APEX：面向无线边缘运维的预测与异常检测的网络原生时间序列基础模型
APEX: A Network-Native Time-Series Foundation Model for Forecasting and Anomaly Detection for Wireless Edge Operations
AI总结：
提出网络原生解码器Transformer APEX，针对企业AP遥测数据预训练，在DHCP退化基准上MAE比最强基线降低18%，异常检测F1=0.93，边缘版本实现亚秒级隐私保护推理。
摘要：
通用时间序列基础模型对无线网络遥测数据的迁移效果较差，因为这些信号具有突发性、零膨胀性且跨协议层耦合。我们提出APEX，一个网络原生的、仅解码器的Transformer，用于预测企业AP遥测数据，并以DHCP退化作为代表性网络任务进行评估。APEX在来自约4500个生产无线网络的10通道多变量遥测数据（约10万AP时间序列，每个AP 34个指标）上预训练，并提供APEX-Large（269M参数，云端）和APEX-Edge（10.5M参数，边缘）两个版本。在192步（4天）的DHCP退化基准上，APEX-Large比最强的基础模型基线（Toto）MAE降低18%，比SARIMA降低38%，异常检测F1=0.93，而APEX-Edge能够在AP级边缘硬件上实现亚秒级、保护隐私的推理。这些结果表明，网络原生预训练是主动无线运维的实用基础。
英文摘要：
Generic time-series foundation models transfer poorly to wireless network telemetry whose signals are bursty, zero-inflated, and coupled across protocol layers. We present APEX, a network-native, decoder-only transformer for forecasting enterprise AP telemetry, and evaluate it on DHCP degradation as a representative network task. APEX is pre-trained on 10-channel multivariate telemetry from ~4,500 production wireless networks (~100K AP time series, 34 metrics per AP), and is available as APEX-Large (269M, cloud) and APEX-Edge (10.5M, edge). On a 192-step (4-day) DHCP degradation benchmark, APEX-Large reduces MAE by 18% over the strongest foundation-model baseline (Toto) and 38% over SARIMA, with anomaly-detection F1 = 0.93, while APEX-Edge enables sub-second, privacy-preserving inference on AP-class edge hardware. These results suggest network-native pre-training is a practical foundation for proactive wireless operations.
机构、总结、中文摘要均由AI分析而来，可能存在错误，请以论文正文为准。
