Fault detection over signal-to-noise ratio constrained channels

In this paper, the fault detection problem of feedback control systems with Signal-to-Noise Ratio (SNR) constrained channels is studied. A residual generation based on system model is designed to guarantee the best trade-off between sensitivity to fault and robustness to system disturbance as well as channel noise. A norm-based residual evaluation function is used and an on-line dynamic threshold is provided. Then the false alarm rate of designed threshold can be computed with the assistance of Chebyshev's inequality. Simulation results show the effectiveness of the proposed method.