Real-Time Wireless Control Networks

From Cyber-Physical Systems Laboratory
Revision as of 07:24, 29 November 2015 by Chengjie (talk | contribs) (→‎Team)
Jump to navigationJump to search

Team

Faculty: Chenyang Lu, Yixin Chen

PhD Students: Rahav Dor, Dolvara Gunatilaka, Bo Li

Alumni: Octav Chipara, Abusayeed Saifullah, Chengjie Wu, You Xu


Wireless sensor-actuator networks represent a new generation of communication technology for industrial process monitoring and control. With the adoption of WirelessHART, an open wireless sensor-actuator network standard, recent years have seen successful real-world deployment of wireless control in process industries. Industrial control systems impose stringent real-time and reliability requirements on wireless control networks. We are developing a new real-time scheduling theory and network protocols for real-time wireless control networks. Our research addresses both practical problems in current WirelessHART networks and fundamental challenges faced by future wireless control networks.

This work is sponsored by NSF through grant CNS-1320921 (NeTS).

Real-time Scheduling Theory for WirelessHART

We established a novel real-time scheduling framework for wireless control networks based on WirelessHART by bridging real-time scheduling theory and wireless networking.

Dynamic transmission scheduling: We devised both optimal and near optimal policies for dynamic priority scheduling of transmissions for real-time flows between sensors and actuators. We observed that transmission conflict (due to half-duplex radio of the nodes) plays a major role in communication delays, making the traditional real-time scheduling policies less effective for transmission scheduling in WirelessHART networks. Using this key observation, we designed an optimal algorithm based on branch and bound, and a heuristic called Conflict-aware Least Laxity First (C-LLF) for dynamic priority scheduling. C-LLF integrates the degree of conflicts associated with a transmission into LLF, and outperforms traditional real-time scheduling policies. This work was presented at [RTSS'10].

Fixed-priority transmission scheduling: For wireless control with firm requirements on network latency, a delay analysis is required to quickly assess the schedulability of the real-tme flows, specially for online admission control and workload adjustment in response to network dynamics. We provided an efficient analysis of the worst-case communication delays of periodic real-time flows that are scheduled based on fixed priorities in a WirelessHART network. The key insight in the analysis is to map real-time multi-channel transmission scheduling to real-time multiprocessor scheduling. This approach allows us to build on existing real-time scheduling theory while focusing on incorporating unqiue features of wireless communication in the analysis. Our analysis establishes safe upper bounds on end-to-end delays, thereby enabling effective schedulability tests for WirelessHART networks. We also proposed optimal and near-optimal priority assignment algorithms based on local search and heuristic search, respectively. Our search approach leverages the lower and upper delay bounds provided by our delay analysis to reduce the search space. The delay analysis and priority assignments were presented at [RTAS'11] and [ECRTS'11], respectively.

Cyber-Physical Co-Design for Wireless Control Systems

In a wireless control system, the choice of sampling rates of the feedback control loops must balance control performance and communication delays. A low sampling rate usually degrades the control performance while a high one may cause excessive communication delays also leading to degraded control performance. We addressed the scheduling-control co-design problem of sampling rate selection to optimize the overall control cost in a WirelessHART network. The resulting constrained optimization is challenging since it is non-differentiable, non-linear, and not in closed-form. We proposed and evaluated five optimization methods including greedy heuristic, subgradient method, simulated annealing based penalty method, and gradient descent method and interior point method upon a differentiable convex relaxation. Our result in this work has drawn some interesting conclusions towards co-design. In particular, we have shown the interior point method and the simulated annealing based adaptive penalty method as the two most effective approaches for rate selection. They represent the opposite ends of the tradeoff between control cost and execution time, while the interior method is likely the most effective approach in practice due to its run time efficiency. This work shows the promise of cyber-physical co-design, where tailoring real-time scheduling analysis allows for an elegant and efficient optimization approach for wireless control systems. This work was nominated for Best Paper Award at [RTAS'12].

CapNet: Real-time Wireless Management Network for Data Center Power Capping

Data center management (DCM) is increasingly a significant challenge for enterprises hosting large scale online and cloud services. Machines need to be monitored, and the scale of operations mandates an automated management with high reliability and real-time performance. Existing wired networking solutions for DCM come with high cost. Wireless sensor networks provide a cost-effective networking solution for DCM while satisfying the reliability and latency performance requirements of DCM. We have developed CapNet, a real-time wireless sensor network for power capping, a time-critical DCM function for power management in a cluster of servers. CapNet employs an efficient event-driven protocol that triggers data collection only upon the detection of a potential power capping event. We deploy and evaluate CapNet in a data center. Using server power traces, our experimental results on a cluster of 480 servers inside the data center show that CapNet can meet the real-time requirements of power capping. CapNet demonstrates the feasibility and efficacy of wireless sensor networks for time-critical DCM applications. This work was reported at RTSS'14.

Publications

  • M. Sha, D. Gunatilaka, C. Wu and C. Lu, Implementation and Experimentation of Industrial Wireless Sensor-Actuator Network Protocols, European Conference on Wireless Sensor Networks (EWSN'15), February 2015. PDF
  • A. Saifullah, Y. Xu, C. Lu and Y. Chen, End-to-End Communication Delay Analysis in Industrial Wireless Networks, IEEE Transactions on Computers, accepted. PDF
  • A. Saifullah, S. Sankar, J. Liu, C. Lu, R. Chandra and B. Priyantha, CapNet: a Real-Time Wireless Management Network for Data Center Power Capping, IEEE Real-Time Systems Symposium (RTSS'14), December 2014. PDF
  • C. Wu, M. Sha, D. Gunatilaka, A. Saifullah, C. Lu and Y. Chen; Analysis of EDF Scheduling for Wireless Sensor-Actuator Networks, ACM/IEEE International Symposium on Quality of Service (IWQoS'14), May 2014. PDF
  • A. Saifullah, C. Wu, P. Tiwari, Y. Xu, Y. Fu, C. Lu and Y. Chen, Near Optimal Rate Selection for Wireless Control Systems, ACM Transactions on Embedded Computing Systems, Special Issue on Real-Time and Embedded Technology and Applications, 13(4s), Article 128, April 2014. PDF
  • O. Chipara, C. Lu and G.-C. Roman, Real-time Query Scheduling for Wireless Sensor Networks, IEEE Transactions on Computers, 62(9): 1850-1865, September 2013. PDF
  • A. Saifullah, C. Wu, P. Tiwari, Y. Xu, Y. Fu, C. Lu and Y. Chen; Near Optimal Rate Selection for Wireless Control Systems, IEEE Real-Time and Embedded Technology and Applications Symposium (RTAS '12), April 2012. (Best Paper Nominee) PDF
  • A. Saifullah, Y. Xu, C. Lu and Y. Chen; Priority Assignment for Real-time Flows in WirelessHART Networks, Euromicro Conference on Real-Time Systems (ECRTS '11), July 2011. PDF
  • O. Chipara, C. Wu, C. Lu and W.G. Griswold, Interference-Aware Real-Time Flow Scheduling for Wireless Sensor Networks, Euromicro Conference on Real-Time Systems (ECRTS '11), July 2011. PDF
  • A. Saifullah, Y. Xu, C. Lu and Y. Chen; End-to-End Delay Analysis for Fixed Priority Scheduling in WirelessHART Networks, IEEE Real-Time and Embedded Technology and Applications Symposium (RTAS '11), April 2011. PDF
  • A. Saifullah, Y. Xu, C. Lu, and Y. Chen; Real-time Scheduling for WirelessHART Networks; IEEE Real-Time Systems Symposium (RTSS '10), December 2010. PDF

Selected Talks

  • Real-Time Wireless Control Networks for Cyber-Physical Systems, University College Cork, Ireland, July 2014. PDF
  • Challenges in Wireless Control Networks for Cyber-Physical Systems, Panel on Networking Challenges for Cyber-Physical Systems, INFOCOM, May 2014. PDF
  • Real-Time Wireless Control Networks for Cyber-Physical Systems, Keynote, International Conference on Pervasive and Embedded Computing and Communication Systems (PECCS'14), January 2014. PDF
  • Real-Time Wireless Control Networks for Cyber-Physical Systems, Keynote, IEEE International Symposium on Industrial Embedded Systems (SIES'13), June 2013. PDF
  • Real-Time Wireless Control Networks: Challenges and Directions, NITRD National Workshop on the New Clockwork for Time-Critical Systems, October 2012. PDF
  • Real-Time Wireless Sensor Networks, Royal Institute of Technology (KTH), Stockholm, Sweden, April 2010. PDF