Indoor positioning system using hybrid method of fingerprinting and pedestrian dead reckoning
Alvin Riady, Gede Putra Kusuma
2021Journal of King Saud University - Computer and Information Sciences21 citationsDOIOpen Access PDF
Abstract
Indoor Positioning System has been one of the most attractive research after Bluetooth Low Energy (BLE) was introduced. BLE technology is mainly used because of the reduction of material and energy cost over time that has huge impact compared to other technology which is more costly. In this research, we propose a new hybrid method to increase the accuracy of indoor positioning system using hybrid of BLE fingerprinting and PDR. Kalman Filter is used as benchmark of this experiment and ANN and SVR are proposed as a new method to combine both measurements. On benchmark, Kalman filter achieve positioning root-mean-squared error 212.21 cm and the proposed SVR 149.12 cm and ANN 111.78 cm.
Topics & Concepts
Benchmark (surveying)Kalman filterComputer scienceDead reckoningBluetoothIndoor positioning systemBluetooth Low EnergyReal-time computingMean squared errorPositioning systemReduction (mathematics)Artificial intelligenceGlobal Positioning SystemWirelessTelecommunicationsEngineeringAccelerometerMathematicsStatisticsGeographyOperating systemGeometryStructural engineeringNode (physics)GeodesyIndoor and Outdoor Localization TechnologiesBluetooth and Wireless Communication TechnologiesRadio Wave Propagation Studies