Terminal CSI Compression with Selectable ML Models for NR Uplinks

Resolve Bottlenecks,
Find Innovative SolutionsGenerate Solutions

Solution Overview

Problem

Existing wireless communication systems lack flexibility and efficiency in their communication structures, particularly in the context of New Radio (NR) technology, which aims to support diverse scenarios such as enhanced Mobile BroadBand (eMBB), massive Machine Type Communication (mMTC), and Ultra Reliable and Low Latency Communication (URLLC).

Innovation Solution

The implementation of Orthogonal Frequency Division Multiplex (OFDM) with CP-OFDM and DFT-s-OFDM, along with resource grid configurations and carrier aggregation, enhances communication flexibility and efficiency by optimizing the use of subcarrier-spacing, OFDM symbols, and channel properties.

Engineering Contradictions & Design Principles

VSEngineering Contradiction Analysis

1Adaptability or versatility

If a traditional communication structure is used, then the system is simple to implement, but the flexibility and efficiency of wireless communication are limited

Engineering Contradiction:

Improvecommunication flexibilityVSAvoidcommunication structure complexity

Core Design Contradiction:

Adaptability or versatilityVSDevice complexity

Solution Approach 1:

The patent implements dynamic communication structure by allowing the network to switch between different OFDM waveform types (CP-OFDM and DFT-s-OFDM) based on service requirements. The terminal device can be configured to use different waveforms for different uplink channels and signals, enabling the system to adapt flexibly to varying communication demands while maintaining structured organization through configuration parameters and signaling mechanisms

Inventive Principle:

Principle #15Dynamics

Solution Approach 2:

The patent segments the uplink transmission by applying different waveform types to different logical channels. Control channels use CP-OFDM for robustness, while data channels can use DFT-s-OFDM for efficiency. This segmentation allows each channel type to be optimized independently, improving overall system flexibility without creating a monolithic complex structure

Inventive Principle:

Principle #1Segmentation

2Productivity

If resource allocation is optimized for specific scenarios, then the efficiency for those scenarios improves, but the system becomes less adaptable to other scenarios

Engineering Contradiction:

Improvecommunication efficiencyVSAvoidscenario adaptability

Core Design Contradiction:

ProductivityVSAdaptability or versatility

Solution Approach 1:

The patent creates a universal communication framework where a single terminal device can operate with multiple waveform configurations. The device is capable of using both CP-OFDM and DFT-s-OFDM waveforms depending on the service type, making it universally applicable to eMBB, mMTC, and URLLC scenarios without requiring separate hardware or protocol stacks for each scenario

Inventive Principle:

Principle #6Universality (Multi-functionality)

Solution Approach 2:

The patent utilizes parameter-based configuration to enable scenario-specific optimization. By changing waveform type parameters, subcarrier spacing parameters, and resource allocation parameters based on the active service scenario, the system achieves high efficiency for the current scenario while maintaining the capability to adapt to other scenarios through parameter reconfiguration

Inventive Principle:

Principle #35Parameter changes

Data Source

PatentUS12531704B2Terminal devices and base station devices

Publication Date: 2026.01.20 SHARP KK

AI summary

A terminal device receives a reference signal, a first machine learning model, and a second machine learning method, and transmits a CSI report. The terminal device selects a machine learning model that is used to compress a CSI related data obtained from the reference signal. The terminal device includes the compressed CSI data and an indication indicating the selected machine learning model in the CSI report.