Radio Channel State Prediction Using Machine-Trained Models
Find Innovative SolutionsGenerate Solutions
Solution Overview
Problem
Current methods for assessing radio channel states in wireless communication networks are inefficient, as they often require lengthy simulations using raytracing, which can slow down decision-making processes and increase processing time, especially in dynamic industrial environments where rapid predictions are necessary.
Innovation Solution
A device and method that utilize a machine-trained model to predict radio channel states, combining digital representations of the surroundings with raytracing, allowing for accelerated predictions and improved decision-making by selecting optimal candidate trajectories and radio lobe coordinates, and coordinating the operation between model-based ascertainment and raytracing units based on confidence levels and accuracy requirements.
Engineering Contradictions & Design Principles
Engineering Contradiction Analysis
1Measurement precision
If raytracing is used to simulate channel propagation, then measurement precision is improved, but processing time increases
Solution Approach 1:
The patent creates a digital twin (virtual copy) of the physical environment including buildings, objects, and spatial arrangements. This digital representation is used to train machine learning models that can predict channel propagation characteristics without performing time-consuming raytracing simulations for each prediction, thus maintaining accuracy while reducing processing time.
Solution Approach 2:
The patent transforms the raytracing simulation process into a machine learning prediction process by changing the fundamental approach from physical simulation to data-driven parameter estimation. The machine learning model learns the relationship between environmental parameters (digital twin data) and channel propagation characteristics, enabling fast predictions without re-running raytracing simulations.
2Loss of time
If machine-trained model is used for prediction, then processing time is reduced, but device complexity increases
Solution Approach 1:
The patent performs preliminary actions by training the machine learning model offline using extensive raytracing simulation data and digital twin representations. Once trained, the model can make predictions in real-time without requiring complex computational resources during actual deployment, thus reducing processing time while the complexity is contained in the pre-trained model parameters.
Solution Approach 2:
The patent introduces a machine learning model as an intermediary between the digital twin environment and the channel propagation prediction. This intermediary component processes environmental parameters and outputs channel state information efficiently, bridging the gap between detailed environmental modeling and fast prediction requirements without requiring direct complex raytracing computations during prediction.
3Adaptability or versatility
If multiple candidate trajectories are assessed, then adaptability is improved, but processing time increases
Solution Approach 1:
The patent replaces the mechanical raytracing simulation process with a machine learning-based prediction system. Instead of performing computationally intensive raytracing for each candidate trajectory, the system uses pre-trained machine learning models that can rapidly predict channel propagation characteristics for multiple trajectories, thus maintaining adaptability while significantly reducing processing time.
Data Source
AI summary
A method for assessing a state of a radio channel. The method includes: ascertaining or providing a piece of state information, which characterizes a simulated state of a spatial arrangement of components of the surroundings of the wireless communication network; ascertaining at least one prediction based on a machine-trained model, the state information being provided as an input parameter in an input section of a machine-trained model, the state information being propagated by the machine-trained model, and the at least one prediction of a piece of channel state information based on the machine-trained model, which characterizes a state of at least one radio channel between two communication modules, being provided in an output section of the machine-trained model.