Strengthening Security in the Internet of Things (IoT): Integrated Approach of Intrusion Detection Systems (IDS) and Edge Computing

The rapid growth of Internet of Things (IoT) applications has led to an increase in sophisticated and malicious attacks, threatening network security and privacy in sectors such as healthcare, industry, and home automation. To address these challenges, The implementation of Intrusion Detection Systems (IDS) has become crucial for detecting and mitigating these threats. At the same time, the emergence of Edge Computing technology offers significant advantages in terms of privacy and performance for IoT infrastructures. However, deploying IDS on resource-limited Edge devices requires innovative solutions to ensure effective protection against cyber threats.By integrating IDS on IoT devices and strengthening security at Edge points, it becomes possible to effectively detect and mitigate malicious attacks while optimizing resource usage and ensuring the protection of sensitive data. By placing security at the core of IoT and Edge network design and deployment, it is possible to guarantee a robust and resilient infrastructure capable of addressing both current and future security challenges.This paper focuses on the study and implementation of machine learning algorithms applied to Intrusion Detection Systems (IDS) within IoT networks. Specifically, we explore the use of real-world data IOT-23 dataset to improve the detection capabilities of IDS. The results show that boosting models, such as XGBoost, CatBoost, and Gradient Boosting, as well as Random Forests, deliver the highest performance with an F1-score of 0.84. These models demonstrate strong capabilities in detecting intrusions while maintaining an ideal balance between precision and recall. These results underscore the effectiveness of these algorithms in improving the security of IoT and Edge networks, paving the way for innovative strategies that allow for optimal resource utilization while ensuring robust protection of sensitive data.