Predictive Channel State Information Reporting for Non-Terrestrial Networks
Find Innovative SolutionsGenerate Solutions
Solution Overview
Problem
Non-terrestrial networks (NTNs) face challenges with accurate channel state information (CSI) reporting due to long transmission delays, Doppler effects, and UE movement, leading to outdated CSI feedback that affects network resource optimization.
Innovation Solution
Implementing prediction-based techniques using AI/ML models and Kalman filtering to enhance CSI reporting, allowing UEs to provide predictive outputs for channel quantities, including aperiodic, periodic, and semi-persistent reporting formats.
Engineering Contradictions & Design Principles
Engineering Contradiction Analysis
1Measurement precision
If prediction-based techniques are implemented to compensate for channel aging, then CSI reporting accuracy is improved, but device complexity increases
Solution Approach 1:
The patent applies preliminary action by having the UE perform prediction computations in advance to generate predicted CSI values before actual channel conditions change. The UE uses prediction models to proactively calculate future channel states, allowing the network to receive forward-looking CSI information that compensates for transmission delays and channel aging effects.
2Adaptability or versatility
If multiple reporting formats (aperiodic, periodic, semi-persistent) are supported, then adaptability is improved, but device complexity increases
Solution Approach 1:
The patent implements dynamics by enabling the UE to dynamically switch between different reporting formats (aperiodic, periodic, semi-persistent) based on current channel conditions and network requirements. The prediction mechanism adapts its operation mode according to the specific reporting format being used, allowing flexible adjustment of reporting behavior without requiring separate complex processing paths for each format type.
3Productivity
If prediction models are applied at the UE, then network resource management is improved, but use of energy increases
Solution Approach 1:
The patent applies partial action by implementing prediction models selectively rather than continuously. The UE performs prediction computations only when necessary based on channel condition changes, movement detection, or network triggering events. This selective application of prediction reduces unnecessary energy consumption while still providing improved network resource management when predictions are actually performed.
Data Source
AI summary
Various aspects of the present disclosure relate to configuring and/or enhancing the reporting of channel state information (CSI) and other channel properties to apply prediction models or other prediction techniques for a network, such as at one or more associated UEs and/or at the network. The updated reporting may enable the network to perform and report predictive outputs for CSI quantities of a network, such as an NTN that has inherent channel aging during do the movement of satellites and/or communications delays between UEs and the network, among other benefits.