Skip to main content

Seminar:  Toward Self-Managing, Context-Aware Networked Systems

Date:February 3, 2014 
Talk Title:Toward Self-Managing, Context-Aware Networked Systems
Speaker:Patrick Tague
Time & Location:12:00pm - 1:00pm
CIC Building, Pittsburgh

Abstract

Recent growth of wireless and mobile networking has been staggering, with system evolution largely being driven by customer demands for support of newer technologies to provide improved performance.

However, much of the system architecture supporting devices and services is based on decades-old design features and constraints.

While there is a path forward using the systems that have emerged, there is a significant opportunity to re-imagine networked system architectures to enable new services and capabilities without a clean-slate re-design.  Our approach is based on deeply integrating computation into network components system-wide and introducing flexibility in device operating characteristics, enabling near-real-time network situational awareness and allowing devices and services to self-adapt based on their collective view of the system.

By combining explicit cross-layer and cross-device information sharing with modern capabilities of software-defined radio (SDR), cloud computing, and machine learning, we can effectively train network components to learn and adapt to varying network conditions, application context, user demands, and device capabilities.  In this talk we'll describe how this deeply integrated context-awareness can be applied to robust wireless communication, efficient mobile/cellular networking, privacy-preserving sensing in smart environments, and adversarial settings.

Speaker Bio

Patrick Tague is a member of the research faculty at Carnegie Mellon Silicon Valley. He received MS and PhD degrees in Electrical Engineering from the University of Washington and BS degrees in Mathematics and Computer Engineering from the University of Minnesota. His PhD research focused on modeling and evaluating vulnerabilities in wireless ad-hoc and sensor networks and designing networking protocols that are robust to failure and attack. His general research interests include mobile communication systems, statistical modeling and inference, and network security.