Abstract
This article discusses the accuracy performance of the Weighted K Nearest Neighbors (WKNN) algorithm which has been widely used by the Received Signal Strength (RSS) fingerprint-based localization in WLAN networks. The basic idea of WKNN algorithm is to estimate the people’s position at the weighted geometric center of the K nearest neighbors. However, the placement of Reference Points (RPs) has been recognized as one of the most significant factors to effect the localization accuracy. Therefore, in this paper, we carefully analyze the localization accuracy of WKNN algorithm and illustrate the analytical relations among localization accuracy, physical interval of RPs and environmental dimensions, which provides a guideline to the design of the WLAN fingerprint-based localization systems. Furthermore, the analytical relations obtained from this paper are verified by a series of experiments conducted in the real WLAN networks.
Sign-in/Register to access full text options 
Free) 
Talk to us
Join us for a 30 min session where you can share your feedback and ask us any queries you have
Schedule a call
