Optimizing Resource Management in 5G Heterogeneous Edge Computing with Blockchain and Deep Learning

The introduction of smart edge computing (SEC), cloud-based apps might be installed on the network's edge, freeing up server resources for more computationally expensive services such as facial recognition and natural language processing. The user interface and device efficiency can be improved by offloading operations normally done by the device's core CPU to the edge. Better coverage for the SEC could benefit Internet of Things (IoT) apps that require a lot of resources. Consistent, steady, and rapid interaction between edge devices and central processing units is difficult to achieve, particularly in 5G Heterogeneous Network (HN) environments where access is dynamic and network devices have varied levels of available resources. As a result, this study presents a paradigm for intelligent resource management in smart edge computing (IMRSEC) that combines blockchain, edge computing, and AI with 5G HNs. A novel dual schedule deep reinforcement learning (DS-DRL) technique was developed, which combines quick and gradual learning schedules. Reduced emptying delay and increased system reserves are major priorities, which can be done by enhancing computation unloading, resource transit, and software caching deployment. Simulation results reveal that the proposed IMRSEC is a successful DS-DRL implementation, as well as a 32% reduction in job completion time using the suggested technique.