Novel fingerprinting mechanisms for indoor positioning

Summary As wireless communications and microelectronic technology rapidly develop, diverse applications and services based on smart handheld devices have drawn the attention of researchers. The popularity of Indoor Location Based services and applications has also gradually increased. Therefore, how to improve indoor positioning accuracy becomes a very important issue. Although indoor positioning has been performed using various techniques in recent years, the computational complexity of ensuring positioning accuracy and positioning is an unsolved problem. Current indoor positioning systems typically use only the receiver or the transmitter to obtain the reference point data, and only the K‐Nearest Neighbors (KNN) or Trilateration algorithm is used to perform positioning. Therefore, positioning accuracy is limited by the use of reference point data from a single source and by the positioning algorithm used. The Novel Fingerprinting Mechanisms (NFM) indoor positioning system proposed in this study, however, uses both the receiver and transmitter to obtain positioning data and employs six positioning mechanisms to improve the current positioning accuracy. The experimental results show that the average error distance is 1.18 m in the NFM indoor positioning system. That is the system outperforms both KNN and Trilateration systems, which have average error distances of 1.35 m and 2.23 m, respectively. This study proves that the positioning accuracy is actually improved. Copyright © 2015 John Wiley & Sons, Ltd.