Bio-inspired Decentralized Rogue Node Detection in Fair Dynamic Spectrum Access Networks

Abstract

The rapid growth of wireless devices as societies adapt to the Internet of Everything (IoE) has led to saturation of spectrum resources. Dynamic spectrum access has been considered a promising solution to alleviate congested channels by allowing unlicensed users to access licensed channels when the licensed users are idle. Various coexistence challenges arise as unlicensed users compete over a limited amount of channel resources. In this article, we build on a previously defined bio-social inspired dynamic spectrum access coexistence scheme where unlicensed users achieve fair sharing of resources by choosing to defer to nodes with more urgent transmission needs. To prevent selfish nodes from taking advantage of the deference mechanism, we propose a decentralized rogue node detection behavioral model. While foraging for resources, each node performs rogue node detection using hardware fingerprinting. We show that we can achieve 99% rogue node detection accuracy with fast detection convergence time and low communication/coordination overhead.