Pulse amplitude modulation PAM-N signal optical performance monitoring method and device

Abstract

The embodiment of the invention provides a pulse amplitude modulation PAM-N signal optical performance monitoring method and device and belongs to the technical field of mobile communication. The method comprises the steps of extracting target features of a PAM-N signal from an eye diagram according to the eye diagram of the obtained PAM-N signal, wherein the target features comprise the maximum voltage value, the minimum voltage value, a Q value, N standard levels, eye height, cross point amplitude and mean square root jitter; and inputting the target features into a preset target neural network model and monitoring output target performance indexes, wherein the output target performance indexes comprise chromatic dispersion, an optical signal noise ratio and polarization mode dispersion,and according to the target neural network model, first quantity of eye diagram data of the PAM-N signal is obtained as a training set, a neural network model is trained through utilization of the training set and the trained model is obtained. According to the scheme provided by the embodiment of the invention, the PAM-N signal optical performance is monitored.

Description

technical field [0001] The invention relates to the technical field of mobile communication, in particular to a method and device for monitoring the optical performance of a pulse amplitude modulated PAM-N signal. Background technique [0002] In 2015, the PAM (Pulse Amplitude Modulation: Pulse Amplitude Modulation) N signal replaced the original OOK (On-Off Keying, binary on-off keying) signal, and was determined by the 400Gbps Internet Working Group as the only data center DataCenter transmission standard, so it is accurate It becomes important to monitor the optical properties of the PAM-N signal. [0003] However, the performance monitoring of the original OOK signal is relatively perfect in the existing technology, such as asynchronous sampling method, eye histogram analysis method, etc. are used to monitor the optical performance of the OOK signal, but the methods applied to OOK are not suitable for PAM -N. [0004] At present, for optical performance monitoring of O...

AI Technical Summary

Problems solved by technology

[0003] However, the performance monitoring of the original OOK signal is relatively perfect in the existing technology, such as asynchronous sampling method, eye histogram analysis method, etc. are used to monitor the optical performance of the OOK signal, but the methods applied to OOK are not suitable for PAM -N

[0004] At present, for optical performance monitoring of OOK signals, there are asynchronous sampling methods, eye histogram analysis methods, etc. to monitor optical performance. These features are input to the single hidden layer neural network for optical performance monitoring of OOK signals, but this method is not suitable for optical performance monitoring of PAM-N signals, such as eye height, Q value, cross point amplitude, root mean square jitter can be The OSNR (Optical Signal Noise Ratio), CD (Chromatic Dispersion), and DGD (Polarization Mode Dispersion, Differential Group Delay) of the OOK signal are better estimated, but for the PAM-4 signal, the OOK signal 2 for level, only 1 for eye height, cross point amplitude, and rms jitter; and 4 for PAM-4 signal level, 3 for eye height, 6 each for cross point amplitude and rms jitter

Therefore, the existing optical performance monitoring of the OOK signal is not suitable for the optical performance monitoring of the PAM-N signal

Images