FUZZY PROTECTION METHOD FOR FLOOD ATTACKS IN SOFTWARE DEFINED NETWORKING (SDN)

Abstract

Flood attacks (FA) are a type of distributed denial of service (DDoS) at-tacks.  In FA, an attacker sends massive floods of packets to consume all resources.  Hierarchical architecture and numerous weaknesses in the structure of communication protocols in conventional networks lead to the fact that firewalls are incapable to provide an integrated and effective mechanism against these attacks.  With the emergence of Software Defined Networking (SDN), there are new prospects for solving structural and security problems in conventional networks.  This study investigates some ideas for protecting against distributed FA using SDN.  Later, by analyzing the strengths and weaknesses of these ideas, a heterogeneous method is proposed based on a combination of conventional service provider and the Software Defined controller.  In the proposed method, the attack detection and the fuzzy decision modules are located in the service provider and the controller (i.e. SDN), respectively.  In order to simulate the heterogeneous method the MiniNet emulator is applied in combination with the Pox controller.  Afterwards, the simulated model is evaluated.  Results show that besides protecting against attacks in conventional networks, the proposed method provides other benefits including the extent of computational load and the response time in comparision to other Software Defined methods.