Harnessing Constructive Interference for Localization in Indoor Environments
A THESIS Presented to the Faculty of The Graduate College at the University of Nebraska In Partial Fulfillment of Requirements For the Degree of Master of Science, Major: Computer Science, Under the Supervision of Professor Ziguo Zhong. Lincoln, Nebraska: May, 2014
Copyright (c) 2014 Jihoon Yun
Note: This thesis has been embargoed and moved to
Wireless network localization using Radio Signal Strength (RSS) is considered a classical option in many scenarios because of its convenience and economic feasibility. However, this localization method is easily affected by various influences, such as unknown radio path loss factors, multi-path effects, hardware discrepancies, antenna orientation, and so forth. Among these, multi-path effects have a profound effect on uni-channel RSS measurements in the “microscale”, defined as a small region within the range of several wavelengths of the radio signal and thus degrade to the accuracy of transitional localization methods based on uni-channel fingerprinting. In response to the above limitation, this thesis introduces Harnessing Constructive Interference (HCI), a system for indoor positioning using the phenomenon of constructive interference. The key observation behind the HCI is that peak RSS measurements in the frequency domain, which are results of the constructive interference, feature a relatively small dynamic range. Utilizing this observation, we propose the HCI design to improve the performance of the location estimate. In addition, we provided three profiling methods: symmetric, asymmetric with Spline interpolation and Whittaker- Shannon interpolation to reduce the system overhead. To evaluate, the proposed designs have been implemented on the NU platform and experiments were carried out with 56 sampling positions, 5 anchor nodes, and a collective number of 1,447,800 RSS measurements. By comparing with a representative algorithms, e.g., RADAR , it is shown that the proposed HCI system significantly reduces localization errors.
Adviser: Ziguo Zhong