Titre: Antenna selection in large arrays
Conférencier: Sébastien Roy , Université de Sherbrooke, Canada
Lieu: ZOOM / https://us02web.zoom.us/j/83596206522?pwd=Z3d5ZXBBWENBejZlQmhkeUo5NmtPZz0 ,
Date et heure: jeudi le 19 mai 2022 de 14:00 à 15:00

Résumé: Nearly all modern high-throughput digital communication devices include multiple antennas, including cellular base stations, WiFi access points, WiFi terminals, cellular phones, etc. In fact, much of the promised capacity of 5G relies on the use of massive MIMO (multi-input, multi-output) techniques at the base station requiring a large number of antennas (typically 64 on current systems). Historically, a sharp inflection point in the system spectral efficiency trend of broadband wireless communications can be traced back to 2005, when multi-antenna techniques began to be widely deployed. However, as the number of antennas grows large, the replication of RF chains becomes problematic. Indeed, RF and analog components do not follow Moore’s law and are typically expensive and power-hungry. High-speed A/D and D/A converters in each branch also constitute cost and power bottlenecks. Since the antenna themselves are typically cheap, it makes sense to reduce the number of RF chains and select one or a subset of antennas through switching means. If the selection is appropriate, it can be shown that all the diversity gain and much of the capacity gain of a large antenna array can be captured, at a greatly reduced cost. Such a selection can be performed at the transmitter, the receiver, or both. From the receiver standpoint, various forms of generalized selection combining (GSC - selecting a subset of K antennas in an L-antenna array) have been extensively studied and the literature on the topic is abundant. However, most contributions overlook practical constraints imposed by the hardware switching matrix required for selection. It will be shown that imposing constraints on the selection process, such that the selected subset is theoretically suboptimal, actually leads to superior performance in practice when such hardware aspects are taken into account.

Note biographique: Sébastien Roy (M’00) received the B.Sc.A. and M.Sc. degrees in electrical engineering from Université Laval, Québec, QC, Canada, in 1991 and 1993, respectively, and the Ph. D. degree from Carleton University, Ottawa, ON, Canada in 2000. From 2000 to 2002, he was a National Sciences and Engineering Research Council of Canada (NSERC) Postdoctoral Fellow at Université Laval. In 2007, 2009, 2011, and 2018, he was invited professor with École Nationale Supérieure de Sciences Appliquées et de Technologie (ENSSAT), Lannion, France. In 2015, he was invited professor with Institut National de Sciences Appliquées (INSA), Rennes, France. From 2002 to 2012, he was sucessively Assistant, Associate (from 2005), and Full Professor (from 2010) at Université Laval. In 2012, he joined the Electrical and Computer Engineering Department, Université de Sherbrooke, Sherbrooke, QC, where he is currently Full Professor, was Department Chair from 2016 to 2017, and where he is working on system-oriented and implementation aspects of digital intelligence, signal processing, advanced wireless communications, and cybersecurity. He is co-founder and co-leader of Université de Sherbrooke’s Center of Excellence in Integrated Intelligent Systems (CESIIUS). He is also a co-founder of the Digital Distributed Intelligence Network (D2IN), a multi-university consortium currently being established in Quebec with funding from the Canadian Foundation for Innovation. Dr. Roy is active in industrial consulting and was involved in the organization of several international conferences. He is also recipient of multiple teaching, research, and technology transfer awards.

Voyez tous les séminaires >>>