95% of mobile phone users leave location data more unique than fingerprints
We all know that people can be uniquely identified by their fingerprints. Now, a study reveals that 95% of mobile phone users leave location data that can be even more unique than fingerprints. Last week, a scientific paper titled Unique in the Crowd: The privacy bounds of human mobility was released and offers a study of 1.5 million mobile users between April 2006 and June 2007 in an unnamed western country.
Because mobile phones ping cell phone antennas automatically as they are carried from place to place, a database was built up. And by tracking connecting antennas when a mobile phone received a call or SMS message, particular individual patterns could be discovered. Using spatial and temporal aggregation, a technique used by law enforcement, a highly accurate prediction could be made as to where specific mobile phone users would be at a particular time of day. This is not to say that these people were identified as they did remain anonymous to the researchers. But with a warrant, we would imagine that individual names could be discovered. Apple has recently updated its privacy policy and will share spatio-temporal location of Apple iPhone users with "partners and licensees".
source: Nature.com, FastCompany via textually.org
One important point to understand is that this study was done before the smartphone became part of everyone's daily arsenal. It also was done in the days before App Stores and apps that required your location. Apps like Twitter, Foursquare and Facebook to name a few, use geo-tagging. This data not only allows you to use certain features of these applications,it also helps advertisers focus a specific campaign on a certain area. It's all benign to be sure, but there is always the fear of abuse. If this study is right, location data could some day provide the "aha" moment in a courtroom. And by the way, this was not a paper done by a couple of morons. This was a project developed by MIT and Belgium's Louvain University.
Using a spatio-temporal filter, a mobile phone user's location can be accurately predicted as to place and time
Things that are NOT allowed: