<![CDATA[The internet of things (IoT) necessitates efficient real-time data transfer protocols to support its vast array of interconnected devices. This study presents an optimized framework for resource allocation and link reliability in IoT–fog–cloud networks by integrating an enhanced support vector machine (ESVM) for link stability prediction with a Communication and Energy Integration for latency improvement (CAELI) algorithm for multi-objective optimization. The proposed system improves the quality of service (QoS) by dynamically selecting energy-efficient, low-latency paths while accounting for network conditions and resource constraints. The ESVM leverages historical link characteristics to assess reliability, whereas CAELI minimizes communication delay and energy usage through adaptive optimization. The simulation results indicate that the model achieves consistent improvements across metrics such as link reliability, end-to-end delay, energy consumption, throughput, and packet delivery ratio (PDR), maintaining a PDR above 94%, which is particularly significant in real-time control systems where even minor packet loss can compromise operational integrity. A comparative analysis with existing baseline and recent optimization approaches demonstrated superior performance in both static and moderately dynamic network environments. However, the model’s effectiveness may be influenced by factors such as network scale, node mobility, and the complexity of parameter tuning in CAELI, which can affect the convergence rate and computational efficiency. These limitations suggest the need for further validation in large-scale heterogeneous IoT deployments. The proposed framework underscores the viability of combining predictive modeling with multi-objective optimization to enhance responsiveness, energy efficiency, and reliability in distributed fog-assisted architectures for time-sensitive IoT applications.]]>
Copyrights © 2025
