MagnifiSense uses electromagnetic signatures to keep tabs on your energy use

3 pictures

The MagnifiSense prototype detects what electronic devices and motor vehicles an individual is interacting from their unique electromagnetic signatures(Credit: University of Washington)

View gallery - 3 images

From the Fitbit to the Apple Watch, there's no shortage of wearable devices that track your daily activity with an eye on your personal health and wellbeing, but a new device developed at the University of Washington (UW) can track your activity as it pertains to the health of the planet. Called MagnifiSense, the wrist-worn prototype detects what devices and vehicles the wearer interacts with throughout the day to help track their carbon footprint.

MagnifiSense senses what electronic devices and motor vehicles an individual is interacting with by detecting their unique electromagnetic signatures.

"When a blender turns on, for instance, modulators change the current profile of the device and create something similar to a vocal cord pattern," explains study lead author and UW electrical engineering student Edward Wang. "A blender 'sings' quite differently than a hair dryer even though to our ears they sound similar."

The prototype combines three basic, off-the-shelf sensors that use coils of wire around magnets (inductors), to accurately capture a broad frequency range of electromagnetic radiation without being too power hungry. Signal processing and machine learning algorithms developed by the team match the electromagnetic signatures to a particular type of device.

Without calibration, the team says the MagnifiSense correctly identified 12 common devices, including microwaves, blenders, remote controls, toothbrushes, laptops, light dimmers, and cars and buses, with 83 percent accuracy. However, a quick one-time calibration upped the accuracy to 94 percent.

"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, an Associate Professor in UW's departments of Computer Science & Engineering and Electrical Engineering. "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."

Over a 24-hour test, in which a single person wearing the prototype went about their day performing everyday tasks such as using a laptop, cooking dinner and riding a bus, the research team says the system correctly identified 25 out of 29 interactions with various devices and vehicles.

In addition to tracking an individual's carbon footprint through the devices they interact with throughout the day, the UW team says the MagnifiSense system could also find use in smart home and aged care applications. For example, it could identify when an individual is interacting with a specific device, such as a tablet, so the experience can be customized appropriately, or track how elderly people are faring with cooking and grooming in an assisted living or nursing home setting. It could also identify if an appliance, such as a stove, has been left on for an extended period.

"The nice thing with MagnifiSense is that you don’t have to instrument every single appliance in your house, which gets expensive and cumbersome," says Wang. "It can also sense some of the blank spots that other technologies can’t, like battery-powered devices."

The team plans to continue development of the Magnifiense, with the aim of expanding the variety of devices it can detect and improve its ability to distinguish between multiple devices in close proximity. Additionally, they plan to miniaturize the technology so that it could be embedded within a smartwatch or wristband – this could be relatively simple, with the team believing that a small improvement in the update rate of magnetic sensors already found in such devices could enable MagnifiSense functionality with nothing more than a software update.

The team will present their study this week at the 2015 ACM International Joint Conference on Pervasive and Ubiquitous Computing (UbiComp 2015).

View gallery - 3 images

Top stories

Recommended for you

Latest in Wearables

Editors Choice