Securing Time in Energy IoT: A Clock-Dynamics-Aware Spatio-Temporal Graph Attention Network for Clock Drift Attacks and Y2K38 Failures

TL;DR

STGAT presents a clock-dynamics-aware approach for securing temporal integrity in energy IoT by jointly modeling clock drift, synchronization offsets, jitter, and Y2K38 overflow as deformable temporal processes. It fuses drift-aware temporal embeddings with transformer-style temporal attention and graph attention across device interconnections, guided by curvature-regularized latent geometry to separate nominal clock evolution from anomalies. The framework supports online detection via a likelihood-based detector with adaptive thresholds and physics-informed checks, and introduces drift-aware data generation to simulate realistic timing faults. On augmented Edge-IIoTset data, STGAT achieves superior accuracy and much faster detection than baselines, with statistically robust improvements and stable latent-space separability, promising practical resilience against long-term clock drift and overflow threats in energy IoT systems.

Abstract

The integrity of time in distributed Internet of Things (IoT) devices is crucial for reliable operation in energy cyber-physical systems, such as smart grids and microgrids. However, IoT systems are vulnerable to clock drift, time-synchronization manipulation, and timestamp discontinuities, such as the Year 2038 (Y2K38) Unix overflow, all of which disrupt temporal ordering. Conventional anomaly-detection models, which assume reliable timestamps, fail to capture temporal inconsistencies. This paper introduces STGAT (Spatio-Temporal Graph Attention Network), a framework that models both temporal distortion and inter-device consistency in energy IoT systems. STGAT combines drift-aware temporal embeddings and temporal self-attention to capture corrupted time evolution at individual devices, and uses graph attention to model spatial propagation of timing errors. A curvature-regularized latent representation geometrically separates normal clock evolution from anomalies caused by drift, synchronization offsets, and overflow events. Experimental results on energy IoT telemetry with controlled timing perturbations show that STGAT achieves 95.7% accuracy, outperforming recurrent, transformer, and graph-based baselines with significant improvements (d > 1.8, p < 0.001). Additionally, STGAT reduces detection delay by 26%, achieving a 2.3-time-step delay while maintaining stable performance under overflow, drift, and physical inconsistencies.

Figures (10)

• Figure 1: Overview of the proposed STGAT architecture.
• Figure 2: Schematic of the edge deployment testbed.
• Figure 3: Detection delay comparison across models.
• Figure 4: Timestamp delta ($\Delta_t$) versus anomaly score.
• Figure 5: Inter-sample jitter ($J_t$) versus anomaly score.