Logo
ScienceStack
Table of Contents
Fetching ...
Sign In

ML KPI Prediction in 5G and B5G Networks

Nguyen Phuc Tran, Oscar Delgado, Brigitte Jaumard, Fadi Bishay

TL;DR

This work tackles KPI prognostication for 5G/B5G network slices by forecasting short-term throughput with a slice-aware LSTM model (LSTM-FSD) and deriving additional KPIs via a lightweight optimization-based estimator (LP-KPI) that leverages predicted throughput and current state. A novel performance metric $ ho$ and overall score $\ P$ are defined to evaluate estimation accuracy, including penalties for under- and over-estimation. Empirical results using an OMNET++-based 5G simulation and a Montreal traffic dataset show that LSTM-FSD attains competitive short-term throughput forecasts (MAPE around 5.5% on training day and 18.66% across slices) and that LP-KPI delivers accurate delay and packet loss estimates with low computational latency (approximately $80\\,\\mu$s). The contributions enable more reliable, proactive service assurance for multi-slice networks, with potential extension to zero-touch management and orchestration in future work.

Abstract

Network operators are facing new challenges when meeting the needs of their customers. The challenges arise due to the rise of new services, such as HD video streaming, IoT, autonomous driving, etc., and the exponential growth of network traffic. In this context, 5G and B5G networks have been evolving to accommodate a wide range of applications and use cases. Additionally, this evolution brings new features, like the ability to create multiple end-to-end isolated virtual networks using network slicing. Nevertheless, to ensure the quality of service, operators must maintain and optimize their networks in accordance with the key performance indicators (KPIs) and the slice service-level agreements (SLAs). In this paper, we introduce a machine learning (ML) model used to estimate throughput in 5G and B5G networks with end-to-end (E2E) network slices. Then, we combine the predicted throughput with the current network state to derive an estimate of other network KPIs, which can be used to further improve service assurance. To assess the efficiency of our solution, a performance metric was proposed. Numerical evaluations demonstrate that our KPI prediction model outperforms those derived from other methods with the same or nearly the same computational time.

ML KPI Prediction in 5G and B5G Networks

TL;DR

This work tackles KPI prognostication for 5G/B5G network slices by forecasting short-term throughput with a slice-aware LSTM model (LSTM-FSD) and deriving additional KPIs via a lightweight optimization-based estimator (LP-KPI) that leverages predicted throughput and current state. A novel performance metric and overall score are defined to evaluate estimation accuracy, including penalties for under- and over-estimation. Empirical results using an OMNET++-based 5G simulation and a Montreal traffic dataset show that LSTM-FSD attains competitive short-term throughput forecasts (MAPE around 5.5% on training day and 18.66% across slices) and that LP-KPI delivers accurate delay and packet loss estimates with low computational latency (approximately s). The contributions enable more reliable, proactive service assurance for multi-slice networks, with potential extension to zero-touch management and orchestration in future work.

Abstract

Network operators are facing new challenges when meeting the needs of their customers. The challenges arise due to the rise of new services, such as HD video streaming, IoT, autonomous driving, etc., and the exponential growth of network traffic. In this context, 5G and B5G networks have been evolving to accommodate a wide range of applications and use cases. Additionally, this evolution brings new features, like the ability to create multiple end-to-end isolated virtual networks using network slicing. Nevertheless, to ensure the quality of service, operators must maintain and optimize their networks in accordance with the key performance indicators (KPIs) and the slice service-level agreements (SLAs). In this paper, we introduce a machine learning (ML) model used to estimate throughput in 5G and B5G networks with end-to-end (E2E) network slices. Then, we combine the predicted throughput with the current network state to derive an estimate of other network KPIs, which can be used to further improve service assurance. To assess the efficiency of our solution, a performance metric was proposed. Numerical evaluations demonstrate that our KPI prediction model outperforms those derived from other methods with the same or nearly the same computational time.
Paper Structure (13 sections, 4 equations, 6 figures, 3 tables)

This paper contains 13 sections, 4 equations, 6 figures, 3 tables.

Table of Contents

  1. Introduction
  2. Literature Review
  3. 5G Traffic Prediction
  4. Machine learning techniques and service assurance
  5. Machine learning-assisted system design
  6. Throughput KPI prediction
  7. Other KPI predictions
  8. Performance metric
  9. Experiments
  10. 5G environment and dataset
  11. Traffic forecast evaluation
  12. LP-KPI model evaluation
  13. Conclusion and future work

Figures (6)

  • Figure 1: Traffic forecasting model (LSTM-FSD ).
  • Figure 2: ML-assisted infrastructure workflow.
  • Figure 3: 5G end-to-end network slicing.
  • Figure 4: Data visualization when using LSTM-FSD on traffic from all slices with T = 5 minutes.
  • Figure 5: Gap comparison for each model (LR: linear regression, ADA: AdaBoost regressor, BAG: Bagging regressor, RIG: Ridge (Linear least squares with L2 regularization)
  • ...and 1 more figures