A White Shark Equilibrium Optimizer with a Hybrid Deep-Learning-Based Cybersecurity Solution for a Smart City...

Area of Science:

• Cybersecurity
• Smart Grids
• Artificial Intelligence

Background:

• Smart grids are crucial for smart cities, managing energy efficiently.
• Distributed Denial of Service (DDoS) attacks pose a significant cybersecurity threat to smart grid stability.
• Existing methods struggle with the complexity and volume of data in smart grid cybersecurity.

Purpose of the Study:

• To develop an advanced cybersecurity solution for smart grids to detect and mitigate DDoS attacks.
• To enhance the reliability and stability of smart grids within smart city infrastructures.
• To address the challenge of high-dimensionality data in identifying cyber threats.

Main Methods:

• A novel White Shark Equilibrium Optimizer with a Hybrid Deep-Learning-based Cybersecurity Solution (WSEO-HDLCS) is proposed.
• WSEO-based feature selection (WSEO-FS) is employed to handle high-dimensional data.
• A stacked deep autoencoder (SDAE) model is utilized for DDoS attack detection, with hyperparameters optimized by the Gravitational Search Algorithm (GSA).

Main Results:

• The WSEO-HDLCS technique effectively identifies the presence of DDoS attacks in smart grids.
• The WSEO-FS approach successfully resolves high-dimensionality data challenges.
• Simulations on the CICIDS-2017 dataset demonstrated superior performance compared to existing methodologies.

Conclusions:

• The WSEO-HDLCS technique offers a promising and effective solution for smart grid cybersecurity against DDoS attacks.
• This approach enhances the security and reliability of smart grids in smart city environments.
• The study highlights the potential of hybrid deep learning and optimization algorithms for advanced cyber threat detection.