Abstract:As the underlying foundation of a digital twin network (DTN), a digital twin channel (DTC) can accurately depict the process of radio propagation in the air interface to support the DTN-based 6G wireless network. Since radio propagation is affected by the environment, constructing the relationship between the environment and radio wave propagation is the key to improving the accuracy of DTC, and the construction method based on artificial intelligence (AI) is the most concentrated. However, in the existing methods, the environment information input into the neural network (NN) has many dimensions, and the correlation between the environment and the channel relationship is unclear, resulting in a highly complex relationship construction process. To solve this issue, in this paper, we propose a construction method of radio environment knowledge (REK) inspired by the electromagnetic wave property to quantify the contribution of radio propagation. Specifically, a range selection scheme for effective environment information based on random geometry is proposed to reduce the redundancy of environment information. We quantify the contribution of radio propagation reflection, diffraction and scatterer blockage using environment information and propose a flow chart of REK construction to replace the feature extraction process partially based on NN. To validate REK's effectiveness, we conduct a path loss prediction task based on a lightweight convolutional neural network (CNN) employing a simple two-layer convolutional structure. The results show that the accuracy of the range selection method reaches 90\%; the constructed REK maintains the prediction error of 0.3 and only needs 0.04 seconds of testing time, effectively reducing the network complexity.

What problem does this paper attempt to address?

The problem that this paper attempts to solve is how to effectively reduce the redundancy of environmental information and clearly establish the relationship between the environment and radio wave propagation when constructing the Digital Twin Channel (DTC) in the Digital Twin Network (DTN). Specifically, the existing methods have problems of high dimension and unclear relationship between the environment and the channel when inputting environmental information into the neural network, resulting in a high - complexity relationship construction process. To solve these problems, this paper proposes a method of constructing Radio Environment Knowledge (REK) based on the characteristics of electromagnetic waves to quantify the contribution of radio propagation. This method reduces the redundancy of environmental information by randomly geometrically selecting the range of effective environmental information and proposes the REK construction process, partially replacing the feature extraction process based on the neural network. To verify the effectiveness of REK, the author uses a lightweight Convolutional Neural Network (CNN) for the path loss prediction task, and the results show that this method not only improves the prediction accuracy but also significantly reduces the test time and simplifies the network structure. ### Main contributions 1. **Selection scheme of effective environmental information**: A selection scheme of effective environmental information is proposed based on random geometry, and its selection accuracy reaches 90% in the cases of impending blockage and complete blockage. 2. **Reflection point acquisition algorithm**: A reflection point acquisition algorithm based on the shortest path of random scatterers is proposed, and the contributions of reflection, diffraction and blockage are quantified using environmental information. 3. **Analysis of the influence of geometric topology on electromagnetic wave propagation**: The influence of environmental geometric topology on electromagnetic wave propagation is analyzed, and the implementation process of REK is proposed, including the cases of completely open, impending blockage and complete blockage. 4. **Lightweight CNN path loss prediction method**: A path loss prediction method based on a simple CNN structure is proposed, and the results show that the path loss prediction error is only 0.3, and the training and test times are 5 seconds and 0.004 seconds respectively. ### Method overview 1. **Range selection of environmental information**: The range of effective scatterers is determined by the random geometric method to reduce the redundancy of environmental information. 2. **Scatterer classification**: The blocking scatterers and non - blocking scatterers are determined according to the user location, and the scatterers with impending blockage are further determined. 3. **Reflection contribution calculation**: The reflection contribution of the scatterers with impending blockage is calculated based on the geometric optics theory, and weights are assigned to the reflection, diffraction and blockage contributions in different scenarios. 4. **Diffraction and blockage contribution calculation**: The diffraction propagation loss is quantified based on the Uniform Geometric Diffraction Theory and Fresnel principle, and the blockage contribution is quantified based on the geometric relationship. 5. **Knowledge coefficient setting**: The knowledge coefficient is set by combining prior experience and the grid - search method to enhance the interpretability between the environment and reflection, diffraction and blockage. ### Verification method To verify the effectiveness of the constructed REK, the author uses a simple two - layer Convolutional Neural Network (CNN) for the path loss prediction task. The experimental results show that this method performs well in path loss prediction, with not only a small prediction error but also a short test time and a simple network structure. In conclusion, this paper solves the problems of environmental information redundancy and complex relationship construction in the existing methods by proposing a method of constructing radio environment knowledge based on the characteristics of electromagnetic waves, providing new ideas and technical support for the construction of digital twin channels.

Electromagnetic Wave Property Inspired Radio Environment Knowledge Construction and AI-based Verification for 6G Digital Twin Channel

Towards 6G Digital Twin Channel Using Radio Environment Knowledge Pool

Learnable Wireless Digital Twins: Reconstructing Electromagnetic Field with Neural Representations

A Digital Twin for Reconfigurable Intelligent Surface Assisted Wireless Communication

ROM-Based Real-Time Analysis of Electromagnetic Performance for APAA in a Digital Twin System

A Machine Learning Based Channel Parameter Prediction Method Utilizing Ray Tracing Simulation Dataset

ChannelGPT: A Large Model to Generate Digital Twin Channel for 6G Environment Intelligence

Reactive Near-Field to 3-Meter Far-Field Transformation Based on Deep Convolutional Neural Networks and Plane Wave Spectrum

Diffraction and Scattering Aware Radio Map and Environment Reconstruction using Geometry Model-Assisted Deep Learning

An End-to-End Neural Network for Complex Electromagnetic Simulations

Spatial Channel State Information Prediction with Generative AI: Towards Holographic Communication and Digital Radio Twin

Wireless Environment Information Sensing, Feature, Semantic, and Knowledge: Four Steps Towards 6G AI-Enabled Air Interface

Digital Twin Channel for 6G: Concepts, Architectures and Potential Applications

AI-assisted Agile Propagation Modeling for Real-time Digital Twin Wireless Networks

Deep Learning-based Channel Estimation Approach for 3D Massive MIMO Millimeter-Wave System in Time-Varying Environments.

Building a Digital Twin for Large-Scale and Dynamic C+L-band Optical Networks

Vision-Assisted Digital Twin Creation for mmWave Beam Management

Outdoor Environment Reconstruction with Deep Learning on Radio Propagation Paths

Wireless Network Digital Twin for 6G: Generative AI as A Key Enabler

Generative adversarial networks based digital twin channel modeling for intelligent communication networks

A Deep-Learning Approach to a Volumetric Radio Environment Map Construction for UAV-Assisted Networks