AI/ML Data Collection Entities in 5G Wireless Networks
Find Innovative SolutionsGenerate Solutions
Solution Overview
Problem
Current data collection mechanisms and procedures in 3GPP networks are inadequate for addressing the unique requirements of AI/ML operations, and there is a lack of frameworks and network entities to handle model monitoring effectively, leading to performance degradation and reduced quality of experience.
Innovation Solution
The implementation of novel mechanisms, frameworks, signaling procedures, and new network entities and functions for AI/ML data collection and model monitoring in 3GPP networks, including methods for transmitting data collection and monitoring requests, receiving responses, and performing operations such as model registration, activation, and deactivation.
Engineering Contradictions & Design Principles
Engineering Contradiction Analysis
1Reliability
If existing data collection mechanisms are used, then basic network operations are maintained, but AI/ML specific requirements are not met leading to performance degradation
Solution Approach 1:
The patent segments the data collection functionality into specialized network entities (Data Collection Entity, Model Monitoring Entity) that are specifically designed for AI/ML operations. This segmentation allows the system to collect and monitor AI/ML-specific data separately from general network operations, ensuring that AI/ML models receive the appropriate data quality and types needed for reliable performance.
Solution Approach 2:
The patent introduces intermediary network entities that act as mediators between the radio access network and AI/ML models. The Data Collection Entity intermediates data collection from multiple sources, while the Model Monitoring Entity intermediates model performance monitoring. These intermediaries adapt and transform data to meet AI/ML specific requirements, resolving the contradiction between general data collection and AI/ML specific needs.
2Reliability
If new network entities for AI/ML monitoring are introduced, then model performance can be monitored effectively, but device complexity increases
Solution Approach 1:
The patent designs the new network entities with multi-functionality to reduce overall system complexity. The Data Collection Entity can collect data for multiple AI/ML models simultaneously and perform various data processing functions. The Model Monitoring Entity can monitor multiple models and perform different monitoring tasks. This universality allows a single entity to handle multiple functions, reducing the number of separate components needed.
Solution Approach 2:
The patent merges related functions into unified network entities. Instead of having separate entities for data collection, data processing, and model monitoring, the patent combines these functions into integrated entities that perform multiple related operations. This merging reduces the overall number of network entities and simplifies the network architecture while maintaining effective model monitoring capability.
Data Source
AI summary
The disclosure relates to a 5th generation (5G) or 6th generation (6G) communication system for supporting a higher data transmission rate. A method for supporting Artificial Intelligence/Machine Learning (AI/ML) by a first network entity in a wireless communication system is provided. The method includes transmitting, to a second network entity, a data collection request message related to data collection for supporting the AI/ML, and receiving, from the second network entity, a data collection response message based on a successful operation related to the data collection.