Wireless Coverage Indicator Prediction with Environmental Path Features
Find Innovative SolutionsGenerate Solutions
Solution Overview
Problem
Existing propagation model technologies for predicting wireless network coverage indicators suffer from low efficiency and poor precision due to the complexity and variability of wireless environments.
Innovation Solution
A method and apparatus for predicting coverage indicators using a trained model that incorporates wireless cell, geographical, and propagation path features, employing a gradient-boosted tree model to improve prediction accuracy and efficiency.
Engineering Contradictions & Design Principles
Engineering Contradiction Analysis
1Measurement precision
If propagation model technology is used to estimate propagation path loss, then coverage indicator can be obtained, but prediction precision is poor
Solution Approach 1:
The patent introduces a machine learning model as an intermediary between the complex wireless environment and the coverage indicator prediction. Instead of directly using propagation models that struggle with environmental complexity, the system uses ML models trained on actual measurement data to bridge the gap, achieving higher prediction precision while maintaining reliability.
Solution Approach 2:
The patent transforms the prediction approach by changing from model-based parameter estimation to data-driven parameter learning. The system collects actual coverage measurement data and uses it to train ML models, allowing the prediction parameters to be optimized based on real-world conditions rather than theoretical assumptions, thereby improving both precision and reliability.
2Productivity
If propagation model technology is used to estimate propagation path loss, then coverage indicator can be obtained, but prediction efficiency is low
Solution Approach 1:
The patent applies preliminary action by collecting and storing coverage measurement data in advance, then using this pre-collected data to train machine learning models. This allows the system to perform rapid predictions without repeatedly executing complex propagation model calculations, significantly improving prediction efficiency while maintaining or enhancing precision through the trained models.
Solution Approach 2:
The patent creates a digital copy of the wireless environment characteristics through ML models trained on measurement data. Instead of repeatedly running computationally intensive propagation models, the system uses the trained model copies to make rapid predictions, achieving high efficiency without sacrificing the precision that would come from detailed environmental analysis.
3Measurement precision
If complex environmental features are considered for accurate prediction, then prediction precision improves, but system complexity increases
Solution Approach 1:
The patent segments the complex environmental features into distinct input categories for the machine learning model, such as geographical coordinates, cell identifiers, and measurement conditions. This segmentation allows the system to handle complex environmental factors systematically through structured data inputs, improving prediction precision while managing system complexity through organized feature processing.
Data Source
AI summary
Provided is a coverage indicator prediction method. The method includes: obtaining a wireless cell feature of a wireless cell to be predicted, a geographical environment feature of the wireless cell to be predicted, a grid geographical environment feature, and a feature of a wireless propagation path from the wireless cell to be predicted to a corresponding grid, where grids are obtained by dividing a designated region; and predicting, according to the wireless cell feature of the wireless cell to be predicted, the geographical environment feature of the wireless cell to be predicted, the grid geographical environment feature, and the feature of the wireless propagation path from the wireless cell to be predicted to the corresponding grid, a coverage indicator value of the grids using a trained coverage indicator prediction model. Coverage indicator prediction apparatus, model training method and apparatus, electronic device, and computer-readable storage medium are also provided.