Resilient Optimal Estimation Using Measurement Prior

Anubi, Olugbenga Moses; Konstantinou, Charalambos; Roberts, Rodney

doi:10.48550/arxiv.1907.13102

Cited by 4 publications

(10 citation statements)

References 34 publications

Supporting

Mentioning

Contrasting

Order By: Relevance

“…Lemma 3.1: Given a support prior T generated by any underlying localization algorithm with associated uncertainty model in (11), if a pruned support prior Tc η is generated by ( 14), then it satisfies…”

Section: A Prior Pruningmentioning

confidence: 99%

“…for all sparse vector h ∈ S T m km . Given a support prior Tc generated by an underlying localization algorithm with associated uncertainty model in (11). Let Tc η be the pruned support prior given by ( 14) with a parameter η ∈ (0, 1).…”

Section: A Prior Pruningmentioning

confidence: 99%

“…Due to sparsity assumption of the attack vector |supp(e)| ≤ k, the resilient estimation problem has been cast as a classical error correction problem [5,11]. Consider a linear observation model y = Hx + e, where H ∈ R N×n denotes an observation matrix, then the resilient estimation is formulated as 0-norm minimization problem [12].…”

Section: Introductionmentioning

confidence: 99%

See 2 more Smart Citations

Attack-Resilient Weighted $\ell_1$ Observer with Prior Pruning

Anubi

2021

Preprint

Self Cite

View full text Add to dashboard Cite

Security related questions for Cyber Physical Systems (CPS) have attracted much research attention in searching for novel methods for attack-resilient control and/or estimation. Specifically, false data injection attacks (FDIAs) have been shown to be capable of bypassing bad data detection (BDD), while arbitrarily compromising the integrity of state estimators and robust controller even with very sparse measurements corruption. Moreover, based on the inherent sparsity of pragmatic attack signals, 1 -minimization scheme has been used extensively to improve the design of attackresilient estimators. For this, the theoretical maximum for the percentage of compromised nodes that can be accommodated has been shown to be 50%. In order to guarantee correct state recoveries for larger percentage of attacked nodes, researchers have begun to incorporate prior information into the underlying resilient observer design framework. For the most pragmatic cases, this prior information is often obtained through some data-driven machine learning process. Existing results have shown strong positive correlation between the tolerated attack percentages and the precision of the prior information. In this paper, we present a pruning method to improve the precision of the prior information, given corresponding stochastic uncertainty characteristics of the underlying machine learning model. Then a weighted 1 -minimization is proposed based on the pruned prior. The theoretical and simulation results show that the pruning method significantly improves the observer performance for much larger attack percentages, even when moderately accurate machine learning model used.

show abstract

Section: A Prior Pruningmentioning

confidence: 99%

Section: A Prior Pruningmentioning

confidence: 99%

See 1 more Smart Citation

Attack-Resilient Weighted $\ell_1$ Observer with Prior Pruning

Anubi

2021

Preprint

Self Cite

View full text Add to dashboard Cite

show abstract

“…These estimators have been validated using cruise control of autonomous ground vehicle, electrical power systems, industrial control systems. However, with the exception of [13], none of the estimators were validated against large percentage FDIA. Also, in [14], a robot intrusion detection system (RIDS) is designed by leveraging physical dynamics of mobile robots.…”

Section: Introductionmentioning

confidence: 97%

“…Many authors [9,12,13] have proposed an 0 -based resilient state estimators with different modifications or under different scenarios. These estimators have been validated using cruise control of autonomous ground vehicle, electrical power systems, industrial control systems.…”

Section: Introductionmentioning

confidence: 99%

Attack-resilient observer pruning for path-tracking control of Wheeled Mobile Robot

Yu¹,

Anubi²

2020

Preprint

Self Cite

View full text Add to dashboard Cite

Path-tracking control of wheeled mobile robot (WMR) has gained a lot of research attention, primarily because of its wide applicability -for example intelligent wheelchairs, explorationassistant remote WMR. Recent increase in remote and autonomous operations/requirements for WMR has led to more and more use of IoT devices within the control loop. Consequently, providing interfaces for malicious interactions through false data injection attacks (FDIA). Moreover, optimization-based FDIAs have been shown to cause catastrophic consequences in feedback control systems while by-passing any residual-based monitoring system. Since these attacks target system measurement process, this paper focuses on the problem of improving the resiliency of dynamical observers against FDIA. Specifically, we propose an attack-resilient pruning algorithm which attempts to exclude compromised channels from being processed by the observer. The proposed pruning algorithm improves attacklocalization precision to 100% with high probability, which correspondingly improves the resiliency of the underlying UKF to FDIA. The improvements due to the developed resilient pruningbased observer is validated through a numerical simulation of a two-layer path-tracking control platform of differential-driven wheeled mobile robot (DDWMR) under FDIA. NOMENCLATUREThe following notations and definitions are used throughout the whole paper: R, R n , R n×m denote the space of real numbers, real vectors of length n and real matrices of n rows and m columns respectively. R + denotes positive real numbers. Normal-face lower-case letters (e.g. x ∈ R) are used to represent real scalar, bold-face lower-case letter (e.g. x ∈ R n ) represents vectors, while normal-face upper case (e.g. X ∈ R n×m ) represents matrices. 1 represents all-ones vector. Let T ⊆ {1, . . . , n}, then for a matrix X ∈ R n×m , X T ∈ R |T |×m is the sub-matrix obtained by extracting the rows of X corresponding to the indices in T . T c denotes the complement of a set T and the universal set on which it is defined will be clear from the context. The symbol • denotes element-wise multiplication of two vectors and is defined as z = x • y, where z i = x i • y i . The symbol * denotes the convolution operator for vectors. supp(x) donotes the support of the vector x given by the set T = supp(x) = {i|x i = 0}. argsort ↓ (x) denotes a function that returns the sorted indices of vector x in descending order. The space of all square integrable signals is denoted by L 2 . The space of all point-wise bounded signals is denoted by L ∞ .

show abstract

A Survey of Machine Learning Methods for Detecting False Data Injection Attacks in Power Systems

Sayghe¹,

Hu²,

Zografopoulos³

et al. 2020

Preprint

Self Cite

View full text Add to dashboard Cite

Over the last decade, the number of cyberattacks targeting power systems and causing physical and economic damages has increased rapidly. Among them, False Data Injection Attacks (FDIAs) is a class of cyberattacks against power grid monitoring systems. Adversaries can successfully perform FDIAs in order to manipulate the power system State Estimation (SE) by compromising sensors or modifying system data. SE is an essential process performed by the Energy Management System (EMS) towards estimating unknown state variables based on system redundant measurements and network topology. SE routines include Bad Data Detection (BDD) algorithms to eliminate errors from the acquired measurements, e.g., in case of sensor failures. FDIAs can bypass BDD modules to inject malicious data vectors into a subset of measurements without being detected, and thus manipulate the results of the SE process. In order to overcome the limitations of traditional residual-based BDD approaches, datadriven solutions based on machine learning algorithms have been widely adopted for detecting malicious manipulation of sensor data due to their fast execution times and accurate results. This paper provides a comprehensive review of the most up-to-date machine learning methods for detecting FDIAs against power system SE algorithms.

show abstract

Resilient Optimal Estimation Using Measurement Prior

Cited by 4 publications

References 34 publications

Attack-Resilient Weighted $\ell_1$ Observer with Prior Pruning

Attack-Resilient Weighted $\ell_1$ Observer with Prior Pruning

Attack-resilient observer pruning for path-tracking control of Wheeled Mobile Robot

A Survey of Machine Learning Methods for Detecting False Data Injection Attacks in Power Systems

Contact Info

Product

Resources

About