RL-Based Adaptive Task-Offloading in Mobile-Edge Computing for IoT Networks

The Internet of Things (IoT) has been increasingly used in our daily lives and in industry. However, due to limitations in computing and power capabilities, IoT devices that need to send their respective tasks to the cloud service stations, which are usually located at far distances. Having to transmit data over long distances presents challenges for services that require low latency, such as industrial control in factories, as well as autonomous driving assisted by artificial intelligence. To solve this issue, mobile edge computing (MEC) is deployed at the network’s edge to reduce transmission time. This study proposes a new offloading scheme for MEC-assisted ultra-dense cellular networks using reinforcement learning (RL) techniques. The RL algorithm learns from the historical data of the network and adapts the offloading decisions to optimize the overall performance of the network. Non-orthogonal multiple access is also adopted to improve resource utilization among IoT devices. The simulation results demonstrate that the proposed scheme outperforms other state-of-the-art offloading algorithms in terms of energy efficiency, network throughput, and user satisfaction.