Reinforcement Learning Beam Pair Selection for 5G Handover

Resolve Bottlenecks,
Find Innovative SolutionsGenerate Solutions

Solution Overview

Problem

The complexity of 5G mobile wireless networks, with their configurable multi-element antennas and variable beam numerology, poses challenges in managing connections and optimizing radio resource configuration for reliable and efficient network connectivity.

Innovation Solution

A system and method that utilize reinforcement learning to dynamically select beam pairs and end-to-end network slices, based on desired service levels, to optimize connectivity and resource utilization in 5G mobile wireless networks.

Engineering Contradictions & Design Principles

VSEngineering Contradiction Analysis

1Adaptability or versatility

If configurable multi-element antenna arrangements with variable beam configuration are implemented, then spectrum utilization flexibility is enhanced, but connection management complexity increases

Engineering Contradiction:

Improvespectrum utilization flexibilityVSAvoidconnection management complexity

Core Design Contradiction:

Adaptability or versatilityVSDevice complexity

Solution Approach 1:

The system employs machine learning models that autonomously perform beam pair selection and network slice selection without requiring manual configuration or complex management intervention. The ML models self-learn optimal configurations based on network conditions, service requirements, and mobility patterns, thereby reducing management complexity while maintaining flexibility.

Inventive Principle:

Principle #25Self-service

Solution Approach 2:

The patent dynamically adjusts beam configuration parameters (beam width, direction, frequency) and network slice parameters based on real-time conditions and service level agreements. This adaptive parameter adjustment allows the system to optimize spectrum utilization for different services and mobility scenarios without requiring complex manual reconfiguration.

Inventive Principle:

Principle #35Parameter changes

2Productivity

If enhanced configurability of 5G radio access is implemented, then resource utilization optimization is improved, but decision-making complexity increases

Engineering Contradiction:

Improveresource utilization optimizationVSAvoiddecision-making complexity

Core Design Contradiction:

ProductivityVSDevice complexity

Solution Approach 1:

The patent replaces traditional rule-based decision-making mechanisms with machine learning-based intelligent decision-making systems. The ML models analyze network conditions, service requirements, and mobility patterns to automatically determine optimal beam pairs and network slices, simplifying the decision-making process while improving resource utilization efficiency.

Inventive Principle:

Principle #28Mechanics substitution (Replace mechanical system)

Solution Approach 2:

The system implements closed-loop feedback mechanisms where the ML models continuously learn from network performance measurements, service quality metrics, and mobility patterns. This feedback enables the system to adapt and optimize resource allocation decisions dynamically, improving productivity without requiring complex manual decision-making processes.

Inventive Principle:

Principle #23Feedback

3Device complexity

If beam pair selection is performed without machine learning assistance, then system complexity is reduced, but connectivity reliability and latency performance deteriorate

Engineering Contradiction:

Improvesystem complexityVSAvoidconnectivity reliability

Core Design Contradiction:

Device complexityVSReliability

Solution Approach 1:

The machine learning models operate autonomously to select optimal beam pairs based on learned patterns from mobility data and network conditions. This self-service capability enables the system to maintain high connectivity reliability and low latency without requiring complex manual configuration or intervention, as the ML models automatically adapt to changing conditions.

Inventive Principle:

Principle #25Self-service

4Device complexity

If traditional network connectivity management components are used, then system complexity is reduced, but ability to manage availability and reliability of network connectivity deteriorates

Engineering Contradiction:

Improvesystem complexityVSAvoidnetwork connectivity availability

Core Design Contradiction:

Device complexityVSReliability

Solution Approach 1:

The patent replaces traditional rule-based connectivity management components with machine learning-based intelligent management systems. These ML systems analyze network conditions, service level agreements, and mobility patterns to dynamically select optimal beam pairs and network slices, thereby improving connectivity availability and reliability without requiring proportionally increased system complexity.

Inventive Principle:

Principle #28Mechanics substitution (Replace mechanical system)

Data Source

PatentUS12308922B2Reinforcement learning machine learning-assisted beam pair selection for handover in radio access networks

Publication Date: 2025.05.20 T MOBILE INNOVATIONS LLC

AI summary

A system and method carried out over a mobile wireless network are described for performing beam pair (BP) and end-to-end (E2E) network slice selection for supporting an invoked service on a mobile equipment (ME). The method includes establishing an initial BP with a radio access network (RAN) node, using an available link policy, enabling communicating a request to the RAN node including an indication of a desired service level for a service invoked on the ME. The method further includes updating, in accordance with the indication of a desired service level, a link policy and an E2E slice policy by performing a reinforcement learning, wherein the link policy is used to select a BP for the ME for a given ME mobility pattern, and wherein the E2E network slice policy is used to select an E2E network slice for the desired service level for the service invoked on the ME.