Subspace-confined QAOA with generalized dicke states for multi-channel allocation in 5G CBRS networks

Abstract

Abstract Efficient spectrum sharing in the Citizens Broadband Radio Service (CBRS) band is essential for maximizing 5G network capacity, particularly when high-traffic base stations require simultaneous access to multiple channels. Standard Quantum Approximate Optimization Algorithm (QAOA) formulations impose such multi-channel constraints through penalty terms, so most of the explored Hilbert space corresponds to invalid assignments. We propose a subspace-confined QAOA tailored to CBRS multi-channel allocation, in which each node-wise channel register is initialized in a Generalized Dicke state and evolved under an intra-register XY mixer. This ansatz confines the dynamics to a tensor product of Johnson-graph subspaces that exactly encode heterogeneous per-node Hamming-weight constraints. Rather than claiming near-term quantum advantage over mature classical solvers on small instances, we emphasize a constraint-preserving ansatz design that keeps the variational dynamics inside the valid allocation manifold. We complement this design with larger-scale and topology-diverse evaluations on structured and Erdős–Rényi interference graphs for $$n=10,12,14$$ nodes (up to 42 qubits). We also include stronger classical heuristics, an exact ILP benchmark, a QAOA depth study for $$p=1,2,3$$ , a small-scale dual-constraint validation, explicit circuit-resource analysis, and search-space scaling analysis. Across all tested instances, the proposed ansatz maintains unit feasibility ratio and zero mean demand deviation, while producing conflict levels that remain competitive with the strongest classical baselines. We further present a dual-constraint extension that simultaneously preserves node-wise demands and per-channel capacities, and we frame this extension as a proof-of-principle construction rather than a hardware-optimized near-term primitive.