Adversarial-Aware Deep Learning System Based on a Secondary Classical Machine Learning Verification Approach.

Deep learning models have been used in creating various effective image classification applications. However, they are vulnerable to adversarial attacks that seek to misguide the models into predicting incorrect classes. Our study of major adversarial attack models shows that they all specifically target and exploit the neural networking structures in their designs. This understanding led us to develop a hypothesis that most classical machine learning models, such as random forest (RF), are immune to adversarial attack models because they do not rely on neural network design at all. Our experimental study of classical machine learning models against popular adversarial attacks supports this hypothesis. Based on this hypothesis, we propose a new adversarial-aware deep learning system by using a classical machine learning model as the secondary verification system to complement the primary deep learning model in image classification. Although the secondary classical machine learning model has less accurate output, it is only used for verification purposes, which does not impact the output accuracy of the primary deep learning model, and, at the same time, can effectively detect an adversarial attack when a clear mismatch occurs. Our experiments based on the CIFAR-100 dataset show that our proposed approach outperforms current state-of-the-art adversarial defense systems.

Multi-Layer Attack Graph Analysis in the 5G Edge Network Using a Dynamic Hexagonal Fuzzy Method.

Overall, 5G networks are expected to become the backbone of many critical IT applications. With 5G, new tech advancements and innovation are expected; 5G currently operates on software-defined networking. This enables 5G to implement network slicing to meet the unique requirements of every application. As a result, 5G is more flexible and scalable than 4G LTE and previous generations. To avoid the growing risks of hacking, 5G cybersecurity needs some significant improvements. Some security concerns involve the network itself, while others focus on the devices connected to 5G. Both aspects present a risk to consumers, governments, and businesses alike. There is currently no real-time vulnerability assessment framework that specifically addresses 5G Edge networks, with regard to their real-time scalability and dynamic nature. This paper studies the vulnerability assessment in the 5G networks and develops an optimized dynamic method that integrates the Technique for Order of Preference by Similarity to Ideal Solution (TOPSIS) with the hexagonal fuzzy numbers to accurately analyze the vulnerabilities in 5G networks. The proposed method considers both the vulnerability and 5G network dynamic factors such as latency and accessibility to find the potential attack graph paths where the attack might propagate in the network and quantifies the attack cost and security level of the network. We test and validate the proposed method using our 5G testbed and we compare the optimized method to the classical TOPSIS and the known vulnerability scanner tool, Nessus.