Researchers have used smartphones, laptops, wireless routers and wired networks to track building occupancy and manage lighting, environmental controls and other services.
It's Monday morning. You've just parked your car and you're approaching the entrance to your office building. The Wi-Fi router in the lobby detects your smartphone as you come within range, and it sends commands that turn on the computer in your office as well as the lights and heat. By the time you reach your desk, everything is cozy and your computer is ready for work. Just as important, your company is saving big bucks on its utility bills.
No, this capability isn't quite available, except perhaps in Bill Gates' house. But we can expect it soon. Researchers have tested using existing IT infrastructure — including smartphones, laptops, wireless routers and wired networks — to track building occupancy in real time and use that information to manage lighting, environmental controls and other services in buildings.
Originally proposed in 2008 by Bruce Nordman and Alan Meier, researchers at the Lawrence Berkeley National Laboratory, the idea is called "implicit occupancy sensing," and according to Nordman, it promises huge savings.
In some offices with occupancy sensors over every cubicle, they found that “many people are not in their office 50 percent of the time," Nordman said. "They may be on vacation, sick, traveling, in a meeting, lunch, who knows? But if you could have a light on only when someone is sitting there, than you would save 50 percent. And when you save on lighting you also reduce your air conditioning load."
Nordman, Ken Christensen of the Department of Computer Science and Engineering at the University of South Florida, and other colleagues from those institutions and the University of Puerto Rico at Arecibo, recently published their findings from testing parts of such a system in a building at the Lawrence Berkeley National Laboratory complex.
The infrastructure in their test included smartphones, networked computers, routers and other devices. By monitoring network addresses of devices and requests as well as automatic polling sent across the network, the software developed by the team was able to determine in real time the occupancy of any location in the building.
The data, Nordman said, showed that the number of spikes in the network peaked around noon. Activity rose in the morning and fell in the afternoon, revealing patterns of people coming to work, powering up their computers, using them, then powering off. “Not only do we know the number, but we knew exactly which computers were on because we also had their IP and MAC addresses," Nordman added.
The system was also able to triangulate the location of specific cell phones by detecting the wireless access points reported as available to the device in addition to its actual connections.
One of the major benefits of implicit occupancy sensing is that, unlike dedicated occupancy sensors, it runs on infrastructure that is already in place for other purposes, Nordman said. "You get the data essentially for free, he said. "There is no cost to install or maintain this network, and you can get highly granular data."
Compared to buying and installing sensors that typically connect to only a single device — such as a lighting control — and that are not connected to the network, Nordman said, "this is a much more powerful and less expensive way to do things."
What's more, the system could be extended to include data from any peripherals connected to networked devices. You might, Nordman suggested, schedule computer cameras to take a photograph of all the lights in the ceiling once a week and automatically analyze the images to see if any lights are out and need to replaced.
Nordman acknowledged that the system isn't yet ready for deployment. "I've had no research funding for this topic yet," he said. "But eventually this will be done just because it makes sense."
The major challenge in developing more robust systems, Nordman said, is developing standard communications protocols "so that the devices that are producing information can send it to the occupancy engine in the building. Then that engine can distribute the information back to devices, which then can utilize the information to change operations."