Abstract
Multi-tenancy is essential for scalable IoT–Cloud systems; however, it introduces complex security vulnerabilities at the intersection of shared cloud infrastructures and resource-constrained IoT environments. This systematic review evaluates next-generation security frameworks designed to enforce tenant isolation without violating the strict latency (<10 ms) and energy bounds of lightweight sensors. Adhering to PRISMA guidelines, we analyze selected high-quality studies to categorize intersectional threats, including cross-tenant data leakage, side-channel attacks, and privilege escalation. Our analysis identifies a critical, unresolved conflict: existing mitigation strategies often incur a 12% computational and communication overhead, creating a significant barrier for real-time applications. Furthermore, we critically analyze emerging technologies, including Zero Trust Architectures (ZTA), adaptive Artificial Intelligence (AI), blockchain, and Post-Quantum Cryptography (PQC). We find that direct PQC deployment is currently infeasible for LPWAN protocols due to key-size constraints (1.6 KB) that exceed typical payload limits. To address these challenges, we propose a novel multi-layer security design principle that offloads heavy isolation and cryptographic workloads to hardware-accelerated edge gateways, thereby maintaining tenant isolation without compromising real-time performance. Finally, this review serves as a roadmap for future research, highlighting federated learning and hardware enclaves as essential pathways for securing next-generation multi-tenant IoT ecosystems.