The study of machine Learning scenarios for the Internet of arctic things

• Published in: 2022 Moscow Workshop on Electronic and Networking Technologies (MWENT)
• Main Authors: Rolich, Alexey, Alexander, Ilyin, Voskov, Leonid
• Format: Conference Object
• Language: English
• Published: Institute of Electrical and Electronics Engineers Inc. 2022
• Subjects: sensor network; terminal data collection devices; energy efficiency; simultaneously connected devices; machine learning methods; Internet of Arctic Things; LoRaWAN data transmission network; IoT devices; Arctic
• Online Access: https://hdl.handle.net/11573/1672874; https://doi.org/10.1109/MWENT55238.2022.9802182

The paper investigated the problem of using IoT data transmission technologies in the absence or underdeveloped network infrastructure. As a result of a study of the technologies used in the IoT for data transmission, the LoRaWAN data transmission network was selected. A model of the functioning of IoT devices of a sensor network and a method for increasing the efficiency of data transmission using machine learning methods on terminal data collection devices to reduce the amount of transmitted data and increase the energy efficiency of systems are proposed. The proposed method was evaluated. The proposed method of using machine learning methods significantly increases the lifetime of terminal devices with certain strategies for collecting and processing data. The method allows to increase the maximum number of simultaneously connected devices, by reducing the use of the radio channel, since only processed information is sent. Processing data on edge devices using machine learning methods increases the autonomy of the IoT system, thereby increasing its reliability and providing increased data protection. With the development of computing systems, the use of machine learning on terminal devices will become more widespread.