Real-World RF-Based Localization: Design, Implementation and Applications

Abstract

The rapid proliferation of Internet of Things (IoT) technologies has revolutionized the way we address safety, efficiency, and resource management challenges in urban environments. Among these technologies, Radio Frequency (RF)-based intelligent positioning systems have emerged as a powerful tool for real-time monitoring and localization in complex indoor and outdoor settings. Despite decades of research on RF-based localization, most studies have been confined to controlled laboratory environments, leaving a significant gap between academic advancements and real-world, large-scale commercial deployments. This thesis bridges this gap by presenting the design, implementation, and evaluation of two large-scale RF-based localization systems—DeMo and DECS—that leverage Bluetooth Low Energy (BLE) technology to address critical urban safety challenges. These systems have been successfully deployed in Hong Kong, demonstrating the potential of intelligent positioning systems to enhance safety and efficiency in urban environments.

The first system, DeMo, focuses on monitoring indoor delivery operations in subway stations, where heavy, high-stacked packages carried on delivery trolleys pose significant safety risks to pedestrians. DeMo utilizes a combination of Inertial Measurement Unit (IMU) sensors and BLE technology to detect delivery violations such as speeding and the use of non-designated paths. A key innovation of DeMo is its fingerprint-free localization approach, which eliminates the need for costly and time-consuming fingerprinting processes. Instead, DeMo adopts a simple RSSI-distance model, drastically reducing deployment and maintenance costs while achieving high accuracy. Since its deployment in May 2020, DeMo has monitored over 74,537 deliveries across 12 subway stations, reducing violation rates from 19% to 0.9%. This significant improvement underscores the system’s effectiveness in ensuring safe and compliant delivery operations in crowded urban environments.

The second system, DECS, addresses the safety of People with Dementia (PwD), who are at high risk of getting lost due to cognitive deterioration. DECS employs a crowdsourcing platform where PwD carry lightweight BLE tags that broadcast signals detected by volunteers’ smartphones and customized BLE gateways (angel boxes). By analyzing PwD’s daily spatial-temporal mobility patterns, DECS enhances search efficiency and has successfully resolved 254 missing cases since its deployment in 2019. With over 45,000 app downloads and support for 3,100+ families, DECS demonstrates the power of community-driven IoT solutions. The system’s success highlights the importance of leveraging existing infrastructure and mobilizing community resources to address critical societal challenges.

This thesis makes several key contributions to the field of RF-based localization. First, it provides detailed insights into the challenges and solutions associated with deploying large-scale intelligent positioning systems in real-world urban environments. Second, it presents novel technical approaches, such as fingerprint-free localization and crowdsourced detection, that address the limitations of existing systems. Third, it demonstrates the societal impact of these systems through measurable improvements in safety and efficiency, including reduced delivery violations and successful searches for missing PwD. Finally, it offers practical lessons and recommendations for future deployments, emphasizing the importance of cost-effective design, user engagement, and privacy preservation. Through extensive evaluations and real-world deployments, this thesis validates the effectiveness of DeMo and DECS in enhancing safety and efficiency in urban environments. It also highlights the broader implications of intelligent positioning systems for improving quality of life and reducing public safety risks. By bridging the gap between academic research and real-world applications, this thesis lays the foundation for future advancements in RF-based localization and IoT technologies, paving the way for smarter, safer, and more sustainable urban environments.

Date of Award	11 Jun 2025
Original language	English
Awarding Institution	City University of Hong Kong
Supervisor	Zhimeng YIN (Supervisor)