SEMINAR:Deep Reinforcement Learning for Resource Allocation..
Speaker: Ajib Wessam
Title: Deep Reinforcement Learning for Resource Allocation in Multi-Band and Hybrid Wireless Networks
Date/Time: Nov 22, 2023, 18:00-19:00
Meeting ID: 918 5845 9130
Abstract: To address the challenge of radio spectrum scarcity and the more-and-more challenging requirements of emerging mobile applications, the use of multi-band systems is a promising solution for future cellular networks. In this seminar, we present a new measure of energy efficiency, and we address the problem of joint optimization of resource allocation in sub-6 GHz, millimeter wave and terahertz coexistence networks. The problem formulated is a nonlinear mixed-integer programming problem and it turns out to be an NP-hard problem. The objective is to maximize the system energy efficiency while respecting the QoS (throughput and signal-to-interference plus noise) constraints of each user. To find a solution within a reasonable amount of time, the problem is divided into two sub-problems to be solved iteratively. First, we solve the joint problem of user-base station association and channel allocation. Secondly, the power allocation sub-problem is solved. Next, we propose a multi-agent resource allocation solution based on deep reinforcement learning. Simulation results reveal the effectiveness of the proposed iterative and reinforcement-learning solutions compared to other solutions and illustrate the impact of multi-band communications on improving the system energy efficiency .
Bio:Wessam Ajib has been a professor at the Université du Québec à Montréal (UQAM) since June 2005. Prior to that, he gained academic and industrial experience working at Nortel Networks from 2000 to 2004 and a postdoctoral fellowship at École Polytechnique de Montréal from 2004 to 2005. He obtained his PhD and DEA from the École Nationale Supérieure des Télécommunications (ENST), Paris, France in January 2001 and September 1997. He also holds an engineering degree from the Institut Nationale Polytechnique de Grenoble (INPG) in physical instrumentation. His research interests are in the different areas of wireless networks, optimization and algorithms, resource allocation, multiple access, traffic scheduling and machine learning algorithms for wireless networks.