Efficient Task Offloading in IoV Using DDPG and MEC with RIS Support

The Internet of Vehicles (IoV) represents a subset of the Internet of Things (IoT) devices specifically designed for vehicles to seamlessly connect with their diverse surroundings and engage in data exchange. However, one persistent challenge within IoV is the ability to perform intricate mathematical computations locally. To address this challenge, Mobile Edge Computing (MEC) has been introduced, enabling Mobile Devices (MDs) to offload resource-intensive computational tasks onto MEC resources. In this research paper, we present a novel Deep Reinforcement Learning (DRL) algorithm that leverages rewards, offloading ratios, and latency considerations to optimize task allocation. We employ the Deep Deterministic Policy Gradient (DDPG) algorithm, which excels in handling high-dimensional state spaces and continuous action spaces, making it well-suited for our IoV context. Furthermore, our IoV network benefits from the integration of Reconfigurable Intelligent Surface (RIS) technology, enhancing communication capabilities among vehicles. Through comprehensive simulations, we demonstrate the effectiveness of our proposed algorithm for managing computation and reducing computational delays in IoV scenarios.