Method and apparatus for compressing channel state information using a latent representation having multiple parts associated with different time intervals.
By generating multi-part CSI feedback transmission patterns using AI/ML models, the problem of high CSI report overhead is solved, achieving efficient CSI compression and improved reconstruction quality.
CN122139320APending Publication Date: 2026-06-02INTERDIGITAL PATENT HOLDINGS INC
View PDF 0 Cites 0 Cited by
Patent Information
- Application Number
- CN202480068854.5
- Authority / Receiving Office
- CN · China
- Patent Type
- Applications(China)
- Current Assignee / Owner
- Priority Date
- 2023-10-30
- Filing Date
- 2024-10-25
- Publication Date
- 2026-06-02
AI Technical Summary
Technical Problem
Existing Channel State Information (CSI) reporting overhead is high, and existing Space Frequency (SF) CSI compression techniques have not yet been able to approach the performance limit of uncompressed CSI.
Method used
Multiple latent structures are generated using an artificial intelligence/machine learning (AI/ML) model. Channel state information feedback is encoded based on configuration information and compressed using CSI feedback transmission modes with time-varying multipart (TDMP) latent features, including single-part and multipart feedback modes.
Benefits of technology
It effectively reduces CSI reporting overhead, approaches the performance limit of uncompressed CSI, and improves the reconstruction quality of channel state information.
✦ Generated by Eureka AI based on patent content.
Smart Images
Figure CN122139320A_ABST
Abstract
A wireless transmit / receive unit (WTRU) can receive configuration information indicating one or more parameters associated with channel state information (CSI) feedback of time-varying multipart (TDMP) latent features. The one or more parameters include a set of TDMP feedback transmission patterns, including single-part TDMP feedback patterns and multipart TDMP feedback patterns. The WTRU can receive one or more transmissions of one or more CSI reference signals (CSI-RS). The WTRU can generate a set of TDMP latent features using an artificial intelligence / machine learning (AI / ML) model based on measurement information associated with the received one or more CSI-RS. The set of TDMP latent features includes multiple parts. The WTRU can determine a TDMP feedback transmission pattern from the set of TDMP feedback transmission patterns. The WTRU can send report information indicating at least a portion of the set of TDMP latent features and the determined TDMP feedback transmission pattern.
Need to check novelty before this filing date? Find Prior Art