Real-Time Jamming Detection in Wireless IoT Networks

Abstract

IoT-based networks are vulnerable to jamming attacks due to their large-scale deployment and shared communication environment. Resource constraints and the low computational power of IoT devices make it harder to implement high-performance ML-based architectures for jamming detection. In this work, the effects of jamming attacks on a Wi-Fi network are presented and a novel real-time jamming detection mechanism is devised which can identify attacks on multiple channels in 2.4 GHz bandwidth simultaneously. The experiments are conducted in the lab environment by generating the jamming attacks with a Software Defined Radio. Certain QoS parameters in an end-to-end wireless IoT system are collected during normal operating conditions and during jamming attacks. The detection mechanism is implemented on IoT devices by employing the effects of jamming on wireless communication. The proposed real-time jamming detection method has an accuracy of 99% with zero false alarms. It benefits from the communication profile of a wireless network to detect jamming and requires minimal computational resources regarding memory and CPU usage which makes it a low-cost and easily deployable solution for IoT devices.