A Novel Approach to Wireless Network Traffic Prediction using CNN-LSTM Attention Model with Prophet Model

This paper presents a novel methodology for predicting network traffic in complex and dynamic environments. In order to tackle the difficulties associated with precise forecasting of network traffic within such settings, our proposed solution involves a model architecture that integrates a CNN-LSTM attention model and Prophet model. This approach enables the capture of both local and global patterns inherent in complex time series data. The experimental results indicate that our proposed methodology surpasses the current state-of-the-art models in accuracy. Our proposed methodology shows promise for improving network traffic prediction in complex and dynamic environments, with potential for further enhancements through incorporating additional data sources or optimizing model parameters.