Judges’ Queries and Presenter’s Replies

  • Members may log in to read judges’ queries and presenters’ replies.

Presentation Discussion

  • Icon for: Emily Mugler

    Emily Mugler

    Trainee
    May 22, 2012 | 04:23 p.m.

    Great job, Troy!

  • Icon for: Troy Hernandez

    Troy Hernandez

    Presenter
    May 24, 2012 | 12:48 a.m.

    Thanks Emily :)

  • Icon for: Lee McDavid

    Lee McDavid

    Coordinator
    May 23, 2012 | 05:00 p.m.

    Do you have any advice for improving GPS devices as they current operate?

  • Icon for: Troy Hernandez

    Troy Hernandez

    Presenter
    May 24, 2012 | 12:58 a.m.

    The GPS device operates independently of our system. There is a deep literature on improving the energy consumption of devices using GPS that appears to have been adopted, at least on my phone.
    Our advice would for improving the usage of data bandwidth for cellular devices rather than improving the actual GPS device. Using a system like ours, a location dependent app could use significantly less bandwidth improving a cellular company’s service and saving a consumer money.

  • Icon for: Bo Xu

    Bo Xu

    Trainee
    May 24, 2012 | 09:38 a.m.

    Great job, Troy! What tool did you use to make the video?

  • Further posting is closed as the competition has ended.

Icon for: Troy Hernandez

TROY HERNANDEZ

University of Illinois at Chicago
Years in Grad School: 6

Thrifty Tracking: Smooth and Accurate Online GPS Tracking with Low Data Uplink Utilization

An online GPS tracking system typically consists of a GPS-equipped mobile device, a wireless uplink, a receiving server and a final trace consumer. A trade-off exists between the accuracy of the location perceived by the final consumer, and the cost incurred at the uplink, or “data usage” in current industry terms. Intuitively, we may reduce data usage by transmitting fewer reports over the uplink, at the cost of reducing accuracy at the consumer.


We present an end-to-end framework for improving the cost efficiency of online GPS tracking systems. Here, we consider one system more efficient than another if it offers improved accuracy at the same cost, or lower cost at the same accuracy. We implement 46 potential thrifty tracking methods, and evaluate these on a collection of GPS traces comprising 1,200 hours of data from 1,000 individuals, for a total of 3.6 million GPS points.
We also perform experiments to characterize the data usage billing policy of a major wireless provider. We find that depending on the the protocol used, bytes of data usage billed per transmission can be up to 1200 the actual bytes of data sent, underscoring the importance of data efficiency.