Performance Analysis of Closed-Form Least-Squares TDOA Location Methods in Multi-Sensor Environments

• Main Authors: Wen-chin Ou, 歐文欽
• Other Authors: Chin-der Wann
• Format: Others
• Language: zh-TW
• Published: 2006
• Online Access: http://ndltd.ncl.edu.tw/handle/59818155782008751796

碩士 === 國立中山大學 === 電機工程學系研究所 === 94 === In indoor environment, the multi-sensor system has been proved to be an efficient solution for target locating process in terms of lower estimation cost. However, the placement of designed multi-sensor has great impact on the location performance in an indoor environment. Based on the time difference of arrival (TDOA), closed-form least-square location methods, including the spherical-interpolation (SI) and the spherical-intersection (SX) methods, are used in the estimation of target locations. The two methods are apart from the usual process of iterative and nonlinear minimization. Consequently, under the influence of noise interference, the performance of the two methods also produce different results. In addition to the above issues, the limitation of these methods will also be examined. The geometric dilution of precision (GDOP) effects of TDOA location on location performance of both inside and outside of the multi-sensor environment in the 2-D scenario have been studied in the past. This thesis aims to further advance the performance of GDOP in 3-D scenarios, analyze the differences, and propose the suitable needs. Programmed 3-D scenario simulations are used in this research, designed according to multiple sensor arrays and the moving latitude of a target. The Setup interprets the degree of multi-sensor separation, and distances from targets to the sensor array. A suitable location algorithm and optimal multi-sensor deployments in an indoor environment were proposed according to the simulation results.