無線射頻辨識技術(RFID)可彌補GPS無法應用於室內建築的缺點,本研究即係利用主動式RFID設備,藉由接收不同訊號強度來決定標籤點與讀取器之間的距離,透過標籤點已知坐標之相對距離建構量測模型,並將RFID 標準訊號強度與理想值進行比對,結合迴歸演算法進行訊號修正以提升主動式RFID室內定位的準確度。 本研究之測試結果發現,利用迴歸演算法進行訊號值修正,最佳之迴歸修正為三次迴歸方程式,可將距離估測誤差由修正前之3.28 m降低為0.49 m;另於測試點上進行定位精度檢核可知,在95%機率下,透過三次迴歸方程式修正後之估測距離,其平面定位精度則為1.06 m。
Radio-frequency identification (RFID) is a technology that uses communication through the use of radio waves to transfer data between a reader and an electronic tag attached to an object for the purpose of identification and tracking, particularly in an indoor environment. This research applys active RFID to receive signal and measure the distance from tag to the reader.The range estimated from the signal strength received is modeling and corrected by using the regression models. The experimental results showed that the best regression function is a cubic curve regression model, in which the range estimates are effectively reducted from 3.28 m to 0.49m. The positioning accuracy made at the test sites revealed that the 2D coordinate accuracy is good to be 1.06 m under the probability of 95% with the corrected ranges.