A novel orientation-based FSPL model parameter optimization method using PSO for indoor localization

Indoor localization plays a very important role both in mobile robotics and Wireless Sensor Networks (WSNs). With the spread of the Internet of Things (IoT), different technologies using radio waves are playing an increasingly crucial function. Among them, the most used technology is WiFi. Usually the Received Signal Strength Indicator (RSSI) is used to determine the distance between two units. The relationship between the distance and the RSSI value is determined by the Free Space Path Loss (FSPL) model. The parameters included in this model affect the distance estimation and, indirectly, the localization accuracy. Therefore, a method that can characterize the model well is crucial. In this paper, a novel orientation-based parameter optimization approach is proposed. Two parameters of the FSPL model, i.e., the environmental factor and the reference RSSI, were considered. Measurements were performed in different orientations between the two ESP32 units, and optimal parameters were obtained for each orientation. The optimization was executed with the Particle Swarm Optimization (PSO) algorithm. The obtained results show that the fine-tuned orientation-dependent parameters significantly increase the measurement accuracy compared to the conventional, orientation-independent one parameter pair-based approach.