Sleep-Aware Intrusion Detection for Edge Devices: Balancing Power and Protection

Abstract

Because there are now many edge devices in IoT, their restricted power and resources have led to additional security challenges. Most traditional IDS have been designed to work on big networks at a central location, so they are inappropriate for smaller, energy-focused edge devices. This paper describes a new way to design sleep-aware intrusion detection for edge devices that addresses the conflict between power use and security. The system introduced here lowers energy waste while ensuring accurate detection using smart sleep and detection tools. The framework uses real-time threats and battery levels to determine which state the device should use. It is observed from testing that using the sleep-aware control, the IDS imposes a power reduction of up to 35% over continuous operation and continues to detect more than 92% of threats. Moreover, the adaptive thresholding process in the system significantly decreases the number of false alarms, ensuring the system is more effective in real life. The results add to the existing knowledge on making security solutions consume less energy in IoT and edge computing environments. These results demonstrate that planning for power is necessary to help devices last long and remain secure. Additional work will focus on using machine learning to create flexible sleep policies and adding hardware support to detect intrusions to improve performance. The framework forms a strong base for building IDS that are energy efficient and can deal with the strict demands of the next generation of edge computing.