This project proposes a deep learning approach for forensic-based SDN in data centers. The SDN controller separates the data and control planes, implementing instructions instead of protocols. The project focuses on identifying and mitigating network threats using a network with mobile nodes, a head controller, a detection engine, and more. Attacks are classified as serious (requiring criminal evidence collection, IP blacklisting, and packet elimination) or not-serious (session-based blocking). Non-malicious packets are transmitted using a proposed TTL protocol. The approach is tested in a simulation environment, demonstrating improved results compared to existing methods. The project includes a comprehensive literature review, discussing the challenges of network forensics, the benefits of SDN, and the use of forensic tools and data provenance. The experimental setup details simulation parameters, and key metrics such as execution time, throughput, and packet loss are analyzed, with graphical representations and formulas provided. The project aims to enhance network security through AI-driven forensic analysis.
![[object Object]](/_next/static/media/star-bottom.7253800d.svg)
