发表机构
The Hong Kong Polytechnic University; Xi’an Jiaotong University; The Hang Seng University of Hong Kong; Hong Kong Shue Yan University（香港理工大学; 西安交通大学; 香港恒生大学; 香港树仁大学）
机构由 AI 辅助整理，请以论文原文为准。AI 中文总结
针对灾难导致蜂窝信令过载的监控问题，提出以互联网CDR为代理，用扩散合成激增进行压力测试，并通过困难样本适应提升检测器在默认阈值下的F1和AUC性能，构成可复用的部署前测试方法。
AI 中文摘要
灾难可能在几分钟内使蜂窝控制平面信令过载，然而细粒度的无线资源控制（RRC）或下一代（NG）应用协议（NGAP）遥测数据涉及隐私且采集成本高昂，难以用于分析。因此，许多应急监控流程依赖粗粒度的呼叫详细记录（CDR）聚合数据。在此约束下，我们将CDR网格中的互联网活动视为隐藏信令压力的实用代理。我们训练了一个轻量级卷积神经网络（CNN），使用风格化的过载注入数据进行训练，并通过扩散合成生成的、保留正常流量结构的激增数据进行压力测试，同时通过在困难合成样本上重新训练来调整检测器。在压力测试条件下，默认警报阈值失效，尽管接收者操作特征（ROC）曲线仍然表现强劲：检测器仍会为过载小区分配比正常小区更高的过载概率，但这些概率低于默认阈值0.5，因此被标记为正常，导致F1最大化阈值——在同一压力测试网格上事后选择（oracle $\tau^*$）——偏移了$0.32 \pm 0.03$（工作点漂移）。在三个随机种子下，困难样本适应将阈值化性能（F1）从0%（在任何种子上默认阈值0.5下均无警报）提升至$85.67 \pm 14.37$%，并将排序性能从ROC-AUC $0.886 \pm 0.040$提升至$0.99996 \pm 0.00007$。扩散合成激增暴露了阈值脆弱性，而匹配条件训练——在相同的风格化注入上训练和测试——则掩盖了这一问题，困难样本适应在默认阈值下恢复了可用警报。这些步骤共同构成了一个可复用的应急监控器部署前压力测试。仅使用互联网CDR输入进一步支持轻量级AI原生工作流，将监控、重新校准和适应相结合。
英文摘要
Disasters can overload cellular control-plane signaling within minutes, yet fine-grained Radio Resource Control (RRC) or Next Generation (NG) Application Protocol (NGAP) telemetry is privacy-sensitive and costly to collect for analytics. Many emergency monitoring pipelines therefore rely on coarse Call Detail Record (CDR) aggregates. We treat Internet activity in CDR grids as a practical proxy for hidden signaling stress under that constraint. We train a lightweight convolutional neural network (CNN) on stylized overload injections, stress-test it with diffusion-synthesized surges that preserve normal traffic structure, and adapt the detector by retraining on hard synthetic samples. Under stress-test conditions, the default alert threshold fails even though receiver operating characteristic (ROC) curves stay strong: the detector still assigns overloaded cells a larger overload probability than normal cells, but those probabilities fall below the default cutoff 0.5 and are labeled normal, so the F1-maximizing threshold -- selected post hoc on the same stress-test grids (oracle $τ^*$) -- shifts by $0.32 \pm 0.03$ (operating-point drift). Across three random seeds, hard-sample adaptation raises thresholded performance (F1) from 0% (no alerts at the default cutoff 0.5 on any seed) to $85.67 \pm 14.37$% and ranking from ROC-AUC $0.886 \pm 0.040$ to $0.99996 \pm 0.00007$. Diffusion-synthesized surges expose threshold fragility that matched-condition training -- training and testing on the same stylized injections -- hides, and hard-sample adaptation restores usable alerts at the default cutoff. Together, these steps define a reusable pre-deployment stress test for emergency monitors. Internet-only CDR input further supports lightweight AI-native workflows that combine monitoring, recalibration, and adaptation.
Comments6 pages, 5 figures. Accepted to the 5th Workshop on Next Generation Intelligent Wireless Emergency Communications, IEEE GLOBECOM 2026, Macau
