Print Email Facebook Twitter LoRa Localisation in Cities with Neural Networks Title LoRa Localisation in Cities with Neural Networks Author Nguyen, Tuan Anh (TU Delft Electrical Engineering, Mathematics and Computer Science) Contributor Zuniga, Marco (mentor) Kuipers, F.A. (graduation committee) Hauff, C. (graduation committee) Degree granting institution Delft University of Technology Programme Electrical Engineering | Embedded Systems Date 2019-10-03 Abstract Billions of wireless devices are interconnected to provide services to manyaspects of life and form The Internet of Things. These devices which areoften battery-powered and energy efficient can benefit greatly from an ac-curate localisation service that does not consume extra energy. Several loc-alisation methods have been developed for Low-Power Wide-Area Networks(LPWANs), with LoRa being of particular interest thanks to its long rangeand cost effectiveness. Time Difference of Arrival (TDoA) is a common wayto find location in a LoRa network which works well in open areas but poorlyin the harsh radio environment of cities. In indoor settings where the radioenvironment is more complicated than outdoor, RSSI fingerprinting tech-niques have been sucessfully used for positioning using WiFi and Bluetooth,with state-of-the-art solutions employing Artificial Neural Networks (ANN).This work aims to provide accurate localisation in an urban LoRa network,using an ANN-based fingerprinting approach. Two publicly available datasets collected in the cities of Utrecht and Antwerp are used to evaluate ourmethod. We show that the ANN model can be trained on these data sets topredict location with mean errors between 411m and 581m. We determinethat the presence of gateways in the fingerprint plays a major role in theANN’s estimation but RSSI information is crucial in improving the accuracy.To realistically compare the ANN approach to TDoA, we train and test theneural network with chronologically split data. Our ANN approach achievesa mean error of 500m with 90% of cases having errors below 1070m. ThisRSSI fingerprinting method is more effective than TDoA at limiting largelocalisation errors in cities. Subject Internet of Things (IoT)localisationRSSI Fingerprintingartificial neural networks To reference this document use: http://resolver.tudelft.nl/uuid:01129207-47ee-47ba-bb05-4417c79ccfb9 Part of collection Student theses Document type master thesis Rights © 2019 Tuan Anh Nguyen Files PDF thesis_final_Tank.pdf 3.59 MB Close viewer /islandora/object/uuid:01129207-47ee-47ba-bb05-4417c79ccfb9/datastream/OBJ/view