AI/ML Prediction for Channel Aging in Non-Terrestrial Networks
Find Innovative SolutionsGenerate Solutions
Solution Overview
Problem
Non-terrestrial networks (NTNs) face challenges in accurately predicting channel state information (CSI) due to factors like long transmission delays, Doppler effects, and aging CSI feedback, which can lead to performance losses and resource optimization issues.
Innovation Solution
The use of AI/ML prediction models that are selected based on specific conditions or attributes of the NTN, such as satellite parameters and weather conditions, to predict CSI quantities and compensate for channel aging.
Engineering Contradictions & Design Principles
Engineering Contradiction Analysis
1Measurement precision
If traditional prediction methods are used for CSI feedback in NTNs, then the system complexity is low, but the prediction accuracy deteriorates due to long transmission delays and Doppler effects
Solution Approach 1:
The patent applies parameter changes by selecting different AI/ML prediction models based on specific NTN conditions such as satellite orbit parameters (GEO, LEO, MEO), transmission delay characteristics, Doppler effect magnitude, and weather conditions. This allows the system to adapt the complexity of the prediction model to match the specific channel aging characteristics of different NTN scenarios, improving prediction accuracy without unnecessarily increasing system complexity for all cases.
Solution Approach 2:
The patent implements dynamics by making the prediction model selection dynamic and adaptive to changing NTN conditions. The system continuously monitors parameters such as satellite position, velocity, transmission delay, and weather conditions to dynamically select the most appropriate prediction model, enabling the system to respond to varying channel aging rates and maintain optimal prediction accuracy throughout different operational phases.
2Reliability
If AI/ML prediction models are selected based on specific NTN conditions, then the CSI feedback prediction accuracy is improved, but the system complexity increases due to model selection and configuration
Solution Approach 1:
The patent applies self-service by enabling the UE to autonomously select and configure the appropriate AI/ML prediction model based on received NTN parameters without requiring complex network-side configuration or manual intervention. The UE independently evaluates conditions such as satellite orbit type, transmission delay, and weather conditions to self-determine the optimal prediction model, reducing the configuration burden on the network while improving prediction reliability.
Solution Approach 2:
The patent implements preliminary action by having the network entity pre-configure multiple AI/ML prediction models with different complexities and characteristics before NTN operation begins. These models are prepared in advance with appropriate hyperparameters and architectures suited for different NTN scenarios, allowing the UE to simply select from pre-prepared options rather than configuring models in real-time, thus reducing operational complexity while maintaining high reliability.
3Productivity
If channel aging is not compensated in NTNs, then the system operation is simple, but performance loss occurs due to outdated CSI feedback
Solution Approach 1:
The patent applies feedback by using the predicted CSI quantities (such as CQI, PMI, RI) to continuously update and refine the resource provisioning decisions. The system uses the AI/ML prediction outputs as feedback to adjust modulation and coding schemes, precoding matrices, and resource allocation, thereby maintaining high resource provisioning efficiency while compensating for channel aging effects through iterative refinement based on predicted channel conditions.
Solution Approach 2:
The patent implements preliminary action by performing AI/ML-based CSI prediction in advance before actual data transmission occurs. This allows the system to proactively compensate for channel aging by having updated CSI estimates ready before the channel conditions deteriorate further, enabling timely and accurate resource provisioning decisions without waiting for outdated CSI feedback to become invalid.
Data Source
AI summary
Various aspects of the present disclosure relate to utilizing artificial intelligence (AI) and/or machine learning (ML) prediction models when compensating for channel aging in non-terrestrial networks (NTNs). The aspects of the present disclosure can facilitate the selection of a prediction model based on specific conditions or attributes of an NTN, such as parameters associated with a satellite of the NTN and/or weather or atmospheric conditions surrounding the NTN.