Weighted Least-Squares Bluetooth Localization Method Based on Data Preprocessing

People’s need for positioning and navigational data has grown over time as communication technology has advanced, becoming an essential and significant aspect of daily life. As a result of the global navigation satellite systems (GNSSs) increasing maturity, satellite positioning technology can now achieve the highest centimeter-level positioning accuracy, effectively meeting the majority of outdoor positioning requirements. However, indoor positioning technology emerged as a solution for locations where satellite signals are obstructed and cannot normally provide positioning information. Researchers have focused on the Bluetooth positioning algorithm among them because it is a wireless positioning method. There are two types of traditional Bluetooth received signal strength indicator (RSSI) positioning methods based on the Internet of Things (IoT): 1) trilateral and 2) multilateral. However, because environmental interference causes the Bluetooth signal’s RSSI value to fluctuate greatly, using either of the above methods will result in significant errors and a decline in positioning accuracy when buildings, objects, or people are in the way. To address this issue, this article proposes preprocessing the data prior to applying the least-squares method. At the same time, based on processing outcomes, the weighted least-squares method is utilized to determine Bluetooth RSSI positioning, resulting in a notable improvement over the accuracy of the traditional RSSI Bluetooth positioning.