Files in this item

FilesDescriptionFormat

application/pdf

application/pdfUILU-ENG-08-2204_DC-234 assembled.pdf (348kB)
(no description provided)PDF

Description

Title:Delayed State Estimation in Discrete Event Systems and Applications to Security Problems
Author(s):Saboori, Anooshiravan; Hadjicostis, Christoforos N.
Subject(s):Discrete event systems
State estimation
Automata
Security
Abstract:Application of discrete event systems in modeling and analyzing security problems has given rise to applications that require keeping track of (part of the) sequence of states that have been visited so far. Specifically, the notion of opacity requires that the truth of a certain predicate on the system state cannot be determined by an outside observer for the duration of a certain time window (or even at all times). Depending on the notion of opacity that is used, this predicate can be defined for states visited in the past (with no bound on how far into the past) or for states which have been visited a fixed number of observations in the past. In this report, motivated by such questions we introduce the problem of delayed estimation in discrete event systems modeled as a finite automaton with a finite number of states, unknown initial state, and partial event observation (but no state observation). Specifically, we consider two estimation problems: (i) initial state estimation which requires the estimate of the initial state following a sequence of observations and, (ii) K- delayed state estimation which requires the estimate of the state the system was in when it generated the Kth to last output (i.e., the state of the system K observations ago). To solve these two problems we construct appropriate state estimators and show that these delay state estimators can be used to verify opacity notions of interest.
Issue Date:2008-03
Publisher:Coordinated Science Laboratory, University of Illinois at Urbana-Champaign
Series/Report:Coordinated Science Laboratory Report no. UILU-ENG-08-2204, DC-234
Genre:Technical Report
Type:Text
Language:English
URI:http://hdl.handle.net/2142/99605
Sponsor:National Science Foundation / NSF ECS 04-26831
Date Available in IDEALS:2018-04-04


This item appears in the following Collection(s)

Item Statistics