Building Machine Learning Based Firewall on Spanning Tree Protocol over Software Defined Networking
Software-Defined Networking (SDN) is the most expected inventions for the network administrators, cloud service providers and businessmen. For its global acceptance, has to undergo security issues and get cover using a proper solution. This study implemen
- PDF / 393,941 Bytes
- 12 Pages / 439.37 x 666.142 pts Page_size
- 57 Downloads / 197 Views
Abstract Software-Defined Networking (SDN) is the most expected inventions for the network administrators, cloud service providers and businessmen. For its global acceptance, has to undergo security issues and get cover using a proper solution. This study implements the firewall on the Spanning Tree Protocol (STP) over SDN using Mininet emulator. The broadcast streams are controlled by the Firewall on STP based SDN without dedicated hardware support. The result focuses the Firewall functionalities over STP based SDN as well as frees the available bandwidth for traffic. Additionally, Machine Learning (ML) techniques have been applied to build up the infrastructure to get the best outcome from a model with top-notch performance. Among different machine learning models using the same NSL-KDD dataset, Random Forest Classifier algorithm stands at the top with highest accuracy in differentiating between regular communication and intrusion on the network. Keywords Software defined networking (SDN) · Spanning tree protocol · Firewall · OpenFlow · Machine learning
N. Islam (B) · S. M. Shamim · Md. Fazla Rabbi Department of Information and Communication Technology, Mawlana Bhashani Science and Technology University, Santosh,Tangail 1902, Bangladesh e-mail: [email protected] S. M. Shamim e-mail: [email protected] Md. Fazla Rabbi e-mail: [email protected] Md. Saikat Islam Khan Department of Computer Science and Engineering, Mawlana Bhashani Science and Technology University, Santosh, Tangail 1902, Bangladesh e-mail: [email protected] M. Abu Yousuf Institute of Information Technology, Jahangirnagar University, Savar, Dhaka 1342, Bangladesh e-mail: [email protected] © The Author(s), under exclusive license to Springer Nature Singapore Pte Ltd. 2021 M. S. Kaiser et al. (eds.), Proceedings of International Conference on Trends in Computational and Cognitive Engineering, Advances in Intelligent Systems and Computing 1309, https://doi.org/10.1007/978-981-33-4673-4_45
557
558
N. Islam et al.
1 Introduction Software Defined Networking (SDN) replaces current network paradigm through the abstraction of central programming interface. SDN is flexible enough to let the network engineers cope up with the changing requirements. Sometimes, for adding new hardware or new program administrator requires numerous individual device configurations and even for adding QoS or security policies. As the switches and routers are from multiple vendors, each vendor has different Operating System (OS) with different configuration methods for their own switches and routers. Despite their comprehensive acceptance, network administrator faces difficulty controlling the traditional IP networks [1]. Use of machine learning in firewall helps to analyze the traffic flow, possible attacks and possibly helps to improve the QoS. Different approaches of Machine learning points to how effective and accurate, these can be when filtering the traffic of the network [2]. Moreover, Internet security and network security are on the rise as quic
Data Loading...
