Energy-Aware Fault Tolerance and Real-time Wireless Sensor Network for Control System

Type PhD Proposal
Start Date April 4, 2017 10:00 AM
End Date April 4, 2017 12:00 PM
Location Eli Lily Room
Organizer Name
Speaker Name Wenchen Wang
Speaker Affiliation Ph.D. Candidate, University of Pittsburgh
Abstract Wireless control systems bring several advantages over traditional wired industrial monitoring and control systems, including self-organization, flexibility, rapid deployment, and maintenance. However, wireless network delay and packet loss can result in two main challenges for the control system: instability and performance degradation. This thesis aims at solving these two major challenges. We develop a fault-tolerance network design and a novel computational model to quantify the control system stability requirement based on the network design. We explore a network reconfiguration framework with time-correlated link failures to improve control system performance. Different techniques have been explored to embody this framework. Accordingly, a precise analytical model and different reconfiguration algorithms have been developed to quantify and improve the performance, respectively. In this document, I summarize my progress in the development of wireless sensor network for the control system and propose to continue the research in two aspects. Firstly, I propose to further explore a network reconfiguration scheme based on a space-correlated link failure model. I will conduct a case study to evaluate the reconfiguration scheme. Then, I propose to develop a cross-layer network flow scheduling algorithm with known application demands. I will study the impact of real-time scheduling on control system performance via another case study.
Committee Rami Melhem, Youtao Zhang, Daniel Cole Advisor:Daniel Mosse