Margaret Martonosi: Collaborative information sharing

Thursday, Nov 29, 2012

As smartphones grow in popularity, Princeton computer science professor Margaret Martonosi and her collaborators are exploring novel ways to apply the devices’ information gathering and sharing abilities to real-world problems. 

 One such project is SignalGuru, an application that shares information gathered from windshield-mounted phones to calculate the optimal speed drivers should maintain to avoid stop-and-go traffic, which would increase fuel efficiency.  Martonosi, the Hugh Trumbull Adams ’35 Professor of Computer Science, worked with Princeton electrical engineering doctoral student Emmanouil Koukoumidis *11 and Li-Shiuan Peh. Peh was an associate professor of electrical engineering at Princeton and is now at the Massachusetts Institute of Technology. 

 In a broader context, SignalGuru is one example of how Martonosi seeks to alleviate customer and industry concerns about privacy, energy use and data-plan limitations. SignalGuru performs traffic light recognition on the phone itself, which not only demonstrates the computational ability of smartphones, but also preserves battery life, reduces data usage and forgoes granting an external computer access to a personal phone.

 “SignalGuru shows that these issues can be addressed, but it also raises questions,” Matonosi said.  “We’re just starting to explore what you can do with mobile devices and the information they can gather.”

 Going forward, Martonosi seeks to determine and reconfigure which smartphone functions can be performed locally or externally in order to better understand — and take advantage of — the computational capability of different mobile devices.