Context-Aware Graph Attention for Unsupervised Telco Anomaly Detection

2026-04-29

AI Summary

This work addresses the challenge of unsupervised anomaly detection in unlabeled multivariate time series within mobile communication networks by proposing C-MTAD-GAT, a novel model that introduces context-aware graph attention mechanisms to telecommunications scenarios for the first time. The approach integrates lightweight contextual embeddings, deterministic reconstruction, and multi-step forecasting, enabling unsupervised threshold calibration through residual validation. Evaluated on the public TELCO dataset, the model outperforms state-of-the-art methods such as MTAD-GAT and DC-VAE in both event-level and point-level F1 scores, significantly enhancing detection accuracy while reducing false positives. Its successful deployment in the core network of a national telecom operator demonstrates strong industrial applicability and practical utility.

Abstract

We propose C-MTAD-GAT, an \emph{unsupervised}, \emph{context-aware} graph-attention model for anomaly detection in multivariate time series from mobile networks. C-MTAD-GAT combines graph attention with lightweight context embeddings, and uses a deterministic reconstruction head and multi-step forecaster to produce anomaly scores. Detection thresholds are calibrated \emph{without labels} from validation residuals, keeping the pipeline fully unsupervised. On the public TELCO dataset, C-MTAD-GAT consistently outperforms MTAD-GAT and the Telco-specific DC-VAE, two state-of-the-art baselines, in both event-level and pointwise F1, while triggering substantially fewer alarms. C-MTAD-GAT is also deployed in the Core network of a national mobile operator, demonstrating its resilience in real industrial settings.

Problem

• unsupervised anomaly detection
• multivariate time series
• mobile networks
• graph attention
• context-aware

Innovation

• unsupervised anomaly detection
• graph attention
• context-aware embedding
• multivariate time series
• threshold calibration