Going into UofThacks, we really wanted to work on a security/privacy related hack, and eventually brainstormed an idea for a location surveilling system based on MAC addresses in WiFi packets that seemed to fit the bill.
What it does
Big Brother consists of a number (currently three) of Raspberry Pis, each with two network adapters, one of which is connected to the University of Toronto's student WiFi, and one of which is set to monitor mode and used by the Kismet packet sniffer to scan for as many 802.11 packets as possible. Packet data is set to a central Kismet server running on a VPS, which combines all the packet data into a single sqlite database.
A straightforward webapp can then be used to comb through the data, narrowing in on particular individuals (see images). Kismet reports the specific RPi that picked up each packet, which can be mapped to a location description (eg. Galbraith building lobby, Sanford Fleming Building Pit, etc.). This enables people's location to be tracked over time based on packets intercepted from their devices.
There are a variety of possible applications for Big Brother, from surveillance for law enforcement purposes, to location based targeted advertising. Our main goal in building it was to raise awareness of the fact that tracking individuals based on wireless devices they carry around is a relatively simple process, and doesn't require any of the target's devices to be compromised.
Challenges we ran into
Getting the extra USB WiFi adapters to work comprised the bulk of the pain involved in this project, it was a long arduous process.
What we learned
Just use Kali Linux, it has the WiFi dongle drivers pre-installed.