# An Analytics Framework for Heuristic Inference Attacks against Industrial Control Systems

**Authors:** Taejun Choi, Guangdong Bai, Ryan K L Ko, Naipeng Dong, Wenlu Zhang, Shunyao Wang  
**arXiv:** [2101.11866](https://arxiv.org/abs/2101.11866)  
**Date:** 2021-01-28  
**Summary page:** https://www.summarizepaper.com/en/paper/2101.11866v1/

## Key points

- The paper addresses cyber attacks on industrial control systems (ICS) within critical infrastructure
- Introduces a vendor-agnostic analytics framework for analyzing attacks on ICS systems without prior knowledge of specific systems or protocols
- Uses 'digital twin' scenarios to execute stealthy deception attacks based on false data injection attacks (FDIA)
- Demonstrates ease in collecting attack datasets and leveraging common penetration testing tools for analysis
- Introduces the concept of 'heuristic inference attacks,' a novel category of attacks on ICS systems independent of specific PLC and HMI models
- Validates efficacy of these attacks using an ICS dataset from a cyber-physical scenario involving water utilities
- Recommends countermeasures to enhance security posture against potential threats, based on findings
- Provides valuable insights into addressing cybersecurity challenges faced by critical infrastructure relying on interconnected industrial control systems

## Summary

The paper "An Analytics Framework for Heuristic Inference Attacks against Industrial Control Systems" by Taejun Choi, Guangdong Bai, Ryan K L Ko, Naipeng Dong, Wenlu Zhang, and Shunyao Wang addresses the growing concern of cyber attacks on industrial control systems (ICS) within critical infrastructure. The authors introduce a vendor-agnostic analytics framework that enables security researchers to analyze attacks on ICS systems without requiring prior knowledge of control automation domains or specific ICS systems. This framework simplifies the process by eliminating the need for familiarity with ICS communication protocols, PLCs, or network penetration testing tools. Through the use of 'digital twin' scenarios replicating industry-standard HMIs, PLCs, and firewalls in a controlled test environment, the framework successfully executes a stealthy deception attack based on false data injection attacks (FDIA). It also demonstrates ease in collecting attack datasets and leveraging common penetration testing tools for analysis. The paper introduces the concept of 'heuristic inference attacks,' a novel category of attacks on ICS systems that are independent of specific PLC and HMI models commonly found in such environments. The efficacy of these attacks is validated using an ICS dataset obtained from a cyber-physical scenario involving water utilities. Additionally, the authors leverage time complexity theory to estimate an attacker's difficulty in conducting packet analyses as part of proposed attacks. Based on their findings, they recommend countermeasures to enhance the security posture of ICS systems against potential threats. This comprehensive approach provides valuable insights into addressing cybersecurity challenges faced by critical infrastructure sectors relying on interconnected industrial control systems.