Traffic Volume Prediction for Dynamic 5G Resource Allocation
Find Innovative SolutionsGenerate Solutions
Solution Overview
Problem
Existing network optimization mechanisms fail to address the dynamic resource allocation needs of 5G services, leading to improper allocation of radio resources, which affects service quality due to excessive or insufficient resources at certain times and locations.
Innovation Solution
A traffic volume prediction method that involves a first management device receiving a request for traffic volume prediction from a second management device, determining predicted traffic volumes based on prediction requirements, and allocating resources accordingly to meet service demands.
Engineering Contradictions & Design Principles
Engineering Contradiction Analysis
1Stability of the object's composition
If radio resources are statically allocated, then resource allocation is simple and stable, but service requirements are affected due to excessive or insufficient resources at certain time points and areas
Solution Approach 1:
The patent applies preliminary action by predicting future traffic volumes before resource allocation is needed. The system forecasts traffic patterns for different time periods and areas, then proactively adjusts radio resource allocation in advance to match predicted demands, avoiding both resource shortages and waste while maintaining allocation stability.
Solution Approach 2:
The patent implements dynamics by transitioning from static to dynamic resource allocation. The system continuously updates resource allocation based on real-time and historical traffic data, enabling flexible adaptation to changing service requirements while maintaining overall stability through controlled adjustment mechanisms.
2Reliability
If network optimization is performed after performance deterioration is detected, then resource allocation can be adjusted, but service requirements are affected during the optimization process due to time-consuming operations
Solution Approach 1:
The patent applies preliminary action by detecting potential performance deterioration trends before actual service quality degradation occurs. The system analyzes traffic patterns and network metrics proactively, triggering optimization adjustments in advance to prevent service quality issues rather than reacting after problems manifest, thereby reducing the time impact on services.
Solution Approach 2:
The patent implements feedback mechanisms that continuously monitor network performance and adjust resource allocation in real-time. The system uses feedback loops to compare actual traffic with predicted traffic, automatically triggering optimization actions when deviations are detected, ensuring service quality is maintained without time-consuming manual intervention.
3Measurement precision
If traffic volume prediction is performed with high precision, then resource allocation can be optimized, but system complexity increases due to data collection and analysis requirements
Solution Approach 1:
The patent applies segmentation by dividing the prediction system into modular components: data collection modules for different network elements, analysis modules for different traffic types, and allocation modules for different time periods. This segmentation enables high-precision predictions while managing system complexity through standardized, reusable prediction units that can be independently configured and scaled.
Solution Approach 2:
The patent implements universality by creating a multi-functional prediction system that handles multiple traffic types, time periods, and network elements through a unified framework. The same core prediction algorithms and data collection mechanisms serve various prediction scenarios, reducing overall system complexity while maintaining high precision across different application cases.
Data Source
AI summary
This application provides a traffic volume prediction method and an apparatus. The method includes: A second management device sends a first message to a first management device, where the first message is used to request the first management device to perform traffic volume prediction, the first message includes a predicted object, a traffic type, and prediction requirement information, and the prediction requirement information includes a prediction granularity and/or a prediction period, where the prediction granularity includes at least one of the following: a prediction area, a prediction service type, a prediction slice, a prediction cell, a prediction public land mobile communications network PLMN, or a prediction tenant. The first management device receives the first message, and determines, for the predicted object based on the prediction requirement information, a traffic volume corresponding to the traffic type.