支援大範圍定位服務的地理指紋社群系統

Abstract

基於位置資訊的服務已成為一個移動計算和無線數據服務的殺手級應用。各定位系統之中，對無線訊號進行樣本比對的定位演算法是最符合成本效益的和有用的。不過在大範圍的環境下，這些定位方式需要消耗人力的訓練過程和比較高的即時運算成本。我們採用一種新型以社群為基礎的系統設計來解決這些問題，並且發展過濾器和反饋的調節機制，以保證數據的可靠性。為運算速度和更精細的位置估算，我們使用了的兩個關鍵的大型算法這個制度。我們也為這個社群定位系統，設計了一個基於位置的服務架構。

Location-based services (LBSs) have emerged as one of the killer applications for mobile computing and wireless data services. Among all localization systems, the RF-based pattern matching schemes are the most cost effective and useful. However, these schemes may incur high calibration efforts and the high comparison cost in the large-scale environment. We adopt a novel community-based system design to solve the calibration problem and develop the confidence filter and feedback tuning mechanisms to guarantee the data reliability. For computation speed and finer location estimation,We apply two critical large-scale algorithms to this system. We also design a location-based service architecture for this community based fingerprinting real time localization system.