An adapted machine learning algorithm based-Fingerprints using RLS to improve indoor Wi-fi localization systems

Indoor localization has gained popularity in recent years. Various technologies have been proposed, but many of them do not give good accuracy without high-cost equipment. However, the Wi-Fi signal-based fingerprinting technique is widely employed for indoor locations because of its simplicity and low hardware requirements. Nevertheless, the Received Signal Strength Indicator (RSSI) values are affected by random fluctuations caused by fading and multi-path phenomena, resulting in decreased accuracy. In this paper, we propose indoor localization using Machine Learning (ML) algorithms such as Random Forest (RF), Extreme Gradient Boosting (XGBoost), K-Nearest Neighbors (KNN), and Support-Vector Machine (SVM) combine with a Recursive Least Squares (RLS) filter to increase the accuracy. The first method involves the use of ML algorithms to build an indoor positioning model. The second approach is to apply the RLS filter to reduce the noise in the data as the environment changes. The performance of these methods is evaluated through extensive real-time indoor experiments. We found that the proposed approach is an improvement over the state-of-the-art and recently published work.