Localizing Wireless Jamming Attacks with Minimal Network Resources

Abstract

Wireless networks such as the wireless sensor networks (WSNs) are increasingly becoming ubiquitous in today’s Internet of Things (IoT) environment. However, wireless networks are very vulnerable to jamming attacks which can be easily carried out by small concealable wireless transmitters. Thus, there is a need to quickly localize and remove the jammer node to prevent further disruptions to the network. In addition, an efficient jammer node localization algorithm should use minimal network resources, e.g., number of reporting nodes. Based on this idea, we propose a maximum likelihood estimator (MLE)-based localization method for the jammer node which allows each node that detected a jammed link to probabilistically transmit a jamming report. The node transmission probabilities are optimized using the cross-entropy (CE) method to ensure that the localization algorithm satisfies a pre-defined reliability threshold. Our simulation results show that the proposed scheme performs significantly better than the simple centroid-based localization methods while using lesser number of reporting nodes and traffic.