A Survey of Advanced Intrusion Detection Systems Using Deep Learning in Cloud-Edge IoT Environments

The Internet of Things (IoT), Cloud Computing (CC), and Edge Computing (EC) represent an important shift from conventional structures on virtualized resource provisioning, introducing better adaptability and transparency for host systems, allowing cloud providers, on-premises sources, and edge nodes to fully realize the "everything-as-a-service" notion. This survey provides a comprehensive review of emerging cybersecurity threats in Cloud-Edge IoT systems and systematically examines AI-driven deep learning approaches for Intrusion Detection Systems (IDSs). The foundational concepts are first presented, including an overview of cybersecurity principles, cloud computing architecture, edge computing paradigms, and the advantages and challenges of IoT integration with cloud and edge infrastructures. Then, it further explores the role of ML and DL models in IDSs, highlighting architectures. A comparative analysis of recent studies is conducted based on datasets, performance metrics, scalability, computational efficiency, and robustness. Key findings are synthesized, identifying research gaps and evaluating practical deployment constraints. Finally, the survey outlines critical challenges and future research directions to support the development of secure, intelligent, and resilient Cloud-Edge IoT ecosystems.