Design and Implementation of Combinatorial Testing based Test suites for Operating Systems used for Internet of Things

Regression test selection, augmentation, prioritizing and pruning are the areas in which significant amount of research work has already been done. In spite of this, in few cases, testing is done in non-methodical manner by selecting a few important parameters and their combinations. It is possible to take a more rigorous approach to this problem using the combinatorial testing approach. Despite the advantages of such an approach, this approach has not been used in many cases where it is needed. This Thesis presents a practical approach to applying combinatorial testing and augment the base regression test suite of Contiki and Cooja with National Institute of Standards Technology (NIST) tools. We share the results of use of combinatorial testing for Contiki and Cooja regression test suite. Our work looks at the existing regression test suite and studies its inadequacies. We then incorporate additional tests as part of re-engineering of the existing test suite using Automated Combinatorial Testing for Software (ACTS) tool. Test suite was written from scratch for the Cooja simulator of Contiki Operating system. In each case we measured the coverage in Contiki simulator. We report on the automation of test case generation for Contiki and Cooja as well.This thesis studies the following research problems:• Study of generic regression test suites with respect to execution time• Design and implementation of combinatorial testing based test suites for Contiki operating system and its simulator Cooja• Measuring the effectiveness of designed test suites using the traditional coverage techniques like code coverage• Automation of test scripts generation from combinatorial testing design model and analyzing coverage. Further refinement of the combinatorial testing design model for Contiki and Cooja till the desired results are achieved.• Propose an integrated test environment for combinatorial testing• Enhancement of test prioritization algorithm viz. Residual test coverage algorithm taking execution time and code coverage as parameters.