Optical network routing method and system based on physical layer impairment constraints based on machine learning

Abstract

The invention discloses an optical network routing method and system based on machine learning-based physical layer damage constraints, and belongs to the field of optical networks. The present invention comprehensively considers the physical layer and the network layer, realizes the performance optimization of the physical layer and reduces the blocking probability; realizes high precision and high speed at the same time by means of machine learning; selects the baud rate, bit rate, channel input optical power and the number of spans of the link , the type of optical amplifier used in each span, the noise figure of the optical amplifier used, the dispersion coefficient, the cumulative value of dispersion, the span length and the nonlinear coefficient are used as the physical layer parameters of the optical network, which are related to the signal transmission quality during the transmission process. The mapping relationship between the affected physical layer damage is found by machine learning training; the use of artificially synthesized data obtained by solving the optical fiber transmission equation has higher reliability and authenticity.

Description

technical field [0001] The invention belongs to the field of optical network routing, and more particularly relates to an optical network routing method and system based on machine learning-based physical layer damage constraints. Background technique [0002] Since the mid-1990s, IP services have experienced explosive growth, and communication networks have entered a "service-driven" era from a "technology-driven" era. The development direction of future networks will be full-service operations. In order to meet customers' ever-increasing service quality requirements, the current transport network must also meet the ever-increasing service quality requirements while providing customers with various services, that is, the physical layer transmission quality requirements. The emergence of wavelength division multiplexing technology has well solved the huge bandwidth transmission requirements in optical networks. However, in the process of continuous evolution of optical netwo...

AI Technical Summary

Problems solved by technology

So far, the traditional techniques used to evaluate the transmission quality of the physical layer of the optical path can be roughly divided into two categories: (1) Accurate analysis models estimate the damage of the physical layer and provide accurate results, but they need to bear a huge amount of calculation, and the real-time performance is relatively low. Poor; (2) Approximate formula, the calculation speed is fast, but the accuracy is not high, usually the introduction of high margin leads to underutilized network resources

Images