The indoor positioning algorithm research based on improved location fingerprinting

• Published in: The 27th Chinese Control and Decision Conference (2015 CCDC) pp. 5736 - 5739
• Main Authors: Xia Mingzhe, Chen Jiabin, Song Chunlei, Li Nan, Chen Kong
• Format: Conference Proceeding
• Language: English
• Published: IEEE 01.05.2015
• Subjects: Accuracy; Fingerprint recognition; improved KNN algorithm; Indoor positioning; location fingerprinting; mean smoothing algorithm; Smoothing methods; Wireless communication; Wireless LAN
• ISSN: 1948-9439; 1948-9447
• DOI: 10.1109/CCDC.2015.7161827

It is the key point of the final precise of positioning that whether the positioning fingerprint database created by location fingerprinting can accurately reflect the mapping relationship between the position and the fingerprints signal. In order to improve the accuracy of indoor positioning, the mean smoothing algorithm is used to process the collected data during the building of WLAN indoor fingerprint database rather than mean value. Eliminating the gross error is necessary before processing data with mean smoothing algorithm. Meanwhile, this paper proposes an improved KNN algorithm, which is to weigh the difference of the test point and the reference point, then choose the appropriate value of α. The algorithm is based on the constructing indoor wireless network with wireless routers and collecting the signal strength of the five wireless routers. Through the comparison with the accuracy of the commonly used indoor positioning algorithms, the results show that the positioning accuracy of the error distance within 3.6m can reach 90%, and within 4.8m can reach 97%.