A Deep Reinforcement Learning-Based Offloading Scheme for Multi-Access Edge Computing-Supported eXtended Reality Systems

In recent years, eXtended Reality (XR) applications have been widely employed in various scenarios, e.g., health care, education, manufacturing, etc. Such application are now easily accessible via mobile phones, tablets, or wearable devices. However, such devices normally suffer from constraints terms of battery capacity and processing power, limiting the range supported lowering Quality Experience. One effective way to address these issues is offload computation tasks edge servers that deployed at network edges, base stations WiFi access point, This communication fashion, also named as Multi-access Edge Computing (MEC), proposed overcome limitations long latency due propagation distance traditional cloud computing approach. XR devices, limited resources energy, can then benefit offloading intensive MEC servers. comprised multiple with variety requirements energy consumption, it important make decision whether one task should be offloaded server not. paper proposes a Deep Reinforcement Learning-based scheme for (DRLXR). The used train derive close-to optimal whereas optimizing utility function optimization equation considers both consumption execution delay simulation results show how our outperforms other counterparts total consumption.