Parallel ML-Based CSI Reporting Under UE CPU Constraints

Resolve Bottlenecks,
Find Innovative SolutionsGenerate Solutions

Solution Overview

Problem

The increasing complexity and size of channel state information (CSI) reports in wireless communication systems, particularly in 5G NR, pose challenges for user equipment (UE) processing capabilities, leading to limitations in parallel processing and reporting of ML-based and non-ML-based CSI reports.

Innovation Solution

Implementing a UE capability for parallel processing of CSI reports using dedicated Type 2 CPUs for ML-based CSI report processing, allowing for separate management of ML-based and non-ML-based CSI reports, and employing a CPU management framework to ensure compliance with maximum CPU usage limits.

Engineering Contradictions & Design Principles

VSEngineering Contradiction Analysis

1Productivity

If the UE processes multiple ML-based and non-ML-based CSI reports in parallel, then the reporting capability and data rate are improved, but the device complexity and CPU resource consumption increase

Engineering Contradiction:

ImproveCSI report processing capabilityVSAvoidCPU resource management complexity

Core Design Contradiction:

ProductivityVSDevice complexity

Solution Approach 1:

The patent segments CPU resources into dedicated Type 2 CPUs for ML-based CSI report processing and Type 1 CPUs for non-ML-based processing. This segmentation allows independent management of different processing workloads, enabling parallel processing of multiple CSI reports while maintaining clear resource boundaries and reducing management complexity.

Inventive Principle:

Principle #1Segmentation

Solution Approach 2:

The patent introduces a new dimension of CPU classification by creating Type 1 and Type 2 CPU categories. This dimensional change in resource organization enables the system to handle both ML-based and non-ML-based CSI reports simultaneously without resource conflicts, thereby improving productivity while managing complexity through structured resource allocation.

Inventive Principle:

Principle #17Another dimension (Dimensionality change)

2Speed

If the UE uses dedicated Type 2 CPUs for ML-based CSI report processing, then the processing speed and accuracy are improved, but the device complexity increases

Engineering Contradiction:

ImproveML-based CSI report processing speedVSAvoidCPU architecture complexity

Core Design Contradiction:

SpeedVSDevice complexity

Solution Approach 1:

The patent segments CPU functionality into dedicated Type 2 CPUs specifically for ML-based CSI processing tasks. This segmentation provides specialized hardware acceleration for computationally intensive ML operations, improving processing speed while isolating the complexity to dedicated units rather than requiring complex general-purpose resource management.

Inventive Principle:

Principle #1Segmentation

Solution Approach 2:

The patent creates a universal CPU management framework that handles both Type 1 (non-ML) and Type 2 (ML-dedicated) CPUs through unified resource allocation mechanisms. This multi-functionality allows the system to manage diverse processing requirements through a single management interface, improving speed through specialized processing while avoiding proportional increases in overall system complexity.

Inventive Principle:

Principle #6Universality (Multi-functionality)

3Productivity

If the UE increases the number of CPUs for parallel CSI report processing, then the productivity is improved, but the loss of time for resource management and coordination increases

Engineering Contradiction:

ImproveParallel CSI report processing throughputVSAvoidCPU resource coordination time

Core Design Contradiction:

ProductivityVSLoss of time

Solution Approach 1:

The patent segments CPU resources into distinct Type 1 and Type 2 categories with dedicated management protocols. This segmentation reduces coordination overhead by establishing clear resource boundaries and allocation rules, allowing parallel processing of multiple CSI reports while minimizing the time lost to resource management through pre-defined resource allocation strategies.

Inventive Principle:

Principle #1Segmentation

4Reliability

If the UE reports high priority CSI reports with measured CSI, then the reliability is improved, but the loss of time for re-measuring low priority reports increases

Engineering Contradiction:

ImproveCSI report accuracyVSAvoidTime for CSI re-measurement

Core Design Contradiction:

ReliabilityVSLoss of time

Solution Approach 1:

The patent applies local quality by allocating dedicated Type 2 CPUs specifically for high-priority ML-based CSI report processing. This localized resource allocation ensures that high-priority reports receive sufficient computational resources for accurate processing, while low-priority reports can utilize remaining resources or previously cached measurements, thereby maintaining reliability for critical reports without proportionally increasing time loss across all reports.

Inventive Principle:

Principle #3Local quality

Data Source

PatentUS20260031947A1Parallel processing for machine learning-based channel state information reports

Publication Date: 2026.01.29 GOOGLE LLC

AI summary

This disclosure provides systems, devices, apparatus, and methods, including computer programs encoded on storage media, for parallel processing and processing delays of ML-based CSI reports. A UE receives (404, 804), from a network entity (104), a configuration for an ML-based CSI report associated with a first CSI-RS. The UE (102) receives (407, 807), from the network entity (104), the first CSI-RS and transmits (412, 812), to the network entity (104), the ML-based CSI report including CSI that is based on a UE capability for processing the ML-based CSI report in association with the first CSI-RS.