5G Radio Link Failure Prediction

A European cellular company uses machine learning to predict radio link failures five days in advance to maintain a reliable, high-speed 5G network. Operators are required to minimize service disruption, maintain reliability, and comply with industry standards such as ultra-reliable low latency communication (URLLC).

Contact form

 

Challenge

Unpredictable weather-related disruptions

Lack of visibility into radio link failure caused by weather conditions such as clouds, rain, and snow can pose major threats to reliability and latency of a 5G network, resulting in poor customer experience.

Solution

Predicting failures to enable interventions

RapidCanvas worked with the company’s teams to create a predictive model.

ICON RC 01
data 01

Use of historical data, weather data, and tower configuration data

ICON RC 02
predict 02

Create an ML model to predict RLF events five days into the future

ICON RP 03
flexible 03

Flexible model to update the configuration for each tower and receive predictions even in a live production setting.

ICON RP 04
configuration 04

The company is able to make hardware and software configuration changes, as needed

Other related case studies

Wind Turbine Predictive Maintenance

A Latin American renewable energy company optimized its maintenance schedules and reduced its risks and costs, extending the useful life of each turbine

Solar Farm Geolocation Optimization

A Brazilian company that builds solar farms identified the most optimum land parcels for new farm development to be evaluated quickly and efficiently.

5G Radio Link Failure Prediction

A European telecommunications company leveraged ML to predict RLF caused by weather conditions and maintain consistent, reliable 5G service.