ENHANCING IOT SECURITY: INTEGRATING MACHINE LEARNING INTO STATIC AND STREAMING IDS
Amrani Ayoub, Ettazi Haitam · Proceedings on Engineering Sciences · 2025
The Internet of Things (IoT) has brought incredible advancements in healthcare, industry, smart cities, and home automation.But with these innovations come serious security challenges.Many IoT devices, which often have limited resources, are attractive targets for cyberattacks due to their extensive connectivity.This situation puts the confidentiality, integrity, and availability of data at risk.To tackle these issues, technologies like artificial intelligence (AI) and machine learning (ML) are becoming increasingly important.AI, especially through ML and deep learning (DL), provides powerful tools for spotting and preventing attacks.By analyzing the massive amounts of data generated by IoT devices, ML algorithms can learn to identify normal behavior patterns and detect any anomalies that might indicate malicious activity.Intrusion detection systems are a key part of this security strategy, spotting suspicious behaviors and supporting traditional measures.For example, using different ML algorithms on a database of IoT communication frames (IOTID20) highlights how effective these approaches can be in protecting IoT systems from emerging threats.