Edge Computing in 5G for Mobile AR/VR Data Prediction and Slicing Model

Abstract

To reduce computational connectivity issues, AR/VR data necessitates vast computational capabilities, tremendous transmission bandwidth, and ultra-low latency. AR/VR data can process data at the Mobile Edge computation (MEC) reducing the latencies in crucial decisions. Data slicing and edge computing are envisioned as critical enabler technologies for prioritizing data download and upload. Edge computing provides storage and processing resources at the network’s edge. The devices mimic a framework for data prediction as well as a slicing model to slice AR/VR data streaming. AR/VR slicing model requires uploading and downloading streams speed limit, connectivity time, bandwidth, and user pattern as its parameters to predict data slicing model in edge computing to improvise the network utilization. MEC uses ML in 3 ways.(1) ML-based task offloading techniques; (2) ML-based task scheduling methods; and (3) ML-based joint resource allocation methods.