An Innovative Weighted KNN Indoor Location Technology

Aiming at the problem of large fluctuation and low precision of the positioning method based on wireless fingerprint matching, we proposed an improved weighted K nearest neighbor algorithm and compared it with the commonly used machine learning algorithm. At the same time, we designed an innovative fingerprint database construction method and a new matching strategy. We used the particle filter algorithm to realize the fusion of the fingerprint matching localization algorithm and the pedestrian dead reckoning (PDR) algorithm, and eliminated the outliers, thus improving the positioning accuracy. The experimental results show that the average positioning accuracy after fusion is 0.512 m, and the positioning error within 1 m is 93.88%. It satisfies the accuracy requirements of indoor positioning and also verifies the effectiveness of the algorithm.