$\pi$-ROAD: a Learn-as-You-Go Framework for On-Demand Emergency Slices in V2X Scenarios
Abstract
Vehicle-to-everything (V2X) is expected to become one of the main drivers of 5G business in the near future. Dedicated \emph{network slices} are envisioned to satisfy the stringent requirements of advanced V2X services, such as autonomous driving, aimed at drastically reducing road casualties. However, as V2X services become more mission-critical, new solutions need to be devised to guarantee their successful service delivery even in exceptional situations, e.g. road accidents, congestion, etc. In this context, we propose -ROAD, a \emph{deep learning} framework to automatically learn regular mobile traffic patterns along roads, detect non-recurring events and classify them by severity level. -ROAD enables operators to \emph{proactively} instantiate dedicated \emph{Emergency Network Slices (ENS)} as needed while re-dimensioning the existing slices according to their service criticality level. Our framework is validated by means of real mobile network traces collected within of a highway in Europe and augmented with publicly available information on related road events. Our results show that -ROAD successfully detects and classifies non-recurring road events and reduces up to the impact of ENS on already running services.
Keywords
Cite
@article{arxiv.2012.06208,
title = {$\pi$-ROAD: a Learn-as-You-Go Framework for On-Demand Emergency Slices in V2X Scenarios},
author = {Armin Okic and Lanfranco Zanzi and Vincenzo Sciancalepore and Alessandro Redondi and Xavier Costa-Perez},
journal= {arXiv preprint arXiv:2012.06208},
year = {2020}
}