I remember when I first started to use Bluetooth (BT) on my phone years ago, I thought it was too much of a hassle to get started and that it was a drain on my battery so I never really used it. In my role here at Connectivity Research Center, over the past year, I have had to work with Bluetooth a lot and more specifically Bluetooth 4.0 which is more commonly referred to as Bluetooth Smart or Bluetooth Low Energy. Bluetooth Smart has become my go to solution for projects that don’t require a constant stream of communication and have to maintain a low power profile.
One useful feature of Bluetooth Smart is that it is backwards compatible so it can work with older Bluetooth devices for its original purpose but it adds in functionality for developers to use it in extremely low power situations to save battery life. It also has specific profiles to make it easier for developers to communicate with certain sensors in an Internet of Things environment. Some examples of these are a blood pressure profile, fitness machine profile, and heart rate profile. Profiles like these are just a few examples of the flexibility of Bluetooth Smart and how it caters to Internet of Things and low power applications.
Another benefit of choosing Bluetooth Smart for a solution to a project is that it is almost everywhere. The Bluetooth Special Interest Group predicts that by 2018 more than 90 percent of Bluetooth enabled smartphones will be compatible with Bluetooth 4.0. Because there are so many devices that implement it, it is a great way to reach a vast majority of people.
Currently I am working on a project that takes advantage of this characteristic. I am using Bluetooth to make an estimation of the amount of people in an outdoor area. Bluetooth devices advertise themselves to other devices so they can pair with each other. My device searches for these advertisements and counts how many unique devices are within a 10 meter radius. Each device has a unique MAC address that is constant so the central device can count these and store these for population estimation and tracking.
The drawback to this idea is that not everyone that has a Bluetooth capable device has Bluetooth activated. And on top of that even if it is activated, some devices won’t broadcast themselves unless told to by the user. Phones are a perfect example of this due to security concerns. Over the course of my experiments I found that Fitbit®, headphones, and things such as the Apple Pencil are always broadcasting. Because of the inconsistency of Bluetooth being ‘on’, a lot of people will be invisible to the scan and those carrying more than one broadcasting device will be seen as more than one person. In a big enough crowd, enough devices should be able to be counted to get an accurate proportion of the crowd that could be directly related to the actual number of people.
This project can be very useful for the University of New Hampshire because it will allow people all over campus to better understand traffic through campus. This would give the ability to get real time data on traffic on campus and allow people to avoid high traffic zones for a quicker commute. Emergency services could also use this to quickly pick the quickest route to the location they need to get to as fast as possible.
Bluetooth has come a long way and can be used in almost any IoT situation. Most devices already have Bluetooth enabled and any DIY project can easily have Bluetooth capability added to it. It is consistent across all devices and is easy to work with while being almost guaranteed to reach a vast majority of users.