DeepGuard: Backdoor Attack Detection and Identification Schemes in Privacy-Preserving Deep Neural Networks

Chen, Congcong; Wei, Lifei; Zhang, Lei; Peng, Ya; Ning, Jianting

doi:10.1155/2022/2985308

Cited by 5 publications

(2 citation statements)

References 29 publications

Supporting

Mentioning

Contrasting

Order By: Relevance

“…By manipulating the weight of the adversarial updates, the backdoor pattern is bound to be injected into the converged global model. However, the attack cost is also the highest among all attack methodologies [185].…”

Section: Discussionmentioning

confidence: 99%

Defense Strategies Toward Model Poisoning Attacks in Federated Learning: A Survey

Wang

Qiao

Zhang

et al. 2022

2022 IEEE Wireless Communications and Networking Conference (WCNC)

View full text Add to dashboard Cite

Due to the greatly improved capabilities of devices, massive data, and increasing concern about data privacy, Federated Learning (FL) has been increasingly considered for applications to wireless communication networks (WCNs). Wireless FL (WFL) is a distributed method of training a global deep learning model in which a large number of participants each train a local model on their training datasets and then upload the local model updates to a central server. However, in general, nonindependent and identically distributed (non-IID) data of WCNs raises concerns about robustness, as a malicious participant could potentially inject a "backdoor" into the global model by uploading poisoned data or models over WCN. This could cause the model to misclassify malicious inputs as a specific target class while behaving normally with benign inputs. This survey provides a comprehensive review of the latest backdoor attacks and defense mechanisms. It classifies them according to their targets (data poisoning or model poisoning), the attack phase (local data collection, training, or aggregation), and defense stage (local training, before aggregation, during aggregation, or after aggregation). The strengths and limitations of existing attack strategies and defense mechanisms are analyzed in detail. Comparisons of existing attack methods and defense designs are carried out, pointing to noteworthy findings, open challenges, and potential future research directions related to security and privacy of WFL.

show abstract

Section: Discussionmentioning

confidence: 99%

Defense Strategies Toward Model Poisoning Attacks in Federated Learning: A Survey

Wang

Qiao

Zhang

et al. 2022

2022 IEEE Wireless Communications and Networking Conference (WCNC)

View full text Add to dashboard Cite

show abstract

“…So, security cannot be dealt with independently [12]. It provides an inclusive analysis of intrusion detection on the basis of deep learning techniques followed by diferent intrusion detection systems [13,14]. DeepGuard is proposed, which is a framework of privacy-preserving backdoor detection and identifcation in an outsourced cloud environment for multiparticipant computation.…”

Section: Introductionmentioning

confidence: 99%

Deception Attacks against Android-Based Smart Bracelets

gao

Xie

2023

Security and Communication Networks

View full text Add to dashboard Cite

The rapid development of the intelligent equipment industry has promoted the emergence of a series of emerging industries, effectively improving the technical level of society and people’s quality of life. Due to the inherent characteristics of smart devices, smart devices face serious privacy disclosure risks. Therefore, we have studied the security problem of the widely used smart device smart bracelet in this paper. We have adopted the attack method from easy to difficult to obtain the reading and writing instructions and timing characteristics of the intelligent devices to attack and steal private information. Specifically, we first attack through the strategy of log analysis; if this method is unable to obtain effective information, then we further acquire sensitive information on the basis of hook technology; and if the method on the basis of hook technology is still unable to obtain relevant information, then we will further use reverse engineering to conduct reverse analysis on the app to obtain sensitive information. Second, we develop a fake app on the basis of sensitive information and use it as a bridge to attack intelligent devices. In order to verify the effectiveness of the method, we successfully attacked and stole information from three popular business intelligence bracelets of different brands on the basis of the proposed method. The first step is to develop a fake app on the basis of the identified vulnerabilities. This app can bypass the protection measures of confusion and forced pairing and resetting to cheat the smart bracelet and can successfully enable or disable the jitters function remotely to modify the time and to obtain the sensitive data of the smart bracelet owner. In our attack process, we do not need the cooperation of the owner of the smart bracelet, nor do we need the target smart bracelet to match with our app.

show abstract

X-HDNN: Explainable Hybrid DNN for Industrial Internet of Things Backdoor Attack Detection

Ahakonye,

Nwakanma,

Lee

et al. 2023

2023 14th International Conference on Information and Communication Technology Convergence (ICTC)

View full text Add to dashboard Cite

DeepGuard: Backdoor Attack Detection and Identification Schemes in Privacy-Preserving Deep Neural Networks

Cited by 5 publications

References 29 publications

Defense Strategies Toward Model Poisoning Attacks in Federated Learning: A Survey

Defense Strategies Toward Model Poisoning Attacks in Federated Learning: A Survey

Deception Attacks against Android-Based Smart Bracelets

X-HDNN: Explainable Hybrid DNN for Industrial Internet of Things Backdoor Attack Detection

Contact Info

Product

Resources

About