Variational Quantum Algorithm for User Scheduling in Broadband MU-MIMO Systems
Chih-Ho Hsu, Zhi Jun Ding
Multiuser MIMO (MU-MIMO) deployment jointly serves multiple users over shared time-frequency Resource Blocks (RBs) to achieve substantial spectral efficiency improvement. To exploit the full potential of MU-MIMO, users should be scheduled on RBs with strong channels and low Co-Channel Interference (CCI). This paper leverages Variational Quantum Algorithm (VQA) framework, tailored for near-term Noisy Intermediate-Scale Quantum (NISQ) devices, for efficient user scheduling in broadband MU-MIMO systems, which is NP-hard. Specifically, our designed quantum circuit incorporates parameterized entangling gates to create controllable correlations between quantum states, which may explore the solution space more efficiently. To facilitate the challenging variational parameters optimization in VQAs, exacerbated by entangling operations, our framework adopts a two-stage approach. First, we utilize Bayesian Optimization (BO) which constructs a global surrogate model to find sets of promising parameters. Second, we perform local optimization to fine-tune the result of BO by leveraging Gaussian kernels to approximate the noisy optimization landscape with a limited amount of quantum shots. Through extensive simulations, we demonstrate the efficacy of our VQA in achieving near-optimal performance for broadband MU-MIMO user scheduling problems.
IEEE Transactions on Wireless Communications•Jul 23, 2025•1 citation