Experimental Study of a LoRa Based Wireless Sensor Network for Low Power Localization in an Outdoor Environment

The Internet of Things (IoT) and Wireless Sensor Network (WSN) have transformed industries by highlighting their crucial roles in sensing, communication, and data-driven decisions. One paramount feature in WSN is localization, enabling precise positioning of devices in real-world applications. Global Positioning Systems (GPS) have long been the mainstay of high-precision localization, but their high energy consumption is a significant obstacle to their seamless integration with low-power devices. In the quest for suitable solutions, Low Power Wide Area Networks (LPWAN) have emerged as key players, providing energy-efficient connectivity. However, these advances raise questions about the localization accuracy within these networks. In this paper, we present the implementation of WSN based on Long Range (LoRa) technology to estimate the position of an object in an outdoor environment. We took advantage of LoRa technology to collect Received Signal Strength Indicator (RSSI) measurements over a long distance. An Artificial Neural Network (ANN) is then trained on a collected dataset to estimate the 2-dimensional position of the object. The results show an estimate of the position with an average absolute error of 40.5 meters in an area of 360 meters by 360 meters. A web interface is made with Node-Red to monitor the position of the mobile object.