報告題目：A Fresh Look at Feedback in Wireless Network Capacity
主講嘉賓： Jinyuan Chen Louisiana Tech University
時間：2016年 9月2日（周五）下午4:00 -- 5:00
邀請人：張勝利 副教授
地點：深圳大學南區(qū)基礎實驗樓北座信息工程學院N710會議室
報告摘要:
The fundamental capacity limits of communication systems with feedback was first studied by Shannon in 1956. Since then, results have shown that for some channels, feedback does not increase capacity at all, whereas for others it can lead to significant gains. Yet, despite many decades of research, the capacity of wireless networks with feedback is open for most networks of interests, particularly when the feedback is delayed, quantized, noisy or intermittent. This talk presents new capacity results that exploit non-ideal feedback to significantly improve the performance of wireless networks. In particular, we show that for a certain heterogeneous wireless networks, completely outdated feedback of the channel state information (CSI) can be as good as instantaneous transmitter CSI in terms of network capacity degrees of freedom. We also derive the exact degrees of freedom for broadcast channels under feedback of delayed and noisy CSI. Finally, we show how implicit feedback, where a node overhears a signal designated for a different node, can increase capacity in certain two-user relay networks. We close the talk with a discussion of the impact of imperfect feedback on the capacity of massive MIMO systems.
嘉賓簡介：
Jinyuan Chen is currently an Assistant Professor at Louisiana Tech University, US. He was a postdoc scholar at Stanford University from 2014 to 2016. He received the PhD degree from Telecom ParisTech in 2013, the M.Sc. degree from Beijing University of Posts and Telecommunications in 2010, and the B.Sc. degree from Tianjin University in 2007. He was a visiting scholar and Assistant Specialist at the University of California, Irvine during the summers 2013, 2016. His research interests include information theory and communication theory, as well as their applications in wireless networks, physical layer security, distributed storage, distributed computing, big data and bioinformatics.
