DESSERT'2018

9th International IEEE Conference
Dependable Systems, Services and Technologies
UKRAINE, KYIV, MAY 24-27, 2018

IEEE
  • Information

  • Deadlines

    (23:59 UTC+0)

    Paper submission: March 4, 2018 March 14, 2018 March 19, 2018 (Extra deadline for update only)
    Notification of paper acceptance: April 25, 2018
    Final manuscript: April 30, 2018
    Registration and payment: May 3, 2018

    Program draft publication: May 7, 2018
    Conference date: May 24-27, 2018

  • Contacts

    Department 503, DESSERT’2018 Organizing Committee,
    National Aerospace University n. a. N. E. Zhukovsky “KhAI”,
    Chkalov str., 17, Kharkiv, 61070, Ukraine
    Phone: +38 (095) 564 76 69
    (contact person – Anastasiia Strielkina)
    e-mail: dessert@csn.khai.edu

Jüri Vain

Markov Model Guided Online Testing of Mission Critical Applications

Abstract:

This talk is about a novel approach to online conformance testing of mission critical applications during their operation. The reliability modes and quality of service of critical systems need to be monitored throughout the mission. This is to pre-empt software runtime errors and hardware faults that can evolve to a fatal mission failure. The difficulty in monitoring lies in identifying the system’s reliability mode online under dynamic time and resource constraints of the mission. When the decision making on reliability modes has severe time constraints the passive monitoring should be switched over to active testing whenever possible. To guide the testing strategy in those cases an integrated model-based approach is proposed. The Multi-Fragment Markov Models (MFMM) are used for specifying the system reliability and quality related behaviour on high level of abstraction. The more concrete state and timing constraints related to MFMM are specified explicitly using Uppaal Probabilistic Timed Automata (UPTA). The test hypothesis are selected based on the Markov model of the system’s reliability modes. It is demonstrated how predefined conformance test cases specified as UPTA models are extracted and their serialization strategy for mode identification generated. The theoretical results are illustrated with a System on Chip application example.

Jüri Vain graduated System Engineering speciality from Tallinn Polytechnic Institute, Estonia in 1979. He received his PhD in computer science from the Institute of Cybernetics at Estonian Academy of Sciences in 1987. Professor Vain is the founder of the Department of Computer Science at Tallinn University of Technology and has been director of the Department over sixteen years. Currently, he is Professor of Computer Science at the Department of Software Science, Tallinn University of Technology. He has been visiting Professor at Abo Akademi University (Finland), visiting researcher at Oslo University (Norway), Danish Technical University, Tampere University of Technology (Finland), and Eindhoven University of Technology (The Netherland). His current research interests include formal methods, model-based testing, cyber physical systems, human-computer interaction, autonomous robotics, and artificial intelligence. He has been leading researcher in several international projects under EU framework programmes and the coordinator of Estonian-Japan medical robotics collaboration programme. He has published over 160 scientific articles including journal papers, book chapters and conference papers. He has been an invited lecturer in several international conferences, summer and winter schools.