尹海帆博士
地 点：武汉光电国家实验室F401
时 间：2016年9月12日下午3点
内容简介：
Massive MIMO isexpected to enable much higher throughput and energy efficiency compared totraditional MIMO systems. It is considered to be a potential key technology infuture 5G standards. Despite the promise, there are still open problems thatlimit the potential of massive MIMO. Among them, this talk focuses on some ofthe challenges related to the acquisition of Channel State Information (CSI) inboth Time-Division Duplex (TDD) mode and Frequency Division Duplex (FDD) mode.
In channel estimationphase of TDD mode, the pilot contamination effect constitutes a bottleneck foroverall performance. We present novel approaches that tackle this problem byexploiting second-order statistics of the user channels. We demonstrateanalytically that in the large-number-of-antennas regime, the pilotcontamination effect completely vanishes under a certain condition on thechannel covariance. This condition states that the support of multipathangle-of-arrival (AoA) of interference is non-overlapping with the AoA supportof the desired channel. This condition is tightly related to the low-ranknessproperty of channel covariance. Furthermore, we show that such a low-rankproperty is not inherently related to ULA. It can be generalized to non-uniformarray, and more surprisingly, to two-dimensional distributed arrays.
Although the proposedMMSE-based estimator leads to full pilot decontamination under the non-overlappingcondition in angular domain, in practice this condition is unlikely to hold atall times, owing to the random user location and scattering effects. To thisend, we propose novel robust channel estimation schemes that combine the meritsof MMSE estimator and the known amplitude based projection method. Asymptoticanalysis shows that the proposed methods require weaker conditions compared tothe known methods to achieve full decontamination.
Finally, we tackle theCSI feedback problem for massive MIMO operating in FDD mode by novelcooperative feedback mechanisms which are enabled by Device-to-Device (D2D)communications. The exchange of local CSI among users allows to construct moreinformative forms of feedback based on this shared knowledge. For a givenfeedback overhead, the sum-rate performance is assessed and the gains of ournovel methods compared to a conventional massive MIMO setup without D2D areshown.
尹海帆博士简介
尹海帆博士：Haifan Yin received thePh.D. degree from Telecom ParisTech and the B.Sc. degree from HuazhongUniversity of Science and Technology, in 2015 and 2012 respectively. From 2009to 2011, he had been with Wuhan National Laboratory for Optoelectronics, China,working on the implementation of TD-LTE systems as an R&D engineer. In January2016, he joined Sequans Communications, France, as a DSP engineer. His currentresearch interests include signal processing, channel estimation, cooperativenetworks, and large-scale antenna systems. He won the Chinese Government Awardfor Outstanding Students Abroad (ranked 1st in France) in 2015 and IEEE BestReadings on Massive MIMO in 2014
