“Generating software tests to check for flaws and functionalities”
From Navigators
Francisco Araujo (advised by Nuno Ferreira Neves, Ibéria Medeiros)
Master’s thesis, Mestrado em Engenharia Informática, Departamento de Informática, Faculdade de Ciências da Universidade de Lisboa, Sept. 2019
Abstract: Industrial products, like vehicles and trains, integrate embedded systems implementing diverse and complicated functionalities. Such functionalities are programmable by software containing a multitude of parameters necessary for their configuration, which have been increasing due to the market diversification and customer demand. However, the increasing functionality and complexity of such systems make the validation and testing of the software highly complex. The complexity inherent to software nowadays has a direct relationship with the rising number of vulnerabilities found in the software itself due to the increased attack surface. A vulnerability is defined as a weakness in the application that if exploitable can cause serious damages and great financial impact. Products with such variability need to be tested adequately, looking for security flaws to guarantee public safety and quality assurance of the application. While efficient automated testing systems already exist, such as fuzzing, no tool is able to use results of a previous testable programme to more efficiently test the next piece of software that shares certain functionalities. The objective of this dissertation is to implement such a tool that can ignore already covered functionalities that have been seen and tested before in a previously tested program and give more importance to block codes that have yet to been tested, detect security vulnerabilities and to avoid repeating work when it is not necessary, hence increasing the speed and the coverage in the new program. The approach was implemented in a tool based on the American Fuzzy Lop (AFL) fuzzing application and was validated with a set of programs of different versions. The experimental results showed that the tool can perform better than AFL.
Export citation
Project(s): Project:Xivt, Project:SEAL
Research line(s): Fault and Intrusion Tolerance in Open Distributed Systems (FIT)