A Survey of Artificial Intelligence and Metaheuristic Techniques for Enhancing loT Security

The extensive use of the Internet of Things (IoT) in many different fields has brought about important advancements, but it has also created intricate and ever-changing security issues. This study examines cutting-edge approaches to improving IoT security, with an emphasis on combining metaheuristic algorithms and artificial intelligence (AI) techniques. The goal of the project is to find out how learning-based and optimization-driven methods may identify and lessen security risks including DDoS assaults, data breaches, and encryption flaws. Common AI techniques, such as machine learning and deep learning, as well as metaheuristic algorithms like genetic algorithms, particle swarm optimization, ant colony optimization, and grey wolf optimizer, are reviewed in an organized manner. Performance in intrusion detection, feature selection, cryptographic optimization, and threat prediction is also examined for hybrid models that include AI and metaheuristics. To show the usefulness of these methods in actual IoT environments, a number of successful implementations are highlighted. Problems including scalability problems, lack of interpretability, and computational limitations are also covered. Federated learning, explainable AI, and lightweight architectures are highlighted as viable avenues for developing safe, distributed, and intelligent IoT settings in the paper's final section, which also identifies future research priorities.

کلیدواژه ها:

IoT security ، artificial intelligence ، metaheuristic algorithms ، intrusion detection ، hybrid models ، optimization