Content Caching-Assisted Vehicular Edge Computing Using Multi-Agent Graph Attention Reinforcement Learning

TL;DR

A novel content caching-assisted vehicular edge computing (VEC) framework and a multi-agent graph attention reinforcement learning (MGARL) based edge caching scheme, which utilizes the graph attention convolution kernel to integrate the neighboring nodes' features of each agent and further enhance the cooperation among agents are constructed.

Abstract

In order to avoid repeated task offloading and realize the reuse of popular task computing results, we construct a novel content caching-assisted vehicular edge computing (VEC) framework. In the face of irregular network topology and unknown environmental dynamics, we further propose a multi-agent graph attention reinforcement learning (MGARL) based edge caching scheme, which utilizes the graph attention convolution kernel to integrate the neighboring nodes' features of each agent and further enhance the cooperation among agents. Our simulation results show that our proposed scheme is capable of improving the utilization of caching resources while reducing the long-term task computing latency compared to the baselines.

Figures (5)

• Figure 1: Illustration of the content caching-assisted VEC system.
• Figure 2: Illustration of the proposed MGARL-based content caching-assisted VEC scheme.
• Figure 3: Comparison of the convergence.
• Figure 4: The converged content hit ratio versus the number of VUs.
• Figure 5: The converged total system latency versus the number of RSUs.