Title: 無線感測網路之修正蒙地卡羅定位研究
Modified Monte Carlo localization with in/out range constraints for wireless sensor networks
Authors: 黃柏益
Huang, Po-Yi
謝世福
Hsieh, Shih-Fu
電信工程研究所
Keywords: 蒙地卡羅;定位;Monte Carlo;localization
Issue Date: 2010
Abstract: 隨著無線感知網路技術的發展,其應用的研究在最近也成為熱門的話題。其中以定位最為重要。因為要提供相關的服務或是應用,必須先知道適用對象的位置。而在無線定位技術中可分為兩類:range-based 和range-free。Range- based的技術是去推算出待測物跟已知物之間的距離。但是此種方法有兩種缺點。第一,必須有額外的硬體設施去量測。第二,接受到的資訊很容易受雜訊的干擾。Range-free的方法則是去量測待測物是否在已知物的範圍內。此種方法可以避免前面兩項缺點。Range-free的方法中又以蒙地卡羅定位法最為常見,它利用亂數模擬產生待測物的樣本區,取用樣本區的中心點當作估計位置。在蒙地卡羅定位法中若可以增加限制進而縮小樣本區便可以提高準確度。在過去,有人提出合作式定位來增加限制。在本論文中,我們多加了界內與界外限制並賦予適當的權重,此修正的蒙地卡羅定位法可以有效縮減樣本區的大小,進而提高定位的準確度。在最後的電腦模擬中,我們不只考慮樣本區的大小,我們也會考慮待測點的個數、已知物的個數及移動的速度是否會影響到最後的結果。最後,藉由電腦模擬可以證明所提出的修正方法是有效的。
Localization is an important topic in wireless sensor networks. In wireless localization, there are two categories : range-based and range-free. The range-based technique is to estimate the distance between normal nodes and anchor nodes. The range-free technique simply detects whether the normal node falls in the range of anchor nodes. Monte Carlo localization is a commonly used range-free method by randomly simulating the sample area of normal nodes. If extra constraints can be imposed, we can improve its localization accuracy. In the past, cooperative localization has been proposed. In this thesis, additional constraints, both range-in and range-out, will be considered to further enhance accuracy. Computer simulations will validate the proposed algorithms can improve the localization accuracy.
URI: http://140.113.39.130/cdrfb3/record/nctu/#GT079813566
http://hdl.handle.net/11536/47048
Appears in Collections:Thesis


Files in This Item:

  1. 356601.pdf
  2. 356601.pdf

If it is a zip file, please download the file and unzip it, then open index.html in a browser to view the full text content.