APEX:用于无线边缘运维的网络原生时间序列基础模型
阅读原文· arxiv.orgAPEX 是一个网络原生、仅解码器的 Transformer 模型,专用于企业无线接入点(AP)遥测的预测与异常检测。它在约 4,500 个生产无线网络的 10 通道多元遥测数据上预训练,涵盖约 10 万条 AP 时间序列、每 AP 34 个指标。提供两个版本:APEX-Large(269M 参数,云部署)和 APEX-Edge(10.5M 参数,边缘部署)。在 192 步(4 天)的 DHCP 退化基准上,APEX-Large 比最强基础模型基线 Toto 降低 MAE 18%,比 SARIMA 降低 38%,异常检测 F1 达 0.93;APEX-Edge 可在 AP 级边缘硬件上实现亚秒级、保护隐私的推理。结果表明网络原生预训练是主动无线运维的实用基础。
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.