Wireless sensor networks location method using linear FMCW and TDOA

This thesis presents a self-positioning wireless sensor networks with Linear Frequency Modulation Continuous Wave (FMCW) systems. It also analyzes the performance by using Time Difference of Arrive (TDOA) techniques with three different positioning methods: Taylor-iteration method, Fang's method and Chan's method in this wireless sensor network.;The FMCW generator has been designed. The computer simulation result approves FMCW system's high accuracy performance and capability to implement on the self-position wireless sensor network. A wireless sensor network environment has been set up by assigning FMCW system at reference nodes (RNs) and general nodes (GNs). The view volume obtained for different RNs have been analyzed based on the geometric dilution of precision (GDOP) theory and the optimum method for the GNs positioning has been found. Based on the experimental results, the system model has been implemented by TDOA positioning technologies. Three different positioning algorithms, Taylor-iteration method, Fang's method and Chan's method, have been compared separately. The simulation has been implemented in different number of RNs, noise levels and sweep frequencies. Large numbers of GNs have been employed in the system to obtain a Monte-Carlo statistic result. The simulation result has been shown that the designed wireless sensor network has multiple applications and robust performance for various of environments.;Three-dimension model also has been built by using entirely different positioning algorithm from two-dimension model. The level of error attenuation (in decibels) obtained using the designed wireless sensor network has been shown to be considerably lower than the level of error attenuation (in decibels) obtained using a commercial GPS. More over, the resource cost of the proposed positioning methods for wireless sensor network is much lower than the resource cost of commercial GPS.