Predictive autoscaler for cellular communication networks

A predictive autoscaler using machine learning models in 5G networks addresses inefficient scaling by anticipating future loading, optimizing resource allocation and improving network performance through proactive scaling of microservices.

WO2026117239A1PCT designated stage Publication Date: 2026-06-04RAKUTEN SYMPHONY INC +1
View PDF 5 Cites 0 Cited by

Patent Information

Application Number
PCT/US2024/057745
Authority / Receiving Office
WO · WO
Patent Type
Applications
Current Assignee / Owner
Filing Date
2024-11-27
Publication Date
2026-06-04

AI Technical Summary

Technical Problem

Existing cellular communication networks struggle with inefficient scaling of software components due to reactive scaling methods that do not anticipate future loading demands, leading to potential overloading or underutilization.

Method used

Implementing a predictive autoscaler system that uses trained machine learning models to process network signals, extract relevant features, and predict future loading of microservices, enabling proactive scaling of software components in 5G networks.

Benefits of technology

The system enables efficient and timely scaling of microservices based on predicted loading, optimizing resource utilization and reducing the risk of overloading or underutilization, thereby enhancing network performance and efficiency.

✦ Generated by Eureka AI based on patent content.

Smart Images

  • Figure US2024057745_04062026_PF_FP_ABST
    Figure US2024057745_04062026_PF_FP_ABST
Patent Text Reader

Abstract

A cellular communication network includes one or more computing devices and configured to process connections to user equipment, each computing device of the one or more computing devices executing one or more software components. Each computing device of the one or more computing devices is configured to: receive a plurality of signals from the one or more microservices; process the plurality of signals with a machine learning model to obtain a predicted loading by the user equipment; and scale up or scale down at least one of the microservices according to the predicted loading.
Need to check novelty before this filing date? Find Prior Art

Citation Information

Patent Citations

  • Predictive execution of distributed game engines

    US20230211233A1

  • Techniques for adaptive pipelining composition for machine learning (ML)

    US20230336340A1

  • Network segmentation for container orchestration platforms

    US20240129161A1

  • Machine learning for detecting and modifying faulty controls

    US20240171474A1

  • Automated orchestration of containers by assessing microservices

    WO2022111156A1