An Investigation of Close-Range Localization using Bluetooth Low Energy

In this undergraduate thesis, I explore various factors affecting localization accuracy when the receiver is within one meter of the beacon. First, I test the effects of Bluetooth Low Energy's multiple advertising channels on signal strength. I show that measuring received signal strength separately on each channel enables more accurate localization by compensating for multipath and fading. I evaluate an ordering algorithm on single channels and show that it is more accurate than the standard ranking method. Next, I determine that relative orientation and azimuth affect signal strength enough that ignoring them leads to substantially lower accuracy. I propose and evaluate a vector matching algorithm for predicting absolute distance using knowledge of the orientation and azimuth. I show that knowledge of the azimuth is not necessary if there are multiple receivers. Finally, I propose future research areas, including more sophisticated algorithms to further improve the accuracy and efficiency of the system. I also consider the real-world feasibility of my techniques.Ope