Optimal Virtualization Framework for Cellular Networks with Downlink Rate Coverage Probability Constraints

TL;DR

This paper presents a stochastic optimization framework for wireless network virtualization in 5G, ensuring QoS guarantees for users while minimizing provider costs amidst network uncertainties.

Contribution

It introduces a novel stochastic optimization approach for resource sharing in virtualized cellular networks with probabilistic QoS guarantees.

Findings

Framework effectively guarantees QoS under uncertainty.

Reduces operational costs for service providers.

Maintains computational efficiency and network overhead.

Abstract

Wireless network virtualization is emerging as an important technology for next-generation (5G) wireless networks. A key advantage of introducing virtualization in cellular networks is that service providers can robustly share virtualized network resources (e.g., infrastructure and spectrum) to extend coverage, increase capacity, and reduce costs. {However, the inherent features of wireless networks, i.e., the uncertainty in user equipment (UE) locations and channel conditions impose significant challenges on virtualization and sharing of the network resources.} In this context, we propose a stochastic optimization-based virtualization framework that enables robust sharing of network resources. Our proposed scheme aims at probabilistically guaranteeing UEs' Quality of Service (QoS) demand satisfaction, while minimizing the cost for service providers, with reasonable computational complexity and affordable network overhead.