Mobile Terminal Machine Learning Model Delivery via Control Plane
Find Innovative SolutionsGenerate Solutions
Solution Overview
Problem
The challenge in mobile communication systems is efficiently providing machine learning models to mobile terminals, especially due to constrained power and computational resources on the terminal side, which makes on-device training difficult.
Innovation Solution
A method is proposed where a mobile radio communication network determines the need to deliver a machine learning model to a mobile terminal and provides it via control plane communication, allowing the terminal to use the model for machine learning tasks.
Engineering Contradictions & Design Principles
Engineering Contradiction Analysis
1Reliability
If machine learning model training is performed on the mobile terminal side, then model customization and performance can be improved, but power consumption and computational resource requirements increase significantly
Solution Approach 1:
The patent extracts the model training function from the mobile terminal and relocates it to the network side. The network entity performs model training using collected data, then delivers the trained model to the terminal. This separation allows the terminal to benefit from customized models without bearing the computational and energy costs of training.
Solution Approach 2:
The patent introduces a network entity as an intermediary between data collection and model deployment. This intermediary collects training data from multiple terminals, performs centralized model training, and distributes the trained model back to terminals. This mediator approach enables efficient resource utilization and reduces individual terminal power consumption.
2Use of energy by moving object
If machine learning models are trained on the network side, then power consumption on terminals is reduced, but model delivery complexity and network resource requirements increase
Solution Approach 1:
The patent creates a universal model delivery mechanism that handles multiple functions: model training, model validation, model selection, and model distribution through a single integrated system. The network entity serves multiple terminals simultaneously, reducing overall system complexity compared to individual terminal-based training.
Solution Approach 2:
The patent performs model training in advance on the network side before terminals need the models. Training data is collected beforehand, models are trained and validated in advance, and ready-to-deploy models are stored in the network entity. This preliminary action eliminates the need for complex real-time training coordination.
3Adaptability or versatility
If machine learning models are delivered via user plane communication, then data transmission flexibility is improved, but control and security management become more difficult
Solution Approach 1:
The patent implements a feedback mechanism where the network entity receives confirmation from terminals about model reception and deployment status. This feedback loop enables the network to monitor model distribution, ensure proper delivery, and maintain control over the model lifecycle while preserving transmission flexibility.
Data Source
Figure 1
Figure 2
Figure 3
AI summary
A method for providing a machine learning model to a mobile terminal of a mobile communication system is described comprising determining, by a mobile radio communication network of the communication system, that a machine learning model should be delivered to a mobile terminal to be used by the mobile terminal for a machine learning task and providing the requested machine learning model to the mobile terminal via control plane communication.