Enhanced Position Estimation Based on Hybrid Bluetooth and Pedestrian Dead Reckoning using Sensor Fusion Algorithm in Indoor Navigation

Complex sensor-network infrastructure in state-of-The-Art local positioning systems (LPS) helps to locate the position with enough accuracy and coverage. The framework of the proposed model presents a generic infrastructure as well as a detailed infrastructure for the proposed position estimation via hybrid Bluetooth and pedestrian dead reckoning using a sensor fusion algorithm in an indoor navigation system. We present a position estimation system based on combined trilateration localization and fingerprinting-based localization by using the BLE RSSI values model. An extended Kalman filter is used to fuse the data from different sensors to enhance the position accuracy. Furthermore, a Butterworth digital filter is used as a low-pass filter to remove the higher value of noise. Weighted K-nearest neighbor (KNN) is used to estimate the distance between BLE beacon devices and the tagging devices. For higher accuracy reference points mark at an angle of 45 degrees during the offline mode of the fingerprinting localization. The proposed hybrid localization technique combined with EKF has produced significantly better results as compared to the existing localization algorithm.