The Internet of Things (IoT) and massive IoT systems are key to sixth-generation (6G) networks due to dense connectivity, ultra-reliability, low latency, and high throughput. Artificial intelligence, including deep learning and machine learning, offers solutions for optimizing and deploying cutting-edge technologies for future radio communications. However, these techniques are vulnerable to adversarial attacks, leading to degraded performance and erroneous predictions, outcomes unacceptable for ubiquitous networks. This survey extensively addresses adversarial attacks and defense methods in 6G network-assisted IoT systems. The theoretical background and up-to-date research on adversarial attacks and defenses are discussed. Furthermore, we provide Monte Carlo simulations to validate the effectiveness of adversarial attacks compared to jamming attacks. Additionally, we examine the vulnerability of 6G IoT systems by demonstrating attack strategies applicable to key technologies, including reconfigurable intelligent surfaces, massive multiple-input multiple-output (MIMO)/cell-free massive MIMO, satellites, the metaverse, and semantic communications. Finally, we outline the challenges and future developments associated with adversarial attacks and defenses in 6G IoT systems.

翻译：物联网（IoT）及海量物联网系统因具备密集连接、超高可靠性、低时延与高吞吐量等特性，成为第六代（6G）网络的关键组成部分。包括深度学习与机器学习在内的人工智能技术，为未来无线电通信中尖端技术的优化与部署提供了解决方案。然而，这些技术易受对抗攻击影响，导致性能下降和错误预测——这对泛在网络而言是不可接受的。本综述全面探讨了6G网络辅助物联网系统中的对抗攻击与防御方法，系统论述了对抗攻击及防御的理论背景与最新研究进展。此外，我们通过蒙特卡洛仿真验证了对抗攻击相较于干扰攻击的有效性。进一步地，通过演示适用于可重构智能表面、大规模多输入多输出（MIMO）/无小区大规模MIMO、卫星通信、元宇宙及语义通信等关键技术的攻击策略，深入分析了6G物联网系统的脆弱性。最后，我们总结了6G物联网系统中对抗攻击与防御面临的挑战及未来发展方向。