Comparative Evaluation of Neural Network Models for Optimal Computational Offloading in Mobile Edge Computing

Mobile Edge Computing (MEC) has been promising in reducing latency for computational offloading to nearby edge servers and enhancing user experience. However, selecting the most suitable Machine Learning (ML) method for optimal task-offloading decisions remains a chal-lenge. This study proposes the most effective Neural Network Model (NNM) for making optimal offloading decisions. It has conducted a comparative analysis of several NNMs based on their performance metrics. The simulation results demonstrate that the Gated Recurrent Unit (GRU) delivers the best performance by achieving the accuracy of 95.71 %, including other parameter metrices.