Protocol re-testing and diagnostic testing methods.

• Main Author: El-Fakih, Khaled Abdul-Ghani.
• Other Authors: Bochmann, Gregor
• Format: Others
• Published: University of Ottawa (Canada) 2009
• Subjects: Computer Science.
• Online Access: http://hdl.handle.net/10393/6429; http://dx.doi.org/10.20381/ruor-11270

Many test selection methods have been developed for deriving tests when a system specification is represented in the form of a Finite State Machine (FSM). In the first part of this thesis, we present test generation methods that select tests for testing the modified parts of the system specification, in order to check that these modifications were correctly implemented in the system implementation. These methods are based on well-known test derivation methods called the W, Wp, HIS and distinguishing sequence methods. As the purpose of conformance testing is to check whether an implementation conforms to its specification, an interesting complementary problem is to locate the differences between a specification and its implementation when the implementation is found to be nonconforming. In the second part of this thesis, we consider a system consisting of two communicating FSMs, called components. First, we show that it is not always possible to locate a fault within the given system, once a fault has been detected in its implementation (called System Under Test (SUT)). Accordingly, we present two new two-level approaches for fault localization within the given system. The first method assumes that the SUT has a single fault in one of its components. Consequently, at the first diagnostic level the methods decide whether it is possible to identify the faulty component in the given system. If this is possible, the faulty component is identified, and if desired, at the second level, the methods determine whether it is possible to locate the fault within the faulty component. If that is possible, the methods provide additional test cases to locate the fault. The second method considers the case when the SUT may have multiple faults in at most one of its components. At the machine level diagnosis, the method decides whether it is possible to identify the faulty component machine in the given system, once faults have been detected in a system implementation. If this is possible, it provides tests for identifying the faulty component machine, and if desired, the method can be used to determine whether it is possible to locate the faults within the faulty component. If that is possible, he method provides additional test cases to locate the faults. (Abstract shortened by UMI.)