ECE is pleased to announce two talks by Prof. Wei Yu, the inaugural speaker for the Ian F. Blake Lectureship, on Friday February 21, 2020.
The Ian F. Blake Lectureship was established in July 2019 with a gift made to UBC by Dr. Vijay Bhargava. This series of lectures honours the work of Ian F. Blake, a Canadian pioneer in the field of Information and Communications Theory and Honorary Professor in ECE.
Talk 1: Spatial Deep Learning for Wireless Scheduling
February 21, 11 am, Kaiser 2020
What is the role of machine learning in the design and optimization of communication systems? In this talk, we examine the well-known challenging problem of optimal scheduling of interfering links in a dense wireless network, and point out that the traditional optimization approach of first estimating all the interfering channel strengths then optimizing the scheduling based on the model may not always the best. This is because channel estimation is resource intensive, especially in a dense network. To address this issue, we investigate the possibility of using a deep learning approach to bypass channel estimation and to schedule links efficiently based solely on the geographic locations of transmitters and receivers. This can be accomplished either by supervised learning using locally optimal schedules generated from fractional programming for randomly deployed device-to-device networks as training data, or by unsupervised learning. In both cases, we use a novel neural network architecture that takes the geographic spatial convolutions of the interfering or interfered neighboring nodes as input over multiple feedback stages to learn the optimum solution. The resulting neural network gives excellent performance for sum-rate maximization and is capable of generalizing to larger deployment areas and to deployments of different link densities. Further, we propose a novel approach of utilizing the sum-rate optimal scheduling heuristics over judiciously chosen subsets of links to provide fair scheduling across the network, thereby showing the promise of using deep learning to solve discrete optimization problems in wireless networking. (Joint work with Wei Cui and Kaiming Shen)
Talk 2: Perfect Hashing, Hypergraph Covering, Identification Capacity, and Collision-Free Feedback for Massive Random Access
February 21, 3 pm, Kaiser 2020
Designing multiple-access protocols capable of supporting massive but sporadic machine-type communications is a key requirement for the future integration of wireless cellular communication systems with Internet-of-Things. In this talk, we consider a massive random access network in which a small random subset of K active users, out of a large number of N total potential users, seek to communicate with a base station. We examine an approach in which the base station first determines the user activities based on an uplink pilot phase, then broadcasts a common feedback message to all the active users for the scheduling of their subsequent data transmissions. Our main question is: What is the minimum amount of common feedback needed to schedule K users in K transmission slots while completely avoiding collisions? Instead of a naive scheme of using K log(N) feedback bits, this talk presents upper and lower bounds to show that the minimum number of required common feedback bits scales linearly in K, plus an additive term that scales only as O(log log(N)). The solution to this problem is closely related to that of constructing a minimal family of perfect hash functions and also that of constructing a minimal covering of a complete hypergraph. The solution has a curious resemblance to the notion of identification capacity. (Joint work with Justin Kang)
Wei Yu received the B.A.Sc. degree in Computer Engineering and Mathematics from the University of Waterloo, and M.S. and Ph.D. degrees in Electrical Engineering from Stanford University. He has been with the Electrical and Computer Engineering Department at the University of Toronto since 2002, where he is now Professor and holds a Canada Research Chair (Tier 1) in Information Theory and Wireless Communications. Prof. Wei Yu currently serves as a Vice President of the IEEE Information Theory Society, and has served on its Board of Governors since 2015. He was an IEEE Communications Society Distinguished Lecturer (2015-16), an Area Editor for the IEEE Transactions on Wireless Communications (2017-20), and chaired the Signal Processing for Communications and Networking Technical Committee of the IEEE Signal Processing Society (2017-18). He received the IEEE Communications Society Award for Advances in Communication in 2019, the IEEE Marconi Prize Paper Award in Wireless Communications in 2019, the IEEE Signal Processing Society Best Paper Award in 2017 and 2008, the Journal of Communications and Networks Best Paper Award in 2017, an E.W.R. Steacie Memorial Fellowship in 2015, and an IEEE Communications Society Best Tutorial Paper Award in 2015. Prof. Wei Yu is a Fellow of IEEE, a Fellow of Canadian Academy of Engineering, and a member of the Royal Society of Canada’s College of New Scholars, Artists and Scientists.