5G Usage Data Forecasting via Multi-Model Segmentation
Find Innovative SolutionsGenerate Solutions
Solution Overview
Problem
There is a need for an improved method to forecast 5G usage data in areas without 5G connectivity, as existing methods are not effective due to the lack of historical 5G usage data and significant differences in data distributions between 5G and non-5G areas.
Innovation Solution
A computer-implemented method that uses a multi-model approach to predict 5G usage data. This involves generating intermediate predictions using models trained on target and reference geographical areas, combining non-network, non-5G network, and 5G network data to forecast future 5G usage.
Engineering Contradictions & Design Principles
Engineering Contradiction Analysis
1Measurement precision
If traditional forecasting methods are used in areas without 5G connectivity, then the forecasting process is simple, but the accuracy of 5G usage data prediction is poor due to lack of historical data and data distribution differences
Solution Approach 1:
The forecasting process is divided into three distinct models: a first model that generates intermediate predictions using target area data, a second model that generates intermediate predictions using reference area data, and a third model that combines these intermediate predictions to produce the final 5G usage forecast. This segmentation allows each model to specialize in specific data transformations while collectively achieving high forecasting accuracy without requiring a single overly complex model
Solution Approach 2:
Intermediate predictions are introduced as mediating elements between the input data and the final 5G usage forecast. The first model generates intermediate predictions from target area non-5G data, the second model generates intermediate predictions from reference area data, and the third model combines these intermediate predictions to produce the final forecast. This intermediary approach enables accurate forecasting by breaking down the complex transformation into manageable steps
2Measurement precision
If multiple models are used to generate intermediate predictions, then the forecasting accuracy is improved, but the computational resources and processing time increase
Solution Approach 1:
The first model performs preliminary action by generating intermediate predictions from target area non-5G data before the final forecasting step. This intermediate prediction captures local patterns and characteristics that would be difficult to extract directly in the final model, thereby improving overall forecasting accuracy while distributing computational work across multiple specialized models
3Measurement precision
If reference geographical areas with 5G connectivity are used for training, then the forecasting model can leverage existing data patterns, but the model may not accurately predict usage in target areas without 5G connectivity due to data distribution differences
Solution Approach 1:
The model applies local quality by training different models on different data characteristics: the first model is trained on target area non-5G data to capture local patterns specific to the target area, while the second model is trained on reference area data to capture general 5G usage patterns. The third model then combines these locally-optimized predictions, allowing the system to leverage both local and general patterns for improved forecasting accuracy while adapting to different data distributions
Data Source
AI summary
A computer-implemented method comprising predicting 5G usage data including generating at least one of first to third 5G usage data predictions, wherein generating the first 5G usage data prediction comprises using a third model to generate the first 5G usage data prediction based on predicted non-network data of a first intermediate prediction and predicted non-5G network usage data of a second intermediate prediction, wherein generating the second 5G usage data prediction comprises using the third model to generate the second 5G usage data prediction based on predicted non-network data of the first intermediate prediction and predicted non-5G network usage data of a third intermediate prediction, and wherein generating the third 5G usage data prediction comprises using the third model to generate the third 5G usage data prediction based on the predicted non-network data of the first intermediate prediction and predicted non-5G network usage data of a sixth intermediate prediction.