Joint optimization secure and energy-efficient computation offloading framework IoT-enabled edge networks

Area of Science:

• Computer Science
• Network Engineering
• Artificial Intelligence

Background:

• Smart cities rely on real-time systems integrating Internet of Things (IoT), edge computing, and emerging technologies.
• Existing systems face challenges in security, unpredictable communication, load balancing, and device heterogeneity.
• Optimal offloading in edge-driven schemes is complicated by unreliable links and device diversity.

Purpose of the Study:

• To introduce a joint optimization framework for energy-efficient, secured computation offloading and dynamic resource allocation in Edge-IoT infrastructure.
• To address challenges of intelligence, security, load balancing, and optimal offloading in dynamic Edge-IoT environments.
• To enhance network stability and improve workload management in distributed IoT ecosystems.

Main Methods:

• Utilizes a Multi-Agent Reinforcement Learning approach for adaptive offloading decisions.
• Establishes secure communication paths and effective resource allocation under dynamic network interactions.
• Leverages trusted, decentralized processing points at the network edge for enhanced stability.

Main Results:

• The proposed framework demonstrates substantial improvements in security and energy efficiency.
• Achieves enhanced network stability by mitigating energy holes and improving workload management.
• Outperforms recent state-of-the-art solutions in simulations for Edge-IoT infrastructure.

Conclusions:

• The joint optimization framework effectively addresses key challenges in Edge-IoT systems.
• The Multi-Agent Reinforcement Learning approach enables adaptive and secure computation offloading.
• The research contributes to more stable, secure, and energy-efficient smart city infrastructures.