Network Congestion Control via Adaptive Neural Model Selection
Find Innovative SolutionsGenerate Solutions
Solution Overview
Problem
Existing network congestion control methods are ineffective in managing network congestion, leading to reduced data packet transmission efficiency.
Innovation Solution
A method and apparatus for controlling data packet sending, which involves obtaining multiple control models corresponding to different conditions, selecting the appropriate control model based on the current congestion control requirement, processing network transmission parameters to obtain control parameters, and adjusting data packet sending accordingly to meet the congestion control requirements.
Engineering Contradictions & Design Principles
Engineering Contradiction Analysis
1Reliability
If a congestion control algorithm is used to obtain control parameters, then data packet sending can be controlled, but network congestion cannot be effectively resolved
Solution Approach 1:
The patent transforms the congestion control problem from traditional parameter adjustment to a machine learning prediction problem. Multiple neural network models are trained to predict optimal control parameters (congestion window, transmit rate) based on different network conditions, enabling dynamic adaptation to changing network states and effectively resolving congestion while maintaining high transmission efficiency.
Solution Approach 2:
The patent divides the congestion control problem into multiple segments by training separate neural network models for different network conditions and scenarios. Each model is specialized for specific conditions, and the appropriate model is selected based on current network state, allowing precise control without compromising overall transmission efficiency.
2Measurement precision
If multiple control models are used for different conditions, then congestion control accuracy is improved, but system complexity increases
Solution Approach 1:
The system implements self-service through automated model selection based on current network conditions. The end node automatically selects the appropriate pre-trained model according to detected network parameters, eliminating the need for manual model management and reducing operational complexity while maintaining high control accuracy.
Solution Approach 2:
Multiple neural network models are pre-trained offline for different network conditions before deployment. This preliminary action allows the system to have ready-to-use models for various scenarios, reducing runtime complexity as the system only needs to select from pre-trained models rather than train them in real-time.
Data Source
AI summary
A controlling data packet sending method. The method for controlling data packet sending includes: obtaining a plurality of control models, where the plurality of control models corresponds to a plurality of conditions; selecting, based on a first condition, a first control model corresponding to the first condition from the plurality of control models, the first control model is used to process a transmission parameter of any network in at least one network; processing, by using the first control model, a transmission parameter of a first network accessed by an end node, to obtain a first control parameter required for sending the data packet of the first application by the end node by using the first network; and controlling sending of the data packet of the first application based on the first control parameter.