A New Model-Based Method to Detect Radio Jamming Attack to Wireless Networks

Wireless communications are subjected to radio jamming attacks due to the fact that the transmission medium is open to any hostile attackers. In a radio jamming, an attacker can jam a channel by emitting a radio signal in certain strength without understanding the protocol. It may also try to disrupt the RF operations of a communication protocol if it understands the protocol. Detection of the jamming has been very challenging, particularly in adversarial environment. Current detection mechanisms using signal strength related metrics rely on extensive measurements that require hardware implementation due to lack of appropriate analytical models. In this work, we propose a novel detection method based on the decomposition relationship between effective channel utilization (ECU) and the interference power from undesirable nodes. This method detects power-based jamming attacks against wireless networks. A new metric using ECU is integrated into the algorithm of the detection of radio jamming. In addition, the method is implemented in OPNET with a dynamic radio channel allocation (RCA) module that monitors and reallocates the channel if being detected as jammed. We have thoroughly demonstrated the effectiveness of the proposed method by extensive experiment and simulation results.