Reflection on WiFi Tracking
Brooke and Hans
Successes and Failures:
Our product didn’t work as well as we would have liked. This was due in part to the high level of security on Tufts’ networks, a Tufts refusal to allow us any temporary admin permissions or allow for the creation of a subnetwork of routers for use in controlled experimentation, and difficulties associated with networking as a whole. These complications led to incredible delays in developing the initial MAC address module and later made it impossible for us to finish the trilateration component of our project. However, a solution for some networking issues was managed with the use of Raspberry Pis and the creation of a private hotspot network. This allowed us to successfully sweep the network and collect the MAC addresses of the devices connected to it.
What we Learned:
Networking sucks. Whether it’s poking around a secure school network, a private home network, or a temporary ad hoc network, the inconsistencies in hardware, software, and interface make it incredibly difficult to develop a generic, modular solution that can be applied to any network. It is through dealing with these inconsistencies and attempting to find solutions to them that we learned the most in this project. WiFi sniffing, screen scraping, trilateration, bash-executed networking packages, and interfacing with RESTful APIs were all employed heavily throughout development, and we became very familiar with them through the process. We also learned that we definitely underestimated many of the components and technical challenges that are necessary to achieving our desired final project. Upon further research, it was revealed to us that our goal is one that IT companies still haven’t accomplished and are currently in the process of working on as well… however they’re afforded the luxuries of applying hundreds of computer scientists and network engineers and having a great deal more time than our two weeks. Once a proper solution is developed, it’d be interesting to take a look and see how they best solved the solution. My guess is it might be fairly similar to our rough concept! All told, though our final product did not greatly reflect our broader aspirations, we learned a tremendous amount by struggling through it.
Network is read properly and MAC addresses are paired and pushed to cloud.
Trilateration didn’t work.
Script requires some higher level permissions to execute. Cannot be executed on a secure network.
Trilateration requires a consistent measurement of signal strength/time of flight to gauge distance. This is difficult to accomplish indoors.
Script requires manual username input for MAC association.
Broad professional applications for network security.
Individual tracking systems already exist (RFID tracking, 3rd parties are developing WiFi solutions).
Walls are difficult to work with for signal measurement!
If put in the wrong hands, this technology could be used inappropriately.