Combined ML Block Configuration for Wireless Signaling Reduction
Find Innovative SolutionsGenerate Solutions
Solution Overview
Problem
Current wireless communication systems, particularly in 5G NR, face challenges in efficiently configuring machine learning (ML) models for user equipment (UE) to balance performance and complexity, leading to potential degraded performance due to unbalanced model complexity and signaling costs.
Innovation Solution
The solution involves configuring a combined ML model by associating a backbone/general block with a specific/dedicated block, where the UE receives separate configurations for each block type, allowing for flexible task association and reduced signaling costs, and enabling the UE to indicate its capability for optimal parameter configuration.
Engineering Contradictions & Design Principles
Engineering Contradiction Analysis
1Reliability
If a single complete ML model is configured for each task, then task performance is improved, but device complexity and signaling overhead increase
Solution Approach 1:
The patent segments a complete ML model into two independent blocks: a backbone block that performs general processing and a dedicated block that performs task-specific processing. This segmentation allows the system to configure only the necessary dedicated block for each task rather than deploying complete separate models, thereby reducing device complexity and signaling overhead while maintaining task performance.
2Reliability
If multiple complete ML models are configured for different tasks, then task-specific performance is improved, but signaling overhead and configuration complexity increase
Solution Approach 1:
The backbone block serves as a universal component that can be shared across multiple tasks, while only the dedicated blocks need to be reconfigured for task-specific requirements. This multi-functionality approach reduces signaling overhead because the same backbone block configuration can be reused for different tasks, eliminating the need to signal complete model configurations for each task.
3Measurement precision
If ML model parameters are optimized for each task, then processing accuracy is improved, but configuration complexity and processing time increase
Solution Approach 1:
The backbone block is pre-configured with general processing parameters that can be reused across multiple tasks. When a new task is introduced, only the dedicated block parameters need to be configured, rather than configuring the entire model from scratch. This preliminary configuration of the backbone block significantly reduces configuration time while maintaining processing accuracy through task-specific dedicated block optimization.
Data Source
AI summary
A UE may receive a first configuration for at least one first ML block and a second configuration for at least one second ML block. The at least one first ML block may be configured with at least one first parameter for a first procedure and the at least one second ML block may be configured with at least one second parameter for a second procedure. The at least one second ML block may be dedicated to a task included in a plurality of tasks associated with the at least one first ML block. The UE may activate an ML model based on an association of the at least one second ML block configured with the at least one second parameter with the at least one first ML block configured with the at least one first parameter.