RadiOrchestra: Proactive Management of Millimeter-wave Self-backhauled Small Cells via Joint Optimization of Beamforming, User Association, Rate Selection, and Admission Control

TL;DR

RadiOrchestra is a comprehensive resource management framework for millimeter-wave self-backhauled small cells, jointly optimizing beamforming, user association, rate selection, and admission control to enhance network performance.

Contribution

It introduces a holistic RRM framework that jointly optimizes multiple aspects of self-backhauled networks considering practical hardware and interference constraints.

Findings

Algorithms offer a trade-off between complexity and optimality.

Performance bounds characterize system limits.

Practical deployment scenarios demonstrate feasibility.

Abstract

Millimeter-wave self-backhauled small cells are a key component of next-generation wireless networks. Their dense deployment will increase data rates, reduce latency, and enable efficient data transport between the access and backhaul networks, providing greater flexibility not previously possible with optical fiber. Despite their high potential, operating dense self-backhauled networks optimally is an open challenge, particularly for radio resource management (RRM). This paper presents, RadiOrchestra, a holistic RRM framework that models and optimizes beamforming, rate selection as well as user association and admission control for self-backhauled networks. The framework is designed to account for practical challenges such as hardware limitations of base stations (e.g., computational capacity, discrete rates), the need for adaptability of backhaul links, and the presence of interference. Our framework is formulated as a nonconvex mixed-integer nonlinear program, which is challenging to solve. To approach this problem, we propose three algorithms that provide a trade-off between complexity and optimality. Furthermore, we derive upper and lower bounds to characterize the performance limits of the system. We evaluate the developed strategies in various scenarios, showing the feasibility of deploying practical self-backhauling in future networks.