Adaptive Channel Detection for Non-Periodic Wireless Services
Find Innovative SolutionsGenerate Solutions
Solution Overview
Problem
Existing channel detection methods in wireless communication systems fail to adapt to non-periodic service generation, leading to inefficiencies in power consumption and signaling overhead due to delayed parameter switching and inaccurate tracking of service changes.
Innovation Solution
Employing machine learning-based prediction models to predict service execution times, allowing for adaptive and personalized channel detection occasions, thereby optimizing power saving schemes and reducing signaling overhead.
Engineering Contradictions & Design Principles
Engineering Contradiction Analysis
1Use of energy by moving object
If discontinuous reception (DRX) is configured for PDCCH detection, then power consumption is reduced, but service tracking accuracy deteriorates due to inability to adapt to non-periodic service patterns
Solution Approach 1:
The patent applies dynamics by transitioning from static DRX periodic detection to dynamic prediction-based detection. The terminal uses machine learning models to dynamically adjust detection timing based on predicted service execution times, allowing the system to adapt detection behavior to actual service patterns rather than following fixed periodic intervals.
Solution Approach 2:
The terminal performs self-service by autonomously predicting its own service execution times using locally deployed machine learning models. This eliminates the need for network-side configuration adjustments and enables the terminal to independently optimize detection timing according to its actual service patterns, resolving the contradiction between power saving and tracking accuracy.
2Loss of information
If semi-static signaling is used to configure PDCCH detection parameters, then signaling overhead is reduced, but adaptability to service changes deteriorates due to delayed parameter switching
Solution Approach 1:
The patent applies preliminary action by pre-training machine learning models offline with historical service data. These pre-trained models are then deployed at the terminal, enabling it to predict future service execution times without requiring real-time network signaling. This preliminary preparation allows the system to adapt quickly to service changes while minimizing ongoing signaling overhead.
Solution Approach 2:
The patent substitutes the mechanical signaling-based parameter configuration system with an intelligent prediction-based system. Instead of relying on network-side semi-static signaling to adjust detection parameters, the terminal uses machine learning models to autonomously determine optimal detection timing, replacing the signaling mechanism with an intelligent decision-making mechanism.
3Reliability
If periodic PDCCH detection is implemented, then service coverage is improved, but power consumption increases due to unnecessary detections during idle periods
Solution Approach 1:
The patent applies parameter changes by dynamically adjusting detection timing parameters based on predicted service execution times. Instead of using fixed periodic intervals, the system modifies detection timing parameters according to learned service patterns, concentrating detections during high-probability service periods and reducing or eliminating detections during idle periods, thus maintaining service coverage while reducing power consumption.
Data Source
AI summary
A channel detection method is applied to a terminal, and includes: determining a first prediction model, the first prediction model being used for predicting a service execution time of a terminal; and according to the service execution time predicted by the first prediction model, determining a channel detection occasion.