source: papers/FDL2012/FDL2012.tex @ 89

Last change on this file since 89 was 89, checked in by cecile, 12 years ago

corrections typos and co

File size: 4.2 KB
Line 
1 \documentclass{article}
2
3  \usepackage{spconf}
4\usepackage{cmap}
5\usepackage[utf8]{inputenc}
6\usepackage[T1]{fontenc}
7\usepackage{lmodern}
8\usepackage{textcomp}
9\usepackage{amsmath}
10\usepackage{amssymb}
11\usepackage{amsfonts}
12\usepackage{bbm}
13\usepackage{bbold}
14\usepackage{amsthm}
15\usepackage{array}
16\usepackage{verbatim}
17\usepackage{hyperref}
18\usepackage{booktabs}
19\usepackage{graphicx}
20\usepackage{tikz}
21\usepackage{pgf}
22\usetikzlibrary{arrows,automata}
23\usepackage[francais, american]{babel}
24\usepackage[babel=true,kerning=true]{microtype}
25\usepackage[ruled]{algorithm2e}
26
27
28\graphicspath{{schema/}}
29\newtheorem{definition}{Definition}
30\newtheorem{property}{Property}
31\newcommand{\TODO}[1] {\textcolor{red}{TODO : #1}}
32\newcommand{\remark}[2]{\textcolor{blue}{#1: #2}}
33
34
35 \title{ An efficient refinement strategy exploiting components' properties in a CEGAR process}
36 \name{Syed Hussein S. ALWI, Emmanuelle ENCRENAZ and C\'{e}cile BRAUNSTEIN}
37% \thanks{This work was supported by...}}
38 \address{Universit\'{e} Pierre et Marie Curie Paris 6, \\
39                 LIP6-SOC (CNRS UMR 7606), \\
40                            4, place Jussieu, \\
41                75005 Paris, FRANCE. }
42
43\begin{document}
44 % OPTIONAL -->   \ninept            <-- OPTIONAL, for nine pt only
45
46\maketitle
47
48\begin{abstract}
49Embedded systems are usually composed of several components and in practice, these components generally have been independently verified to ensure that they respect their specifications before being integrated into a larger system. Therefore, we would like to exploit the specification (i.e. verified CTL properties) of the components in the objective of verifying a global property of the system. A complete concrete system may not be directly verifiable due to the state explosion problem, thus abstraction and eventually refinement process are required. In this paper, we propose a technique to select properties in order to generate a good abstraction and reduce refinement iterations. We have conducted several preliminary experimentations which shows that our approach is promising in comparison to other abstraction-refinement techniques implemented in VIS \cite{ucberkeley96vis}.
50\end{abstract}
51
52\begin{keywords}
53Compositional verification, CTL properties, CEGAR, model-checking
54\end{keywords}
55
56
57%\def\abstract{\begin{center}
58%{\bf ABSTRACT\vspace{-.5em}\vspace{0pt}}
59%\end{center}}
60%\def\endabstract{\par}
61
62\section{Introduction}
63
64\input{introduction}
65
66
67\section{Our Framework}
68
69\input{framework}
70\section{Refinement}
71
72\input{abstraction_refinement}
73
74
75
76\section{Experimental results}
77
78 \input{exp_results}
79
80
81
82
83\section{Conclusion and Future Works}
84
85%\section*{Drawbacks}
86
87We have presented a new strategy in the abstraction generation and refinement
88which is well adapted for compositional embedded systems. This verification
89technique is compatible and suits well in the natural development process of
90complex systems. Our preliminary experimental results shows an interesting
91performance in terms of duration of abstraction generation and the number of refinement iteration. Futhermore, this technique enables us to overcome repetitive counterexamples due to the presence of cycles in the system's graph.
92
93Nevertheless, in order to function well, this refinement technique requires a
94complete specification of every components of the concrete model. Futhermore,
95it may be possible that none of the properties available is capable of
96eliminating the counterexample which is probably due to an uncomplete
97specification or a counterexample that should be eliminated by the product of
98local properties. In this case, other refinement techniques such as the
99refinement by eliminating the counterexample only, or the identification of a
100good set of local properties to be integreted simultaneously, should be considered.
101We are currently investigating other complementary techniques to overcome these particular cases.
102A complementary approach consists in oimproving the specification of the
103model~: at the component level, or for groups of components. The work of
104Kroening \cite{pwk2009-date} could help us in this direction.
105
106
107
108%\begin{thebibliography}
109 \ninept
110%  <-- OPTIONAL, for nine pt only
111%\bibliographystyle{plain}
112\bibliographystyle{IEEEbib}
113\bibliography{myBib}
114
115%\end{thebibliography}
116
117
118
119\end{document} 
Note: See TracBrowser for help on using the repository browser.