Data injection attacks in randomized gossiping

Gentz, Reinhard; Wu, Sissi Xiaoxiao; Wai, Hoi-To; Scaglione, Anna; Leshem, Amir

doi:10.1109/tsipn.2016.2614898

Cited by 33 publications

(36 citation statements)

References 18 publications

Supporting

Mentioning

Contrasting

Order By: Relevance

“…By Lemma 1, it can be shown thatŨ andṼ are sparse matrices. Then, (19) implies that (24). This completes the proof.…”

Section: B Design Of Resilient Attack Detection Estimatorssupporting

confidence: 51%

“…The statistical distribution of the nodes' data was then exploited to devise techniques for mitigating the influence of data falsifying on the detection system. In [19], the problem of detecting and mitigating data injection attacks was studied in randomized gossip-based sensor networks. By analyzing the statistics of the sensors' states, decentralized consensus strategies were designed to detect and localize insider attackers.…”

Section: B Relevant Work On Detection And/or Identification Against mentioning

confidence: 99%

See 1 more Smart Citation

Distributed Attack Detection and Secure Estimation of Networked Cyber-Physical Systems Against False Data Injection Attacks and Jamming Attacks

Guan

2018

IEEE Trans. on Signal and Inf. Process. over Networks

278

122

View full text Add to dashboard Cite

This paper is concerned with the problem of joint distributed attack detection and distributed secure estimation for a networked cyber-physical system under physical and cyber attacks. The system is monitored by a wireless sensor network in which a group of sensors is spatially distributed and the sensors' measurements are broadcast to remote estimators via a wireless network medium. A malicious adversary simultaneously launches a false data injection attack at the physical system layer to intentionally modify the system's state and jamming attacks at the cyber layer to block the wireless transmission channels between sensors and remote estimators. The sensors' measurements can be randomly dropped with mathematical probability if the corresponding transmission channels are deliberately jammed by the adversary. Resilient attack detection estimators are delicately constructed to provide locally reliable state estimations and detect the false data injection attack. Then, criteria for analyzing the estimation performance and designing the desired estimators are derived to guarantee the solvability of the problem. Finally, the effectiveness of the proposed approach is shown through an illustrative example.

show abstract

“…By Lemma 1, it can be shown thatŨ andṼ are sparse matrices. Then, (19) implies that (24). This completes the proof.…”

Section: B Design Of Resilient Attack Detection Estimatorssupporting

confidence: 51%

Section: B Relevant Work On Detection And/or Identification Against mentioning

confidence: 99%

Distributed Attack Detection and Secure Estimation of Networked Cyber-Physical Systems Against False Data Injection Attacks and Jamming Attacks

Guan

2018

IEEE Trans. on Signal and Inf. Process. over Networks

278

122

View full text Add to dashboard Cite

show abstract

“…This behavior at the application or at the physical communication layer depart significantly from the norm. In an insider attack, however, nodes can collude to force the system to converge to their target outcome following strategies that are harder to discriminate [22], [26], [27].…”

Section: Introductionmentioning

confidence: 99%

Neural Networks-Aided Insider Attack Detection for the Average Consensus Algorithm

Zhang

et al. 2020

IEEE Access

View full text Add to dashboard Cite

To support the big-data processing needs of large-scale deployments of smart devices, there is significant interest in moving from cloud-computing to multi-agent (fog-computing) models, given these algorithms scalability and self-healing properties with respect to nodes and link failures. However, these algorithms are often based on the average consensus primitive, which is, unfortunately, vulnerable to data injection attacks. Recognizing this challenge, this work proposes three novel methods for detecting and localizing adversarial nodes using neural network (NN) models. The methods proposed are based on fully distributed algorithms, wherein each node locally updates its local states by exchanging information with its neighbors without supervision. Compared to the state-of-the-art, the proposed approach leverages the automatic learning characteristics of NN to reduce the dependence on pre-designed models and human expertise in complex internal attack scenarios. Simulation results show that the NN-based methods can significantly improve the attacker detection and localization performance. INDEX TERMS Gossip algorithm, average consensus, neural network (NN), insider attack, detection and localization.

show abstract

“…For instance, we might end up excluding SNs contributing towards the FC global decision that might have high local signal-tonoise-ratios (SNRs). Recently, the authors in [19], [28] both consider a decentralized network in the presence of compromised SNs while in this paper we consider a centralized scheme. The authors in [19] propose a synchronous distributed weighted average consensus algorithm that is claimed to be robust to Byzantine attacks while reference [28] considers the detection and mitigation of data injection attacks in a randomized average consensus.…”

Section: Introductionmentioning

confidence: 99%

A Secure Optimum Distributed Detection Scheme in Under-Attack Wireless Sensor Networks

Nurellari

McLernon

Ghogho

2018

IEEE Trans. on Signal and Inf. Process. over Networks

View full text Add to dashboard Cite

Abstract-We address the problem of centralized detection of a binary event in the presence of β fraction falsifiable sensor nodes (SNs) (i.e., controlled by an attacker) for a bandwidthconstrained under − attack spatially uncorrelated distributed wireless sensor network (WSN). The SNs send their one-bit test statistics over orthogonal channels to the fusion center (FC), which linearly combines them to reach to a final decision. Adopting the modified deflection coefficient as an alternative function to be optimized, we first derive in a closed-form the FC optimal weights combining. But as these optimal weights require a − priori knowledge that cannot be attained in practice, this optimal weighted linear FC rule is not implementable. We also derive in a closed-form the expressions for the attacker "flipping probability" (defined in paper) and the minimum fraction of compromised SNs that makes the FC incapable of detecting. Next, based on the insights gained from these expressions, we propose a novel and non-complex reliability-based strategy to identify the compromised SNs and then adapt the weights combining proportional to their assigned reliability metric. In this way, the FC identifies the compromised SNs and decreases their weights in order to reduce their contributions towards its final decision. Finally, simulation results illustrate that the proposed strategy significantly outperforms (in terms of FC's detection capability) the existing compromised SNs identification and mitigation schemes.

show abstract

Data injection attacks in randomized gossiping

Cited by 33 publications

References 18 publications

Distributed Attack Detection and Secure Estimation of Networked Cyber-Physical Systems Against False Data Injection Attacks and Jamming Attacks

Distributed Attack Detection and Secure Estimation of Networked Cyber-Physical Systems Against False Data Injection Attacks and Jamming Attacks

Neural Networks-Aided Insider Attack Detection for the Average Consensus Algorithm

A Secure Optimum Distributed Detection Scheme in Under-Attack Wireless Sensor Networks

Contact Info

Product

Resources

About