Tradeoffs between sensing quality and energy efficiency for context monitoring applications

Mobile devices have gone far beyond their basic usage like calling, texting, etc. New generation mobile devices like smartphones contain a rich set of embedded sensors which enable them to perform sensing operations in different domains. To get user contextual information, both human centric & participatory sensing can be helpful. In order to accurately capture, recognize and classify various user states, sensors need to operate more in active mode. However, continuous sensing results in huge energy consumption, decreasing the battery lifetime. Hence, a tradeoff in between the sensing accuracy and energyefficiency is required. In this paper, we have developed novel strategies to make this tradeoff. User contexts are monitored and multiplicative increase and multiplicative decrease (MIMD) approach is exploited to dynamically adjust the sensing frequency following the sensing quality requirements of the applications. We have developed a real-time android application to evaluate the effectiveness of the proposed method in terms of sensing accuracy and energy-efficiency. The results show a good level of improvement compared to a state-of-the-art work.