MDP and deep reinforcement learning for optimized task off-loading in blockchain-enabled edge-cloud systems

The development of IoT and edge-cloud systems, task offloading is becoming more and more essential for devices with limited resources. The use of blockchain improves security. An optimization-oriented offloading strategy with a hybrid MDP and DRL approach is suggested in this research.MDP and DRL for task offloading decisions Increase task-processing adaptability and efficiency. Combine blockchain to secure edge-cloud systemsWe used MDP combined with DRL techniques for intelligent task offloading in blockchain-based edge-cloud systems to increase the performance of computation efficiency and utilization of resources.This proposed approach is more accurate, has better precision, and recall than existing approaches. Relative mean error (RME), due to variations in change points, also demonstrates a significant decrease at 3-RRN, highlighting improved performance and adaptability against varying conditionsThis method is also suggested to provide more accuracy, precision and recall rates than the current methods. It further successfully reduces the relative mean error (RME) demonstrating improved performance and adaptability to changing conditions