Congestion-Based Transfer with Real-Time Network Prediction
Find Innovative SolutionsGenerate Solutions
Solution Overview
Problem
Communication systems face challenges in predicting network performance due to varying conditions, leading to suboptimal application recommendations and degraded system performance, as applications often operate under the assumption of consistent network availability.
Innovation Solution
A network congestion reporting (NCR) system utilizing artificial intelligence and machine learning models to provide real-time network congestion predictions and exposure, allowing applications to make informed decisions based on actual network conditions.
Engineering Contradictions & Design Principles
Engineering Contradiction Analysis
1Ease of operation
If applications operate under the assumption of consistent network availability, then application simplicity is maintained, but system performance degrades due to suboptimal recommendations
Solution Approach 1:
The patent introduces a network congestion reporting system as an intermediary between the network infrastructure and applications. This mediator collects network performance data, processes it through machine learning models to predict congestion, and exposes this information to applications via a standardized interface. Applications can then make informed decisions about task scheduling without needing to understand complex network internals, resolving the contradiction between operational simplicity and performance optimization.
Solution Approach 2:
The system implements a feedback loop where network congestion data is continuously collected, analyzed, and fed back to applications. The machine learning models process historical and real-time network data to generate predictions about future congestion states. This feedback mechanism enables applications to adapt their behavior based on predicted network conditions, improving system performance while maintaining application simplicity through automated decision support.
2Measurement precision
If real-time network congestion prediction is implemented, then task scheduling accuracy is improved, but device complexity increases
Solution Approach 1:
The patent segments the network congestion prediction functionality into distinct modular components: data collection modules that gather network performance metrics, machine learning model modules that process the data and generate predictions, and exposure modules that interface with applications. This segmentation allows each component to be independently developed, optimized, and maintained, reducing overall system complexity while maintaining high prediction accuracy.
Solution Approach 2:
The network congestion reporting system operates autonomously, automatically collecting network data, training and updating machine learning models, generating predictions, and exposing this information to applications without requiring manual intervention. The system self-manages the complexity of implementing real-time prediction, allowing applications to benefit from accurate measurements without bearing the burden of system complexity.
3Measurement precision
If network congestion data is collected and processed, then bandwidth prediction accuracy is improved, but loss of time increases due to data processing requirements
Solution Approach 1:
The system performs preliminary actions by continuously collecting and preprocessing network congestion data in the background, even when not immediately needed for predictions. Machine learning models are trained offline on historical data, and prediction models are pre-computed for upcoming time windows. This preliminary processing reduces the computational burden during real-time decision-making, maintaining high prediction accuracy while minimizing the time loss during actual task scheduling operations.
Data Source
AI summary
In some implementations, a device may receive a request as to whether a task is able to be fulfilled while satisfying one or more conditions, wherein the task is associated with a user equipment (UE) in a wireless network. The device may determine, based on the request, a network congestion state at a location associated with the UE. The device may determine, based on the network congestion state, whether the task is able to be fulfilled while satisfying the one or more conditions. The device provides a response to the request.