学术报告 | Rate Splitting for MIMO Wireless Networks: A Promising PHY-Layer Strategy for 5G and Beyond

发布者:孙威发布时间:2019-09-11浏览次数:765

题目:Rate Splitting for MIMO Wireless Networks: A Promising PHY-Layer Strategy for 5G and Beyond

时间:9月12日,14:00

地点:无线谷1402会议室



演讲人:Yijie (Lina) Mao


Presentation Title

Rate Splitting for MIMO Wireless Networks: A Promising PHY-Layer Strategy for 5G and Beyond


Presenters

Dr. Yijie (Lina) Mao, Postdoctoral Research Associate

Imperial College London, London, United Kingdom

Email: ym2516@ic.ac.uk

Phone: +44 7421817446.


Yijie (Lina) Mao is currently a postdoctoral research associate with the Communications and Signal Processing Group (CSP), Department of the Electrical and Electronic Engineering at the Imperial College London (London, United Kingdom). She received the B.Eng. degree from the Beijing University of Posts and Telecommunications, and the B.Eng. (Hons.) degree from the Queen Mary University of London (London, United Kingdom) in 2014. She received the Ph.D. degree in the Electrical and Electronic Engineering Department from the University of Hong Kong (Hong Kong, China) in 2018. She was a Postdoctoral Research Fellow at the University of Hong Kong (Hong Kong, China) from Oct. 2018 to Jul. 2019. Her research interests include Multiple Input Multiple Output (MIMO) communication networks, rate-splitting and non-orthogonal multiple access for 5G and beyond.


Research Area: MIMO communication networks, rate-splitting and non-orthogonal multiple access



Presentation Abstract

MIMO has grown beyond the original point-to-point channel and nowadays refers to a diverse range of centralized and distributed deployments. Numerous techniques have been developed in the last decade for MIMO wireless networks, including among others MU-MIMO, CoMP, Massive MIMO, NOMA, millimetre wave MIMO. All those techniques rely on two extreme interference management strategies, namely fully decode interference and treat interference as noise. Indeed, while NOMA based on superposition coding with successive interference cancellation relies on strong users to fully decode and cancel interference created by weaker users, MU-MIMO/Massive MIMO/CoMP/millimetre wave MIMO based on linear precoding rely on fully treating any multi-user interference as noise. In the presence of imperfect channel state information at the transmitter (CSIT), CSIT inaccuracy results in additional multi-user interference that is treated as noise by all those techniques.

In this talk, we depart from those two extremes of fully decode interference and treat interference as noise and introduce the audience to a more general and more powerful transmission framework based on Rate-Splitting (RS) that consists in decoding part of the interference and in treating the remaining part of the interference as noise. This capability of RS to partially decode interference and partially treat interference as noise enables to softly bridge and therefore reconcile the two extreme strategies of fully decode interference and treat interference as noise.

In order to partially decode interference and partially treat interference as noise, RS relies on the transmission of common (degraded) messages decoded by multiple users, and private (nondegraded) messages decoded by their corresponding users. As a result, RS pushes multiuser transmission away from conventional unicast-only transmission to superimposed unicast multicast transmission and leads to a more general class/framework of strategies. For instance, in a MISO Broadcast Channel, RS is shown to encompass NOMA and MU-MIMO with linear precoding as special cases. Through information and communication theoretic analysis, RS is shown to be optimal (from a Degrees-of-Freedom region perspective) in a number of scenarios and provide significant benefits in terms of spectral efficiencies, reliability and CSI feedback overhead reduction over conventional strategies used/envisioned in LTE-A/5G that rely on fully treat interference as noise or fully decode interference. The gains of RS will be demonstrated in a wide range of scenarios: multi-user MIMO, massive MIMO, multi-cell MIMO/CoMP, overloaded systems, NOMA, multigroup multicasting, mmwave communications, communications in the presence of RF impairments and coded caching. Signal processing and optimization techniques used to achieve the fundamentally promised gains are further presented and elaborated. Open problems and challenges will also be discussed.