Mobile pollution sensors to be developed at Vanderbilt using Microsoft grant

Vanderbilt engineers have won an award from Microsoft Corp. to develop a real-time, online, detailed and accurate picture of air quality in large metropolitan areas like Nashville. The mobile air quality monitoring system will make it possible to monitor air quality more accurately than the current system of fixed stations performing low-resolution sampling by including car-mounted sensors that measure, process and report emission levels.

Engineers in the Vanderbilt Institute for Software Integrated Systems will adapt Microsoft SensorMap technology for this purpose. SensorMap is a software platform designed to integrate and publish various types of sensor data in real time on the Internet.

The Vanderbilt team is one of 21 winners chosen from more than 140 university applicants worldwide in response to Microsoft Research’s SensorMap and Virtual Earth(TM) request for proposals.

“The ability to search for and analyze information within the context of location is a field with great potential,” said Sailesh Chutani, director of external research and programs, the arm of Microsoft Research that works closely with academic institutions. “These researchers are using a powerful new approach to solve fundamental problems, and our programs are designed to help them in a number of ways — funding to bring in additional resources, software and data to use in experiments, access to top researchers at Microsoft and collaboration with the broader, global research community.”

According to principal investigator Akos Ledeczi, the system being built with the $70,000 grant will include five prototype sensors that can be mounted on vehicles. “We will develop the sensors as well as the necessary infrastructure to measure the pollutants, gather, process and visualize the data and to deploy the system in the Nashville metropolitan area to provide a continuous live data feed on the SensorMap portal,” he said.

Ledeczi, who is a research assistant professor of electrical engineering at Vanderbilt, explained that when the sensor-carrying car is in motion, the sensor will sample the pollutants every few seconds, noting time and location of each sample. When the car comes into contact with a WiFi hotspot, it will upload the data to the SensorMap portal, where a detailed picture of the air quality in the area will be displayed.

“This will push the limits of the SensorMap project by proposing ways to handle mobile sensors, developing more advanced data aggregation algorithms and new visualization methods,” Ledeczi said.

Assistant Professor of Computer Science Xenofon Koutsoukos and Research Scientist Engineer Peter Volgyesi will serve as co-principal investigators on the project.