Quantum-enabled Solution for Tracking Area Planning based on NISQ-optimized Quantum Encoding for Urban Cellular Systems

Sammanfattning

Tracking area (TA) planning is a critical component of mobility management in 5G and 6G networks, influencing handover efficiency, signaling overhead, and resource utilization. Existing approaches often rely on classical clustering and graph partitioning techniques, which may not fully capture complex network constraints or dynamic mobility patterns. In this thesis, we present two hybrid quantum-classical solutions for optimizing TA planning in 5G and 6G networks: a Laplacian-based spectral clustering algorithm adapted to a variational quantum framework and a quadratic unconstrained binary optimization formulation solved using a variational algorithm with a qubitefficient encoding. Both are integrated into a workflow where classical algorithms generate initial partitions and enforce constraints after the quantum stage. Experiments on a dataset of 1,503 pre-clustered network nodes, simulated on classical hardware, show that the Laplacian-based method balances load but significantly increases inter-area handovers after refinement, whereas the quadratic unconstrained binary optimization method reduces handovers by 49.18% compared to the classical baseline while maintaining similar load distribution. Scalability analysis reveals circuit depth as the main bottleneck, underscoring the need for shallow, hardware-aware designs. These results highlight the potential of quantum-enabled methods for practical TA planning in next-generation mobile networks.