Channel Estimation Algorithm of OFDM-RoF System in 5G Mobile Front-end Network Based on Artificial Neural Network

Abstract

In the environment of the 5G era, with the advancement of communication technology and the continuous improvement of people's living and work needs, users' demand for network access bandwidth is increasing. Orthogonal Frequency Division Multiplexing-Radio Frequency over Optical (OFDM-RoF) system is an Internet solution with high spectrum utilization, large bandwidth and fast transmission data rate. The chromatic dispersion (CD) and polarization mode dispersion (PMD) existing in the system will affect the transmission performance of the OFDM-RoF system. In this paper, the artificial neural network algorithm is applied to the field of channel estimation. Reduce the effect of dispersion on the system by estimating the activation function of the channel. Simulation results show that compared with the frequency domain least squares (FDLS) method, this algorithm can improve the system performance and improve the bit error rate optimization ability by an order of magnitude.