Methods for temporal spatial frequency (TSF) channel state information (CSI) compression and for TSF parameter determination

The use of AI/ML models in WTRUs to determine TSF parameters for CSI compression addresses the inefficiencies in existing CSI reporting methods, reducing overhead and enhancing data transmission efficiency through dynamic adaptation to channel conditions.

WO2024233618A1PCT designated stage expired Publication Date: 2024-11-14INTERDIGITAL PATENT HOLDINGS INC

Patent Information

Application Number
PCT/US2024/028283
Authority / Receiving Office
WO · WO
Patent Type
Applications
Current Assignee / Owner
Priority Date
2023-05-09
Filing Date
2024-05-08
Publication Date
2024-11-14

AI Technical Summary

Technical Problem

Current wireless communication systems face challenges in reducing channel state information (CSI) reporting overhead, particularly due to the lack of effective methods to leverage temporal correlation properties, leading to inefficient data compression and increased processing times.

Method used

A wireless transmit/receive unit (WTRU) employs artificial intelligence/machine learning models to compress CSI by determining temporal-spatial-frequency (TSF) parameters based on configuration information and measured CSI, using techniques such as recurrent neural networks (RNNs) to calculate a compressed CSI report, which includes TSF parameters like buffer size and compression rate, optimized by metrics like squared generalized cosine similarity (SGCS).

Benefits of technology

This approach significantly reduces CSI reporting overhead while maintaining reconstruction quality, enabling more efficient data transmission and improved channel state estimation by dynamically adapting to changing channel conditions.

✦ Generated by Eureka AI based on patent content.

Smart Images

  • Figure US2024028283_14112024_PF_FP_ABST
    Figure US2024028283_14112024_PF_FP_ABST
Patent Text Reader

Abstract

A wireless transmit / receive unit (WTRU) receives configuration information. The configuration information is associated with temporal-spatial-frequency (TSF) compression. The WTRU measures channel state information (CSI). The WTRU determines TSF parameters based on the configuration information and the measured CSI. The WTRU calculates a compressed CSI based on the measured CSI and the determined TSF parameters. The WTRU sends the determined TSF parameters and the compressed CSI in a CSI measurement report.
Need to check novelty before this filing date? Find Prior Art

Citation Information

Patent Citations

  • Channel state information feedback using channel compression and reconstruction

    US20210273706A1

  • Architectures for temporal processing associated with wireless transmission of encoded data

    US20220284267A1

  • Reporting configurations for neural network-based processing at a ue

    WO2022040046A1

  • Neural network architecture for CSI feedback

    WO2024064541A1

Cited By

  • Methods to handle variable rank reporting for eigenvector-based temporal-spatial-frequency compression

    WO2026136232A1