DOI: 10.23883/IJRTER.2017.3252.M8YED 475 Software Defined Networking based Intrusion Detection System Payal Kapre 1 , Riya Shreshthi 2 , Madhuri Kalgane 3 , Kalyani Shekatkar 4 ,Yogita Hande 5 1,2,3,4,5 Dept.Of Comp. Engg.Sinhgad Institutes Of Technology Narhe,Pune. Abstract:With the rapid expansion of computer networks during the past decade, security has become a crucial issue for computer systems. The goal of a security system is to protect the most valuable assets of an organization like data and information. This paper state the method of learning the Intrusion Detection, rules based on genetic algorithms. The implementation is experimented using data sets on intrusions, which has become the de facto standard for testing intrusion detection systems. The paper talks about using Genetic Algorithm (GA) in Network Security. Specially, it describes a technique of applying GA to network Intrusion Detection Systems (IDSs). GA is one of the commonly used approaches on data mining. It presents a brief overview of the Intrusion Detection System, GA, and related detection techniques. Our experiments detect different Denial of Service (DoS) attack & network scanning attacks such as Neptune attack etc. Keywords:– Software Defined Network (SDN), OpenFlow, Genetic Algorithm. I. INTRODUCTION Increasing access to the networks leads to further knowledge traffic and these demands needs a lot of numbers of repose connected servers for multiprocessing. Security is that the main issue in networking wherever existing security mechanism are deployed in an exceedingly static manner. Managing static security configurations in giant networks is incredibly time intense. software package outlined Networking (SDN) is supposed to handle the matter of static networks. SDN is associate approach to the pc networking that permits network administrator to manage all the network services providing abstraction of low-level practicality. SDN decouples management plane from knowledge plane. on top of things plane, selections regarding wherever to send the traffic from underlying systems are taken and therefore the knowledge plane that is liable for forwarding the information traffic to the destination. SDN is dynamic, manageable, convertible and cost-efficient network. SDN has programmable and centrally managed infrastructure. because the transmission of information over the net will increase, the requirement to shield connected systems additionally will increase. Intrusion Detection Systems (IDSs) are the most recent technology used for this purpose. though the sphere of IDS continues to be developing, the systems that do exist are still not complete, within the sense that they're not capable to notice all sorts of intrusions. Some attacks that are detected by numerous tools obtainable these days can not be detected by different merchandise, looking on the kinds and ways that they're designed on. employing a Genetic algorithmic rule (GA) is one among the ways that IDSs use to notice intrusions. . They incorporate the conception of Darwin’s theory and survival of the fittest to notice intrusions. Not a lot of analysis has been conducted during this space besides the Genetic algorithmic rule as another Tool for Security Audit Trails Analysis (GASSATA) tool; [16] [23] there are only a few IDSs that are fully developed from victimisation GAs. The genetic algorithmic rule is utilized to derive a group of classification rules from network audit knowledge, and therefore the support-confidence framework is used as fitness perform to guage the standard of every rule. The generated rules are then accustomed notice or classify network intrusions in an exceedingly time period setting. in contrast to most existing GA- based approaches, owing to the easy illustration of rules and therefore the effective fitness perform, the planned methodology is less complicated to implement whereas providing the flexibleness to