Open RAN Slicing with Quantum Optimization

TL;DR

This paper introduces a quantum optimization framework for resource allocation in Open RAN slicing, aiming to improve efficiency and adaptability for diverse 5G services like URLLC and eMBB.

Contribution

It presents a novel quantum-based approach to RAN slicing resource allocation, formulated as a constrained quadratic model and implemented on a quantum annealing platform.

Findings

Quantum optimization can provide real-time solutions for RAN slicing.

The approach addresses limitations of heuristic and RL methods.

Potential for improved scalability and adaptability in 5G networks.

Abstract

RAN slicing technology is a key aspect of the Open RAN paradigm, allowing simultaneous and independent provision of various services such as ultra-reliable low-latency communications (URLLC), enhanced mobile broadband (eMBB), and massive machine-type communications (mMTC) through virtual networks that share a single radio access infrastructure. Efficient resource allocation is crucial for RAN slicing, as each service has specific quality of service (QoS) requirements, and a balance between different services must be maintained. Although heuristic and reinforcement learning (RL) techniques have been explored to achieve efficient resource allocation, these approaches face notable limitations: heuristic algorithms face complexity issues that limit their effectiveness in large networks, RL solutions are constrained by their dependency on training data and struggle to adapt to new scenarios and environments. This paper proposes a framework that leverages quantum optimization techniques to optimize radio resource blocks allocation in Open RAN slicing for URLLC and eMBB services. We provide a classical problem formulation and the quantum implementation using the constrained quadratic model on Dwave quantum annealing platform, showcasing the potential of quantum optimization techniques to deliver in real-time optimal solutions for optimization problems in 5G and beyond networks.