18255486. DEVICES, METHODS, AND SYSTEMS FOR PREDICTING RADIO ACCESS NETWORK PERFORMANCE simplified abstract (Telefonaktiebolaget LM Ericsson (publ))

DEVICES, METHODS, AND SYSTEMS FOR PREDICTING RADIO ACCESS NETWORK PERFORMANCE

Organization Name

Telefonaktiebolaget LM Ericsson (publ)

Inventor(s)

Samara Keshavarz Hedayati of Stockholm (SE)

Rerngvit Yanggratoke of Järfälla (SE)

[[:Category:Mattias Lidstr�m of Stockholm (SE)|Mattias Lidstr�m of Stockholm (SE)]][[Category:Mattias Lidstr�m of Stockholm (SE)]]

Marius Sorin Sirbu of Åkersberga (SE)

DEVICES, METHODS, AND SYSTEMS FOR PREDICTING RADIO ACCESS NETWORK PERFORMANCE - A simplified explanation of the abstract

This abstract first appeared for US patent application 18255486 titled 'DEVICES, METHODS, AND SYSTEMS FOR PREDICTING RADIO ACCESS NETWORK PERFORMANCE

Simplified Explanation

The abstract describes methods and devices for predicting Radio Access Network performance by training a prediction model using information associated with network elements and performance indicators.

• Prediction model training device obtains information on network elements
• Determines tags based on network environment properties
• Associates network elements with clusters based on tags
• Calculates performance indicator values for clusters
• Creates training set with performance indicator values
• Trains prediction model using training set

Potential Applications

This technology can be applied in telecommunications to predict and optimize Radio Access Network performance, leading to improved network efficiency and user experience.

Problems Solved

1. Predicting Radio Access Network performance accurately 2. Optimizing network elements based on performance indicators

Benefits

1. Enhanced network performance 2. Improved user experience 3. Efficient resource allocation

Potential Commercial Applications

Optimizing Radio Access Network performance can benefit telecommunication companies by reducing costs, improving service quality, and increasing customer satisfaction.

Possible Prior Art

One possible prior art could be the use of machine learning algorithms to predict network performance in telecommunications systems.

How does this technology handle real-time network performance prediction?

This article does not specify how the prediction model adapts to real-time changes in network performance.

What are the limitations of this technology in predicting network performance in complex environments?

The article does not address the potential challenges or limitations of the prediction model in predicting network performance in complex network environments.

Original Abstract Submitted

Methods, devices, and a system for prediction of Radio Access Network performance. A device for training a prediction model for predicting Radio Access Network performance obtains information associated with at least one network element; determines a tag for the at least one network element, wherein the tag is based on a network environment property of the at least one network element; and determines, based on the tag, a cluster to be associated with the at least one network element. The device determines a performance indicator value for the at least one cluster, creates a training set comprising the performance indicator value of at least one cluster; and trains a prediction model using the training set. The prediction model trained by the device can be used by a device for prediction of Radio Access Network performance.