Optical path transmission quality prediction method based on ANN loss function optimization

A loss function and transmission quality technology, applied in the field of optical communication, can solve problems such as network capacity reduction, achieve the effects of reducing the maximum positive deviation, increasing optical network capacity, and improving distribution efficiency

Inactive Publication Date: 2021-12-24
BEIJING UNIV OF POSTS & TELECOMM
View PDF5 Cites 0 Cited by
  • Summary
  • Abstract
  • Description
  • Claims
  • Application Information

AI Technical Summary

Problems solved by technology

[0010] Existing literature uses mean square error (MSE) and absolute error (MAE) as loss functions to train neural networks, both of which are symmetric loss functions that give the same penalty for overestimation and underestimation, focusing on the average of the model Accuracy, which is the average error over all samples, without additional attention to the maximum positive deviation of the model, which is the design margin of the network
In order to ensure the reliability of the optical path, operators and equipment manufacturers will use the difference between the predicted value of the optical path signal quality and the design margin to allocate modulation formats to them when deploying the optical path. If the design margin is too high, it will lead to low selection during optical path deployment. Modulation format, resulting in reduced network capacity

Method used

the structure of the environmentally friendly knitted fabric provided by the present invention; figure 2 Flow chart of the yarn wrapping machine for environmentally friendly knitted fabrics and storage devices; image 3 Is the parameter map of the yarn covering machine
View more

Image

Smart Image Click on the blue labels to locate them in the text.
Viewing Examples
Smart Image
  • Optical path transmission quality prediction method based on ANN loss function optimization
  • Optical path transmission quality prediction method based on ANN loss function optimization
  • Optical path transmission quality prediction method based on ANN loss function optimization

Examples

Experimental program
Comparison scheme
Effect test

Embodiment Construction

[0028] In order to better understand the technical solution, the method of the present invention will be described in detail below in conjunction with the accompanying drawings.

[0029] 1. Loss function design

[0030] At present, most of the loss functions in the research literature are symmetric loss functions, such as: MSE mean square error function, MAE, both of which have the same penalty for overestimation and underestimation.

[0031]

[0032]

[0033] In order to reduce the maximum positive deviation of the model, this paper introduces a regularization term I(x) based on MAE and MSE, which is expressed as formula (3):

[0034]

[0035] Among them, y represents the real value, Indicates the predicted value, and α is the penalty coefficient to achieve the purpose of reducing the maximum overestimation. When the predicted value is greater than the real value, that is, when it is overestimated, we give an additional penalty by adjusting the value of α. When the...

the structure of the environmentally friendly knitted fabric provided by the present invention; figure 2 Flow chart of the yarn wrapping machine for environmentally friendly knitted fabrics and storage devices; image 3 Is the parameter map of the yarn covering machine
Login to View More

PUM

No PUM Login to View More

Abstract

The invention discloses an optical path transmission quality prediction method based on ANN loss function optimization. The method introduces a regularization term based on traditional MSE and MAE loss functions, and reduces an overestimation value and proportion of a model by giving greater punishment to overestimation, thereby achieving the purpose of reducing the maximum positive deviation of the model. Compared with a traditional QoT estimation model based on MSE and MAE, the method provided by the invention can greatly reduce the maximum positive deviation value of the model while paying attention to the accuracy of the model, thereby achieving the purposes of improving the distribution efficiency of the optical path modulation format and improving the optical network capacity.

Description

technical field [0001] The invention relates to the technical field of optical communication, in particular to an optical path transmission quality prediction method based on ANN loss function optimization. Background technique [0002] The emergence of new technologies such as cloud computing, edge computing, Internet of Things, virtual reality, artificial intelligence, and 5G has led to explosive growth of network data. As one of the most important infrastructures for network data transmission, the optical transport network has a sharp increase in capacity requirements. It is necessary to further develop large-capacity transmission systems and improve network utilization. [0003] At present, in the traditional optical transmission network, in order to meet the smooth deployment of optical network services in its life cycle, it is necessary to reserve sufficient margin during the deployment phase of the optical network. The margin mainly includes system margin and unalloca...

Claims

the structure of the environmentally friendly knitted fabric provided by the present invention; figure 2 Flow chart of the yarn wrapping machine for environmentally friendly knitted fabrics and storage devices; image 3 Is the parameter map of the yarn covering machine
Login to View More

Application Information

Patent Timeline
no application Login to View More
IPC IPC(8): H04Q11/00G06N3/04G06N3/08G06Q10/04
CPCH04Q11/0062G06Q10/04G06N3/084H04Q2011/0086H04Q2011/0084G06N3/045
Inventor谷志群纪越峰张佳玮史亚男
OwnerBEIJING UNIV OF POSTS & TELECOMM