Nowadays, mobile devices incorporate many sensors to monitor operational parameters, so that possible failures in the systems can be detected and prevented. Therefore, failure detection has become crucial to ensure the automation of certain applications, such as, health monitoring or unmanned aerial vehicles. On the other side, fuzzy models perfectly fit when the input-output relationships use categorical values and they are not deterministic. However, find a feasible model is not a trivial task due to the interaction of many variables at time. In this work, we propose a fault detection model based on fuzzy logic to early detect potential fault in mobile devices. Our approach considers the interaction of four variables, all of them could be measured from sensors of the device. As a proof of concept, we have tested our model in a simulated scenario with random values taken from all the possible combinations of the input fuzzy sets. The mapping into the fuzzy output called, risk of fault, shows accordance with the expected values in literature. Finally, results show that our model can distinguish four levels of failure risk and it is able to be implemented in a production environment.
|Title of host publication||Communications in Computer and Information Science|
|Editors||Miguel Botto-Tobar, Mayra D’Armas, Miguel Zúñiga Sánchez, Miguel Zúñiga-Prieto, Guillermo Pizarro|
|Number of pages||12|
|State||Published - 1 Jan 2019|
|Event||4th International Conference on Technology Trends, CITT 2018 - EC, Babahoyo, Ecuador|
Duration: 29 Aug 2018 → 31 Aug 2018
|Name||Communications in Computer and Information Science|
|Conference||4th International Conference on Technology Trends, CITT 2018|
|Period||29/08/18 → 31/08/18|
Bibliographical noteFunding Information:
Acknowledgments. This work was supported by IDEIAGEOCA Research Group of Universidad Politécnica Salesiana in Quito, Ecuador.
© Springer Nature Switzerland AG 2019.
- Fault tolerance
- Fuzzy logic
- Mobile devices