Trending: Warren denounces Amazon for ‘trying to tilt the Seattle City Council elections’ with $1.45M
University of Washington electrical engineering graduate student Edward Wang wears a MagnifiSense sensor. The sensor can be used to measure someone's carbon footprint. Photo: Lisa Stiffler.
University of Washington electrical engineering graduate student Edward Wang wears a MagnifiSense sensor. The device can be used to measure someone’s carbon footprint. Photo: Lisa Stiffler.

The coolest thing about a new electromagnetic-radiation sensing device from the University of Washington might be the elegant simplicity of its design.

Or maybe it’s the fact that the wearable tech gadget could be put to important uses, such as measuring someone’s carbon footprint, helping prevent injuries for older people with dementia, and blocking inappropriate content from kids when they turn on a computer or TV.

Or perhaps the best thing about the MagnifiSense sensor, which was built from off-the-shelf materials bought at RadioShack, is that its potential uses are still being discovered.

“That’s kind of the whole point about our work,” said Edward Wang, an electrical engineering doctoral student at the UW’s ubiquitous computing (UbiComp) research lab who helped create the device. “We provide this insight on some new sensing technology and some other people who are a little better at coming [up with] more useful applications say, ‘It would be really cool if you could do this.’”

UbiCompLab logoLast week Wang presented the lab’s research on the MagnifiSense system at the 2015 ACM International Joint Conference on Pervasive and Ubiquitous Computing (UbiComp 2015) held in Osaka, Japan.

The MagnifiSense sensor is built from three wire coils that each surround a magnet, all tucked into a plastic box about half the size of an Altoids’ mint tin. These sensor coils can detect changes in the nearby magnetic fields, which fluctuates when someone turns on a microwave, blow dries their hair, flips a light switch or starts a car.

That information is recorded by an audio card and then software on a computer or smartphone can read the fluctuations and identify what kind of appliance or vehicle, including transit, is being used and who it’s being used by.

“It’s another way to log what you’re interacting with so at the end of the day or month you can see how much energy you used,” said Shwetak Patel, UW professor of Computer Science & Engineering and Electrical Engineering and director of the UbiComp Lab, in a press release.

“Right now, we can know that lights are 20 percent of your energy use. With this, we divvy it up and say who consumed that energy,” said Patel.

“A wearable carbon counter would make visible to people their global-warming footprint,” said Alan Durning, director of the Seattle-based think tank Sightline Institute, which helped develop the Walk Score tool. “It could empower citizens to live more sustainably.”

In the application for older people with memory problems, the MagnifiSense would detect when someone turned on a stove or oven, and then trigger an alarm or warn a caregiver if the appliance was left on for an extended period or if the person left the room for a long time.

Similarly, the sensor could be personalized to figure out who was turning on a TV or other digital device and display the shows or games that they like, and prevent children from accessing violent or sexual content.

Radiation patterns from common electronic devices. Figure: Edward Wang, University of Washington.
Radiation patterns from common electronic devices. Figure: Edward Wang, University of Washington.

The MagnifiSense project began as something seemingly less ambitious than a tool for a greener-living or making grandma and the kids safer.

“Our original problem we were trying to solve was figuring out what side of the car you were sitting on,” Wang said.

The idea was that if your smart phone, based on the sound coming from car speakers, could figure out if you were the driver or a passenger, it could change how it displays an app. If the app was being used by the driver, it could launch verbal directions or information, or it could use visual displays if a passenger was using it.

So the researchers built their sensor and recorded all the signals it received while sitting in the car as it was running.

Back in lab, they realized “there was something interesting not in the sound, but in the magnetic field,” Wang said. “It gave us this idea that there is this signal that I don’t think anyone is using.”

The team then began recording the signals from a variety of common plug-in and battery-operated devices, including light bulbs, refrigerators, hairdryers, ovens and ranges, TV remote controls, computers and a variety of gas, hybrid and electric cars. They linked the unique signature from each recording with the type of item or vehicle.

To test the technology, they used the MagnifiSense system in 16 homes in four cities, while interacting with 12 different types of devices. In their trial of more than 500 minutes of recorded activity, MagnifiSense correctly recognized 94 percent of the devices when the system was calibrated, and 83 percent without calibration.

In a more organic trial, a single user wore the device for 24 hours and engaged in normal activities such as reading on a laptop, cooking and taking the bus. The system correctly identified 25 out of 29 of the interactions with electronics and motors without calibration.

The results and a demo of MagnifiSense were presented at the September conference in Osaka. In addition to Wang and Patel, the team working on the project includes UW electrical engineering doctoral student Tien-Jui Lee, computer science and engineering doctoral students Alex Mariakakis and Mayank Goel, and Sidhant Gupta of Microsoft Research.

Wang said it would be easy to make the sensor much smaller and faster, and incorporate it into a phone or other wearable device. Where it goes next is unclear.

“Hopefully,” Wang said, “someone picks it up and says, ‘This is interesting. We want to move forward with it.’ ”

Like what you're reading? Subscribe to GeekWire's free newsletters to catch every headline

Comments

Job Listings on GeekWork

CTO-in-Residence // Co-FounderALLEN INSTITUTE FOR ARTIFICIAL INTELLIGENCE (AI2)
Deep Learning Engineer // Co-FounderALLEN INSTITUTE FOR ARTIFICIAL INTELLIGENCE (AI2)
Application SpecialistCity of Missoula
Find more jobs on GeekWork. Employers, post a job here.