Paper Contents
Abstract
The increasing sophistication of cyber threats in modern networks necessitates intelligent and resilient security solutions. Conventional intrusion detection systems (IDS) are often hindered by high computational requirements and limited scalability, making them less effective for real-time use in resource-constrained settings. To overcome these limitations, this research introduces energy-efficient IDS leveraging the CICIDS2017 dataset, which offers a diverse mix of benign and malicious traffic for robust training and evaluation. The proposed method employs a hybrid deep learning framework that combines Convolutional Neural Networks (CNNs) and Long Short-Term Memory (LSTM) networks to extract both spatial and temporal characteristics from network flows. Pruning and quantization enhance energy efficiency without reducing accuracy. The model achieves strong performance while lowering energy use, making it fit for real-time, energy-sensitive environments.
Copyright
Copyright © 2025 Rahul Senthiya. This is an open access article distributed under the Creative Commons Attribution License.
