Probabilistic Approaches to Fault Detection in Networked Discrete Event Systems

Abstract

In this paper, we consider distributed systems that can be modeled as finite state machines with known behavior under fault-free conditions, and we study the detection of a general class of faults that manifest themselves as permanent changes in the next-state transition functionality of the system. This scenario could arise in a variety of situations encountered in communication networks, including faults occurred due to design or implementation errors during the execution of communication protocols. In our approach, fault diagnosis is performed by an external observer/diagnoser that functions as a finite state machine and which has access to the input sequence applied to the system but has only limited access to the system state or output. In particular, we assume that the observer/diagnoser is only able to obtain partial information regarding the state of the given system at intermittent time intervals that are determined by certain synchronizing conditions between the system and the observer/diagnoser. By adopting a probabilistic framework, we analyze ways to optimally choose these synchronizing conditions and develop adaptive strategies that achieve a low probability of aliasing, i.e., a low probability that the external observer/diagnoser incorrectly declares the system as fault-free. An application of these ideas in the context of protocol testing/classification is provided as an example.

Distributed Fault Detection and Isolation of Continuous-Time Non-Linear Systems

“…Fault diagnosis is a crucial problem in the safe operation of distributed systems, and in making possible the existence and proliferation of "societies" made of networks of autonomous software agents, mobile sensors and robots [10], but the number of works addressing it is still quite limited. It is true that a considerable effort was aimed at developing distributed fault diagnosis algorithms suited to discrete event systems (see, for instance, [8,50,118,91,3,54,115]), especially in the Computer Science literature where the problem of fault diagnosis for multi-processor systems is of great importance [85,51,48,16,81,91,121]. A notable contribution in the field of decentralized hybrid systems fault diagnosis is [27], although the fault detection scheme implemented does take into account only the jump part of the model, and not the continuous flow.…”

“…The structural graph and its decomposition are detailed in the Appendix 3. In the paper, a capital-case index denotes a specific sub-system.…”

Detecting and Reacting to Changes in Sensing Units: The Active Classifier Case

“…Clearly, by considering a CDT-based solution we construct a general approach whose validity is well behind that of a specific application, chosen in the sequel of classification type for its relevance in industrial applications. Assessing the quality T of the outcome product [3], validating a sensor [4], and designing a fault diagnosis system with the fault identification ability [5] are some immediate examples of industrial applications requiring classification systems. The focus is then more on applications requiring the inspection of changes affecting the sensing datastream and its impact on the classifier output than those affecting the relationship between sensing data and the output of the classifier only.…”

Computation of Diagnosable Fault-Occurrence Indices for Systems With Repeatable Faults

“…The approach has been extended to timed setting [5], [10], [17], [25] and probabilistic setting [1], [22]. To facilitate generalization of fault specifications, linear-time temporal logic (LTL) based specification and diagnosis of its fault was proposed in [9].…”