School of Computer Science and Engineering, University of Electronic Science and Technology of China, China

Biography: Guolin Sun received his B.S., M.S. and Ph.D. degrees all in Communication and Information Systems from the University of Electronic Science and Technology of China (UESTC), Chengdu, China, in 2000, 2003 and 2005 respectively.
After Ph.D. graduation in 2005, he has got eight years industrial work experience on wireless research and development for LTE, Wi-Fi, Internet of Things (ZIGBEE and RFID, etc.), Cognitive radio, Localization and navigation. Before he joined the School of Computer Science and Engineering, University of Electronic Science and Technology of China as an Associate Professor in Aug. 2012, he worked in Huawei Technologies Sweden.
Dr. Guolin Sun has filed over 40 patents, and published over 40 scientific conference and journal papers, acts as TPC member of conferences. Currently, he serves as a vice-chair of the 5G oriented cognitive radio special interest group (SIG) of the IEEE Technical Committee on Cognitive Networks (TCCN) of the IEEE Communication Society. His general research interests include software defined networks, network function virtualization, radio resource management.

Speech Title: Resource Slicing and Customization for Heterogeneous Traffics in Virtualized Radio Access Network with Deep Reinforcement Learning

Abstract: The emerging future generation 5G technology is expected to support service-oriented virtualized networks where different network applications provide unique services. 5G networks are envisioned to allow different slices to co-exist in a common network and satisfy the requirements of diverse users. In networks with heterogeneous traffics, service operators are required to provide services in isolation since each operator has its own defined performance requirements. However, achieving an efficient resource provisioning mechanism for such traffics is very challenging. This paper proposes a coarse resource provisioning scheme and dueling deep reinforcement learning based dynamic resource slicing refinement algorithm for virtualized radio access network. Firstly, coarse resource provisioning scheme provisions and allocates radio resource to slices based on preferences and weights at different base stations. Secondly, reinforcement learning based slicing refinement autonomously adjusts the resource allocated to slices in order to balance satisfaction and resource utilization. Then, a shape-based resource allocation algorithm is proposed to customize the diverse requirements of users to improve satisfaction and resource utilization. A comprehensive performance evaluation is conducted against state-of-the-art solutions based on 5G air-interface design. The results reveal that the proposed algorithm balances satisfaction and resource utilization with 80% of the available resources. The algorithm also provides performance isolation such that, a sudden change in user population in one slice does not affect the others.