Path-Aware Cognitive Handover Optimization for 5G Networks

Resolve Bottlenecks,
Find Innovative Solutions
Generate Solutions

Solution Overview

Problem

Current mobile and wireless telecommunication systems, such as 5G and LTE, face inefficiencies in handover processes, leading to unwanted handover events like failures and ping pongs, due to suboptimal cell individual offset (CIO) and time-to-trigger (TTT) settings, which affect user experience and network performance.

Innovation Solution

The implementation of path-aware cognitive handover optimization (PACHO) using machine learning models that learn user paths and optimal handover settings, such as CIO and TTT, to evaluate target candidates and points, trained with anonymized and encrypted user data, and deployed on user equipment for inference, minimizing computational and network resource usage.

Engineering Contradictions & Design Principles

VSEngineering Contradiction Analysis

1Measurement precision

If machine learning models are deployed on user equipment for path-aware cognitive handover optimization, then handover decision accuracy is improved, but device complexity increases

Engineering Contradiction:
Improvehandover decision accuracyVSAvoiddevice complexity
Core Design Contradiction:
Measurement precisionVSDevice complexity

Solution Approach 1:

The patent segments the handover optimization system into two parts: a training phase that occurs in the network infrastructure (separated from user equipment) and an inference phase that runs on user equipment. This segmentation allows complex machine learning model training to occur in the network while only deploying lighter inference engines on devices, thus improving handover decision accuracy without excessively increasing user equipment complexity.

Inventive Principle:
Principle #1Segmentation

Solution Approach 2:

The patent implements preliminary action by pre-training machine learning models in the network infrastructure before deployment to user equipment. The models are trained offline using aggregated path data from multiple users, and then the trained models are deployed for inference. This preliminary training approach allows complex computations to occur before deployment, reducing the computational burden on user equipment while maintaining high decision accuracy.

Inventive Principle:
Principle #10Preliminary action

2Productivity

If path data is collected and processed for machine learning training, then handover optimization performance is improved, but user privacy protection becomes more challenging

Engineering Contradiction:
Improvehandover optimization performanceVSAvoiduser privacy risk
Core Design Contradiction:
ProductivityVSObject-affected harmful factors

Solution Approach 1:

The patent applies preliminary action by implementing data anonymization and encryption before the path data is used for machine learning training. User path data is processed and anonymized in advance, removing personally identifiable information while retaining the spatial-temporal patterns needed for handover optimization. This preliminary privacy protection measures allow the system to collect and process path data for improved handover performance while safeguarding user privacy from the outset.

Inventive Principle:
Principle #10Preliminary action

Data Source

PatentEP4197219B1Path-aware cognitive handover optimization
Publication Date: 2024.12.04 NOKIA TECHNOLOGIES OY
  • EP4197219B1 patent drawingFigure 1
  • EP4197219B1 patent drawingFigure 2
  • EP4197219B1 patent drawingFigure 3

AI summary

A method and apparatus may include receiving, by a network element from at least one user equipment, one or more of at least one path, at least one observed reference signal received power associated with at least one cell, and at least one global positioning system coordinate. The method may further include computing, by the network element, at least one loss function based upon at least one handover event evaluation rule of the received data. The method may further include training, by the network element, at least one handover optimization model.