|
|
Correlation of Inbound and Outbound Traffic of Tor Network
Coufal, Zdeněk ; Veselý, Vladimír (referee) ; Polčák, Libor (advisor)
Communication in public networks based on the IP protocol is not really anonymous because it is possible to determine the source and destination IP address of each packet. Users who want to be anonymous are forced to use anonymization networks, such as Tor. In case such a user is target of lawful interception, it presents a problem for those systems because they only see that the user communicated with anonymization network and have a suspicion that the data stream at the output of anonymization network belong to the same user. The aim of this master thesis was to design a correlation method to determine the dependence of the data stream at the input and the output of the Tor network. The proposed method analysis network traffic and compares characteristics of data streams extracted from metadata, such as time of occurence and the size of packets. This method specializes in correlating data flows of protocol HTTP, specifically web server responses. It was tested on real data from the Tor network and successfully recognized dependency of data flows.
|
|
|
Application for Hand Corrections of Automatically Processed Data
Coufal, Zdeněk ; Beran, Vítězslav (referee) ; Szőke, Igor (advisor)
The objective of this bachelor thesis is to create application for hand corrections of automatically processed data. Thanks to this application it will be allowed to visitors of portal SuperLectures.com to correct text transcripts of slides or audio of the current segment of the lecture. If they decide to help, they will be redirected to the Annotation portal as this application was called and they will be assigned an elementary job to correct representing the given part of the lecture. At first informal analysis is performed, looking for possible tools and procedures for realization. After that requirements are formally specified and later the application design is based on them. Application is implemented according to the design with use of selected tools and procedures. In conclusion, the application is tested and test results are evaluated.
|
|
|
Application for Hand Corrections of Automatically Processed Data
Coufal, Zdeněk ; Beran, Vítězslav (referee) ; Szőke, Igor (advisor)
The objective of this bachelor thesis is to create application for hand corrections of automatically processed data. Thanks to this application it will be allowed to visitors of portal SuperLectures.com to correct text transcripts of slides or audio of the current segment of the lecture. If they decide to help, they will be redirected to the Annotation portal as this application was called and they will be assigned an elementary job to correct representing the given part of the lecture. At first informal analysis is performed, looking for possible tools and procedures for realization. After that requirements are formally specified and later the application design is based on them. Application is implemented according to the design with use of selected tools and procedures. In conclusion, the application is tested and test results are evaluated.
|
|
|
Correlation of Inbound and Outbound Traffic of Tor Network
Coufal, Zdeněk ; Veselý, Vladimír (referee) ; Polčák, Libor (advisor)
Communication in public networks based on the IP protocol is not really anonymous because it is possible to determine the source and destination IP address of each packet. Users who want to be anonymous are forced to use anonymization networks, such as Tor. In case such a user is target of lawful interception, it presents a problem for those systems because they only see that the user communicated with anonymization network and have a suspicion that the data stream at the output of anonymization network belong to the same user. The aim of this master thesis was to design a correlation method to determine the dependence of the data stream at the input and the output of the Tor network. The proposed method analysis network traffic and compares characteristics of data streams extracted from metadata, such as time of occurence and the size of packets. This method specializes in correlating data flows of protocol HTTP, specifically web server responses. It was tested on real data from the Tor network and successfully recognized dependency of data flows.
|
guest ::
