Published: July 19, 2021
Author(s)
Qingtian Zou (Penn State University), Anoop Singhal (NIST), Xiaoyan Sun (California State University), Peng Liu (Penn State University)
Conference
Name: IFIP Annual Conference on Data and Applications Security and Privacy
Dates: 07/19/2021 - 07/20/2021
Location: Calgary, Canada
Citation: DBSec 2021: Data and Applications Security and Privacy XXXV, vol. 12840, pp. 221-234
Network attack is still a major security concern for organizations worldwide. Recently, researchers have started to apply neural networks to detect network attacks by leveraging network traffic data. However, public network data sets have major drawbacks such as limited data sample variations and unbalanced data with respect to malicious and benign samples. In this paper, we present a new end-to-end approach to automatically generate high-quality network data using protocol fuzzing, and train the deep learning models using the fuzzed data to detect the network attacks that exploit the logic flaws within the network protocols. Our findings show that fuzzing generates data samples that cover real-world data and deep learning models trained with fuzzed data can successfully detect real network attacks.
Network attack is still a major security concern for organizations worldwide. Recently, researchers have started to apply neural networks to detect network attacks by leveraging network traffic data. However, public network data sets have major drawbacks such as limited data sample variations and...
See full abstract
Network attack is still a major security concern for organizations worldwide. Recently, researchers have started to apply neural networks to detect network attacks by leveraging network traffic data. However, public network data sets have major drawbacks such as limited data sample variations and unbalanced data with respect to malicious and benign samples. In this paper, we present a new end-to-end approach to automatically generate high-quality network data using protocol fuzzing, and train the deep learning models using the fuzzed data to detect the network attacks that exploit the logic flaws within the network protocols. Our findings show that fuzzing generates data samples that cover real-world data and deep learning models trained with fuzzed data can successfully detect real network attacks.
Hide full abstract
Keywords
network attack; protocol fuzzing; deep learning
Control Families
None selected
) or https:// don't prove our archive is authentic, only that you securely accessed it. Note that we are working to fix that :)
