We propose a novel energy based framework for the validation of smart-spaces. The framework includes, in addition to the given smart-space SS, an exterior environment (Env) that mimics a typical real environment where the SS can be deployed, and a Tester for determining which pre-post condition requirements are satisfied by the SS in the considered Env contexts. The Tester appropriately uses an energy aware simulator to derive device operation sequences (or tests), with minimal power consumptions cost, that can be used to move Env to the intended exterior context and the SS to a context satisfying the given pre-condition. In addition, the Tester monitors the relevant SS context attributes to release verdicts about which pre-post conditions are met under each considered Env context. The framework is deployed in a real SS environment to assess the actual energy consumption of derived tests in practice. Experiments show that the actual power consumption of the derived tests is close to the estimated values. Furthermore, a detailed case study is provided to assess the gains in using energy aware tests in comparison to tests derived using non-energy aware alternatives.
CITATION STYLE
Mizumoto, T., El-Fakih, K., Yasumoto, K., & Higashino, T. (2018). An energy aware testing framework for smart-spaces. In Lecture Notes in Computer Science (including subseries Lecture Notes in Artificial Intelligence and Lecture Notes in Bioinformatics) (Vol. 11146 LNCS, pp. 85–101). Springer Verlag. https://doi.org/10.1007/978-3-319-99927-2_8
Mendeley helps you to discover research relevant for your work.