English (United Kingdom)

https://curated-unify.zendy.io/wp-json/zendy-region/v1/featured_content/oa?rat=en

https://curated-unify.zendy.io/wp-json/zendy-region/v1/highlighted_journal/

Global: Zendy Plus annual

Presents the access of premium content as premium feature

Premium Content

Presents the keyphrase highlighting as premium feature

Keyphrase Highlighting

Presents the summarisation as premium feature

Summarisation

Insights

Presents the pdf analysis as premium feature

PDF Analysis

Presents the zaia usage as premium feature

ZAIA

Global: Zendy Plus monthly

Global: Zendy Tools annual

Global: Zendy Tools monthly

Global: Zendy Free Plan

Network traffic classification is used in various applications such asnetwork traffic management, policy enforcement, and intrusion detectionsystems. Although most applications encrypt their network traffic and some ofthem dynamically change their port numbers, Machine Learning (ML) andespecially Deep Learning (DL)-based classifiers have shown impressiveperformance in network traffic classification. In this paper, we evaluate therobustness of DL-based network traffic classifiers against Adversarial NetworkTraffic (ANT). ANT causes DL-based network traffic classifiers to predictincorrectly using Universal Adversarial Perturbation (UAP) generating methods.Since there is no need to buffer network traffic before sending ANT, it isgenerated live. We partition the input space of the DL-based network trafficclassification into three categories: packet classification, flow contentclassification, and flow time series classification. To generate ANT, wepropose three new attacks injecting UAP into network traffic. AdvPad attackinjects a UAP into the content of packets to evaluate the robustness of packetclassifiers. AdvPay attack injects a UAP into the payload of a dummy packet toevaluate the robustness of flow content classifiers. AdvBurst attack injects aspecific number of dummy packets with crafted statistical features based on aUAP into a selected burst of a flow to evaluate the robustness of flow timeseries classifiers. The results indicate injecting a little UAP into networktraffic, highly decreases the performance of DL-based network trafficclassifiers in all categories.

Adversarial Network Traffic: Towards Evaluating the Robustness of Deep Learning-Based Network Traffic Classification