# Fishy Cyber Attack Detection in Industrial Control Systems

## TL;DR

This paper introduces an automated LSTM-based method for detecting and localizing cyber attacks in industrial control systems, enhancing security and safety in critical infrastructure environments.

## Contribution

It presents a novel automated approach using LSTM networks for attack detection and localization in industrial control systems, improving upon existing zone division and PCA methods.

## Key findings

- Effective detection of cyber attacks in simulated water plant
- Accurate localization of compromised nodes
- Enhanced security and safety in industrial control systems

## Abstract

Cyber attacks have become serious threats to Industrial Control systems as well. It becomes important to develop a serious threat defense system against such vulnerabilities. For such process control systems, safety should also be assured apart from security. As unearthing vulnerabilities and patching them is not a feasible solution, these critical infrastructures need safeguards to prevent accidents, both natural and artificial, that could potentially be hazardous. Morita proposed an effective Zone division, capable of evaluating remote and concealed attacks on the system, coupled with Principal Component Analysis. But the need to analyze the node that has been compromised and stopping any further damages, requires an automated technique. Illustrating the basic ideas we'll simulate a simple Water plant. We propose a new automated approach based on Long Short Term Memory networks capable of detecting attacks and pin point the location of the breach.