Robustness Analysis of Embedded Control Systems with Respect to Signal Perturbations: Finding Minimal Counterexamples Using Fault Injection
Issue No.01 - Jan.-Feb. (2014 vol.11)
Satya Gautam Vadlamudi , Dept. of Comput. Sci. & Eng., Indian Inst. of Technol. Kharagpur, Kharagpur, India
Partha Pratim Chakrabarti , Dept. of Comput. Sci. & Eng., Indian Inst. of Technol. Kharagpur, Kharagpur, India
DOI Bookmark: http://doi.ieeecomputersociety.org/10.1109/TDSC.2013.23
Fault-tolerance of embedded control systems is of great importance, given their wide usage in various domains such as aeronautics, automotive, medical, and so on. Signal perturbations such as small amounts of noise, shift, and spikes, can sometimes severely hamper the performance of the system, apart from complete failure of components and links. Finding minimal counterexamples (perturbations on the system leading to violation of fault-tolerance requirements) can be of great assistance to control system designers in understanding and adjusting the fault-tolerance behavior of the system. Fault injection is an effective method for dependability analysis of such systems. In this paper, we introduce the concept of dominating sets of perturbations, and define a minimal set of counterexamples called the basis. We propose effective methods using a simulation-based fault injection technique on Simulink models for finding the basis set at an early stage of design, given the fault specification and fault-tolerance requirements. Experimental results on two different control system examples from the Simulink automotive library demonstrate the efficacy of the proposed framework.
Embedded system, Fault tolerance, Robustness, Control systems, Signal perturbations,monotonicity, Robustness analysis, embedded control systems, signal perturbations, quality-faults, minimal counterexamples, fault injection
Satya Gautam Vadlamudi, Partha Pratim Chakrabarti, "Robustness Analysis of Embedded Control Systems with Respect to Signal Perturbations: Finding Minimal Counterexamples Using Fault Injection", IEEE Transactions on Dependable and Secure Computing, vol.11, no. 1, pp. 45-58, Jan.-Feb. 2014, doi:10.1109/TDSC.2013.23