Cell State Prediction in Radio Access Networks

Resolve Bottlenecks,
Find Innovative Solutions
Generate Solutions

Solution Overview

Problem

Current methods are unable to predict when a cell in a radio access network (RAN) will enter a sleeping state, leading to delayed detection and resulting in poor network performance, revenue loss, and increased operating expenses.

Innovation Solution

A method and system for predicting a cell's state in a RAN by obtaining cell information, determining sets of conditions indicating a decrease in Random Access Channel (RACH) success rate, and predicting the cell will enter a sleeping state when these conditions are met.

Engineering Contradictions & Design Principles

VSEngineering Contradiction Analysis

1Loss of time

If traditional detection methods are used to identify sleeping cells, then detection is simple, but detection delay is long (24-48 hours)

Engineering Contradiction:
Improvedetection delayVSAvoiddetection system complexity
Core Design Contradiction:
Loss of timeVSDevice complexity

Solution Approach 1:

The system performs preliminary actions by continuously monitoring cell performance parameters and calculating probabilities of cells entering sleeping state before the actual state transition occurs. This allows early detection and preventive measures to be taken, reducing the detection delay from 24-48 hours to a much shorter timeframe while maintaining manageable system complexity through automated algorithms.

Inventive Principle:
Principle #10Preliminary action

Solution Approach 2:

The system implements feedback mechanisms by continuously collecting cell performance data, analyzing trends, and adjusting predictions based on historical patterns. The feedback loop processes performance parameters, calculates sleeping state probabilities, and triggers alerts or preventive actions when thresholds are met, enabling timely detection without requiring overly complex manual monitoring systems.

Inventive Principle:
Principle #23Feedback

2Reliability

If no prediction method is implemented, then system operation is simple, but network performance deteriorates due to undetected sleeping cells

Engineering Contradiction:
Improvenetwork performanceVSAvoidprediction system complexity
Core Design Contradiction:
ReliabilityVSDevice complexity

Solution Approach 1:

The patent replaces manual monitoring and reactive repair mechanisms with an automated prediction system that uses algorithms to analyze performance parameters and calculate sleeping state probabilities. This substitution improves network reliability by enabling proactive detection while keeping the system complexity manageable through automated decision-making processes rather than human intervention.

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

Solution Approach 2:

The prediction system performs self-service by automatically collecting data, analyzing trends, generating predictions, and triggering alerts without requiring continuous human oversight. The system uses historical performance data to train models and improve predictions over time, enhancing network reliability while maintaining operational simplicity through automation.

Inventive Principle:
Principle #25Self-service

3Loss of information

If early detection of sleeping cells is achieved, then revenue loss is reduced, but measurement and detection difficulty increases

Engineering Contradiction:
Improverevenue lossVSAvoidcell state detection difficulty
Core Design Contradiction:
Loss of informationVSDifficulty of detecting and measuring

Solution Approach 1:

The system uses a universal approach by monitoring multiple performance parameters (RACH success rate, handover success rate, throughput, etc.) simultaneously to detect sleeping cells. This multi-functional monitoring framework reduces revenue loss by detecting various indicators of cell degradation while managing detection complexity through integrated analysis of all parameters rather than separate specialized systems.

Inventive Principle:
Principle #6Universality (Multi-functionality)

Solution Approach 2:

The system detects sleeping cells by monitoring changes in performance parameters over time, such as declining RACH success rates or throughput degradation. By tracking parameter trends and calculating probability thresholds, the system achieves early detection that prevents revenue loss while managing measurement difficulty through automated parameter analysis rather than manual inspection.

Inventive Principle:
Principle #35Parameter changes

Data Source

PatentEP3857979B1Method and system for predicting a state of a cell in a radio access network
Publication Date: 2025.06.04 TELEFONAKTIEBOLAGET LM ERICSSON (PUBL)
  • EP3857979B1 patent drawingFigure 1
  • EP3857979B1 patent drawingFigure 2~3
  • EP3857979B1 patent drawingFigure 4

AI summary

A method and system for predicting a state of a cell (11) in a radio access network are described herein. The method comprises obtaining (S210) information of the cell (11), determining (S220) one or more sets of conditions based on the information and predicting (S230) that the cell (11) will enter a sleeping state when at least one set of the one or more sets of conditions is fulfilled. The method further comprises outputting (S240) an action to prevent the cell (11) from entering the sleeping state based on the probability and a number of wireless devices currently connected to the cell (11).