Enhancing IoT security: a hybrid intelligent intrusion detection system integrating machine learning and metaheuristic algorithm

Abstract

The rapid proliferation of the internet of things (IoT) has introduced significant security and privacy challenges. As IoT devices often have limited computational power and memory, they are highly vulnerable to cyber threats. Traditional intrusion detection systems (IDS) struggle to operate efficiently in these constrained environments, necessitating more adaptive and optimized security solutions. To address these challenges, this study proposes an innovative IDS model, MSAMLP, which combines the moth search algorithm (MSA) with a multilayer perceptron (MLP) classifier. The objective is to enhance the classification accuracy of malicious and benign network traffic while maintaining computational efficiency. The model was evaluated using two widely recognized intrusion detection datasets, benchmarking its performance against existing IDS approaches. Experimental results indicate that MSAMLP outperforms conventional classification models, achieving high accuracy, improved detection rates, and reduced false alarm rates. Its adaptive learning capability ensures better anomaly detection in dynamic IoT environments. In conclusion, the proposed MSAMLP model demonstrates superior performance in securing IoT networks, offering an effective solution to mitigate evolving cyber threats. This research contributes to the advancement of IoT security by introducing a robust and scalable intrusion detection approach.