AI-Driven Access Device Traffic Configuration
Find Innovative SolutionsGenerate Solutions
Solution Overview
Problem
Current communication network systems are not adequately equipped to support machine learning-based applications, particularly in terms of data collection, processing, and output distribution, hindering their ability to effectively manage complex network tasks.
Innovation Solution
An extended framework communication protocol is introduced, utilizing an AI/ML model to process traffic statistics data, enabling the access device to transmit and receive traffic configuration information with a computing device, which then configures the network accordingly, optimizing bandwidth allocation and QoS control.
Engineering Contradictions & Design Principles
Engineering Contradiction Analysis
1Adaptability or versatility
If current communication network systems are used, then basic network operations are maintained, but they cannot support machine learning-based applications for data collection, processing, and output distribution
Solution Approach 1:
The system is divided into distinct functional modules: access devices for data collection, computing devices for AI/ML processing, and separate configuration management components. This segmentation allows each module to specialize in specific tasks (data collection, processing, or configuration), enabling machine learning support without requiring complete system redesign.
Solution Approach 2:
The extended framework protocol serves multiple functions: it enables data collection, AI/ML processing, traffic configuration, and network management within a unified architecture. The protocol can handle both traditional network operations and machine learning-based applications, providing versatility without proportionally increasing complexity.
2Productivity
If AI/ML models are integrated into network configuration, then bandwidth allocation and QoS control are optimized, but data processing complexity increases
Solution Approach 1:
The extended framework protocol acts as an intermediary layer between raw traffic statistics data and AI/ML processing. It standardizes data collection, processing, and output distribution formats, reducing the complexity of integrating AI/ML models while enabling optimized bandwidth allocation and QoS control through machine learning.
Solution Approach 2:
The system enables self-service through automated AI/ML-based traffic configuration. The computing device automatically processes traffic statistics, generates optimization configurations, and applies them without manual intervention, improving productivity while the automated nature reduces operational complexity.
3Manufacturing precision
If traffic statistics data is collected and processed through AI/ML models, then network configuration accuracy is improved, but data collection and transmission overhead increases
Solution Approach 1:
The system extracts only the essential traffic statistics data needed for AI/ML processing through the extended framework protocol, rather than transmitting all raw network data. This selective extraction maintains configuration accuracy while reducing data transmission overhead by focusing on relevant metrics.
Data Source
AI summary
Examples of the disclosure relate to an access device, a computing device, a related method, an apparatus, and a medium. In an aspect, the access device transmits traffic statistics data related to traffic passing through the access device to the computing device based on at least one statistics data model for statistics data; receives traffic configuration information for the traffic from the computing device based on at least one configuration data model for traffic configuration, wherein the traffic configuration information is determined by utilizing an artificial intelligence (AI)/machine learning (ML) model to process the traffic statistics data; and configures the traffic based on the traffic configuration information. The examples of the disclosure extend the statistics data model/configuration data model, and can optimize traffic classification, bandwidth allocation, quality of service (QOS) control, etc. based on the AI/ML model.