A Secure Edge-Cloud AI Framework Integrating Computer Vision, Signal Processing, and Iot Analytics for Intelligent Communication Systems

Abstract

The convergence of artificial intelligence, edge-cloud computing, and Internet of Things technologies provide new opportunities for developing responsive, secure, and resource-efficient intelligent communication systems. This study developed a secure edge-cloud AI framework integrating computer vision, signal processing, IoT analytics, adaptive workload allocation, and cybersecurity. The framework was evaluated using 1,200 multimodal observations across cloud-only, edge-only, and adaptive edge-cloud configurations. Statistical, correlation, security, and machine-learning analyses were performed to assess communication latency, throughput, energy consumption, quality of service (QoS), task success, and threat detection. Adaptive edge-cloud processing achieved the lowest mean latency (49.19 ms) and a 92.22% task-success rate while requiring substantially less energy than edge-only processing. Packet loss was negatively associated with QoS (ρ = −0.503), whereas signal quality showed a positive association (ρ = 0.489). The security mechanism achieved 93.42% detection accuracy and an F1-score of 86.77% across multiple cyberattack scenarios. Random Forest predicted QoS with an R² of 0.708, although task-success discrimination remained limited despite high classification accuracy. Overall, the findings demonstrate that security-aware adaptive edge-cloud intelligence can effectively balance latency, resource efficiency, communication quality, and cyber resilience, providing a scalable foundation for next-generation intelligent communication infrastructures and future AI-enabled network applications.