Designing an Authentication Methodology in IoT Using Energy Consumption Patterns
Edi Marian Timofte, Alexandra Ligia Balan, Teodor Iftime · 2024
In this paper we are proposing an authentication method based on the energy usage patterns of connected devices. This approach marks a step up from traditional security protocols, using artificial intelligence and predictive analytics to identify the unique energy behaviors of each device. We implemented this system on Raspberry Pi 5, a small and affordable microcomputer, to show its efficiency and practicality in the real context of IoT applications. The solution we propose provides a new level of security by detecting subtle changes in energy usage that could mean suspicious activity or deviations from normal operation. The adaptability and scalability of this system makes it suitable for a wide range of IoT applications, from smart home devices to complex industrial systems. By closely studying energy data, our work highlights the importance and efficiency of integrating emerging technologies into IoT infrastructures for a safer and more resilient IoT environment.