Multi-conductor noise in power line communication is a complex phenomenon that cannot be analyzed in a simple manner. Data analytics techniques, such as automatic clustering, are a promising methodology to analyze multi conductor noise traces with reduced computational effort and small number of setup parameters. In this article, we propose an approach to uncover the structure of the narrow band multi conductor power line noise with an automatic procedure that exploits the tool self organizing map to extract features, classify and label the noise time series. The proposed algorithms are evaluated through real noise data in the spectrum 3-500kHz.
Automatic Clustering of Noise in Multi-Conductor Narrow Band PLC Channels
Tonello A. M.
2019-01-01
Abstract
Multi-conductor noise in power line communication is a complex phenomenon that cannot be analyzed in a simple manner. Data analytics techniques, such as automatic clustering, are a promising methodology to analyze multi conductor noise traces with reduced computational effort and small number of setup parameters. In this article, we propose an approach to uncover the structure of the narrow band multi conductor power line noise with an automatic procedure that exploits the tool self organizing map to extract features, classify and label the noise time series. The proposed algorithms are evaluated through real noise data in the spectrum 3-500kHz.I documenti in IRIS sono protetti da copyright e tutti i diritti sono riservati, salvo diversa indicazione.