3D Digital Twin for Wireless Network Performance Prediction
Find Innovative SolutionsGenerate Solutions
Solution Overview
Problem
In industrial facilities, the stability of 5G wireless networks is compromised by moving parts, changing layouts, and obstructions, making it difficult to maintain uninterrupted and high-quality data communication, which is essential for industrial automation. Current network testing and validation processes are cumbersome, time-consuming, and lack automation, leading to delays and high costs.
Innovation Solution
A method and apparatus that transform a physical environment into a 3D digital representation, map it to feature vectors, and use machine learning algorithms to predict wireless network performance, enabling the generation of virtual layouts and performance predictions, thereby facilitating continuous testing and adaptation to changing environments.
Engineering Contradictions & Design Principles
Engineering Contradiction Analysis
1Productivity
If traditional network testing and validation processes are used in industrial facilities, then network performance can be measured, but the process is cumbersome, time-consuming, and lacks automation
Solution Approach 1:
The patent creates a digital twin (virtual copy) of the physical industrial environment including 3D models of facilities, equipment, and wireless signal propagation characteristics. This digital replica enables automated simulation and testing of network performance without requiring physical deployment and manual testing in the actual industrial facility, thereby dramatically improving testing efficiency and automation level
Solution Approach 2:
The system performs network performance prediction and validation in advance by simulating various scenarios in the digital twin environment before actual deployment. This preliminary testing allows identification and resolution of potential network issues before they affect real operations, eliminating the need for time-consuming on-site testing and reconfiguration
2Reliability
If physical network deployment is performed in industrial facilities with moving parts and changing layouts, then network coverage can be established, but stability is compromised by environmental changes
Solution Approach 1:
The system continuously monitors changes in the physical industrial environment (such as moving equipment, changing layouts, new obstructions) and feeds this information back to update the digital twin model. The digital twin then re-simulates network performance to predict how these changes affect coverage and stability, enabling proactive network optimization before actual performance degradation occurs
Solution Approach 2:
The digital twin model is designed to be dynamic rather than static, automatically updating its representation of the industrial environment as changes occur. This dynamic modeling allows the system to adapt to moving parts, changing layouts, and new obstructions, maintaining accurate network performance predictions despite environmental variability
3Measurement precision
If comprehensive network validation is performed in physical environments, then accurate performance data can be obtained, but deployment costs increase due to necessary changes and reconfigurations
Solution Approach 1:
By performing all network validation and performance measurement in the digital twin environment rather than physical deployment, the system obtains accurate performance data without incurring the costs of physical installation, testing, and reconfiguration. The digital replica eliminates material costs, equipment deployment costs, and labor costs associated with on-site testing
Data Source
Figure 1
Figure 2
Figure 3
AI summary
The embodiments relate to a method, comprising transforming a physical environment into a three-dimensional (3D) digital representation; mapping the 3D digital representation to a feature vector representation to generate a set of features; determining a performance of a wireless network at the physical environment; enabling a machine learning algorithm to learn mapping between the determined performance and the generated set of features; generating a virtual layout of the environment; and generating performance prediction by means of the machine learning algorithm inferring a performance corresponding to the generated virtual layout. The embodiments also relate to technical equipment for implementing the method.