By Tracy Staedter, Discovery News
Source: http://disc.discovery.com
A network of sensors to be installed on and powered by city streetlights in Cambridge, Mass., could make urban monitoring easier than a doctor's visit.
After all, a city's health, like a person's, can benefit from regular checkups. But monitoring such physical factors as weather and air quality are typically time-consuming and labor-intensive.
The CitySense project, led by researchers from Harvard University and Cambridge-based BBN Technologies, will disperse about 100 wireless devices to collect weather and air quality information about the city, and will also serve as a research platform for other scientists to experiment with sensor networks.
"Our network is a backbone. It's an experimental testbed. Scientists from all over the world will be able to program these nodes," said Matt Welsh, team leader and assistant professor at Harvard.
The network will consist of nodes, each about 6 inches by 6 inches square and 3 inches deep, bearing a processor and a memory board sheathed in a weather-resistant case. They will run on Linux and use a USB memory card for storing data, as opposed to an internal hard drive that could fail in the extreme weather swings of Massachusetts.
Instead of receiving and transmitting data to and from a central hub, each node will communicate with its nearest neighbors to form a mesh network. In these networks, data hops from one node to the next, leapfrogging over those that are busy or damaged.
Several sensors attached to each device will monitor pollution and weather variables such as humidity, barometric pressure, wind speed, wind direction, temperature and rainfall. That's for starters, but there is the potential for expansion.
The idea, said Welsh, is that scientists would be able to log onto an Internet site for the network and request research time. They could upload and test their own software or work with Welsh's team to integrate other sensors into the network.
But before that can happen, the group needs to make the system robust. By the end of this summer, they plan to have about 20 sensors deployed around the Harvard campus as well as on the BBN Technologies building, a couple of miles away. In another three years or so, they plan to have the remaining 80 sensors distributed across Cambridge.
"The challenge is to keep it up and running," said Ramesh Rao, professor of electrical and computer engineering and director of the California Institute for Telecommunications and Information Technology at the University of California, San Diego.
And that may require a lot of handholding and some customer support that currently doesn't exist.
"The difficulty should not be understand estimated," he said.
Public perception could also be an issue. People may not approve of the sensors, even if the devices promise not to infringe upon privacy, said Rao.
"When you say the sensors can do 'x' but not 'y,' do people believe you? That sounds like a 'trust me' statement. How do they know that for sure?" he said.
Initial data gleaned from the CitySense project will be incorporated into one of the first high-resolution studies of urban pollution. Welsh is also working with Tufts University to develop a educational curriculum for K-12 students to better understand sensor networks.
Source: http://disc.discovery.com
A network of sensors to be installed on and powered by city streetlights in Cambridge, Mass., could make urban monitoring easier than a doctor's visit.
After all, a city's health, like a person's, can benefit from regular checkups. But monitoring such physical factors as weather and air quality are typically time-consuming and labor-intensive.
The CitySense project, led by researchers from Harvard University and Cambridge-based BBN Technologies, will disperse about 100 wireless devices to collect weather and air quality information about the city, and will also serve as a research platform for other scientists to experiment with sensor networks.
"Our network is a backbone. It's an experimental testbed. Scientists from all over the world will be able to program these nodes," said Matt Welsh, team leader and assistant professor at Harvard.
The network will consist of nodes, each about 6 inches by 6 inches square and 3 inches deep, bearing a processor and a memory board sheathed in a weather-resistant case. They will run on Linux and use a USB memory card for storing data, as opposed to an internal hard drive that could fail in the extreme weather swings of Massachusetts.
Instead of receiving and transmitting data to and from a central hub, each node will communicate with its nearest neighbors to form a mesh network. In these networks, data hops from one node to the next, leapfrogging over those that are busy or damaged.
Several sensors attached to each device will monitor pollution and weather variables such as humidity, barometric pressure, wind speed, wind direction, temperature and rainfall. That's for starters, but there is the potential for expansion.
The idea, said Welsh, is that scientists would be able to log onto an Internet site for the network and request research time. They could upload and test their own software or work with Welsh's team to integrate other sensors into the network.
But before that can happen, the group needs to make the system robust. By the end of this summer, they plan to have about 20 sensors deployed around the Harvard campus as well as on the BBN Technologies building, a couple of miles away. In another three years or so, they plan to have the remaining 80 sensors distributed across Cambridge.
"The challenge is to keep it up and running," said Ramesh Rao, professor of electrical and computer engineering and director of the California Institute for Telecommunications and Information Technology at the University of California, San Diego.
And that may require a lot of handholding and some customer support that currently doesn't exist.
"The difficulty should not be understand estimated," he said.
Public perception could also be an issue. People may not approve of the sensors, even if the devices promise not to infringe upon privacy, said Rao.
"When you say the sensors can do 'x' but not 'y,' do people believe you? That sounds like a 'trust me' statement. How do they know that for sure?" he said.
Initial data gleaned from the CitySense project will be incorporated into one of the first high-resolution studies of urban pollution. Welsh is also working with Tufts University to develop a educational curriculum for K-12 students to better understand sensor networks.
No comments:
Post a Comment