Coding Schemes for Securing Cyber-Physical Systems Against Stealthy Data Injection Attacks

Fei Miao, Quanyan Zhu, Miroslav Pajic, George J. Pappas

Abstract

This paper considers a method of coding the sensor outputs in order to detect stealthy false data injection attacks. An intelligent attacker can design a sequence of data injection to sensors and actuators that pass the state estimator and statistical fault detector, based on knowledge of the system parameters. To stay undetected, the injected data should increase the state estimation errors while keep the estimation residues small. We employ a coding matrix to change the original sensor outputs to increase the estimation residues under intelligent data injection attacks. This is a low cost method compared with encryption schemes over all sensor measurements in communication networks. We show the conditions of a feasible coding matrix under the assumption that the attacker does not have knowledge of the exact coding matrix. An algorithm is developed to compute a feasible coding matrix, and, we show that in general, multiple feasible coding matrices exist. To defend against attackers who estimates the coding matrix via sensor and actuator measurements, time-varying coding matrices are designed according to the detection requirements. A heuristic algorithm to decide the time length of updating a coding matrix is then proposed.

Introduction

Cyber-physical systems (CPS)—such as power grids, autonomous vehicles, and medical devices—rely on sensor data to estimate their physical state and make control decisions. However, these systems are increasingly vulnerable to "stealthy" false data injection attacks. An intelligent attacker who understands the mathematical model of the system can craft data injections that drive state estimation errors to infinity while keeping the "residue" (the difference between expected and actual sensor readings) small enough to bypass standard statistical fault detectors.

System model illustrating an attacker injecting false data into both actuators and sensors to bypass the detector.

While traditional bit-level encryption could prevent such tampering, it is often too computationally expensive and introduces excessive communication overhead for resource-constrained embedded sensors. This paper introduces a lightweight, low-cost coding scheme to protect CPS against these sophisticated threats.

Methodology: Sensor Output Coding

The authors propose a linear algebraic defense mechanism that acts as a lightweight alternative to encryption. The core of this defense is the application of a secret, invertible "coding matrix" ($\Sigma$) to sensor measurements before they are transmitted over the network.

The proposed encoder block applies a coding matrix Σ to sensor outputs before they reach the communication channel.

Breaking the Stealth

Stealthy attacks work by hiding in specific mathematical "blind spots" of the system, specifically aligning with unstable eigenvectors. The coding matrix effectively rotates the sensor outputs. If an attacker injects a signal designed for the original, uncoded system, the secret rotation pushes their signal out of the stealthy subspace. This causes the fault detector's residue to spike, triggering an immediate alarm.

Dynamic Defense and Matrix Generation

To ensure the attacker cannot easily guess the matrix, the authors provide an algorithm using Givens rotations to generate feasible coding matrices. Furthermore, to counter an advanced attacker who might try to reverse-engineer $\Sigma$ by eavesdropping on measurements, the authors propose a time-varying defense. By periodically updating the coding matrix, the system creates a "moving target," forcing the attacker to solve complex, noisy bilinear equations faster than the system updates its security parameters.

Key Insights and Results

  • Active Monitoring Limitations: The research highlights that simply adding random noise or authentication signals to control inputs is insufficient to detect pure sensor data injection; a dedicated sensor-side coding approach is required.
  • Computational Efficiency: The proposed matrix multiplication requires only $O(n^3 + p^3)$ operations (where $n$ is the number of states and $p$ is the number of sensors). This is significantly cheaper than complex non-linear cryptographic primitives.
  • Simulation Performance: In a simulated 2D Linear Time-Invariant (LTI) system, an attack that remained completely undetected on an uncoded system triggered the alarm in just 12 seconds once the coding matrix was applied. Even when an attacker was allowed to eavesdrop for 20 time steps, their newly crafted attack still triggered the alarm within 50 seconds, proving the system has ample time to rotate the matrix and stay ahead.

Conclusion and Future Work

This coding scheme provides a robust, low-overhead defense for embedded architectures where traditional encryption is impractical. Future work will explore coding schemes for systems with structural constraints—such as sparse sensor networks—and distributed coding schemes where multiple matrices are applied across different segments of a large-scale system.

Summary updated

Abstract

This paper considers a method of coding the sensor outputs in order to detect stealthy false data injection attacks. An intelligent attacker can design a sequence of data injection to sensors and actuators that pass the state estimator and statistical fault detector, based on knowledge of the system parameters. To stay undetected, the injected data should increase the state estimation errors while keep the estimation residues small. We employ a coding matrix to change the original sensor outputs to increase the estimation residues under intelligent data injection attacks. This is a low cost method compared with encryption schemes over all sensor measurements in communication networks. We show the conditions of a feasible coding matrix under the assumption that the attacker does not have knowledge of the exact coding matrix. An algorithm is developed to compute a feasible coding matrix, and, we show that in general, multiple feasible coding matrices exist. To defend against attackers who estimates the coding matrix via sensor and actuator measurements, time-varying coding matrices are designed according to the detection requirements. A heuristic algorithm to decide the time length of updating a coding matrix is then proposed.
BibTeX
@article{miao2017coding,
  author = {Fei Miao and Quanyan Zhu and Miroslav Pajic and George J. Pappas},
  title = {{Coding Schemes for Securing Cyber-Physical Systems Against Stealthy Data Injection Attacks}},
  journal = {IEEE Trans. Control. Netw. Syst.},
  year = {2017},
  url = {https://openreview.net/forum?id=_Zs2ktWXuGS},
}
Figure