Developing a BIM-Assisted 3D Access Point Placement Optimization Algorithm for Enhancing Wi-Fi Fingerprint-Based Indoor Positioning

Indoor positioning is pivotal to various promising applications in building construction and facility management. Indoor position information of humans and objects, such as construction workers, materials, vehicles, and tools can largely improve working efficiency and workplace safety. Wi-Fi fingerprinting is a popular technique that collects Wi-Fi signal strengths, which are measured by received signal strength index (RSSI), for indoor positioning. However, Wi-Fi access points (WAPs) are normally placed arbitrarily, which causes poor positioning accuracy. In fact, positioning accuracy can be considerably enhanced by optimizing the access point (AP) placement strategy. In this paper, the objective of the AP placement optimization is to maximize the distinctiveness between individual Wi-Fi fingerprints in a virtual environment. BIM technology provides 3D geometric and semantic information to accurately reproduce the virtual environment for realistic Wi-Fi signal propagation simulation. The Wi-Fi signal propagation is modelled by a modified indoor radio wave path loss model, combined with the ray-tracing method using particle swarm optimization (PSO), to emulate the indoor multipath effect. The optimal AP placement strategy is obtained by using the genetic algorithm (GA). The accuracy test results inside a university library have validated that the developed AP placement optimization algorithm substantially enhances the positioning accuracy by 57.4%.