QUBO-base uplink power control in CF-mMIMO: Exploring classical and quantum approaches

Abstract

Cell-Free Massive Multiple-Input Multiple-Output (CF-mMIMO) has emerged as a promising architecture for beyond-5G systems, offering gains in spectral efficiency, energy efficiency, and coverage uniformity by deploying distributed access points (APs) that jointly serve all users without cell boundaries. A key challenge in CF-mMIMO is uplink power control: under the max-min SINR fairness criterion, the problem is inherently NP-hard due to the strong coupling of user transmissions across APs. This thesis addresses the problem by discretising user power levels and reformulating the optimisation as a Quadratic Unconstrained Binary Optimization (QUBO) model.

The formulation is evaluated in a simulation framework against brute-force optimal solutions and greedy heuristics, using both classical and quantum solvers. The study highlights not only numerical performance of all the methods and solvers but also their optimality gap, allocation correctness, and simulation runtime scaling.

The results show that the QUBO approach achieves near-optimal SINR fairness with substantially better consistency than the greedy heuristic baselines, while highlighting its potential as a new paradigm that bridges classical optimisation with emerging quantum computing technologies.