SECURING SOFTWARE-DEFINED NETWORKS THROUGH DOS ATTACK TYPE CLASSIFICATION AND MITIGATION FRAMEWORK USING LRS2BTM

Authors: Amaresan Venkatesan

DOI: https://doi.org/10.5281/zenodo.14005251

Short DOI: https://doi.org/g8pmdq

Country: USA

Full-text Research PDF File:   View   |   Download

Abstract: Software-Defined Networking (SDN) optimizes network management by detecting and mitigating Denial of Service (DoS) attacks. Traditional techniques fail to address specific DoS attack types like DOS Hulk, Benign, DOS Slow HTTP Test, DOS Slowloris, and DOS GoldenEye, leading to ineffective mitigation and network vulnerabilities. To classify DoS attack types and enhance network security, Lasso-Resilience-Ridge Swishmax-based Bidirectional Long-Short Term Memory (LRS2BTM) and Chaos-Shannon-based Elliptical Curve Cryptography (CS-ECC) are used. Initially, the user registration and key generation are done, followed by logging in to the DoS attack detection phase. In this phase, the attacks are categorized using LRS2BTM. From the categorized attacks, DOS Hulk and GoldenEye attacks due to high traffic rates are mitigated using FIFO-TBA; also, DOS Slow HTTP Test and Slowloris attacks are blocked as they cause service disruption. Non-attacked (benign) and mitigated data are then secured with an Encryption Time (ET) of 2571ms. Thus, the proposed work outperformed the existing methodologies in enhancing network security and efficiency.

Keywords: First In First Out based Token Bucket Algorithm (FIFO-TBA), Diversity-Preserving Perturbation based Secretary Bird Optimization Algorithm (DP2-SBOA), Elliptical Curve Cryptography (ECC), Bidirectional Long Short-Term Memory (BiLSTM), Software Defined-Networking (SDN), Data Security, Attack Detection (AD).

Paper Id: 231388

Published On: 2023-12-05

Published In: Volume 11, Issue 6, November-December 2023

Share this