Enhancing indoor positioning based on partitioning cascade machine learning models

This paper proposes the method, called Partitioning Machine Learning Classifier (PMLC), to enhance accuracy of fingerprinting indoor positioning by using machine learning algorithms. PMLC exploits limited information of the signal strength and combines a clustering task and a classification task. PMLC is compared with well-known machine learning classifiers, i.e. Decision Tree, Naive Bayes, and Artificial Neural Networks. The performance comparison is done in terms of accuracy of position classification and precision of distance classifier. The result of this study shows that PMLC can increase performance for indoor positioning of all classifiers when an appropriate number of clusters is assigned in the clustering process. In addition, PMLC is the most optimized model while having Decision Tree to be its classifier.