Author

Jun Chen

Date of Award

8-2010

Level of Access

Campus-Only Thesis

Degree Name

Master of Science (MS)

Department

Electrical and Computer Engineering

Advisor

Ali Abedi

Second Committee Member

Donald M. Hummels

Third Committee Member

Mauricio Pereira da Cunha

Abstract

Excerpted from Conclusions, page 82: In this thesis, a wide range of wireless channel models are investigated including indoor propagation, building penetration, hilly terrain, rural, urban area and ultra-wideband indoor channels. Wireless channel models help evaluate the positioning accuracy of localization techniques in realistic environments.

Theoretical models are set up to analyze fundamental limits of RSS, TOA, AOA and hybrid localization techniques in passive wireless sensor networks. Numerical results based on proposed mathematical models show that the hybrid localization can achieve high resolution positioning with the accuracy within a centimeter using sensors having wide bandwidth of 1 GHz and high center frequency of 6 GHz and the receiver having a relatively large distance of 1 meter among a number of receiving antennas. Hybrid TOA/AOA and RSS/TOA/AOA schemes demonstrate up to 250 times improvements of the accuracy compared with RSS-only localization, TOA-only localization and AOA-only localization, respectively.

Furthermore, localization in WCDMA networks is shown as an example for active wireless systems. Simulation results illustrate that the hybrid RSS/TOA/AOA scheme in a WCDMA network could achieve approximately 73.6%, 41.5%, 35.8% and 16.2% improvements of the positioning accuracy over AWGN channel compared with RSS-only localization, TOA-only localization and AOA-only localization, respectively. It is shown that the proposed hybrid model over various fading channel models provides more than 20% improvement of the resolution than that of any other localization technique.

Share