Selective Local Model Parameter Communication for 5G Signaling
Find Innovative SolutionsGenerate Solutions
Solution Overview
Problem
The transmission of large-scale model parameters in a 5G network leads to significant signaling overheads, which is a challenge in the context of AI integration within 5G networks.
Innovation Solution
A communication method that involves determining and transmitting only a part of the local model parameters based on received information, reducing the need for full model parameter transmission, thereby minimizing signaling overheads and energy consumption.
Engineering Contradictions & Design Principles
Engineering Contradiction Analysis
1Reliability
If all local model parameters are transmitted over the network, then the model training data is complete, but the signaling overheads become huge
Solution Approach 1:
The patent extracts and transmits only the necessary part of local model parameters (those not yet present in the target model) rather than transmitting all parameters. This selective extraction reduces signaling overhead while maintaining the completeness of training data for the target model.
Solution Approach 2:
The patent applies partial action by transmitting only a subset of model parameters that are needed for training, rather than transmitting the complete set. The first apparatus determines which parameters to send based on what the second apparatus needs, reducing unnecessary transmission overhead.
2Reliability
If all local model parameters are calculated and prepared for transmission, then the model training is thorough, but the calculation amount and energy consumption increase
Solution Approach 1:
The first apparatus extracts only the necessary local model parameters that need to be transmitted to the second apparatus, rather than calculating and preparing all parameters. This reduces the calculation amount and associated energy consumption while maintaining training thoroughness.
Solution Approach 2:
The patent applies partial action by performing calculations only for the subset of parameters that will be transmitted, rather than calculating all parameters. This partial computation reduces energy consumption while ensuring the training process receives all necessary data.
Data Source
AI summary
Embodiments of this disclosure provide a communication method and a related apparatus, to reduce signaling overheads of sending a local model parameter of a first model by a first apparatus. The first apparatus receives first information from a second apparatus, where the first information indicates whether the first apparatus sends each local model parameter of the first model of the first apparatus. The first apparatus determines a part of to-be-sent local model parameters of the first model based on the first information, where the part of the local model parameters is obtained by training the first model. The first apparatus sends the part of the local model parameters to the second apparatus.