Optical network power optimization method and system based on trusted configuration space

Abstract

According to the optical network power optimization method and system based on the trusted configuration space, reliable optimization of the optical network power is achieved by dynamically updating the trusted configuration space. The method comprises the following steps: estimating a Lipschitz constant by using a digital twin system, determining a trusted configuration space according to the estimated constant and a system sampling data set, and executing an optimization algorithm in the space to generate a new configuration. In the optimization process, the hill climbing algorithm is used in the early stage to quickly improve the system margin, and the CMA-ES algorithm is switched to fit the evolutionary strategy in the later stage. And after new configuration deployment, updating the Lipschitz constant estimation value and the sampling data set, and continuing iterative optimization. The optical power optimization method solves the problem that the traditional optimization method may cause significant reduction and even interruption of the system performance, and improves the reliability of optical power optimization.

Description

Technical Field

[0001] The present invention relates to the technical field of optical communication networks, and in particular to an optical network power optimization method and system based on a trusted configuration space. Background Art

[0002] In optical fiber communication networks, optical power optimization is one of the key technologies to ensure the quality of optical transmission. With the rapid development of information technology, the amount of business carried by optical networks continues to increase, which puts higher requirements on the performance and stability of optical transmission. However, for online optical communication networks, especially those that carry business, the adjustment process of optical power optimization becomes particularly complicated due to the limited known information about the optical network.

[0003] Traditional optical power optimization methods are often configured based on static network models and parameters, which are difficult to adapt to dynamically changing network environments. At the same time, these methods may cause system performance to significantly decline or even interrupt during the optimization process, posing a serious threat to business continuity and stability. Therefore, how to optimize optical power while ensuring stable system performance has become a technical problem that needs to be solved urgently in the current optical network field.

[0004] In recent years, some emerging technical methods have been proposed to meet this challenge. For example, at the 2021 "Optical Fiber Communication Conference", Zhizhen Zhong, Manya Ghobadi and others disclosed the "Optical Power Optimization Method Based on Bayesian Optimization". This method uses the Bayesian optimization algorithm to optimize optical power and has achieved certain results. However, this method still has some shortcomings. Because the Bayesian optimization algorithm may adopt unsafe configurations during the exploration process, it leads to unstable system performance and may even cause system interruption. Therefore, although this method can achieve optical power optimization, there are still certain risks in practical applications.

[0005] In addition, some well-known companies and scientific research institutions are also actively exploring new technologies for optical power optimization. For example, companies such as the Massachusetts Institute of Technology and Facebook have deep research experience in the field of optical networks and have conducted in-depth research on optical power optimization technologies. However, these methods are still unable to meet the dual requirements of online optical communication networks for performance stability and optimization efficiency.

[0006] Therefore, it is necessary to propose a new technical solution to improve the above technical problems. Summary of the invention

[0007] In view of the defects in the prior art, the object of the present invention is to provide an optical network power optimization method and system based on a trusted configuration space.

[0008] Compared with the prior art, the present invention has the following beneficial effects:

[0009] 1. The present invention introduces a trusted online optimization strategy based on the Lipschitz constant to construct a dynamically updated trusted configuration space, effectively avoiding the problem that the traditional optimization method may cause significant degradation or even interruption of system performance during the exploration process, thereby significantly improving the reliability and stability of optical network power optimization;

[0010] 2. The present invention can not only conduct efficient exploration under the premise of ensuring system safety, but also continuously optimize system configuration and achieve continuous improvement of optical network power;

[0011] 3. The optimization process of the present invention is a cycle of continuous iteration and updating. By collecting the gradient information and QoT data fed back by the system in real time, the Lipschitz constant estimation value and the sampling data set are dynamically updated, so that the optimization strategy can be adaptively adjusted to better cope with the uncertainties and changes in the optical network. BRIEF DESCRIPTION OF THE DRAWINGS

[0012] Other features, objects and advantages of the present invention will become more apparent from the detailed description of non-limiting embodiments made with reference to the following drawings:

[0013] Figure 1 It is a schematic diagram of the process of the present invention. DETAILED DESCRIPTION

[0014] The present invention is described in detail below in conjunction with specific embodiments. The following embodiments will help those skilled in the art to further understand the present invention, but are not intended to limit the present invention in any form. It should be noted that, for those of ordinary skill in the art, several changes and improvements can also be made without departing from the concept of the present invention. These all belong to the protection scope of the present invention.

[0015] Embodiment 1:

[0016] Reference Figure 1 According to an optical network power optimization method based on a trusted configuration space provided by the present invention, the method comprises the following steps:

[0017] Step S1: Initialize the digital twin system and estimate the Lipschitz constant using the digital twin system;

[0018] Step S1.1: Establish a digital twin system based on the GNPy model and monitoring data;

[0019] The digital twin system includes a fiber transmission model and an amplifier model. The fiber transmission model is based on the Gaussian noise model and the parameters provided by the fiber manufacturer. The GNPy library is called during calculation to estimate the attenuation and nonlinear noise in the optical transmission process. The amplifier model is based on the physical model of the erbium-doped fiber amplifier and the amplifier data provided by the manufacturer. It is used to estimate the amplifier's spontaneous radiation noise during calculation. In the process of building a digital twin system, the optical power values ​​at multiple points in the real optical fiber transmission link can be monitored through the SDN controller. These monitored optical power data are used to calibrate the joint loss and insertion loss values ​​at multiple points of the digital twin system, so that the digital twin system matches the actual optical fiber transmission link and completes the construction of the digital twin system.

[0020] Step S1.2: randomly generate an initial configuration in the digital twin system and calculate the maximum gradient value of the quality-to-transmission QoT pair configuration;

[0021] Step S1.3: The calculated maximum gradient value is used as the initial Lipschitz constant.

[0022] Step S2: determining a credible configuration space based on the estimated Lipschitz constant and the system sampling data set;

[0023] Step S2.1: Obtain current system configuration and its performance indicators;

[0024] Step S2.2: Calculate the credible space range based on the Lipschitz constant and the performance threshold.

[0025] Step S3: Execute the optimization algorithm in the trusted configuration space to generate a new configuration;

[0026] Step S3.1: In the early stage of optimization, use the hill climbing algorithm to improve the system margin and collect performance and gradient data during the optimization process;

[0027] Step S3.2: In the later stage of optimization, switch to the covariance matrix adaptive evolution strategy CMA-ES algorithm and use the data collected in the early stage to fit the evolution strategy.

[0028] Step S4: deploy the generated new configuration, update the Lipschitz constant estimate and the sampling data set, and return to step S2 to continue iterative optimization;

[0029] Step S4.1: deploy the configuration to the online system;

[0030] Step S4.2: collecting gradient information and QoT data fed back by the system;

[0031] Step S4.3: Update the Lipschitz constant estimate and the sampled data set.

[0032] The generated new configuration is deployed to the online system through instructions issued by the optical network software-defined network (SDN) controller. The deployment process includes adjusting the configuration of each optical amplifier. Before and after adjusting each optical amplifier, the SDN controller will collect changes in system performance and calculate the gradient value of system performance changes corresponding to the adjustment. If the maximum value of these gradient information exceeds the original Lipschitz constant estimate, the Lipschitz constant estimate is updated with a larger gradient value. After the new configuration is deployed, the SDN controller will collect the system QoT data corresponding to the new configuration and add it to the sampling data set to update the sampling data set. Then return to step S2 to use the updated Lipschitz constant and the system sampling data set to determine the new trusted configuration space and continue iterative optimization.

[0033] The present invention also provides an optical network power optimization system based on a trusted configuration space. The optical network power optimization system based on a trusted configuration space can be implemented by executing the process steps of the optical network power optimization method based on the trusted configuration space, that is, those skilled in the art can understand the optical network power optimization method based on the trusted configuration space as a preferred implementation of the optical network power optimization system based on the trusted configuration space.

[0034] Embodiment 2:

[0035] The present invention also provides an optical network power optimization system based on a trusted configuration space, the system comprising the following modules:

[0036] Module M1: Initialize the digital twin system and estimate the Lipschitz constant using the digital twin system;

[0037] Module M1.1: Establish a digital twin system based on the GNPy model and monitoring data;

[0038] Module M1.2: Randomly generate initial configurations in the digital twin system and calculate the maximum gradient value of the quality-to-transmission (QoT) pair configuration;

[0039] Module M1.3: The calculated maximum gradient value is used as the initial Lipschitz constant.

[0040] Module M2: Determine the credible configuration space based on the estimated Lipschitz constant and the system sampling data set;

[0041] Module M2.1: Get the current system configuration and its performance indicators;

[0042] Module M2.2: Calculate the credible space range based on the Lipschitz constant and performance threshold.

[0043] Module M3: Executes the optimization algorithm in the trusted configuration space to generate a new configuration;

[0044] Module M3.1: In the early stages of optimization, use the hill climbing algorithm to improve the system margin and collect performance and gradient data during the optimization process;

[0045] Module M3.2: In the later stage of optimization, switch to the covariance matrix adaptive evolution strategy CMA-ES algorithm and use the data collected in the early stage to fit the evolution strategy.

[0046] Module M4: deploys the generated new configuration, updates the Lipschitz constant estimate and the sampling data set, and returns to call module M2 to continue iterative optimization;

[0047] Module M4.1: Deploy the configuration to the online system;

[0048] Module M4.2: Gradient information and QoT data fed back by the acquisition system;

[0049] Module M4.3: Update Lipschitz constant estimates and sampled data sets.

[0050] Embodiment 3:

[0051] The present invention proposes a trusted online optimization strategy based on the Lipschitz constant, which realizes the reliable optimization of optical network power by establishing and dynamically updating a trusted configuration space, including the following steps:

[0052] Step 1: Initialize the digital twin system and use the digital twin system to estimate the Lipschitz constant;

[0053] 1.1 Establish a digital twin system based on the GNPy model and monitoring data;

[0054] 1.2 Randomly generate an initial configuration in the digital twin system and calculate the maximum gradient value of QoT to the configuration;

[0055] 1.3 The calculated maximum gradient value is used as the initial Lipschitz constant;

[0056] Step 2: Determine the credible configuration space using the estimated Lipschitz constant and the system sampling data set;

[0057] 2.1 Obtain the current system configuration and its performance indicators;

[0058] 2.2 Calculate the credible space range based on the Lipschitz constant and performance threshold;

[0059] Step 3: Execute the optimization algorithm in the trusted configuration space to generate a new configuration;

[0060] 3.1 In the early stage of optimization, use the hill climbing algorithm to quickly improve the system margin and collect performance and gradient data during the optimization process;

[0061] 3.2 In the later stage of optimization, switch to the CMA-ES algorithm and use the data collected in the early stage to fit the evolutionary strategy;

[0062] Step 4: Deploy the generated new configuration and update the Lipschitz constant estimate and sampling data set;

[0063] 4.1 Deploy the configuration to the online system;

[0064] 4.2 Collect the gradient information and QoT data fed back by the system;

[0065] 4.3 Update Lipschitz constant estimate and sampling data set;

[0066] 4.4 Using the updated data, return to step 2 to continue iterative optimization;

[0067] The present invention solves the problem that the traditional optimization method may cause a significant decrease or even interruption of system performance during the optimization process by adopting a structure of a trusted configuration space based on the Lipschitz constant, thereby achieving the effect of reliable optical power optimization.

[0068] Those skilled in the art may understand this embodiment as a more specific description of Embodiment 1 and Embodiment 2.

[0069] Those skilled in the art know that, in addition to realizing the system and its various devices, modules, and units provided by the present invention in a purely computer-readable program code, it is entirely possible to realize the same functions in the form of logic gates, switches, application-specific integrated circuits, programmable logic controllers, and embedded microcontrollers by logically programming the method steps. Therefore, the system and its various devices, modules, and units provided by the present invention can be considered as a hardware component, and the devices, modules, and units included therein for realizing various functions can also be regarded as structures within the hardware component; the devices, modules, and units for realizing various functions can also be regarded as both software modules for realizing the method and structures within the hardware component.

[0070] The above describes the specific embodiments of the present invention. It should be understood that the present invention is not limited to the above specific embodiments, and those skilled in the art can make various changes or modifications within the scope of the claims, which does not affect the essence of the present invention. In the absence of conflict, the embodiments of the present application and the features in the embodiments can be combined with each other arbitrarily.

Claims

1. A method for optimizing optical network power based on a trusted configuration space, characterized in that: The method comprises the following steps: Step S1: Initialize the digital twin system and estimate the Lipschitz constant using the digital twin system; Step S2: determining a credible configuration space based on the estimated Lipschitz constant and the system sampling data set; Step S3: Execute the optimization algorithm in the trusted configuration space to generate a new configuration; Step S4: Deploy the generated new configuration, update the Lipschitz constant estimate and the sampling data set, and return to step S2 to continue iterative optimization.

2. The optical network power optimization method based on trusted configuration space according to claim 1 is characterized in that: The step S1 comprises the following steps: Step S1.1: Establish a digital twin system based on the GNPy model and monitoring data; Step S1.2: randomly generate an initial configuration in the digital twin system and calculate the maximum gradient value of the quality-to-transmission QoT pair configuration; Step S1.3: The calculated maximum gradient value is used as the initial Lipschitz constant.

3. The optical network power optimization method based on trusted configuration space according to claim 1 is characterized in that: The step S2 comprises the following steps: Step S2.1: Obtain current system configuration and its performance indicators; Step S2.2: Calculate the credible space range based on the Lipschitz constant and the performance threshold.

4. The optical network power optimization method based on trusted configuration space according to claim 1, characterized in that: The step S3 comprises the following steps: Step S3.1: In the early stage of optimization, use the hill climbing algorithm to improve the system margin and collect performance and gradient data during the optimization process; Step S3.2: In the later stage of optimization, switch to the covariance matrix adaptive evolution strategy CMA-ES algorithm and use the data collected in the early stage to fit the evolution strategy.

5. The optical network power optimization method based on trusted configuration space according to claim 1, characterized in that: The step S4 comprises the following steps: Step S4.1: deploy the configuration to the online system; Step S4.2: collecting gradient information and QoT data fed back by the system; Step S4.3: Update the Lipschitz constant estimate and the sampled data set.

6. An optical network power optimization system based on a trusted configuration space, characterized in that: The system includes the following modules: Module M1: Initialize the digital twin system and estimate the Lipschitz constant using the digital twin system; Module M2: Determine the credible configuration space based on the estimated Lipschitz constant and the system sampling data set; Module M3: Executes the optimization algorithm in the trusted configuration space to generate a new configuration; Module M4: Deploy the generated new configuration, update the Lipschitz constant estimate and sampling data set, and return to module M2 to continue iterative optimization.

7. The optical network power optimization system based on trusted configuration space according to claim 6, characterized in that: The module M1 includes the following modules: Module M1.1: Establish a digital twin system based on the GNPy model and monitoring data; Module M1.2: Randomly generate initial configurations in the digital twin system and calculate the maximum gradient value of the quality-to-transmission (QoT) pair configuration; Module M1.3: The calculated maximum gradient value is used as the initial Lipschitz constant.

8. The optical network power optimization system based on trusted configuration space according to claim 6, characterized in that: The module M2 includes the following modules: Module M2.1: Get the current system configuration and its performance indicators; Module M2.2: Calculate the credible space range based on the Lipschitz constant and performance threshold.

9. The optical network power optimization system based on trusted configuration space according to claim 6, characterized in that: The module M3 includes the following modules: Module M3.1: In the early stages of optimization, use the hill climbing algorithm to improve the system margin and collect performance and gradient data during the optimization process; Module M3.2: In the later stage of optimization, switch to the covariance matrix adaptive evolution strategy CMA-ES algorithm and use the data collected in the early stage to fit the evolution strategy.

10. The optical network power optimization system based on trusted configuration space according to claim 6, characterized in that: The module M4 includes the following modules: Module M4.1: Deploy the configuration to the online system; Module M4.2: Gradient information and QoT data fed back by the acquisition system; Module M4.3: Update Lipschitz constant estimates and sampled data sets.

Images