TOA(Time of Arrival)無線定位法則之研究

Abstract

本研究選用幾何分佈一次彈射Spatial-temporal 通道模型，評估重心法、兩直線交點法、最小平方法等三種傳統TOA定位演算法在蜂巢式網路運作的效能，包括多重路徑干擾效應、非視線內傳播效應、區域散射效應、行動台在細胞內的位置等因素對定位準確度的影響。非視線內傳播效應為定位誤差之主要來源，吾人將使用以下的方式提高傳統TOA定位演算法的定位準確度：(1)利用均方根時間延遲擴展降低非視線內傳播效應對定位準確度之影響，加強傳統演算法的定位效能。(2)多組基地台同時進行定位，並利用第(1)種方式加強整體定位效能。(3)提出混合性定位演算法：結合統計法與幾何法提高定位準確度。修正後的定位演算法平均可降低50~70公尺的定位誤差，但除第(1)種方式外，其它皆需花更多的計算時間。

In this paper we want to validate the performance of traditional TOA (Times of Arrival) position algorithm applying in Cellular Network by numerical simulation, including Multipath effect, NLOS (Non-light of Sight) propagation, local scattering effect, cell coverage radius, the position of Mobile Station (MS) in the cell and the choice of different positioning base station affect the positioning accuracy. The NLOS propagation is the main source of position error. We enhance the traditional algorithms and develop a hybrid algorithm to mitigate the NLOS error in location estimation. These algorithms can reduce averagely 50 to 70 meters of location error, but take more time in execution.