Doctor of Philosophy (Ph.D.)
Pervasive computing systems are comprised of various personal mobile devices connected by the wireless networks. Pervasive computing systems have gained soaring popularity because of the rapid proliferation of the personal mobile devices. The number of personal mobile devices increased steeply over years and will surpass world population by 2016.;However, the fast development of pervasive computing systems is facing two critical issues, energy efficiency and security assurance. Power consumption of personal mobile devices keeps increasing while the battery capacity has been hardly improved over years. at the same time, a lot of private information is stored on and transmitted from personal mobile devices, which are operating in very risky environment. as such, these devices became favorite targets of malicious attacks. Without proper solutions to address these two challenging problems, concerns will keep rising and slow down the advancement of pervasive computing systems.;We select smartphones as the representative devices in our energy study because they are popular in pervasive computing systems and their energy problem concerns users the most in comparison with other devices. We start with the analysis of the power usage pattern of internal system activities, and then identify energy bugs for improving energy efficiency. We also investigate into the external communication methods employed on smartphones, such as cellular networks and wireless LANs, to reduce energy overhead on transmissions.;As to security, we focus on implantable medical devices (IMDs) that are specialized for medical purposes. Malicious attacks on IMDs may lead to serious damages both in the cyber and physical worlds. Unlike smartphones, simply borrowing existing security solutions does not work on IMDs because of their limited resources and high requirement of accessibility. Thus, we introduce an external device to serve as the security proxy for IMDs and ensure that IMDs remain accessible to save patients' lives in certain emergency situations when security credentials are not available.
© The Author
Xu, Fengyuan, "Improving Energy Efficiency and Security for Pervasive Computing Systems" (2013). Dissertations, Theses, and Masters Projects. William & Mary. Paper 1539623629.