ML-Based Beamforming for Massive MIMO Capacity

Resolve Bottlenecks,
Find Innovative SolutionsGenerate Solutions

Solution Overview

Problem

Existing massive MIMO-based LTE systems cannot maximize transmission capacity in the coverage area of an active antenna unit (AAU) due to preset beamforming parameters that do not adapt to changing terminal device directions and cell load conditions.

Innovation Solution

A beamforming method that uses machine learning models to obtain feature data from terminal devices and cells, determining optimal beam directions, widths, and quantities based on azimuth information, load, and distribution, thereby adapting beamforming parameters to actual conditions for improved capacity.

Engineering Contradictions & Design Principles

VSEngineering Contradiction Analysis

1Productivity

If preset beamforming parameters (quantity of beams, beam direction, beam width) are used, then the system complexity is reduced and ease of operation is improved, but the transmission capacity in the coverage area of the AAU cannot be maximized

Engineering Contradiction:

Improvetransmission capacityVSAvoidbeamforming parameter determination complexity

Core Design Contradiction:

ProductivityVSDevice complexity

Solution Approach 1:

The system performs self-service by automatically determining beamforming parameters through machine learning models. The radio access network device uses the machine learning model to autonomously determine the quantity of beams, beam directions, and beam widths based on input data, eliminating the need for manual presetting and optimization while maximizing transmission capacity

Inventive Principle:

Principle #25Self-service

Solution Approach 2:

The invention changes the approach from fixed preset parameters to dynamically determined parameters. The machine learning model processes input data (including terminal device information and cell information) to generate optimized beamforming parameters, allowing the system to adapt parameters such as beam quantity, directions, and widths to current network conditions for maximum transmission capacity

Inventive Principle:

Principle #35Parameter changes

2Adaptability or versatility

If preset beamforming parameters are used, then the ease of operation is improved, but the adaptability to actual conditions (terminal device distribution and cell load) deteriorates

Engineering Contradiction:

Improveadaptability to actual conditionsVSAvoidoperational simplicity

Core Design Contradiction:

Adaptability or versatilityVSEase of operation

Solution Approach 1:

The system implements feedback by using the machine learning model to continuously determine beamforming parameters based on current network conditions. The model processes real-time input data including terminal device distribution and cell load information, generating adapted beamforming parameters that respond to actual network states, thereby improving adaptability while the automated process maintains ease of operation

Inventive Principle:

Principle #23Feedback

Solution Approach 2:

The invention transforms static preset parameters into dynamic parameters that adapt to changing network conditions. The machine learning model enables the beamforming parameters (quantity of beams, beam directions, beam widths) to dynamically adjust based on terminal device distribution and cell load, achieving high adaptability without compromising operational simplicity

Inventive Principle:

Principle #15Dynamics

Applied Scientific Principles

This section explains which scientific principles are used to turn an abstract innovation direction into a practical engineering solution.

Function Achieved in This Case

This approach enhances transmission capacity by accurately directing beams to densely populated areas and optimizing beam quantities, leading to improved user experience and system performance.

Implementation Method 1

obtains a beamforming parameter based on the feature data and a machine learning model, where the beamforming parameter includes at least one of a beam direction, a beam width, and a quantity of beams

Methodology Applied

Scientific EffectMachine learning:

Data Source

PatentUS12089064B2Beamforming method and apparatus, radio access network device, and readable storage medium

Publication Date: 2024.09.10 HUAWEI TECH CO LTD

AI summary

A beamforming method includes obtaining feature data. The feature data includes at least one of feature data of a terminal device or feature data of a cell. The feature data of the terminal device represents azimuth information of the terminal device, and the feature data of the cell represents load information of the cell and distribution information of the terminal device. The beamforming method further includes obtaining a beamforming parameter based on the feature data and a machine learning model. The beamforming parameter includes at least one of a beam direction, a beam width, or a quantity of beams. The beamforming method further includes performing beamforming on a first signal based on the beamforming parameter.