Strategic cost reduction in indoor positioning systems using signal propagation modeling techniques

Abstract

Indoor Positioning Systems (IPSs) are used to estimate the position of mobile devices in indoor environments. Fingerprinting is the most used technique because of its higher accuracy. However, this technique requires a labor-intensive training phase that measures the Received Signal Strength Indicator (RSSI) at all Reference Points (RPs) locations. On the other hand, model-based IPSs use signal propagation models to estimate distances from RSSI. Thus, they do not require expensive training but result in higher positioning errors. To mitigate this problem, this thesis explores improvements in signal propagation modeling as an alternative to reduce data collection efforts focusing on system accuracy. Three novel approaches are proposed. SynTra-IPS (Synthetic Training Indoor Positioning System) is a hybrid approach that generates synthetic training datasets using a log-distance propagation model. Machine learning algorithms process these datasets, and data fusion techniques enhance position estimates. ADAM-IPS (Adaptive Model Indoor Positioning System) improves anchor node selection and applies signal propagation modeling with data fusion to estimate distances, eliminating the need for extensive dataset collection. PSO-MIPS (Particle Swarm Optimization - Model-based Indoor Positioning System) uses particle swarm optimization with signal propagation modeling to refine position estimation without requiring predefined parameters or prior training. These methods were tested in large-scale real-world environments, demonstrating their effectiveness in reducing data collection requirements while maintaining optimal localization accuracy compared to existing indoor positioning systems.