Enhancing security and efficiency in Mobile Ad Hoc Networks using a hybrid deep learning model for flooding attack...

Area of Science:

• Computer Science
• Network Security
• Artificial Intelligence

Background:

• Mobile Ad Hoc Networks (MANETs) offer decentralized communication but are vulnerable to flooding attacks.
• Flooding attacks disrupt MANETs by degrading performance and draining energy resources.

Purpose of the Study:

• To develop a hybrid deep learning model for detecting and mitigating flooding attacks in MANETs.
• To enhance the security, reliability, and energy efficiency of MANETs.

Main Methods:

• Integration of Convolutional Neural Networks (CNN), Long Short-Term Memory (LSTM), and Gated Recurrent Unit (GRU) architectures.
• Optimization using a novel DECEHGS algorithm (Differential Evolution and Evolutionary Population Dynamics).

Main Results:

• Achieved 95% accuracy in detecting malicious nodes.
• Increased packet delivery ratio by 12%.
• Reduced routing overhead by 20% compared to traditional methods.

Conclusions:

• The proposed hybrid deep learning model offers an effective and energy-efficient solution for MANET security.
• Demonstrated significant improvements in network performance and robustness against attacks.