An Autoencoder-OFDM Power Line Carrier Communication System Based on Deep Learning

The energy problem is one of the focuses of contemporary attention. As an energy-saving technology, power line carrier communication technology has been applied in many fields. However, there is some unpredictable noise or interference sources and serious multipath fading in the power line channel. Traditional QAM-ODFM technology usually solves this problem through channel estimation. In this paper, deep learning network is creatively applied to OFDM framework, and a power line carrier communication technology based on autoencoder is proposed. This new technique modulates the subcarriers specifically for different subchannels and optimizes the communication parameters from a global point of view. Itnot only reduce the impact of channel noise and interference on communication efficiency, but also greatly save communication resources and reduce power consumption.