距離估計基於接收訊號強度於視線與非視線傳輸

Abstract

隨著近來提供定位服務的需求，加上行動上網的普及，RSS-based定位模型已經有廣泛的研究和討論，本文將以無線Access Point(AP)作為定位基準，由於行動上網的使用者們可能造成傳輸過程中的屏蔽(NLoS傳輸)，導致一般以RSS-based定位的誤差值增高，本文將定位模型分為二類，分別為LoS和NLoS傳輸，藉由使用者連線到AP的訊號強弱和變動程度來判別。這當中我們將在欲定位的環境中進行模型的建立，並記錄他的變動程度，分析出LoS訊號和NLoS訊號間的不同，有了這些模型和數據，最後應用到判別傳輸模式的方法之中。 更進一步的，我們允許使用者在原地進行轉動，改變當下的傳輸模式，並利用一些偵測機制來感應這一類的變化，如此一來可更加的動態掌握使用者的傳輸狀況，而適時的能更正系統當下判別的傳輸模式，藉由上述這兩種機制的同時運行，我們能得到更精確的距離估計結果。This paper focuses on the development of the received signal strength (RSS)-based model in order to estimate the user distance from the access point to his/her wireless handset. However, the communication between the user’s handset and access point may have a non-line-of-sight (NLoS) propagation. In this case, it may cause a large deviation of measurement, and hence errors in location estimate. In this paper, we build RSS models separately based on line-of-sight (LoS) and NLoS communications and determine which model to use. Furthermore, we propose a method to detect whether the user has changed his/her communication condition in each moment. Simulation and experiment results show that our method is able to select a model with more than 50% of correctness for the user distance estimation when the user's position is static. An even more better performance is obtained when the user changes his/her communication condition.