Toward Energy-Efficient and Low-Carbon Intrusion Detection in Edge and Cloud Computing Based on GreenShield...

Area of Science:

• Cybersecurity
• Edge-Cloud Computing
• Sustainable Computing

Background:

• Edge-cloud computing growth increases cybersecurity demands and the energy consumption of intrusion detection systems (IDSs).
• Traditional IDSs contribute significantly to the carbon footprint of computing infrastructures.
• There is a need for energy-efficient and low-carbon cybersecurity solutions.

Purpose of the Study:

• To introduce GreenShield, a novel framework for low-carbon cybersecurity in edge-cloud environments.
• To develop an energy-efficient and carbon-conscious intrusion detection system.
• To demonstrate the effectiveness of GreenShield in enhancing cybersecurity while reducing environmental impact.

Main Methods:

• Implementation of lightweight cryptography and energy-efficient deep learning.
• Utilizing a hierarchical federated learning architecture with knowledge distillation.
• Employing a carbon-aware scheduling controller for dynamic security response.

Main Results:

• GreenShield achieved 98.73% detection accuracy on UNSW-NB15 and CIC-IDS2017 datasets.
• Demonstrated a 67.4% improvement in energy efficiency compared to traditional deep learning IDSs.
• Reduced operational carbon emissions by up to 97.6% (approx. 2.8 kg CO2-equivalent/hour).

Conclusions:

• GreenShield provides a viable and scalable solution for sustainable cybersecurity.
• The framework effectively balances high detection performance with significant energy and carbon reductions.
• GreenShield supports carbon-conscious security workflows in future edge-cloud architectures.