Context for system resource management: An application in wireless data management
Master of Science
Context information brings new opportunities for efficient and effective system resource management of mobile devices. In this work, we focus on the use of context information to manage wireless data. The findings from our field-collected data show that the energy cost of network interfaces poses a great challenge to ubiquitous connectivity, despite fairly good network availability. Based on our findings, we propose to leverage the complementary strengths of Wi-Fi and cellular networks by automatically selecting the more energy-efficient wireless interface based on context information. We formulate the selection of wireless interfaces as a statistical decision problem. The key challenge is to accurately estimate Wi-Fi network conditions without powering up its network interface. We explore the use of different context information, including time, history, cellular network conditions, and device motion, and devise algorithms that can effectively learn from context information and estimate the probability distribution of Wi-Fi network conditions. Simulations based on field-collected traces show that our algorithms can improve the average battery lifetime of a commercial mobile phone for a three-channel ECG reporting application by 39%, very close to the determined theoretical upper bound of 42%. Finally, a field validation of our most simple algorithm demonstrates a 35% battery lifetime improvement in normal usage.
Electrical engineering; Computer science; Applied sciences