Research on Indoor Position Fingerprint Location Based on Machine Learning combined Particle Filter

With the rapid development of wireless networks, wireless positioning technology has also rapidly gained popularity. At the same time, more and more indoor positioning scenes have also received widespread attention. Therefore, high-precision indoor positioning technology is particularly important. This paper introduces the model of machine learning into indoor location fingerprint positioning. Firstly, the K-Nearest Neighbor (KNN) algorithm is used to study the accuracy of wireless fingerprint location, and compare it with the performance of Support Vector Machine (SVM), Random Forest (RF), and Multi-layer Perceptron (MLP)algorithm in fingerprint positioning system. Finally, combined Particle Filter (PF) algorithm based on machine learning algorithm, and experimental results show that the particle filter algorithm improves the positioning accuracy, and the random forest has the highest positioning accuracy.