A Delay-Aware Energy-Efficient System for a 5G MEC/D2D Network Using a Computational Selection Algorithm

Abstract

The challenge now is to find a way to combine D2D and MEC into a low latency, energy-efficient system. The main goal of this study is to present a delay-aware, energy-efficient system for carrying the D2D capabilities of the MEC network. As these two technologies are built into 5G. In this research first specified the network topology and proposed a computational selection algorithm before formulating user / IoT devices, MEC servers, task parameters, and task execution issues. To find the ideal dump position, two algorithms are provided. One is for time-sensitive data and the other is for devices that have power limitations. As a result of exhaustive planning, problem characterization, and decision-making algorithms. To evaluate the effectiveness implemented a simulation. Also show that the proposed EEDOS The energy saving rate is shown in simulation iterations, but the number of units is limited to 100 nodes. To evaluate the effect of work on energy savings, the probability of task completion was raised in this study from 30% to 70%. Increasing the task allocation rate in the latter reduced the number of units available in the neighborhood. EEDOS, on the other hand, consistently saves energy in the range of 95.8% to 96.9% in both cases. The D2D mitigation is 50% more energy efficient than when working on local hardware. The average energy efficiency of their system is 60%. EEDOS saved 97% of energy compared to local data execution. The results show that EEDOS outperforms existing data relief solutions in terms of energy efficiency and data execution time for mobile / IoT devices when compared to edge computing data relief and D2D collaboration. EEDOS requires significantly less server computer capacity than the MEC unloading schedule. Finally, EEDOS reduced the number of missed deadlines compared to previous research.