Wireless communications networks are facing increasingly complex and constantly changing environments. A new requirement for future wireless networks is the intelligence to dynamically optimize the radio resources in an uncertain and unpredictable environment. Reinforcement learning is an essential tool to respond to network changes in a timely and appropriate way. In this talk, Gan will introduce the multi-armed bandit algorithm, a special form of reinforcement learning, and two design examples in wireless networks. The first one focuses on green communications with renewable energy, and the second one provides an enhanced channel selection method for software defined cognitive radio. We will demonstrate the performance gain using both computer simulations and a real-time testbed for smart grid applications.
Dr. Gan Zheng received the PhD degree in Electrical and Electronic Engineering from The University of Hong Kong in 2008. He is currently a Senior Lecturer in the Wolfson School of Mechanical, Electrical and Manufacturing Engineering, Loughborough University, UK. His research interests focus on 5G and beyond wireless networks, with current emphasis on edge caching, UAV communications, full-duplex radio, and artificial intelligence for wireless communications. He has published over 100 papers in international journals and conferences, which have received more than 4000 citations. He received six international best paper awards, including the 2015 GLOBECOM and the 2013 IEEE Signal Processing Letters Best Paper Award. He serves as an Associate Editor for IEEE Communications Letters.