Network management and control method and system thereof, and storage medium

Patent metadata

Field	Value
Publication number	US-12574299-B2
Application number	US-202218684040-A
Country	US
Kind code	B2
Filing date	Jul 29, 2022
Priority date	Sep 22, 2021
Publication date	Mar 10, 2026
Grant date	Mar 10, 2026

Abstract

A network management and control method and system, and a storage medium are disclosed. The method may include: obtaining a parameter change value of a target object, wherein the parameter change value is from a Digital Twin (DT) virtual model, the DT virtual model is constructed based on a physical model, the physical model comprises entity objects of a physical network, and the parameter change value represents a change in transmission performance of the target object; inputting the parameter change value into a pre-trained perception model, to obtain a state prediction result output by the perception model; inputting the state prediction result into a pre-trained cognitive model, to obtain configuration adjustment information output by the cognitive model; and in response to the configuration adjustment information passing emulation verification, adjusting the physical model according to the configuration adjustment information.

First claim

What is claimed is:

1 . A network management and control method, comprising:
obtaining a parameter change value of a target object, wherein the parameter change value is from a Digital Twin (DT) virtual model, the DT virtual model is constructed based on a physical model, the physical model comprises entity objects of a physical network, and the parameter change value represents a change in transmission performance of the target object; inputting the parameter change value into a pre-trained perception model, to obtain a state prediction result output by the perception model; inputting the state prediction result into a pre-trained cognitive model, to obtain configuration adjustment information output by the cognitive model; and in response to the configuration adjustment information passing emulation verification, adjusting the physical model according to the configuration adjustment information; wherein before inputting the parameter change value into the pre-trained perception model, the method further comprises: obtaining an environment data set corresponding to the physical model from the DT virtual model, wherein the environment data set comprises a performance parameter and a state parameter of each of the entity objects in the physical model; wherein after obtaining the environment data set corresponding to the physical model from the DT virtual model, the method further comprises: obtaining a preset time sequence; determining probability distribution of the performance parameter and the state parameter in the time sequence; and constructing a signal generation model according to the probability distribution and the time sequence.

2 . The method of claim 1 , wherein before obtaining the parameter change value of the target object, the method further comprises:
determining an entity object whose performance parameter changes in the physical model as the target object.

3 . The method of claim 1 , wherein obtaining the parameter change value of the target object comprises:
determining a target analysis time sequence; and constructing the parameter change value of the target object in the target analysis time sequence according to the signal generation model.

4 . The method of claim 1 , wherein inputting the state prediction result into the pre-trained cognitive model, to obtain the configuration adjustment information output by the cognitive model comprises:
obtaining a target state and a constraint condition which are preset; and inputting the environment data set, the target state, the state prediction result, and the constraint condition into the cognitive model, to obtain the configuration adjustment information for the target object which is output by the cognitive model.

5 . The method of claim 4 , wherein after obtaining the environment data set corresponding to the physical model from the DT virtual model, the method comprises:
performing training from the cognitive model to the perception model using a backward propagation algorithm, wherein the performance parameter is a training input of the perception model, the state parameter is a training output of the perception model, the state prediction result and the environment data set are training inputs of the cognitive model, and the target state is determined as an output of the cognitive model.

6 . The method of claim 5 , wherein performing training from the cognitive model to the perception model using the backward propagation algorithm comprises:
determining, in a sequence from an output end to an input end, a gradient value for an output result of each network layer in the cognitive model and the perception model; and in response to the gradient value meeting a preset training condition, determining that the training from the output end of the cognitive model to the input end of the perception model is completed.

7 . The method of claim 1 , wherein in response to the configuration adjustment information passing the emulation verification, the method comprises:
performing the emulation verification according to the configuration adjustment information, to obtain an emulation result; and in response to the emulation result representing that a running status of the physical model meets a preset standard, determining that the configuration adjustment information passes the emulation verification.

8 . The method of claim 1 , wherein before obtaining the parameter change value of the target object, the method further comprises:
obtaining analysis requirement information, and determining a to-be-analyzed object in the analysis requirement information as the target object.

9 . A network management and control system, comprising a memory, a processor, and a computer program stored in the memory and executable by the processor which, when executed by the processor, causes the processor to perform a network management and control method, comprising:
obtaining a parameter change value of a target object, wherein the parameter change value is from a Digital Twin (DT) virtual model, the DT virtual model is constructed based on a physical model, the physical model comprises entity objects of a physical network, and the parameter change value represents a change in transmission performance of the target object; inputting the parameter change value into a pre-trained perception model, to obtain a state prediction result output by the perception model; inputting the state prediction result into a pre-trained cognitive model, to obtain configuration adjustment information output by the cognitive model; and in response to the configuration adjustment information passing emulation verification, adjusting the physical model according to the configuration adjustment information; wherein before inputting the parameter change value into the pre-trained perception model, the method further comprises: obtaining an environment data set corresponding to the physical model from the DT virtual model, wherein the environment data set comprises a performance parameter and a state parameter of each of the entity objects in the physical model; wherein after obtaining the environment data set corresponding to the physical model from the DT virtual model, the method further comprises: obtaining a preset time sequence; determining probability distribution of the performance parameter and the state parameter in the time sequence; and constructing a signal generation model according to the probability distribution and the time sequence.

10 . The network management and control system of claim 9 , wherein before obtaining the parameter change value of the target object, the method further comprises:
determining an entity object whose performance parameter changes in the physical model as the target object.

11 . The network management and control system of claim 9 , wherein before obtaining the parameter change value of the target object, the method further comprises:
obtaining analysis requirement information, and determining a to-be-analyzed object in the analysis requirement information as the target object.

12 . The network management and control system of claim 10 , wherein obtaining the parameter change value of the target object comprises:
determining a target analysis time sequence; and constructing the parameter change value of the target object in the target analysis time sequence according to the signal generation model.

13 . The network management and control system of claim 10 , wherein inputting the state prediction result into the pre-trained cognitive model, to obtain the configuration adjustment information output by the cognitive model com

Assignees

• Zte Corp

Inventors

• Wang Dajiang
• Wang Qilei
• Zhou Jin
• Jia Kui

Classifications

• H04L41/147

  for predicting network behaviour · CPC title

• H04L41/145 Primary

  involving simulating, designing, planning or modelling of a network · CPC title

• H04L41/0816

  the condition being an adaptation, e.g. in response to network events · CPC title

• G06N3/049

  Temporal neural networks, e.g. delay elements, oscillating neurons or pulsed inputs · CPC title

• G06N3/084

  Backpropagation, e.g. using gradient descent · CPC title