Effect of Different Signal Weighting Function of Magnetic Field Using KNN for Indoor Localization

The present work aimed to investigate the signal weighting function based on magnetic field (MF) models to obtain accurate location estimates for indoor positioning system. We compare the state-of-the-art Euclidean distance and three proposed different signal weighting function namely actual weight, square weight and square root weight which used to estimate location using MF. Additionally, the effect of signal weighting function is investigated further using multiple k value of K nearest neighbor (KNN) algorithm. According to the results, the square root weighting function have low position error of 8.156 m than Euclidean distance with improvement of 5.5%. We also found that the use of (k = 5) of KNN for square weight of my distance measure give the lowest mean estimation error of 7.188 m.