Section: New Results
Diagnosis and application to Security
Predictability of Sequence Patterns in Discrete Event Systems
Following our preliminary results on diagnosis of discrete event systems, we studied in [Oops!] the problem of predicting the occurrences of a pattern in a partially-observed discrete-event system. The system is modeled by a labeled transition system. The pattern is a set of event sequences modeled by a finite-state automaton. The occurrences of the pattern are predictable if it is possible to infer about any occurrence of the pattern before the pattern is completely executed by the system. An off-line algorithm to verify the property of predictability is presented. The verification is polynomial in the number of states of the system. An on-line algorithm to track the execution of the pattern during the operation of the system is also presented. This algorithm is based on the use of a diagnoser automaton. The results are illustrated using an example from computer systems. This work has been done in cooperation with S. Lafortune and S. Genc (University of Michigan, USA).
Construction of monitor for the supervision of security properties
Regarding security, besides our work on test generation for security properties, we have been interested in constructing monitors for the detection of confidential information flow in the context of partially observed discrete event systems modelled by finite labelled transitions systems. We focused on the case where the secret information is given as regular languages. First, we characterised the set of observations allowing an attacker to infer secret information. Further, based on the diagnosis of discrete event systems theory, we provided necessary and sufficient conditions under which detection and prediction of secret information flow can be ensured and construct a monitor allowing an administrator to detect it. We consider the general case where the attacker and the administrator have different partial views of the system [Oops!] .