URL: https://infocom2020.ieee-infocom.org/accepted-paper-list-main-conference to
read papers in the field of network traffic classification.
- Autonomous Unknown-Application Filtering and Labeling for DL-based Traffic Classifier Update
Jielun Zhang, Fuhao Li, Feng Ye and Hongyu Wu (University of Dayton, USA)
Paper URL: https://arxiv.org/pdf/2002.06359.pdf
Main content : The
traditional traffic classification models based on machine learning are all experiments done in the close-world. These models cannot classify unknown categories. The author proposes an autonomous machine learning framework to update the traffic classifier. This framework includes a classifier based on deep learning, a self-learning discriminator, and a self-labeled model. The discriminator and self-tagger can generate a new data set and relearn the classifier together. This thing first identifies unknown traffic, and then automatically tags the unknown traffic.
This thing is very similar to Chen's work.
This is a hydrology. - Exposing the Fingerprint: Dissecting the Impact of the Wireless Channel on Radio Fingerprinting
Amani Al-Shawabka, Francesco Restuccia, Salvatore D'Oro, Tong Jian, Bruno Costa Rendon, Nasim Soltani, Jennifer Dy, Stratis Ioannidis, Kaushik Chowdhury and Tommaso Melodia (Northeastern University, USA)
Paper address: https://ece.northeastern.edu/fac-ece/ioannidis/static/pdf/2020/AlShawabka2020Exposing.pdf
Main content:
Large-scale measurement of CNN network through Radio Fingerprinting The ability to identify wireless devices. Their input is a spectrogram. This should be a physical attack. - MagPrint: Deep Based Learning the User Fingerprinting the Using Electromagnetic Elements Signals
Lanqing Yang, the Chao Yi-Chen, Hao Pan, Dian Ding, Guangtao Xue, Linghe Kong, JIADI Yu and Li Minglu (on Shanghai Jiao Tong University, China)
papers connected: http: // www.cs.sjtu.edu.cn/~yichao/pmwiki/assets/publications/Infocom20_Yang.pdf
Main content:
This paper proposes that users' electromagnetic fingerprints are used for continuous identity authentication of users.
Definition of electromagnetic fingerprint:
In this study, we proposed to use electromagnetic (EM)
signals as a side-channel for fingerprinting users. The intensity
of EM signals generated from a mobile device reflects the
computational intensity of the device. For example, a heavy
computational load tends to push up power consumption by
the CPU, which leads to an increase in EM induction. Furthermore, users vary considerably in the way that they use their
devices in terms of key hold duration, typing interval, cursor
speed, pause and click time, etc [5]–[7 ]. The instructions
associated with these actions vary the internal operations of
the device and are thus reflected in the EM signals.
How to collect electromagnetic fingerprints?
Using an electromagnetic sensor, what is collected is an ordinary magnetic field signal. It can also be an electromagnetic sensor built into the device.
Compared with the traditional handprint and front camera, what are the advantages?
Traditional fingerprint collection tools cannot always authenticate users, and the front camera will cause obvious interference to users.