Machine Learning Channel Feedback with Interoperable Model Schemas

Resolve Bottlenecks,
Find Innovative SolutionsGenerate Solutions

Solution Overview

Problem

Existing wireless communication systems face challenges in efficiently transmitting channel condition feedback due to high resource consumption and inadequate accuracy using codebooks, and collaboration between apparatus and vendors is difficult without standardization for model interoperability.

Innovation Solution

Implementing machine learning models for channel information feedback, with schemes for model interoperability and collaboration, including specifying model structures, capabilities, and data transfer protocols to reduce resource usage and improve accuracy.

Engineering Contradictions & Design Principles

VSEngineering Contradiction Analysis

1Quantity of substance

If codebooks are used to transmit channel condition feedback, then the amount of data for feedback transmission is reduced, but the feedback accuracy becomes inadequate and significant overhead data must still be transmitted

Engineering Contradiction:

Improveamount of data for feedback transmissionVSAvoidfeedback accuracy

Core Design Contradiction:

Quantity of substanceVSMeasurement precision

Solution Approach 1:

The patent changes the fundamental parameter of feedback representation from discrete codebook indices to continuous compressed channel information representations generated by machine learning models. This allows transmitting only the essential compressed information rather than full precision data, reducing quantity while maintaining accuracy through intelligent compression and reconstruction models trained to preserve critical channel characteristics.

Inventive Principle:

Principle #35Parameter changes

2Quantity of substance

If machine learning models are used to compress channel information, then resource consumption for feedback transmission is reduced, but model interoperability between different apparatus and vendors becomes difficult

Engineering Contradiction:

Improveresource consumption for feedback transmissionVSAvoidmodel interoperability

Core Design Contradiction:

Quantity of substanceVSAdaptability or versatility

Solution Approach 1:

The patent introduces standardized model information schemas and interfaces as intermediaries between different machine learning models from various vendors. These schemas define common input/output formats, data types, and communication protocols that enable apparatus from different manufacturers to interchange compressed channel information and model parameters seamlessly, solving the interoperability problem while maintaining compression benefits.

Inventive Principle:

Principle #24Intermediary (Mediator)

Solution Approach 2:

The patent creates universal model information structures that can accommodate different machine learning architectures and compression methods from various vendors. By defining standardized interfaces and data formats, the system allows any compliant model to work with any other model, making the feedback transmission system universally applicable across different equipment manufacturers while preserving the resource reduction advantages of ML-based compression.

Inventive Principle:

Principle #6Universality (Multi-functionality)

3Adaptability or versatility

If collaboration between apparatus and vendors is enabled for model training, then model interoperability is improved, but the complexity of information exchange and standardization increases

Engineering Contradiction:

Improvemodel interoperabilityVSAvoidcomplexity of information exchange

Core Design Contradiction:

Adaptability or versatilityVSDevice complexity

Solution Approach 1:

The patent segments the complex model collaboration process into distinct standardized modules: model capability declaration, model information exchange, training data sharing, and verification. Each module operates independently with well-defined interfaces, making the overall complex collaboration manageable and systematic. This segmentation reduces the complexity burden by breaking down the standardization task into smaller, implementable components that can be adopted incrementally.

Inventive Principle:

Principle #1Segmentation

Data Source

PatentUS20250337469A1Model interoperability and collaboration for machine learning based reporting of communication channel information

Publication Date: 2025.10.30 SAMSUNG ELECTRONICS CO LTD

AI summary

An apparatus may include a receiver configured to receive a reference signal using a channel, a transmitter configured to transmit a representation relating to the channel, and a processing circuit configured to determine channel information based on the reference signal, generate, using a model, the representation based on the channel information, and transfer, using the receiver or the transmitter, model information to specify the model. The model information may include an identifier for the model. The model information may include structure information for the model. The model information may include information about a type of input for the model. The model information may include information about a format of input for the model. The model information may include mapping information for mapping channel information to an input of the model. The mapping information may include information for a first subband, and information for a second subband.