Abstract:
The rise of handheld computing devices has inspired a great deal of research aimed at addressing the unique problems posed by their mobile, "always-on" nature. In order to help mobile devices navigate a complex world of overlapping, uneven public wireless coverage, one must be mindful of the distinction between nomadic usage and true mobility. Accordingly, systems research must move beyond simply optimizing for a set of local conditions (e.g., finding the best access point for a laptop user in a stationary location) to considering the "derivative of connectivity" when network conditions are constantly in flux.
This dissertation presents a new paradigm for networking support on mobile devices. This project has several complementary aspects. As devices encounter network connectivity our system both evaluates the application-level quality of WiFi access points and updates a device-centric mobility model. Together, this mobility model and AP quality database yield "connectivity forecasts," which let applications optimize not just for current network conditions but for the expected big picture to come. Results of a prototype deployment in several cities shows that considering the application-level quality of APs (rather than just signal strength) significantly boosts the success rate of finding a usable access point. Furthermore, this dissertation shows how connectivity forecasts—even with minimal model training time—allow several applications commonly found on mobile devices to reap significant benefits, such as extended battery life.
Mobile devices are often within range of multiple connectivity options, however, and choosing just one therefore ignores potential connectivity. This dissertation describes a virtual link layer for Linux, called Juggler, that uses one network card to simultaneously associate with many WiFi APs, ad hoc groups or mesh networks. The results show how Juggler can boost effective bandwidth by striping data across multiple APs, enable seamless 802.11 handoff by preemptively associating with the "next" AP before the current one become unusable, and maintain a modest side-channel to the user's personal area network or mesh network without impacting foreground bandwidth to infrastructure.
| Adviser | Brian D. Noble |
| School | UNIVERSITY OF MICHIGAN |
| Source Type | Dissertation |
| Subjects | Computer science |
| Publication Number | 3343171 |
Use the link below to access a full citation record of this graduate work:
If your library subscribes to the ProQuest Dissertations & Theses (PQDT) database, you may be entitled to a free electronic version of this graduate work. If not, you will have the option to purchase one, and access a 24 page preview for free (if available).
About ProQuest Dissertations & Theses
With nearly 4 million records, the ProQuest Dissertations & Theses (PQDT)
Global database is the most comprehensive collection of dissertations and theses in the
world. It is the database of record for graduate research.
PQDT Global combines content from a range of the world's premier universities - from the Ivy League to the Russell Group. Of the nearly 4 million graduate works included in the database, ProQuest offers more than 2.5 million in full text formats. Of those, over 1.7 million are available in PDF format. More than 90,000 dissertations and theses are added to the database each year.
If you have questions, please feel free to visit the ProQuest Web site - http://www.proquest.com - or contact ProQuest Support.