ML-Based RACH Carrier Selection Under Changing Channel Conditions
Find Innovative SolutionsGenerate Solutions
Solution Overview
Problem
Existing wireless communication systems face challenges in accurately selecting uplink carriers for random access channel procedures due to infrequent evaluation of reference signal reserve power, leading to suboptimal carrier selection based on outdated channel conditions.
Innovation Solution
Adaptive carrier selection using machine learning models that monitor channel conditions during random access channel attempts to generate predictions for selecting the most likely successful carrier, allowing for dynamic switching and early radio link failure declaration.
Engineering Contradictions & Design Principles
Engineering Contradiction Analysis
1Measurement precision
If traditional carrier selection methods are used based on infrequent reference signal reserve power evaluation, then device complexity is reduced, but carrier selection accuracy deteriorates due to outdated channel conditions
Solution Approach 1:
The patent implements dynamic carrier selection by continuously monitoring channel conditions during RACH attempts and using machine learning models to adapt carrier choices in real-time, transforming the static carrier selection process into a dynamic one that responds to changing wireless conditions
Solution Approach 2:
The system establishes a feedback loop where channel conditions are monitored during RACH attempts, ML models predict successful carriers based on this feedback, and subsequent RACH attempts use these predictions, creating a continuous improvement cycle that enhances selection accuracy
2Reliability
If machine learning models are deployed for adaptive carrier selection, then random access success rate is improved, but device complexity increases due to additional monitoring and processing requirements
Solution Approach 1:
The system performs preliminary channel condition monitoring and ML model predictions before executing RACH attempts, preparing carrier selection decisions in advance based on analyzed channel data, which improves success rates by avoiding reactive carrier selection
Solution Approach 2:
The device autonomously performs channel monitoring, data collection, ML model execution, and carrier selection without external intervention, enabling the system to self-optimize carrier choices based on observed channel conditions and improve its own reliability
3Measurement precision
If channel conditions are monitored frequently during each RACH attempt, then carrier selection accuracy is improved, but energy consumption increases
Solution Approach 1:
The system monitors channel conditions continuously during RACH attempts (excessive action) to gather comprehensive data for ML model training, while using these models to reduce the number of actual RACH attempts needed, thereby compensating for the increased monitoring energy through reduced transmission energy
Data Source
AI summary
Methods, systems, and devices for wireless communications are described. In some examples, a UE may select a carrier from a plurality of configured carriers for a random access channel (RACH) procedure on a cell using a machine learning model. The UE may monitor parameters during each RACH attempt of the RACH procedure to determine a channel dataset associated with the cell. Based on a channel dataset for a current RACH procedure, the ML model may predict a likelihood of successfully performing a RACH procedure for each carrier. The UE 115 may select a carrier based on the prediction and may perform the RACH procedure. In some examples, the UE may switch carriers after failing a quantity of RACH attempts of the RACH procedure. Additionally, or alternatively, the UE may declare an early radio link failure based on failing the quantity of RACH attempts and the prediction.