Deep Learning and Optimization Approaches in Secure AI for 6G Mobile Devices: Deep Kronecker Neural Network Optimized with Hybrid Cat Hunting Optimization to Combat Side-Channel Attacks: A Review

Abstract

The advancement of sixth-generation (6G) communication networks is expected to enable intelligent, ultra-low latency, and high-speed mobile systems powered by Artificial Intelligence (AI). However, the integration of AI into resource-constrained mobile devices introduces critical security challenges, particularly vulnerability to side-channel attacks (SCAs), which exploit physical leakages such as power consumption and electromagnetic emissions to extract sensitive information. Traditional cryptographic defenses are insufficient to mitigate these threats, necessitating advanced AI-driven security mechanisms.

This paper presents a comprehensive review of deep learning and optimization approaches for secure AI in 6G mobile devices, focusing on Deep Kronecker Neural Networks (DKNN) optimized using Hybrid Cat Hunting Optimization (HCHO). DKNN architectures leverage Kronecker product representations to reduce model complexity while maintaining high-dimensional learning capability. The integration of HCHO enhances parameter optimization, improving detection accuracy and convergence speed.

The study reviews recent developments from 2020–2023, highlighting the effectiveness of hybrid deep learning and metaheuristic optimization in combating SCAs. Results indicate that DKNN-HCHO models outperform conventional approaches in terms of accuracy, computational efficiency, and robustness. Future directions include lightweight model design, federated learning integration, and explainable AI for secure and scalable 6G systems.