Supporting Energy-Efficient Uploading Strategies for Continuous Sensing Applications on Mobile Phones

Abstract

Abstract. Continuous sensing applications (e.g., mobile social networking applications) are appearing on new sensor-enabled mobile phones such as the Apple iPhone, Nokia and Android phones. These applications present significant challenges to the phone's operations given the phone's limited computational and energy resources and the need for applications to share real-time continuous sensed data with back-end servers. System designers have to deal with a trade-off between data accuracy (i.e., application fidelity) and energy constraints in the design of uploading strategies between phones and back-end servers. In this paper, we present the design, implementation and evaluation of several techniques to optimize the information uploading process for continuous sensing on mobile phones. We analyze the cases of continuous and intermittent connectivity imposed by low-duty cycle design considerations or poor wireless network coverage in order to drive down energy consumption and extend the lifetime of the phone. We also show how location prediction can be integrated into this forecasting framework. We present the implementation and the experimental evaluation of these uploading techniques based on measurements from the deployment of a continuous sensing application on 20 Nokia N95 phones used by 20 people for a period of 2 weeks. Our results show that we can make significant energy savings while limiting the impact on the application fidelity, making continuous sensing a viable application for mobile phones. For example, we show that it is possible to achieve an accuracy of 80% with respect to ground-truth data while saving 60% of the traffic sent over-the-air.

PhoneJoule: An Energy Management System for Android-Based Smartphones

“…Dong et al [10] propose Sesame, which leverages the possibility of self-power measurement through the smart battery interface and employs a suite of methods to achieve accuracy and rate much higher than that of the smart battery interface. Musolesi et al [11] also present the design, implementation and evaluation of several techniques to optimize the information uploading process for continuous sensing on mobile phones. These systems, however, might have their own energy overheads.…”

Where and what: Using smartphones to predict next locations and applications in daily life

“…In addition, many modern mobile sensing systems have integrated inference modules for reasoning about human behavior and context to achieve energy-efficient sensing and high-level representation of context [21,22]. Interestingly, not much work has been done to extend predictions to other facets of users' most common activities such as phone usage or physical interaction with other people.…”

A Picture of Present Ubicomp Research Exploring Publications from Important Events in the Field

“…Musolesi et al (2010) presented the design, implementation and evaluation of several techniques to optimize the information uploading process for continuous sensing on mobile phones. Pyles et al (2011) exploited, during runtime, the silence periods of a VoIP call to save smart phone Wi-Fi energy.…”