Direction finding for UWB RFID system in dense cluttered environments
Date of Issue2014
School of Electrical and Electronic Engineering
Positioning and Wireless Technology Centre
Source direction finding with an antenna array is a major topic in radar-related applications. Typically, an incident radio wave can be detected by measuring the difference of arrival time or phase between spatially spaced sensors of the antenna array. It becomes challenging due to abundant multipath and obstacles in dense cluttered environments, such as enclosed areas, tunnels or under-ground. Ultra-wide band (UWB) radio frequency identification (RFID) system provides a promising solution in these global-positioning-system (GPS-) denied environments due to a number of advantages such as low duty cycle, multipath immunity and low cost. However, the system susceptibility to dense cluttered environments is not well studied, and the potential of UWB signals for direction finding is not well investigated. In this thesis, the performance limits in dense cluttered environments are firstly studied in detail. A channel model of plane waves from far-field reflectors or scatterers superimposed with near-field interferences is proposed, and the Cram´er-Rao bounds for ranging and direction finding are derived to analyze the channel effects. Based on the insights into path-overlapping effects in the first cluster, a joint time-of-arrival (TOA) and angle-of-arrival (AOA) estimator with improved performance is then designed. Secondly, various channel effects, including the signal-to-noise ratio (SNR), the near-field perturbation level, the far-field reflector spatial spacing and uncertain number of plane waves, on the robustness of UWB RFID system for direction finding are addressed. A robust solution for UWB impulse radio signals employing the alternating projection (AP) strategy in the presence of interference is investigated. Detection of the number of plane waves prior to AOA estimation is considered to improve the processing efficiency of AP strategy. Finally, channel identification that includes the line-of-sight (LOS), the non-line-of-sight (NLOS) with detected direct path (DDP) and the NLOS with undetected direct path (UDP) is studied. Measurements of three channel features are modeled as the log-normal random variables with site-specific distribution parameters. Joint log-likelihood ratio test with an empirical threshold selection is designed to make the decision. Accurate AOA estimation is helpful to measure the rise time, the kurtosis of signal envelope and the excess attenuation of the first detected path.
DRNTU::Engineering::Electrical and electronic engineering::Antennas, wave guides, microwaves, radar, radio