Cross‑border FM spectrum coordination must protect foreign broadcasting services while preserving domestic coverage, but real‑world planning data often involve thousands of transmitters and millions of transmitter‑pixel relationships, making conventional power‑control optimization computationally prohibitive due to its high dimensionality.
This paper introduces an interference‑driven clustered optimisation framework for large‑scale FM coordination. The key observation is that violations of foreign‑service protection are usually caused by a limited set of transmitters. Accordingly, protected services are analysed to identify dominant interferers and quantify their impact. An interference graph is then built to represent transmitter‑to‑transmitter interference, from which optimisation‑oriented clusters are extracted.
Within each cluster, the power‑control sub‑problem is solved using clustered simulated annealing, followed by a global refinement step that captures residual inter‑cluster interactions. Coverage and interference are evaluated with frequency‑dependent protection criteria and a dynamic strongest‑service assignment model. To support operational‑scale planning, the framework leverages sparse matrices and GPU‑accelerated computation.
Experiments on realistic cross‑border FM scenarios demonstrate that the clustering strategy dramatically reduces optimisation complexity and runtime while maintaining foreign‑service protection and domestic coverage. The approach also yields an interpretable ranking of transmitters that contribute most to harmful interference, aiding subsequent optimisation and spectrum‑planning decisions.
Review