|
|
System for network device detection and recognition of used protocols
Sasák, Libor ; Fujdiak, Radek (referee) ; Holasová, Eva (advisor)
This master's thesis deals with the recognition of used protocols in a network using machine learning and the creation of a system for this purpose. It focuses on the most widely used industrial and common application protocols and describes selected well-proven machine learning techniques for their recognition. However, priority is given to artificial neural networks. It briefly describes databases and the specific implementation SQLite3 used in the final system implementation. A virtual environment for simulating selected Modbus/TCP, DNP3, HTTPS and FTP protocols is also created and described. Part of the thesis is devoted to the collection, analysis and processing of the data needed to recognize the protocols. Furthermore, it covers the creation and testing of machine learning models for the given protocols. Last but not least, the thesis is devoted to the design of the recognition system and its implementation with a graphical user interface. It also includes testing and evaluation of its advantages and limitations.
|
|
|
Network Supervision via Protocol Identification in the Network
Holasova, E. ; Kuchar, K. ; Fujdiak, R.
This paper is focused on a comparison of ML (Machine Learning) and DNN (Deep Neural Network) techniques in protocol recognition to support network supervision for further proper handling, e.g., detection of a security incident. The DNN approach uses 11 layers and the ML approach is consisting of 28 mutually different predictive models. Both techniques were performed/compared on a freely accessible dataset containing browsing pcap files for further comparison, e.g., with other approaches. The predictive multiclass models were trained (fitted) to be capable of detecting five network protocols. Both approaches were compared by the achieved accuracy (based on testing and validating data), learning time, and predicting the time point of view. Using the ML approach, we were able to recognize the protocol with an accuracy of 1 and using DNN with an accuracy of 0.97.
|
|
|
System for network device detection and recognition of used protocols
Sasák, Libor ; Fujdiak, Radek (referee) ; Holasová, Eva (advisor)
This master's thesis deals with the recognition of used protocols in a network using machine learning and the creation of a system for this purpose. It focuses on the most widely used industrial and common application protocols and describes selected well-proven machine learning techniques for their recognition. However, priority is given to artificial neural networks. It briefly describes databases and the specific implementation SQLite3 used in the final system implementation. A virtual environment for simulating selected Modbus/TCP, DNP3, HTTPS and FTP protocols is also created and described. Part of the thesis is devoted to the collection, analysis and processing of the data needed to recognize the protocols. Furthermore, it covers the creation and testing of machine learning models for the given protocols. Last but not least, the thesis is devoted to the design of the recognition system and its implementation with a graphical user interface. It also includes testing and evaluation of its advantages and limitations.
|
guest ::
