Industrial and Management Systems Engineering


Date of this Version



A DISSERTATION Presented to the Faculty of The Graduate College at the University of Nebraska In Partial Fulfillment of Requirements For the Degree of Doctor of Philosophy, Major: Engineering (Industrial, Management Systems, and Manufacturing Engineering), Under the Supervision of Professor Robert E. Williams. Lincoln, Nebraska: December, 2012

Copyright (c) 2012 Jiaqing Wu


Radio Frequency Identification (RFID) is an information exchange technology based on radio waves communication. It is also a possible solution to indoor localization. Due to multipath propagation and anisotropic interference in the indoor environment, theoretical propagation models are generally not sufficient for RFID-based localization. In fact, the radio frequency (RF) signal distribution may not even be monotonic and this makes range-based localization algorithms less accurate. On the other hand, range free localization algorithms, such as k Nearest-Neighbor (kNN), require reference tags to be spread throughout the whole three-dimensional (3D) space which is simply not practical. In this work, a hybrid real-time localization algorithm that combines reference tags with Received Signal Strength Indicator (RSSI) ranging is introduced to improve RFID-based 3D localization in high-complexity indoor environments. The experiments demonstrate that the proposed system is more accurate than traditional algorithms under real world constraints. The active RFID system includes 4 readers and 24 reference tags deployed in a fully furnished room. The localization algorithm is implemented in MATLAB and is synchronized with RF signal data collection in real-time. The results show that the novel hybrid algorithm achieves an average 3D localization error of 1.08m which represents a significant improvement over kNN and RSSI algorithms under the same circumstance. A battery-assisted passive RFID system was deployed side-by-side to the active system for comparison. Furthermore, the reader and tag performance was evaluated in both high-complexity laboratory environment and International Space Station (ISS) mock-up with high-reflection interior surface. In addition, theoretical models on minimum number of required reference tags and localization error prediction were introduced.

Adviser: Robert E. Williams