Photo: Colourbox

Apps can monitor your movements

Friday 29 May 15

Contact

Sune Lehmann Jørgensen
Professor
DTU Compute
+4545 25 39 04

More information

Read Sune Lehmann’s blog posts about the results.

Read the study ‘Tracking Human Mobility using WiFi signals’ (PDF).

Download the DTU Compute app, which registers your movements on the basis of WiFi information.

WiFi information—data from wireless routers—which is routinely collected by an Android smartphone can provide highly accurate information about your whereabouts 90 per cent of the time. This is the finding of a new study from DTU Compute.

The majority of apps in Android phones have access to a list of routers for which the phones are scanning every 20 seconds, even though you have turned off WiFi on your phone. Based on six-month data from 63 DTU students, a new study from DTU Compute shows that app developers can use WiFi information as localization tools to pinpoint when people leave home, when they are at work, and where they spend their spare time.

“People are very predictable and do many similar things. So the ten places you visit most frequently cover 90 per cent of your time. This makes it possible to map your main movement patterns using very few routers,” explains Associate Professor Sune Lehmann from DTU Compute, and continues:

“Even though you turn WiFi off on your mobile phone, it will still be scanning for wireless access points, and your apps will still have access to this information. We have now demonstrated that WiFi information can easily be converted into information about your geographical position. So even if you don’t allow an app to track your location, it will still be able to perform the exact same tracking of your movements thanks to its access to WiFi information. Without having asked for access to this information, that is.”

Our very predictable movement patterns also mean that unless we move to another address in the meantime, the information gathered will also reveal a lot about our movements after six months.

Movements easy to track
The study is part of the Sensible DTU research project, where 1,000 DTU students have being given a smartphone in return for donating their data to the study. The huge data volumes collected can be used to find patterns in human interaction. But in this case, the massive data collection has also given the researchers an insight into how much you are actually able to find out about a person, explains Sune Lehmann:

“We know that companies know a great deal about us, but not how much. By making a similar monitoring of our students, we are beginning to understand just how many possibilities the companies have to gather knowledge about us—and we can begin to see how much they actually know.”

In February, Sune Lehmann and his colleagues checked the 20 most popular games on Google Play Store and discovered that 17 of them had access to WiFi information. Only six out of the 17 games gave an explanation as to why this was the case. Sune Lehmann emphasizes that the study does not show whether the companies actually utilize the movement tracking capability, which is very easily done. For example, the team behind the study developed a small app for the sole purpose of tracking a person’s movements based on WiFi information.

“We show that everyone is able to build an app that can most likely reconstruct your movements. So we shouldn’t just focus our attention on the large companies here,” says Sune Lehmann, stating that companies, among other things, can exploit the information to target their marketing, for example by displaying context-specific advertisement when you move into a specific area.

“The problem is that we would like to live our lives without being monitored. In my opinion, privacy is a fundamental human right. What we are showing here is yet another opportunity available to companies when they try to describe you with an algorithm in their efforts to manipulate you to buy their products,” says Sune Lehmann.

Related videos  

Show more

News and filters

Get updated on news that match your filter.