The crucial part of our project’s functionality is accurately detecting proximity. There are 4 possible ways of accomplishing this: GPS mapping, ultrasound, RSSI (Received Signal Strength Indication), and signal bouncing. For our purposes, GPS and sonic detection are unfeasible because they are too imprecise. We have chosen to implement RSSI in our prototype because it is the simplest and most ubiquitous proximity detection method. RSSI uses the signal strength of the devices’ bluetooth connection to predict distance. The closer the beacons are to each other, the higher the signal strength, and vice versa. However, it is still highly prone to error due to signal reflection and absorption. It is possible that RSSI readings change depending on the location of the beacon on a user’s person or even the environment. We plan to implement several RSSI vs. distance curves in our code to more accurately predict distance based on these factors.

In the future, we would like to attempt the signal bouncing method. We began using RSSI so that we could focus on establishing a connection between devices, but there are still improvements that can be made to precision. Signal bouncing involves high clock rate timing of when our beacons receive signal. Since our desired range is within 6 feet and EM signals travel at the speed of light, we would need a clock rate of over 85 MHz, but this could require upwards of 1 GHz to accurately time even closer proximities. Once a device confirms that it receives a signal, it will “bounce” the signal back to the initial broadcasting device back and forth. During this process, we may use timers to conclude how long it takes for the signal to bounce a specified amount. With this, we could theoretically calculate the distance between the beacons.