Bluetooth Tracking Reboot

New Team, New Goals, Same Tools

Frequent readers of this blog may remember last semester’s Bluetooth Tracking Project. It focused on developing methods to trilaterate the position of a Bluetooth device based off signal strength.

This semester, the goal has shifted. Our team will build and implement a network of stand-alone Bluetooth monitoring devices. We are basing this off of research and methodologies developed last semester. These devices will use publicly available toolsets and low cost hardware. They will collect and report information on Bluetooth devices to a central hub. This will make the findings easy to view.

Tools used last semester are being used for this project. Last semester the Bluetooth team used  blue_hydra and ubertooth in conjunction. Both provided key information regarding the functioning of the trilateration. Because we no longer need the same feedback, we decided to focus on using blue_hydra as the scanning tool. The hardware Ubertooth, and he underlying software library, will still be utilized. This will simplify the scripting process of our project immensely. Unlike last semester, we’ve decided to use the Raspberry Pi model 3 as the base of our sensor nodes. The current team is more familiar with this type of hardware. It’s also compatible with all the hardware and software required for the project. Plus, we have them handy.


Current Bluetooth Progress

We started out familiarizing ourselves with our new goals and the research from last semester’s Bluetooth project. We used the installation guide from last semester. This ensured that all our hardware is properly configured. Over the past few weeks, we have focused on testing and configuring the hardware and software. We are progressing nicely. When we have a functioning model, future reports will outline the code.

Another big aspect we have been tackling is creating a server infrastructure for collecting all the data. By nature, the sensor nodes distribute across a large physical area. They will need a way of reporting the data they have collected. We decided to take a more structured approach. We implemented an elasticsearch cluster that the nodes can connect to. Elasticsearch has a wealth of built in features, the biggest being an inherent compatibility with kibana dashboards and the elasticsearch python library. The Elasticsearch Python library ships data from our nodes to our Elasticsearch cluster. Kibana allows us to visualize, manipulate, and collect metrics. This makes it easy to view Bluetooth devices that have been collected from our nodes.

Wrap Up

When laying the groundwork for this project, we took into account previous work. It was important to set up the project in a way future teams would be able to build without making radical design changes. Last semester’s Bluetooth team had thorough methodologies and research. This enabled us to start making infrastructure choices immediately. Our next steps will focus on executing our code outlines and building out the server infrastructure.

Stay tuned for more updates to come and follow us on Twitter @ChampForensics, Instagram @ChampForensics, and Facebook @ChamplainLCDI.

More Research Projects
The Leahy Center Inventory Project
Social Media Footprint Awareness
My Experience on The VPN Comparison Team