Open AccessProposed approach to detect distributed denial of service attacks in software defined network using machine learning algorithms
Author(s) -
Shruti Banerjee,
Partha Sarathi Chakraborty
Publication year - 2018
Publication title -
international journal of engineering and technology
Language(s) - English
Resource type - Journals
ISSN - 2227-524X
DOI - 10.14419/ijet.v7i2.8.10488
Subject(s) - computer science , software defined networking , forwarding plane , denial of service attack , controller (irrigation) , algorithm , networking hardware , virtualization , node (physics) , software , malware , computer network , distributed computing , network packet , operating system , cloud computing , the internet , structural engineering , agronomy , biology , engineering
SDN (Software Defined Network) is rapidly gaining importance of ‘programmable network’ infrastructure. The SDN architecture separates the Data plane (forwarding devices) and Control plane (controller of the SDN). This makes it easy to deploy new versions to the infrastructure and provides straightforward network virtualization. Distributed Denial-of-Service attack is a major cyber security threat to the SDN. It is equally vulnerable to both data plane and control plane. In this paper, machine learning algorithms such as Naïve Bayesian, KNN, K Means, K-Medoids, Linear Regression, use to classify the incoming traffic as usual or unusual. Above mentioned algorithms are measured using the two metrics: accuracy and detection rate. The best fit algorithm is applied to implement the signature IDS which forms the module 1 of the proposed IDS. Second Module uses open connections to state the exact node which is an attacker and to block that particular IP address by placing it in Access Control List (ACL), thus increasing the processing speed of SDN as a whole.