LIGHTWEIGHT AND FLEXIBLE INTRUSION DETECTION SYSTEM TO PROTECT INDUSTRIAL IoT SETTINGS: USING A MIXED AI STRATEGY

Hackers are targeting Industrial control systems and Industrial Internet of Things networks more often now.
These systems often rely on outdated protocols and weak security making them vulnerable. They also lack
advanced layers of defense that could adapt to new threats. Basic intrusion detection systems do not work well
in these environments. They struggle to meet the demands of fast communication real-time monitoring, and
limited resources. This research offers a simple and flexible hybrid Intrusion Detection System designed for
industrial use. The system mixes rule-based methods with machine learning to spot unusual activity. It uses tools
like deep packet inspection and analysis of time-based patterns to catch both known and new unknown
threats .Tests on benchmark datasets like NSL-KDD and a created Modbus-TCP traffic set show that detection
accuracy reaches 98.6 percent while keeping false positives low.