Trained Models for Wireless Channel Spread Estimation
Find Innovative SolutionsGenerate Solutions
Solution Overview
Problem
Estimating Doppler spread and delay spread in wireless communication systems is challenging and time-consuming, affecting the performance of these systems.
Innovation Solution
An apparatus and method using trained models, specifically one-dimensional convolutional neural networks with residual neural network blocks, to determine Doppler and delay spread estimations from received data, enabling efficient reconstruction of wireless channel characteristics.
Engineering Contradictions & Design Principles
Engineering Contradiction Analysis
1Measurement precision
If traditional methods are used to estimate Doppler spread and delay spread, then measurement precision can be achieved, but the process becomes time-consuming and reduces productivity
Solution Approach 1:
The system performs preliminary action by training machine learning models offline using historical channel data. The trained models capture complex channel characteristics in advance, enabling rapid real-time spread estimation without repeating computationally intensive analysis during actual operation. This resolves the contradiction by preparing solutions beforehand that deliver both accuracy and speed during deployment.
Solution Approach 2:
The invention creates simplified copies of complex channel behavior through trained ML models that replicate the relationship between received signals and spread parameters. Instead of performing full channel analysis repeatedly, the system uses these learned models to quickly estimate spreads by copying patterns from training data, achieving both precision and computational efficiency.
2Measurement precision
If complex algorithms are used to improve spread estimation accuracy, then measurement precision improves, but device complexity increases
Solution Approach 1:
The system changes parameters by selecting and tuning specific ML model architectures (such as neural network layer configurations, tree ensemble sizes) to achieve optimal balance between accuracy and complexity. By adjusting model parameters and selecting appropriate complexity levels for different deployment scenarios, the system maintains high estimation precision while controlling device resource requirements.
Solution Approach 2:
The invention segments the estimation task by using separate specialized models for delay spread and Doppler spread estimation, and further segments training into offline preparation and online inference phases. This segmentation allows each component to be optimized independently, reducing overall system complexity while maintaining comprehensive estimation accuracy.
Data Source
AI summary
To obtain delay spread estimations and/or Doppler spread estimations, data representing received data is input to at least one trained model, the trained model outputting spread estimations.