Feature Extraction of the Noise in L-PLC

This paper investigates chaotic characteristics of the noise in low voltage power line communication (L-PLC) system. Firstly, the negative effect of measurement noise (M- noise) is studied and a wavelet transform denoise algorithm is adopted to restraint M-noise. Secondly, the largest Lyapunov exponent and the correlation dimension of the denoised and original time series are calculated. The results show that the method of wavelet denoise can restraint M-noise while maintaining the chaotic characteristics of the observed series not changed. Further more, the effect of the sampling method including the sampling circuit design sampling, frequency as well as number of sampling date to the analysis result is discussed. Finally, the meaningful conclusions, L-PLC noise is the short- term predictable and 4 independent variables are at least needed to build a mathematics model of L-PLC system, are given based on the calculating results.