多階訊號強度之感測網路定位技術

Abstract

感測式網路(sensor network)這一項技術讓我們可以深入且貼近地觀察一整片地區，目前討論的課題多半著重在感測式節點的繞路、合作與能量消耗等方面。定位系統是無線感應式網路的重要基本功能，然而卻是比較少被討論到的。在本篇論文中，我們將提出一個適合應用在無線感測式網路上的定位方法。我們利用參考點廣播其多階訊號強度，將整個平面分割成許多小區域，感測式節點依照收到訊號的不同來進行定位。在論文中，我們將會討論這個定位方法的極限，並進行模擬測試。根據模擬結果顯示，我們的方法相較於單一訊號強度方法可提供較好的準確度，且這個方法在訊號傳輸距離不穩定、訊號碰撞以及隨機散佈等狀況下皆有不錯的表現，相當適合運用在感測式網路環境中。

Sensor network technique enhances our ability to monitor and observe the physical world. Many recently researches for sensor network focus on routing, nodes cooperation and energy consumption. In addition to these topics, localization service is an important function in sensor network system. In this thesis, we will present a multiple power-levels localization algorithm for wireless sensor network. We also discuss the boundary of our algorithm and evaluate the performance in different environment. The simulation results show that our algorithm can reach much better accuracy than traditional single power-level methods. In critical situations such as reference node failure, unstable radio transmission range and beacon collision, our algorithm performs well. We believe that this algorithm is suitable for sensor network.