International Journal of Advanced Research in Computer and Communication Engineering

Abstract: As cyber threats rapidly evolve in complexity, Gen- erative Adversarial Networks (GANs) have emerged as trans- formative tools in network security, serving both as formidable defenses and novel avenues for attack. This survey introduces a comprehensive taxonomy of GAN applications in network secu- rity, classifying contemporary research across critical domains such as adversarial sample generation, intrusion detection, syn- thetic traffic modeling, federated security architectures, IoT and edge protection, and encrypted traffic analysis. By systematically mapping these domains, the paper illustrates how diverse GAN variants are leveraged for simulating threats, resolving class imbalance, and circumventing conventional detection strategies. The taxonomy uncovers key trends in the development and deployment of GAN-driven security models, providing a robust framework for assessing progress and identifying persistent challenges. The review concludes by outlining emerging research directions rooted in the taxonomy, and calls for standardized benchmarks and ethical guidelines to support secure, scalable integration of GANs into modern network defenses.
Index Terms: GANs, Network Security, Intrusion detection, Generative AI, Cybersecurity, Adversarial attacks.

Downloads: | DOI: 10.17148/IJARCCE.2025.141113