An improved fingerprint method based on K-NN algorithm for indoor multiple object tracking

Global Positioning System (GPS) is one of famous technology which applied for outdoor localization. Unfortunately, GPS has lacked in accuracy while it uses in indoor environments. Fingerprint is a method enriched with lots of received signal strength. This method is used for mapping the observation area, and it is suitable for indoor application such as crowdsources tracking moving object in indoor environment. Fingerprint method saves lots of offline data which will be compared with online data captured by moving object. The nearest distance data will be selected as the estimated data. In this paper, we utilize a K-Nearest Neighbor (K-NN) algorithm in fingerprint method to improve the accuracy of moving object position. The simulation result shows that Mean Square Error (MSE) of proposed method is reduced up to 66% in 1.5 meter compared to the common fingerprint with 4.5 meter accuracy. Based on the aforementioned result, the proposed method could improve the accuracy of fingerprint localization method.