Building energy consumption models based on smartphone user’s usage patterns

The increasing usage of smartphones in everyday tasks has been motivated many studies on energy consumption characterization aiming to improve smartphone devices’ effectiveness and increase user time. In this scenario, it is essential study mechanisms capable characterizing patterns, so smartphones’ components can be adapted promote the best experience with lower consumption. goal build an model based patterns provide accurate used by application developers automated optimization. To develop models, we established a method identify most influence smartphone’s states each influential device. Besides that, prove robustness models constructed using inaccurate hardware strategy assess accuracy built. After training testing user’s perform Nemenyi test, demonstrated that possible get Mean Absolute Error 158.57 mW when average power 1970.1 mW. Some show leading workload user. Based fact, developed automatic building methodology analyzing data smart estimate pattern. With methodology, adopt strategies minimize drain battery.