Machine Learning Approach towards LoRaWAN Indoor Localization

Abstract

The growth of the Internet of Things (IoT) continues to be rapid, making it an essential part of information technology. As a result, IoT devices must be able to handle data collection, machine-to-machine (M2M) communication, and preprocessing of data, while also considering cost, processing power, and energy consumption. This paper introduces a system for device indoor localization that uses variations in the strength of the wireless signal. The proposed system addresses logistics use cases in which it is imperative to achieve reliable end-to-end delivery, such as pharmaceutic delivery, delivery of confidential documents and court exhibits, and even food, since the same is introduced into human organism and presents a potential risk of terrorist or other attack. This work proposes a concept based on low-power and low-cost LoRaWAN based system that utilizes a Machine Learning technique based on Neural Networks to achieve high accuracy in device indoor localization by measuring the signal strength of a beacon device. Furthermore, using signal strength measurements, that is, RSSI and SNR captured by LoRaWAN gateways, it is possible to estimate the location of the device point with an accuracy of up to 98.8%.