Explainable indoor localization of BLE devices through RSSI using recursive continuous wavelet transformation and XGBoost classifier

Indoor localization systems with higher precision and integrity are being highly demanded because of their numerous applications in superstores, smart homes, smart cities, elderly care, and disaster management. Although there are many technologies for indoor positioning e.g., Wireless Fidelity (Wi-Fi), Bluetooth Low Energy (BLE), and Radio-Frequency Identification (RFID), the high precision localization is still challenging because of the multipath effect and non-line of sight propagation of radio waves in complex indoor environment. This research proposes an explainable indoor localization (EIL) method for higher precision and integrity in IPS. The proposed localization method considered received signal strength indicator (RSSI) from BLE devices for predicting their precise locations. A recursive continuous Wavelet transform (R-CWT) method is proposed to extract discriminative features from the beacon signals for efficient localization. The extracted features are then fed to the extreme gradient boosting machine for the accurate classification of indoor positions. Moreover, to ensure integrity in indoor position classification, the Shapley additive explanations (SHAP) method is introduced to interpret the results obtained from the gradient boosting machine. The proposed method (EIL) precisely localize BLE devices within 1.5 m in a superstore environment with an accuracy of 98.04% which is much higher than the reported accuracies in existing studies.