Modified Firefly Algorithm Multi-Layer Perceptron Optimization Approach for Intrusion Detection System in Mobile Ad Hoc Networks

Mobile Ad Hoc Networks (MANETs) present unique security challenges due to their decentralized architecture, node mobility, and limited computational resources. To address these issues, this paper suggests a very accurate intrusion detection system (IDS) that combines a Modified Firefly Algorithm (MFA) with a simple Multi-Layer Perceptron (MLP) classifier. The system performs decentralized feature selection at each node, guided by trust and energy parameters to ensure contextaware optimization. Evaluated using a filtered version of the UNSW-NB15 dataset, the proposed model consistently achieves a detection accuracy of <tex xmlns:mml="http://www.w3.org/1998/Math/MathML" xmlns:xlink="http://www.w3.org/1999/xlink">$\mathbf{9 9. 0 0} \%$</tex> across multiple simulated nodes. This exceptional accuracy highlights the effectiveness of the framework in identifying malicious activities while minimizing false detections, making it a promising solution for real-time intrusion detection in dynamic and resource-constrained MANET environments.