Indoor positioning systems (IPS) have become critically important in environments where GPS signals are unreliable, such as offices, factories, warehouses, hospitals, and construction sites. The core problem is that traditional GPS performs poorly indoors due to signal attenuation, creating a significant gap in the ability to track people and assets in real time. This gap impedes the potential to improve safety, streamline operations, and enable new applications like automation and augmented reality navigation, necessitating a robust alternative solution.
IPS leverages alternative technologies to achieve location awareness indoors. Geomagnetic positioning relies on unique distortions of the Earth’s magnetic field inside buildings to create a location fingerprint, while inertial navigation uses built-in sensors to track motion and orientation. Wireless signals such as Wi-Fi and Bluetooth Low Energy (BLE) are used to strengthen localization accuracy. To ensure robust performance, techniques such as Kalman filtering and wavelet denoising are applied to remove noise from sensor data, while clustering algorithms like K-Means and DBSCAN help categorize asset states. Furthermore, the integration with AI technologies enables advanced use cases like AR/VR-based guidance and multilingual analysis.
We have focused on a mobile-device-centric solution that synergistically combines inertial, magnetic, and wireless technologies that is expected to support an average accuracy of about one to three meters across different indoor environments. The system demonstrates reliable positioning performance and compatibility with widely available devices, supported by stable data processing through advanced filtering and denoising techniques. We have successfully implemented AI techniques for accurate state detection, such as moving, paused, or resting. Furthermore, we have demonstrated a path for AI integration, enabling future applications in multilingual support and immersive navigation, all within a smartphone-friendly and scalable framework.
Future work includes enhancing the accuracy with high precision that are required in domains such as healthcare or robotics, and the scalability in large, complex environments like multi floor buildings with metallic structures. Robustness must be improved against environmental changes like furniture rearrangement or wireless interference. There is also a pressing need for better deployment tools, including user-friendly calibration methods and visualization dashboards, alongside addressing critical privacy and security concerns for handling sensitive location data. Standardization of protocols and collaboration between academia, industry, and policymakers are also essential to accelerate mainstream adoption.
Indoor positioning with smartphones is already feasible and provides practical benefits. We have built a functional system with reliable accuracy and have integrated key data processing and AI techniques. Moving forward, contributions can be made in developing advanced sensor fusion algorithms, creating adaptive AI models, and building robust visualization platforms. The future success of IPS will depend not only on technical innovation but also on strong collaboration across technology providers, industry users, and research communities to address practical deployment challenges. By focusing on scalability, cost-effectiveness, and
security, IPS can mature into a foundational technology for the next generation of smart buildings, connected factories, and intelligent healthcare facilities.
