Title:  Machine Learning for Traffic Prediction and Communication Efficient Data Analytic in Wireless Networks

Committee: 

Dr. Fekri, Advisor

Dr. Sivakumar, Chair

Dr. AlRegib

Dr. Anderson

Abstract: The objective of the proposed research is to design predictive models by learning the mobile users behaviors in wireless networks and design efficient task-aware models at the wireless-network edge for Internet of Things (IoT). The integrated framework aims to reduce the data costs and network operation costs of mobile users and network operators, respectively, and enhance the communication traffic by minimizing the latency. We propose to tap into machine learning (ML) and probabilistic graphical models to capture the dependency relationships between observed and unobserved variables, and develop efficient computation algorithms that exploit the graph structure.