Master’s Thesis Defense at the College of Information Technology on Enhancing Fog Computing Security Using Deep Learning

Duha Fadill Abbas
The Department of Information Networks at the College of Information Technology, University of Babylon, witnessed the defense of the Master’s thesis submitted by Ali Al-Ridha Khalil Ibrahim, entitled “Enhancing Fog Computing Security Using Deep Learning,” under the supervision of Prof. Dr. Mahdi Abadi Manea. The defense was conducted in a rigorous academic atmosphere that reflected the advanced scientific level of the research and its importance at the intersection of cybersecurity, Internet of Things technologies, fog computing architectures, and modern artificial intelligence approaches.

The thesis highlighted fog computing as a promising paradigm for supporting large-scale and latency-sensitive Internet of Things (IoT) applications by extending computational and storage capabilities toward edge devices. The study demonstrated that fog-based IoT environments remain vulnerable to multiple security threats due to limited resources, decentralized architectures, and the increasing complexity of cyberattacks, particularly in light of the emerging challenges of the post-quantum computing era. These challenges emphasize the need for more intelligent and scalable security mechanisms compared with traditional approaches.

The researcher proposed a lightweight multi-stage security framework integrating secure authentication, post-quantum data encryption, and intelligent intrusion detection mechanisms. In the first stage, an enhanced authentication approach was introduced through the development of the ECDH protocol using the Lorenz chaotic system to improve key randomness and resistance against various attacks while maintaining low computational overhead. The results obtained from NIST SP 800-22 statistical tests confirmed a high level of statistical randomness in the generated keys.

The study concluded that the proposed framework achieves an effective balance between efficiency and security by providing strong cryptographic randomness, efficient performance resistant to post-quantum threats, and highly accurate intrusion detection while maintaining low computational overhead. These findings indicate that the framework represents a promising and scalable security solution suitable for deployment in resource-constrained fog-based Internet of Things environments.