Rule Based Intrusion Detection System By Using Statistical Flow Analysis Technique For Software Defined Networking

Abstract

From the fact of this report our project for intrusion detection in software-defined networking using a technique, rule-based statistics has been illustrated and explained thoroughly. Need for this solution was a major security problem in present networking era. Transferring data and managing it pro grammatically in networks also require confidential, authenticated and reliable communication. Many hardware devices are useful in dealing with these threats towards the security of networks but instead using these costly solutions application program is introduced in our project. The basic methodology that is being implemented in our project is rule-based st.at.istical flow analysis. By this technique, the malware and normal packets are differentiated. Data packets are obtained and studied to know their features that are then feed to the application that uses rules to categorize the traffic as malicious and innocent. These characteristics are also drawn into the evaluation function to evaluate the performance of the application and to make a comparison with other applications based on techniques like machine learning, data mining and genetic algorithm. The program is implemented and installed in SD N RY U controller that uses open flow protocol for navigating the traffic across the networks that are neither costly nor it needs to be replaced to achieve maintenance and durability. This application brings a wide enhancement in programmable networks