An Improved Weighted K-Nearest Neighbor Algorithm for Indoor Positioning

摘要

The weighted K-nearest neighbor algorithm (WKNN) is widely used in indoor positioning based on Wi-Fi. However, the accuracy of this traditional algorithm using Euclidean distance is not high enough due to the ignorance of statistical regularities from the training set. In this paper, the Manhattan distance is introduced to the WKNN algorithm to distinguish the influence of different reference nodes. Simultaneously, a new method is proposed to increase the accuracy of the algorithm by adjusting the weight of adjacent reference nodes. The simulation and experiment results show that the improved algorithm can have a better performance by increasing the accuracy by 33.82%.