Automatic Analysis of a NonRepudiation Protocol.pdfVIP

QAPL 2004 Preliminary Version Automatic Analysis of a Non-Repudiation Protocol Ruggero Lanotte 1 Dipartimento di Scienze della Cultura, Politiche e dell’Informazione, Universit`a dell’Insubria, Via Valleggio 11, 22100 Como, Italy Andrea Maggiolo-Schettini 2 Angelo Troina 3 Dipartimento di Informatica, Universit`a di Pisa, Via F. Buonarroti 2, 56127 Pisa, Italy Abstract We define a probabilistic model for the analysis of a Non-Repudiation protocol that guarantees fairness, without resorting to a trusted third party, by means of a probabilistic algorithm. By using the PRISM model checker, we estimate the probability for a malicious user to break the non-repudiation property, depending on various parameters of the protocol. 1 Introduction Repudiation is defined as denial by one of the entities involved in a communi- cation of having participated in all or part of the communication. One speaks of repudiation of the origin if the originator of a message denies having sent the message, and of repudiation of receipt if the recipient of the message denies having received the message. Protocols have been defined which ensure non-repudiation by making use of a trusted third party in the communication. Protocols involving no third party have been proposed in fault-less scenarios. Markowitch and Roggeman [13] give a probabilistic protocol which achieves fair non-repudiation services with- out the need of a third party and without further assumptions. In particular, the probabilistic protocol is fair up to a given tolerance ε. This tolerance depends on the values chosen for various parameters of the protocol. 1 Email: ruggero.lanotte@uninsubria.it 2 Email: maggiolo@di.unipi.it 3 Email: tro