CS Colloquium 10/14/2011 "From Open-Loop Sensing to Closed-Loop Sensing and Control: Challenges to Embedded Wireless..."Posted Oct. 7, 2011
From Open-Loop Sensing to Closed-Loop Sensing and Control: Challenges to Embedded Wireless Networking
Dr. Hongwei Zhang
Computer Science Department
Wayne State University
4:00 pm, 14 October 2011, 228 MSB
(NOTE: This week's talk starts at 4:00pm instead 3:45pm)
Moving beyond open-loop sensing, embedded wireless networks are increasingly being applied to closed-loop, real-time sensing and control (e.g., for next-generation vehicles/transportation, industrial plants, and smart power grids). In supporting mission-critical tasks, closed-loop, real-time sensing and control requires predictable reliability and real-time in wireless networking. Nonetheless, wireless networking is subject to inherent dynamics and uncertainties within the system and the environment.
Within a system, wireless communication assumes complex spatial and temporal dynamics, network topology may constantly change due to node mobility, network traffic pattern may be dynamic due to event-triggered data traffic and time-varying control strategies, and application requirements on networking quality (e.g., reliability, timeliness, and throughput) may also vary over time and across different applications. Moreover, different dynamics may well interact with one another to yield complex behavior. Within the environment, a wide variety of factors affect wireless networking. Environmental factors such as temperature and humidity can affect wireless communication, electromechanical equipments can introduce complex environmental noise, moving objects or persons may introduce uncertainties to wireless communication, and malicious attackers may try to jam a network.
For predictable reliability and real-time in wireless networking, it is important to address the aforementioned systems and environmental dynamics. Given the potential resource constraints of embedded wireless networks, the solutions have to be light-weight and efficient too. In this talk, we will review these challenges, and we will discuss in detail how to address co-channel interference and how to enable real-time routing in highly-dynamic settings. We will also discuss how to enable measurement-based protocol analysis, as a part of our project in the NSF Global Environment for Network Innovations (GENI) program.
Hongwei Zhang received his Ph.D. in computer science and engineering from The Ohio State University in 2006, and he joined Wayne State University as an assistant professor thereafter. His work explores new theories, methods, and systems building-blocks that address dynamics and uncertainties in networked systems that involve wireless networks, sensor networks, embedded networks, and the Internet.
Presently, with support from the National Science Foundation (NSF) and industry (Ford Research, GM Research, and Detection Innovation Inc.), he is especially interested in the modeling, algorithmic, and systems issues in wireless, vehicular, sensor, control, and embedded networks. For instance, with support from the NSF CAREER program and the NSF CPS program, his group investigate field-deployable mechanisms for reliable, real-time, and secure wireless networking, and they investigate cross-layer approaches to taming cyber-physical uncertainties in wireless networked sensing and control; as a part of the NSF GENI program, they develop the theoretical and systems foundations for experimentation and service provisioning in federated, networked sensing. Their work has also provided foundational components for several large scale wireless network systems including the NetEye experimental infrastructure (which has 130 IEEE 802.15.4 nodes and 15 802.11b/g nodes) and the DARPA sensornet project ExScal (which, with its 200-node 802.11b mesh network and 1,200-node mote network, was the world's largest wireless sensor network and 802.11b mesh network deployed at its time).His work has been published in premier journals/conferences in networking, distributed computing, real-time systems, and dependable systems. His papers have been selected as a Spotlight Paper of the IEEE Transactions on Mobile Computing (TMC) and a Best Paper Candidate for IEEE International Conference on Network Protocols (ICNP) in 2010. He is a recipient of the NSF CAREER Award. More information about his work can be found from his website at http://www.cs.wayne.edu/~hzhang/.