Co-sponsored by: UMD ECE Dept.

This is sponsored by the University of Maryland Booz Allen Hamilton Distinguished Colloquium Series of the Electrical and Computer Engineering at University of Maryland, College Park, and co-sponsored by the IEEE Signal Processing Washington Chapter.

“Low-Energy and Flexible GPS Sensing Through Cloud Offloading”

Speaker: Dr. Jie Liu (Microsoft Research — Redmond, WA)

Speaker(s): Dr. Jie Liu, , Dr. Jie Liu,

Agenda:

3:15 – 3:30pm:  Checking in

3:30 – 4:30pm:  Lecture by Dr. Jie Liu (Microsoft Research — Redmond, WA)

4:30 – 5:00pm:  Reception (with light refreshment)

“Low-Energy and Flexible GPS Sensing Through Cloud Offloading”

Location-based services have become ubiquitous, thanks to sensors like GPS and RF receivers in our smart phones. However, the energy consumption of GPS receiving is a major bottleneck for long-term and frequent location sensing. In this talk, I will take a deep look into the GPS receiving pipeline and present our efforts on a cloud-offloading approach to GPS-based location sensing. The partitioning between device and cloud brings three kinds of benefits. First, we can reduce the device side energy consumption by orders of magnitude. Second, with the computation power in the cloud, we can achieve high sensitivity and make GPS receiver work in many indoor places. Finally, we can leverage opportunistic energy sources on the device side to reduce the amount of data to be transferred between device and cloud. The cloud-offloading architecture help push location sensing and tracking to the next tier of IoT devices.

Location:
Room: 1110
Bldg: Kim Engineering Building
University of Maryland – College Park
8228 Paint Branch Drive
College Park, Maryland
20742