Gramian Angular Field-Based Power Line Communication Sensing for Railway Cable Aging Detection

Aging and degradation of power cables in railway systems can lead to failures, posing significant safety and operational risks. This paper presents a novel Gramian Angular Field (GAF)-based Power Line Communication (PLC) sensing method for detecting cable aging in both in-train wiring and railway longitudinal power lines. By extracting Channel State Information (CSI) from PLC signals and encoding it into structured GAF representations, the method preserves temporal correlations and nonlinear dynamics while enhancing resilience to noise and transient disturbances. A machine learning-based classification model is then applied to analyze these representations, improving the detection of subtle degradation patterns. Experimental results demonstrate that this approach significantly enhances early-stage cable aging detection, offering a robust and adaptive solution for railway maintenance and safety.