Optimal wireless transmitter placement is a central task in radio-network planning, yet exhaustive search becomes prohibitively expensive at scale. This paper studies the single-transmitter setting under a fixed learned propagation surrogate, where exhaustive per-pixel evaluation remains tractable and provides surrogate-exact ground truth. We introduce a dataset of 167,525 urban scenarios (RadioMapSeer-Deployment) with dual surrogate-exact labels for coverage-optimal and power-optimal transmitter locations. Ground-truth analysis reveals an asymmetric coverage-power trade-off: coverage-optimal placement sacrifices 13.86% of received power, whereas power-optimal placement sacrifices only 5.50% of coverage; the best achievable balanced placement lies at $\bar{d}=2.60$ from the ideal point (100%,100%). We evaluate two learning formulations: indirect heatmap-based models that predict received-power radio maps, and direct score-map models that predict the objective landscape over feasible transmitter locations. Within the heatmap family, discriminative models deliver one-shot predictions 1350-2400x faster than exhaustive search, while diffusion models additionally support multi-sample inference that improves single-objective performance and, by reusing the same sample pool under a balanced criterion, recovers strong balanced placements without explicit multi-objective training. Dual score-map strategies combining power and coverage score maps match the exhaustive balanced optimum ($\bar{d}=2.60$) and remain close across smaller candidate budgets, at 14-22x speedups after candidate re-evaluation. Both formulations admit very fast one-shot inference; on this benchmark, dual score-map methods are strongest for balanced placement, whereas heatmap formulations remain attractive for their physically meaningful intermediate maps and, in the diffusion setting, for inference-time search.

翻译：最优无线发射机布局是无线网络规划的核心任务，但穷举搜索在规模扩大时成本过高。本文研究固定学习传播代理下的单发射机场景，其中逐像素穷举评估仍可操作，并能提供代理精确的真实标签。我们构建了一个包含167,525个城市场景的数据集（RadioMapSeer-Deployment），针对覆盖最优和功率最优发射机位置提供双代理精确标签。真实标签分析揭示了非对称的覆盖-功率权衡：覆盖最优布局牺牲13.86%的接收功率，而功率最优布局仅牺牲5.50%的覆盖；可实现的最优均衡布局与理想点（100%，100%）的欧氏距离为$\bar{d}=2.60$。我们评估了两种学习范式：基于间接热图的模型预测接收功率无线电地图，以及直接得分图模型预测可行发射机位置上的目标景观。在热图方法中，判别模型实现单次预测比穷举搜索快1350-2400倍，而扩散模型额外支持多样本推理，能提升单目标性能，并通过在均衡准则下重用相同样本池，无需显式多目标训练即可恢复强均衡布局。结合功率与覆盖得分图的双得分图策略匹配穷举均衡最优值（$\bar{d}=2.60$），且在更小候选预算下保持接近，候选重评估后获得14-22倍加速。两种范式均可实现极快单次推理；在本基准测试中，双得分图方法在均衡布局中最强，而热图方法因其物理可解释的中间地图以及扩散设置下的推理时搜索能力而保持吸引力。