Software-Defined HoneyNet: Towards Mitigating Link Flooding Attacks
Jin-Woo Kim, Seungwon Shin · 2017
In this paper, we present SDHoneyNet, a SDN based system which exposes fake topology to attackers. SDHoneyNet finds potential bottleneck links by computing the static metric, which is Betweenness Centrality (BC) of the network topology given by the SDN controller. It also calculates the dynamic metric, which is Consumed Bandwidth Rate (CBR) of each link by collecting OF's port statistics. It finds a minimum intersection set containing the nodes that have both high static and dynamic metrics. Then, SDHoneyNet deploys the honey topology where the node degree follows a power-law distribution, to mimic complex networks. We leverage software switches to easily build and deploy the honey topology. We implement the prototype of SDHoneyNet as an SDN application (2,000 lines of Java code) on the ONOS controller.