Adaptive data rate control in low power wide area networks for long range IoT services

Internet of Things (IoT) technologies can provide various intelligent services by collecting information from objects. To collect information, Wireless Sensor Networks (WSNs) are exploited. The Low Power Wide Area Network (LPWAN), one type of WSN, has been designed for long-range IoT services. It consumes low power and uses a low data rate for data transmission. The LPWAN includes several communication standards, and Long Range Wide Area Network (LoRaWAN) is the representative standard of the LPWAN. LoRaWAN provides several data rates for transmission and enables adaptive data rate control in order to maintain network connectivity. In the LoRaWAN, the wireless condition is considered by the reception status of the acknowledgement (ACK) message, and adaptive data rate control is performed according to the wireless condition. Because the judgment of the wireless condition by the reception status of ACK messages does not reflect congestion, adaptive data rate control can lead to inefficiency in data transmission. For efficient data transmission in long-range IoT services, this paper proposes a congestion classifier using logistic regression and modified adaptive data rate control. The proposed scheme controls the data rate according to the congestion estimation. Through extensive analysis, we show the proposed scheme's efficiency in data transmission.