AI Service Prediction for Wireless Resource Scheduling
Find Innovative SolutionsGenerate Solutions
Solution Overview
Problem
Current wireless communication systems fail to effectively utilize the temporal correlation of service generation times and volumes, leading to inefficient scheduling and resource allocation.
Innovation Solution
Implement AI models in radio access network devices and terminals to predict channel states and service characteristics, enabling advanced resource reservation and scheduling based on first and second prediction information.
Engineering Contradictions & Design Principles
Engineering Contradiction Analysis
1Productivity
If traditional scheduling methods are used without AI prediction, then device complexity is low, but system efficiency and resource allocation are poor
Solution Approach 1:
The patent applies preliminary action by using AI models to predict future channel states and service characteristics before actual communication occurs. The system performs prediction in advance based on historical data and patterns, enabling proactive resource allocation and scheduling rather than reactive responses, thereby improving system efficiency while managing complexity through automated prediction mechanisms
Solution Approach 2:
The system implements self-service by having terminals and network devices autonomously generate and exchange prediction information. Terminals perform local predictions of channel states and service characteristics, while network devices utilize these predictions for autonomous scheduling decisions, reducing the need for complex centralized control and improving overall system efficiency through distributed intelligence
2Measurement precision
If prediction information is collected and processed from multiple sources, then measurement precision of service characteristics is improved, but loss of time increases
Solution Approach 1:
The patent applies preliminary action by performing predictions in advance based on historical service data and channel state information. The system pre-calculates predicted values for service characteristics and channel conditions, so that when actual communication occurs, the predictions are already ready, reducing processing time while maintaining high measurement precision through sophisticated prediction algorithms
Solution Approach 2:
The system implements feedback mechanisms where prediction results are continuously refined based on actual observed values. The network device receives feedback information about actual service characteristics and channel states, compares these with predicted values, and uses the differences to improve future predictions. This feedback loop enables the system to maintain high precision while reducing time loss by learning from past performance and continuously optimizing prediction accuracy
3Productivity
If AI models are deployed for prediction, then productivity is improved through better scheduling, but device complexity and computational requirements increase
Solution Approach 1:
The patent applies segmentation by dividing the AI prediction system into multiple components distributed across different devices. The complexity is segmented between terminals performing local channel state predictions and network devices performing service characteristic predictions. Each component handles specific prediction tasks independently, reducing the computational burden on any single device while maintaining overall system productivity through coordinated operation
Solution Approach 2:
The system implements parameter changes by dynamically adjusting prediction parameters and model complexity based on actual communication conditions. The AI models adapt their prediction granularity, time horizon, and computational intensity according to current channel conditions, service types, and network load, thereby improving productivity when needed while reducing complexity during normal operations
Data Source
Figure 1~3
Figure 4~5
Figure 6~7
AI summary
The present disclosure relates to the technical field of communications, and relates to a service prediction method and device, and a storage medium, for use in improving system efficiency and reducing ineffective scheduling. The method comprises: receiving prediction information, the prediction information including at least one of first prediction information and second prediction information, wherein the first prediction information is used for indicating a channel state of a terminal at a first moment, and the second prediction information is determined on the basis of service characteristics of the terminal.