Cyber physical attacks with control objectives and detection constraints

Chen, Yuan; Kar, Soummya; Moura, J.M.F.

doi:10.1109/cdc.2016.7798418

Cited by 22 publications

(18 citation statements)

References 17 publications

Supporting

Mentioning

Contrasting

Order By: Relevance

“…Research on attack design includes developing conditions for undetectable FDI attack [10], design of a linear deception attack scheme to fool the popular χ 2 detector (see [6]), optimal attack design for noiseless systems [11], FDI design to penetrate AC-based bad data detection system [12], etc. The paper [13] designs an optimal attack to steer the state of a control system to a desired target under a constraint on the attack detection probability. Stealth attack design on a quantized networked control system has been solved in [14].…”

Section: A Related Literaturementioning

confidence: 99%

See 1 more Smart Citation

Design of False Data Injection Attack on Distributed Process Estimation

Moulik

Chattopadhyay

Mitra

et al. 2022

IEEE Trans.Inform.Forensic Secur.

View full text Add to dashboard Cite

Section: A Related Literaturementioning

confidence: 99%

“…) to solve (UP) at time t by using the set of linear equations ( 10), (11), (12), (13). Then it makes the following update:…”

Section: B Updating λ(T) Iteratively: Olaade-kktmentioning

confidence: 99%

Design of False Data Injection Attack on Distributed Process Estimation

Moulik

Chattopadhyay

Mitra

et al. 2022

IEEE Trans.Inform.Forensic Secur.

View full text Add to dashboard Cite

“…The problem of FDI attack design and its detection has received significant attention in recent times; attack design: conditions for undetectable FDI attack [4], design of a linear deception attack scheme to fool the popular χ 2 detector (see [5]), optimal attack design for noiseless systems [6]. The paper [7] designs an optimal attack to steer the state of a control system to a desired target under a constraint on the attack detection probability. On the other hand, attempts on attack detection includes centralized (and decentralized as well) schemes for noiseless systems [8], coding of sensor output along with χ 2 detector [9], comparing the sensor observations with those coming from from a few known safe sensors [10], and the attack detection and secure estimation schemes based on innovation vectors in [11].…”

Section: A Related Literaturementioning

confidence: 99%

Optimal deception attack on networked vehicular cyber physical systems

Moulik

Chattopadhyay

Mitra

et al. 2019

2019 53rd Asilomar Conference on Signals, Systems, and Computers

View full text Add to dashboard Cite

Herein, design of false data injection attack on a distributed cyber-physical system is considered. A stochastic process with linear dynamics and Gaussian noise is measured by multiple agent nodes, each equipped with multiple sensors. The agent nodes form a multi-hop network among themselves. Each agent node computes an estimate of the process by using its sensor observation and messages obtained from neighbouring nodes, via Kalman-consensus filtering. An external attacker, capable of arbitrarily manipulating the sensor observations of some or all agent nodes, injects errors into those sensor observations. The goal of the attacker is to steer the estimates at the agent nodes as close as possible to a pre-specified value, while respecting a constraint on the attack detection probability. To this end, a constrained optimization problem is formulated to find the optimal parameter values of a certain class of linear attacks. The parameters of linear attack are learnt on-line via a combination of stochastic approximation and online stochastic gradient descent. Numerical results demonstrate the efficacy of the attack.The attacker computes S k (t) . = (U k (t)) U k (t) for all 1 ≤ k ≤ N . 5) The sensors make observations {y k (t)} 1≤k≤N , which are accessed by the attacker. 6) The attacker calculates z k (t) = y k (t)−H k Ax (k) (t−1) for all k ∈ {1, 2, · · · , N }. 7) The attacker calculatesz k (t) = T k (t)z k (t) + b k (t) for all k ∈ {1, 2, · · · , N }, where b k (t) ∼ N (M k (t)θ (k) (t− 1) + d k (t), S k (t)) chosen independently of all other

show abstract

“…The problem of FDI attack design and its detection has received significant attention in recent times; attack design: conditions for undetectable FDI attack [5], design of a linear deception attack scheme to fool the popular χ 2 detector (see [6]), optimal attack design for noiseless systems [7]. The paper [8] designs an optimal attack to steer the state of a control system to a desired target under a constraint on the attack detection probability. On the other hand, attempts on attack detection includes centralized (and decentralized as well) schemes for noiseless systems [9], coding of sensor output along with χ 2 detector [10], comparing the sensor observations with those coming from from a few known safe sensors [11], and the attack detection and secure estimation schemes based on innovation vectors in [12].…”

Section: Introductionmentioning

confidence: 99%

Design of false data injection attack on distributed process estimation

Moulik¹,

Chattopadhyay²,

Mitra³

et al. 2021

Preprint

View full text Add to dashboard Cite

Herein, design of false data injection attack on a distributed cyber-physical system is considered. A stochastic process with linear dynamics and Gaussian noise is measured by multiple agent nodes, each equipped with multiple sensors. The agent nodes form a multi-hop network among themselves. Each agent node computes an estimate of the process by using its sensor observation and messages obtained from neighboring nodes, via Kalman-consensus filtering. An external attacker, capable of arbitrarily manipulating the sensor observations of some or all agent nodes, injects errors into those sensor observations. The goal of the attacker is to steer the estimates at the agent nodes as close as possible to a pre-specified value, while respecting a constraint on the attack detection probability. To this end, a constrained optimization problem is formulated to find the optimal parameter values of a certain class of linear attacks. The parameters of linear attack are learnt on-line via a combination of stochastic approximation based update of a Lagrange multiplier, and an optimization technique involving either the Karush-Kuhn-Tucker (KKT) conditions or online stochastic gradient descent. The problem turns out to be convex for some special cases. Desired convergence of the proposed algorithms are proved by exploiting the convexity and properties of stochastic approximation algorithms. Finally, numerical results demonstrate the efficacy of the attack.

show abstract

Cyber physical attacks with control objectives and detection constraints

Cited by 22 publications

References 17 publications

Design of False Data Injection Attack on Distributed Process Estimation

Design of False Data Injection Attack on Distributed Process Estimation

Optimal deception attack on networked vehicular cyber physical systems

Design of false data injection attack on distributed process estimation

Contact Info

Product

Resources

About