English  |  正體中文  |  简体中文  |  Items with full text/Total items : 2737/2828
Visitors : 341773      Online Users : 31
RC Version 4.0 © Powered By DSPACE, MIT. Enhanced by NTU Library IR team.
Scope Adv. Search
LoginUploadHelpAboutAdminister

Please use this identifier to cite or link to this item: http://ir.lib.stu.edu.tw:80/ir/handle/310903100/1275

Title: 嵌入式系統上實現室內定位之研究
The Research of Implementing Indoor Location in Embedded System
Authors: 徐嘉宏
Hsu Chia Hung
Contributors: 吳鴻志
Wu Hung Chih
資訊工程學系
Keywords: 嵌入式系統;室內定位;無線感測網路;類神經網路
Embedded System;Indoor Location;Wireless Sensor Networks;Artificial Neural Network
Date: 2007
Issue Date: 2011-05-24 15:12:05 (UTC+8)
Publisher: 高雄市:[樹德科技大學資訊工程學系]
Abstract: 近年來,越來越多的技術需要依附著定位技術服務為前提來發展。以往應用在室內定位方面大致有收訊角度法、收訊時間差、時間差與收訊角度混合法、訊號強度法等四種,上述方法皆須要大量演算和需增加額外輔助硬體或短距離誤差較大之缺點。因此本論文採用類神經網路提出一種簡單快速的室內無線感測網路定位方法並實做於嵌入式系統上,透過倒傳遞類神經網路訓練和模擬測試,命中率可達100%,廣義化測試在測試誤差為0.5公尺時命中率為68%,在測試誤差為1公尺時命中率為77%,在測試誤差為1.5公尺時命中率為80%。文中也就Incremental、Batch 兩種基本倒傳遞演算法和使用QuickProp演算法針對訊號接收器的方向,分成有方向性和無方向性的兩種定位方法做整體比較,比較結果顯示採用有方向性的QuickProp演算法的定位方法於廣義化測試時會得到較高的定位命中率。
In recent years , more and more technology needs developed based on location based service. Roughly there are four kinds of accepting the Angle of Arrival (AOA), Time of Arrival/Time Difference of Arrival (TOA/TDOA), Hybrid Angle and Time of Arrival, Received Signal Strength Approach (RSS) of in localization methods to use, shortcoming that above-mentioned methods must all make mathematical calculations in a large amount or the short distance error is greater. This thesis puts forward a kind of simple and fast indoor Wireless Sensor Network (WSN) localization method and implement in Embedded System. Through simulation training , the inside test hit rate can up to 100% and the generalized test while testing the error for 0.5 meters the hit rate can be up to 68% , the error for 1 meters the hit rate can be up to 77%, the error for 1.5 meters the hit rate can be up to 80%.We also make full comparison to two traditional back propagation algorithm and QuickProp back propagation algorithm with directionality and non-directional in the article ,the comparative result shows that one kind of neural networks will get the higher hit rate when generalized test to adopt Quickprop algorithms with directionality.
Appears in Collections:[資訊工程系(所) ] 博碩士論文

Files in This Item:

There are no files associated with this item.



All items in STUAIR are protected by copyright, with all rights reserved.

 


無標題文件

著作權政策宣告:

1.

本網站之數位內容為樹德科技大學所收錄之機構典藏,無償提供學術研究與公眾教育等公益性使用,惟仍請適度,合理使用本網站之內容,以尊重著作權人之權益。商業上之利用,則請先取得著作權人之授權。
 
2. 本網站之製作,已盡力防止侵害著作權人之權益,如仍發現本網站之數位內容有侵害著作權人權益情事者,請權利人通知本校護人員(clairhsu@stu.edu.tw),維護人員將立即採取移除該數位著作等補救措施。
 
DSpace Software Copyright © 2002-2004  MIT &  Hewlett-Packard  /   Enhanced by   NTU Library IR team Copyright ©   - Feedback