Networked Fault Diagnosis & Estimation Using Wireless Sensor Networks

Control systems technology has significantly been influenced by wireless sensors and actuators, such that wireless condition monitoring and control have been  important research topics in control theory. Network capacity, topology  management, security, radio channel variations, packet drop, quantization error, time delay, model uncertainty and non-linearity are of challenging issues in the design and development of networked monitoring and control systems.

By using H-infinity theory and asymptotic mean-square stability criterion, we designed over-network observer-based state estimation and fault diagnosis systems, which are robust to packet drop, quantization error, and model non-linearity. We developed a Matlab-based hardware-in-the-loop test-bed, by using the IEEE 802.15.4 wireless sensors, for the performance evaluation of the proposed techniques under realistic radio channel conditions. 

The following videos demonstrate the results. There are two videos for each experiment, one recording the signals and one recording the sensor movement. Both videos should be played synchronously. 

The results of “Fast State Estimation Subject to Random Data Loss in Discrete-Time Nonlinear Stochastic Systems”, Vol. 86, Issue 12, pp. 2302-2314, 2013.

The results of “Fault Detection in Nonlinear Stable Systems Over Lossy Networks”, IEEE Transactions on Control Systems Technology, Vol. 21, Issue 6, pp. 2129-2142, 2013.

Power Control in Wireless Sensor Networks

Power control is a very important problem in battery-powered wireless sensor networks. It aims to increase the battery life, while the performance of the data communications is preserved. By using H-infinity and quantitative feedback theories and linear matrix inequality, we developed practically implementable robust distributed active power control techniques for IEEE 802.15.4 wireless sensor networks addressing the radio channel uncertainties and time delays, as well as the wind-up issue. 

The effectiveness of the proposed techniques has been verified practically by using the IEEE 802.15.4 wireless sensors. The following videos have been recorded from one of the experiments verifying the performance of anti-windup control. There are two videos for the experiment, one recording the signals and one recording the sensors movement, which should be played synchronously. Thanks to Dr. Michael Walsh for recording the videos.

Selected Publications

  • S.M.M. Alavi, M. Saif, Fault Detection in Nonlinear Stable Systems Over Lossy Networks, IEEE Transactions on Control Systems Technology, Vol. 21, Issue 6, pp. 2129-2142, 2013. 
  • S.M.M. Alavi, M. Saif, Fast State Estimation Subject to Random Data Loss in Discrete-Time Nonlinear Stochastic Systems, International Journal of Control, Vol. 86, Issue 12, pp. 2302-2314, 2013.
  • M.J. Walsh, S.M.M.Alavi, M.J. Hayes, Practical assessment of hardware limitations on power aware wireless sensor networks- an anti-wind up approach, International Journal of Robust and Nonlinear Control, Vol. 20, No. 2, pp. 194-208, 2010. 
  • S.M.M. Alavi, M.J. Walsh, M.J. Hayes, Robust power control for IEEE 802.15.4 wireless sensor networks with round-trip time-delay uncertainty, Wireless Communications and Mobile Computing, Vol. 10, No. 6, pp. 811-825, 2010. 
  • S.M.M. Alavi, M.J. Walsh, M.J. Hayes, Robust distributed active power control technique for IEEE 802.15.4 wireless sensor networks – A quantitative feedback theory approach, Control Engineering Practice, Vol. 17, No. 7, pp. 805-814, 2009. [Download