Skip to Main Content
 

Global Search Box

 
 
 
 

Files

File List


Nick_Perry_Thesis.pdf (1.4 MB)

ETD Abstract Container

Abstract Header

Neural Network-Based Crossfire Attack Detection in SDN-Enabled Cellular Networks

Perry, Nicholas
http://orcid.org/0000-0001-6873-2599
http://rave.ohiolink.edu/etdc/view?acc_num=miami1689073041698329

Abstract Details

2023, Master of Science, Miami University, Computer Science and Software Engineering.
In today's networked world, cybersecurity threats pose a significant challenge to the integrity and reliability of communication networks. One such threat is the crossfire attack, where adversaries exploit network vulnerabilities by injecting malicious packets into traffic flows. To address this, we present a novel crossfire detection scheme that solely inspects packet headers, reducing the computational overhead associated with packet inspection. Our proposed detection scheme includes both analysis of variance (ANOVA) and neural networks to identify anomalous packet behaviors indicative of crossfire attacks. To evaluate the effectiveness of our approach, we conducted experiments on a real ATT backbone topology, simulating a crossfire attack in the Mininet simulation environment. The results demonstrate that our detection scheme achieves an accuracy of 95.3\% in detecting adversarial packets, effectively mitigating the crossfire threat. Furthermore, we introduce a traffic optimization model to adapt routing decisions in response to crossfire or link flooding attacks. Leveraging the detection scheme's real-time analysis, our optimization model dynamically alters routing paths to minimize the impact of attacks on network performance. Overall, our research presents an innovative and comprehensive framework that combines efficient crossfire detection using packet headers, high-accuracy detection using ANOVA and neural networks, and an adaptive traffic optimization model.
Suman Bhunia (Advisor)
Daniela Inclezan (Committee Member)
Vaskar Raychoudhary (Committee Member)
69 p.
Computer Science
crossfire attack; DoS; cybersecurity; SDN; mininet; ryu; open flow; networking; attack;

Recommended Citations

Citations

  • Perry, N. (2023). Neural Network-Based Crossfire Attack Detection in SDN-Enabled Cellular Networks [Master's thesis, Miami University]. OhioLINK Electronic Theses and Dissertations Center. http://rave.ohiolink.edu/etdc/view?acc_num=miami1689073041698329

    APA Style (7th edition)

  • Perry, Nicholas. Neural Network-Based Crossfire Attack Detection in SDN-Enabled Cellular Networks. 2023. Miami University, Master's thesis. OhioLINK Electronic Theses and Dissertations Center, http://rave.ohiolink.edu/etdc/view?acc_num=miami1689073041698329.

    MLA Style (8th edition)

  • Perry, Nicholas. "Neural Network-Based Crossfire Attack Detection in SDN-Enabled Cellular Networks." Master's thesis, Miami University, 2023. http://rave.ohiolink.edu/etdc/view?acc_num=miami1689073041698329

    Chicago Manual of Style (17th edition)

miami1689073041698329
47
© 2023, all rights reserved.
This open access ETD is published by Miami University and OhioLINK.