Radio Network Self-Optimization Using Spatiotemporal Sensor Data

Resolve Bottlenecks,
Find Innovative Solutions
Generate Solutions

Solution Overview

Problem

5G wireless networks face challenges in maintaining performance due to dynamic environmental changes, which can lead to radio network degradation below acceptable levels for critical applications like URLLC, as traditional reactive approaches often fail to prevent performance drops in time.

Innovation Solution

A predictive approach using an AI neural network model that associates sensor data with radio network information to anticipate and preemptively correct performance issues by performing actions such as handovers or adjusting network parameters before significant degradation occurs.

Engineering Contradictions & Design Principles

VSEngineering Contradiction Analysis

1Reliability

If traditional reactive approaches are used to maintain radio network performance, then the system complexity remains low, but the network performance degrades below acceptable levels for critical applications

Engineering Contradiction:
Improveradio network performanceVSAvoidsystem complexity
Core Design Contradiction:
ReliabilityVSDevice complexity

Solution Approach 1:

The patent implements predictive analytics using machine learning models to forecast radio network performance degradation before it occurs. The system analyzes historical network data, environmental conditions, and traffic patterns to predict future performance issues, enabling proactive adjustments to network parameters such as power settings, handover thresholds, and resource allocation. This preliminary action prevents performance degradation rather than merely reacting to it, thereby improving reliability while managing complexity through automated prediction algorithms.

Inventive Principle:
Principle #10Preliminary action

Solution Approach 2:

The patent establishes a closed-loop feedback system where network performance metrics are continuously monitored, analyzed, and used to automatically adjust network parameters. The machine learning models process real-time feedback from the network and environmental sensors, generating optimization recommendations that are implemented by the network management system. This continuous feedback loop enables the system to adapt dynamically to changing conditions, maintaining high reliability through automated self-optimization without requiring manual intervention.

Inventive Principle:
Principle #23Feedback

2Reliability

If predictive analytics with sensor data association are implemented, then network performance is maintained above acceptable levels, but the device complexity increases

Engineering Contradiction:
Improveradio network performanceVSAvoidsystem complexity
Core Design Contradiction:
ReliabilityVSDevice complexity

Solution Approach 1:

The system implements predictive analytics by associating sensor data (environmental conditions, location information, device status) with radio network performance metrics using machine learning models. These models forecast future performance degradation by analyzing patterns in historical data, enabling the system to take preliminary corrective actions before performance drops below acceptable levels. This approach maintains high reliability through proactive optimization while managing complexity through automated data association and prediction algorithms.

Inventive Principle:
Principle #10Preliminary action

Solution Approach 2:

The patent implements self-service automation where the network management system autonomously performs performance optimization without manual intervention. The machine learning models automatically associate sensor data with network metrics, predict performance issues, and generate optimization recommendations that are automatically implemented. This self-service capability maintains high reliability through continuous automated monitoring and adjustment, reducing the need for manual network management while managing complexity through intelligent automation.

Inventive Principle:
Principle #25Self-service

3Reliability

If preemptive corrective actions are performed, then errors are prevented and compliance with 5G requirements is ensured, but the loss of time for data processing and action execution increases

Engineering Contradiction:
Improvecompliance with 5G requirementsVSAvoidprocessing time
Core Design Contradiction:
ReliabilityVSLoss of time

Solution Approach 1:

The system performs preemptive corrective actions by predicting performance degradation before it occurs using machine learning models that analyze historical data and current conditions. The models identify upcoming performance issues and trigger corrective actions (such as adjusting power settings, modifying handover parameters, or reallocating resources) before actual degradation happens. This preliminary action ensures compliance with 5G reliability and latency requirements by preventing errors rather than reacting to them, thereby maintaining high performance while managing processing time through efficient prediction algorithms.

Inventive Principle:
Principle #10Preliminary action

Solution Approach 2:

The patent replaces traditional manual or rule-based network optimization mechanisms with machine learning-based predictive systems. The ML models automatically process sensor data and network metrics, predict performance issues, and generate optimization recommendations without requiring manual analysis or intervention. This substitution of mechanical/manual processes with intelligent automation reduces processing time by eliminating human decision-making delays while ensuring compliance with 5G requirements through continuous automated monitoring and adjustment.

Inventive Principle:
Principle #28Mechanics substitution (Replace mechanical system)

Data Source

PatentEP3857818B1Radio-network self-optimization based on data from radio network and spatiotemporal sensors
Publication Date: 2024.03.06 NOKIA TECHNOLOGIES OY
  • EP3857818B1 patent drawingFigure 1
  • EP3857818B1 patent drawingFigure 2
  • EP3857818B1 patent drawingFigure 3

AI summary

A technique includes receiving, from one or more sensors, sensor data samples; receiving radio network information data samples associated with a radio network; determining, based on an association of one or more received sensor data samples with one or more of the received radio network information data samples, a first set of one or more associated sensor and radio network information data samples; developing a model that is trained based on at least a portion of the first set the associated sensor and radio network information data samples that are relevant to performance of the radio network; and improving performance of the radio network based on at least the model.