A data center optical communication dispersion estimation and management method based on deep learning

Abstract

The invention discloses a data center optical communication dispersion estimation and management method based on deep learning. The equalizer based on artificial neural network is divided into two phases. In the first phase, the artificial neural network is trained with the impulse response data of optical channel, and the model parameters of artificial neural network are optimized, and the nonlinear response model of artificial neural network is established. In the second stage, the trained artificial neural network equalizer is used to process the transmission data of the optical channel, soas to realize the estimation and compensation of the dispersion of the optical channel. The simulation results show that the equalizer based on artificial neural network improves the optical signal-to-noise ratio of optical communication and prolongs the transmission distance of optical communication.

Description

technical field [0001] The invention relates to a data center optical communication dispersion estimation and management method based on deep learning. Background technique [0002] With the development of cloud computing, the scale of data centers has grown rapidly, and optical networks are currently the main communication method for data centers. Dispersion phenomenon exists in optical fiber, which leads to linear or nonlinear distortion during the propagation of light waves. In the coherent optical communication system, the linear equalizer can be used to directly compensate the dispersion, but the linear equalizer scheme cannot be used in the direct detection system, because the phase information of the symbol is lost in the detection process of the direct detection system, so the dispersion phenomenon lead to obvious errors. Among all equalizer technologies, MLSE (Maximum Likelihood Receiver) has the best equalization performance, but its calculation cost for each sym...

AI Technical Summary

Problems solved by technology

[0003] Maximum likelihood estimation is the most accurate estimation algorithm for optical communication dispersion, but the calculation cost of maximum likelihood estimation for each symbol is proportional to the interference between symbols. In the case of high-speed optical communication, the interference between optical communication symbols is extremely large. Therefore, maximum likelihood estimation has extremely high computational complexity and low practicability for high-speed optical communication.

The data center optical rate of cloud computing is generally higher than 10Gbps, and in this case the calculation cost of maximum likelihood estimation is extremely high

Images