A Stacked Ensemble Learning Framework for Accurate Communication Network State Prediction

Accurate prediction of communication network states is critical for ensuring system reliability, performance optimization, and proactive management. However, complex nonlinear dynamics, heterogeneous data sources, and evolving traffic patterns pose significant challenges. In this study, we propose a novel stacked ensemble learning framework integrating Informer, Temporal Convolutional Network (TCN), and CatBoost models, complemented by an attention-based meta-learner for dynamic fusion. Experiments on the CIC-IDS-2017 dataset demonstrate that our method outperforms traditional single models and mainstream deep learning approaches in MAE, MSE, RMSE, and R<sup xmlns:mml="http://www.w3.org/1998/Math/MathML" xmlns:xlink="http://www.w3.org/1999/xlink">2</sup> metrics. Ablation studies further validate the effectiveness of each component. Our framework offers a robust, scalable solution for intelligent communication network state prediction.