Kalman filter for RSSI-based indoor positioning system with min-max technique

Location-based service (LBS) is a digital service that utilizes geographical location to provide various services or information to the user. Although LBS is more commonly used outdoors, demands for indoor LBS have been rising. However, the global positioning system (GPS), the technology widely used in LBS, is unreliable when used for indoor positioning systems (IPS) due to signal attenuation and the lack of line-of-sight (LOS) caused by the walls and roofs. Multiple technologies and methods have been proposed to achieve reliable IPS. The received signal strength indicator (RSSI) is a parameter commonly used in distance-based IPS. A path loss model is used in distance-based IPS using RSSI to find the relationship between power and distance. RSSI can be very noisy, which causes unreliable conversion to distance and eventually leads to unreliable IPS. In this research, we propose an RSSI values improvement by applying a Kalman filter to the raw RSSI values obtained from the Bluetooth low energy (BLE) devices. We applied the min-max technique by using five BLE devices. We conducted actual measurement campaigns for our proposal validation. Measurements results show that the BLE devices have the best condition to obtain RSSI values within a three-meter distance. We employed the Kalman filter for raw RSSI values and improved our IPS system rather than only using raw RSSI for the min-max technique. The filtered RSSI improves accuracy, precision, and resolution of 0.78m, 0.23m, and 1.01m compared to raw RSSI of 0.90m, 0.36m, and 1.50m, respectively. These improvements prove that applying the Kalman filter to our raw RSSI data can improve min-max-based IPS performance.