Detecting and identifying radio jamming attacks in low-power wireless sensor networks

Abstract: Wireless sensor networks (WSNs) are used in all kinds of different sectors ranging from agriculture, environment, healthcare and the military. Embedded systems such as sensor nodes are low-power and consist of low memory, which creates a challenge for its security. One of WSN’s worst enemies is interference radio jamming attacks. They are easy to construct and execute, but hard to detect and identify. In this thesis, we tackle the problems of detecting, but most importantly identifying, and distinguishing the most commonly used interference radio jamming attacks. Presenting SpeckSense++, a firmware that makes it possible for low-power embedded systems to detect, identify and distinguish interference radio jamming attacks and unintentional interference such as Bluetooth and WiFi to a certain degree. Showing an accuracy of 96 to 90 % for proactive jammers, 89% for reactive, and 92 to 85 % for unintentional interference.