BSCSM-GAN: End-to-End Generative Prediction of Base Station Coverage Strength Maps for 6G-Aware Urban Deployment

Efficient base station (BS) deployment in dense urban environments requires rapid and accurate coverage evaluation. Traditional ray-tracing, although precise, is too computationally intensive for real-time or large-scale planning. In this paper, we introduce the Base Station Coverage Strength Map (BSCSM), a novel spatial metric that quantifies deployment suitability for each location at full spatial resolution. We propose an end-to-end conditional Generative Adversarial Network (cGAN) framework that predicts BSCSM directly from building layout images, eliminating the need for expensive per-location simulations. Compared to conventional U-Net and Pix2Pix baselines, our model achieves inference speeds several orders of magnitude faster than ray-tracing, while maintaining superior TopK hotspot prediction accuracy. The generated BSCSM enable advanced 6G network planning, UAV path optimization, and real-time connectivity assessment, providing a scalable and practical solution for next-generation wireless infrastructure design.