5G Throughput Prediction via UE Context
Find Innovative SolutionsGenerate Solutions
Solution Overview
Problem
Commercial 5G networks face challenges in achieving reliable and efficient throughput due to the high variability of mmWave signals, which are sensitive to environmental and mobility factors, leading to fluctuating performance and high energy consumption, and require complex tradeoffs in edge computing applications.
Innovation Solution
A machine learning framework is developed to predict 5G throughput by identifying key user equipment (UE) side factors and constructing a performance model using a wide variety of features, allowing for context-aware predictions and adaptive augmentation with 4G networks to ensure reliable edge offloading.
Engineering Contradictions & Design Principles
Engineering Contradiction Analysis
1Speed
If 5G mmWave radio is used to achieve high throughput, then speed is improved, but reliability deteriorates due to signal attenuation and blockage
Solution Approach 1:
The patent applies dynamics by making the system adaptively switch between 5G mmWave and 4G LTE networks based on real-time throughput predictions and environmental conditions. The machine learning model continuously monitors UE-side features and dynamically selects the optimal network, allowing the system to capitalize on 5G's high speed when conditions permit while falling back to 4G's reliability when mmWave signals are blocked or attenuated.
2Productivity
If 5G mmWave radio is used to achieve high throughput, then productivity is improved, but energy consumption increases
Solution Approach 1:
The patent applies parameter changes by using the machine learning model to predict throughput based on UE-side features such as location, mobility, and environmental conditions. This prediction enables the system to adjust network selection parameters dynamically, choosing 5G mmWave only when predicted throughput justifies the higher energy consumption, and selecting 4G LTE when energy efficiency is prioritized or 5G performance is unlikely to be superior.
3Measurement precision
If machine learning framework is used to predict 5G throughput, then measurement precision is improved, but device complexity increases
Solution Approach 1:
The patent applies the taking out principle by extracting only the most critical UE-side features (location, mobility, environmental conditions) for the machine learning prediction model, rather than processing all possible parameters. This selective extraction maintains high prediction accuracy while minimizing the computational burden and complexity added to the mobile device.
4Adaptability or versatility
If adaptive network selection is implemented, then adaptability is improved, but device complexity increases
Solution Approach 1:
The patent applies self-service by implementing an automated machine learning-based system that autonomously predicts 5G throughput and selects the optimal network without requiring manual configuration or complex user intervention. The system self-manages the complexity of monitoring multiple UE-side features, running predictions, and making network selection decisions, thereby providing high adaptability while keeping the user-facing complexity low.
Data Source
AI summary
A system and method for predicting one or more cellular performance parameters associated with user equipment (UE) within a three-dimensional (3D) space having one or more cellular nodes, the cellular nodes including one or more cellular nodes, including a 5G cellular node. For each of one or more of pieces of UE within the 3D space, determine values associated with one or more UE-side features of each piece of UE. Predict values of the one or more cellular performance parameters for each UE as a function of the values associated with the one or more UE-side features of each respective piece of UE, wherein predicting values of the one or more cellular performance parameters includes applying the values determined for each respective piece of UE to a machine learning module trained using truth data associated with the one or more UE-side features.