"A sufficiently long power measurement (several minutes) enables the learning algorithm to 'see' through the noise," Yan Michalevski, one of the Stanford's researchers, told Wired. "We show that measuring the phone's aggregate power consumption over time completely reveals the phone's location and movement"
"If you take the same ride a couple of times, you'll see a very clear signal profile and power profile," says Michalevsky. "We show that those similarities are enough to recognize among several possible routes that you're taking this route or that one, that you drove from Uptown to Downtown, for instance, and not from Uptown to Queens."
"You could install an application like Angry Birds that communicates over the network but doesn't ask for any location permissions. It gathers information and sends it back to me to track you in real time, to understand what routes you've taken when you drove your car or to know exactly where you are on the route. And it does it all just by reading power consumption," Michalevski concluded.