Modeling the colored background noise of power line communication channel based on artificial neural network

Abstract

Power-line network is not specifically designed for high frequency signal transmission. Some tests and documents verify that power-line channel takes on many adverse characteristics that are not conducive to high frequency signal transmission, and noise characteristics are important parameters to describe the nature of power-line communication channel interference. The paper studied noise classification of the power-line communication channel, and specially analyzed the colored background noise characteristic and its mathematical model. Based on artificial neural network, optimization model is studied for the colored background noise of power-line communication channel. The paper educed the neural network algorithm used for modeling colored background noise. Through simulation, amplitude-frequency response curve is derived for neural network filter. And also time domain waveform and power spectrum density of colored noise are achieved from the neural model. The results indicate that artificial neural network modeling method is effective for modeling the colored background noise of power-line communication channel. It is helpful to construct the basis for simulation actual power-line channel.