Date Awarded


Document Type


Degree Name

Doctor of Philosophy (Ph.D.)


Computer Science


Qun Li


In the post-PC era, mobile devices will replace desktops and become the main personal computer for many people. People rely on mobile devices such as smartphones and tablets for everything in their daily lives. A common requirement for mobile computing is wireless communication. It allows mobile devices to fetch remote resources easily. Unfortunately, the increasing demand of the mobility brings many new wireless management challenges such as security, energy-saving and location-awareness. These challenges have already impeded the advancement of mobile systems. In this dissertation we attempt to discover the guidelines of how to mitigate these problems through three general communication patterns in 802.11 wireless networks. We propose a cross-section of a few interesting and important enhancements to manage wireless connectivity. These enhancements provide useful primitives for the design of next-generation mobile systems in the future.;Specifically, we improve the association mechanism for wireless clients to defend against rogue wireless Access Points (APs) in Wireless LANs (WLANs) and vehicular networks. Real-world prototype systems confirm that our scheme can achieve high accuracy to detect even sophisticated rogue APs under various network conditions. We also develop a power-efficient system to reduce the energy consumption for mobile devices working as software-defined APs. Experimental results show that our system allows the Wi-Fi interface to sleep for up to 88% of the total time in several different applications and reduce the system energy by up to 33%. We achieve this while retaining comparable user experiences. Finally, we design a fine-grained scalable group localization algorithm to enable location-aware wireless communication. Our prototype implemented on commercial smartphones proves that our algorithm can quickly locate a group of mobile devices with centimeter-level accuracy.



© The Author