EECS Seminar: Quality of Time - Enabling Robust, Secure and Efficient IoT
Circuits & Systems, Signals & Systems
Abstract: Nanoscale electronics, pervasive connectivity and cloud computing have together ushered in the internet of things (IoT). Accurate and reliable knowledge of time is essential for IoT systems to perform their tasks via a complex web of feedback loops where data are collected from myriads of sensors, distributed and processed through multitiered networks and distributed computing substrates, and, eventually influences and controls the states of natural, engineered and human systems. Applications depend on precise knowledge of time with a diversity of semantics for purposes such as coordinated sensing, efficient wireless communication, correctly ordered computation, location awareness and appropriately choreographed actuation.
Despite it being so critical, time is taken for granted with little thought given to the uncertainty in the knowledge of time. The uncertainty in the knowledge of time varies across network nodes, hardware and software layers, and over time. Moreover, many of the methods used in modern computing systems for improved performance make uncertainty worse. Oblivious of these uncertainties, system designs typically overcompensate, resulting in systems that are over designed, inefficient and fragile. This talk presents research under Roseline, an NSF CPS Frontier Project led by UCLA with collaborators from CMU, UCSB, UCSD, and the University of Utah, where we formalize uncertainty in the knowledge of time as a quality of time (QoT) metric that is made observable and controllable in order to robustly support time-aware applications across the edge-middle-cloud tiers. QoT is made visible to the applications so that they can adapt, exchanged across the hardware and software layers so as to tune clock generation, OS scheduling etc., and propagated across the network so as to optimize distributed coordination. The talk will describe the enabling system abstractions and run-time mechanisms that we have developed to help realize the QoT concept. Lastly, QoT can also be manipulated by adversarial actors such as a compromised OS and network nodes, causing time-aware applications to fail. The talk will close by describing some of the vulnerabilities and uncertainties relating to time that exist in current systems, and methods to mitigate them.
Bio: Mani Srivastava is on the faculty at UCLA where he is associated with the Department of Electrical and Computer Engineering and has a joint appointment in the Department of Computer Science. His research focuses broadly in the area of human-cyber-physical and IoT systems that are secure, privacy-aware, AI/ML-enabled and application-driven. It spans problems across the entire spectrum of applications, architectures, algorithms and technologies in the context of systems and applications for mHealth and sustainable buildings. He is a fellow of both the ACM and the IEEE. More information about his research is available at his lab’s website: http://www.nesl.ucla.edu and his Google Scholar profile at https://scholar.google.com/citations?user=X2Qs7XYAAAAJ.