Overcoming Data Limitations in Internet Traffic Forecasting: LSTM Models with Transfer Learning and Wavelet Augmentation

TL;DR

This paper demonstrates that transfer learning and wavelet-based data augmentation significantly improve internet traffic forecasting accuracy in small ISP networks, especially for short-term predictions, using LSTM models.

Contribution

It introduces the combined use of transfer learning and wavelet augmentation with LSTM models to address data scarcity in internet traffic prediction.

Findings

Wavelet augmentation improves short-term forecast accuracy.

LSTMSeq2Seq outperforms LSTMSeq2SeqAtn in small datasets.

Data augmentation is essential for limited data scenarios.

Abstract

Effective internet traffic prediction in smaller ISP networks is challenged by limited data availability. This paper explores this issue using transfer learning and data augmentation techniques with two LSTM-based models, LSTMSeq2Seq and LSTMSeq2SeqAtn, initially trained on a comprehensive dataset provided by Juniper Networks and subsequently applied to smaller datasets. The datasets represent real internet traffic telemetry, offering insights into diverse traffic patterns across different network domains. Our study revealed that while both models performed well in single-step predictions, multi-step forecasts were challenging, particularly in terms of long-term accuracy. In smaller datasets, LSTMSeq2Seq generally outperformed LSTMSeq2SeqAtn, indicating that higher model complexity does not necessarily translate to better performance. The models' effectiveness varied across different network domains, reflecting the influence of distinct traffic characteristics. To address data scarcity, Discrete Wavelet Transform was used for data augmentation, leading to significant improvements in model performance, especially in shorter-term forecasts. Our analysis showed that data augmentation is crucial in scenarios with limited data. Additionally, the study included an analysis of the models' variability and consistency, with attention mechanisms in LSTMSeq2SeqAtn providing better short-term forecasting consistency but greater variability in longer forecasts. The results highlight the benefits and limitations of different modeling approaches in traffic prediction. Overall, this research underscores the importance of transfer learning and data augmentation in enhancing the accuracy of traffic prediction models, particularly in smaller ISP networks with limited data availability.