The Internet network has received huge attentions by the research community. At a first glance, the network optimization and scalability issues dominate the efforts of researchers and vendors. Many results have been obtained in the last decades: the Internet’s architecture is optimized to be cheap, robust and ubiquitous. In contrast, such a network has never been perfectly secure. During all its evolution, the security threats of the Internet persist as a transversal and endless topic. Nowadays, the Internet network hosts a multitude of mission critical activities. The electronic voting systems and financial services are carried out through it. Governmental institutions, financial and business organizations depend on the performance and the security of the Internet. This role confers to the Internet network a critical characterization. At the same time, the Internet network is a vector of malicious activities, like Denial of Service attacks; many reports of attacks can be found in both academic outcomes and daily news. In order to mitigate this wide range of issues, many research efforts have been carried out in the past decades; unfortunately, the complex architecture and the scale of the Internet make hard the evaluation and the adoption of such proposals. In order to improve the security of the Internet, the research community can benefit from sharing real network data. Unfortunately, privacy and security concerns inhibit the release of these data: its suffices to imagine the big amount of private information (e.g., political preferences or religious belief) it is possible to get while reading the Internet packets exchanged between users and web services. This scenario motivates my research, and represents the context of this dissertation which contributes to the analysis of the security issues of the Internet infrastructures and describes relevant security proposals. In particular, the main outcomes described in this dissertation are: • the definition of a secure routing protocol for the Internet network able to provide cryptographic guarantees against false route announcement and invalid path attack; • the definition of a new obfuscation technique that allow the research community to publicly release their real network flows with formal guarantees of security and privacy; • the evidence of a new kind of leakage of sensitive informations obtained hacking the models used by sundry Machine Learning Algorithms.
On the cyber security issues of the internet infrastructure / Vitali, Domenico. - (2013 Feb 18).
On the cyber security issues of the internet infrastructure
VITALI, DOMENICO
18/02/2013
Abstract
The Internet network has received huge attentions by the research community. At a first glance, the network optimization and scalability issues dominate the efforts of researchers and vendors. Many results have been obtained in the last decades: the Internet’s architecture is optimized to be cheap, robust and ubiquitous. In contrast, such a network has never been perfectly secure. During all its evolution, the security threats of the Internet persist as a transversal and endless topic. Nowadays, the Internet network hosts a multitude of mission critical activities. The electronic voting systems and financial services are carried out through it. Governmental institutions, financial and business organizations depend on the performance and the security of the Internet. This role confers to the Internet network a critical characterization. At the same time, the Internet network is a vector of malicious activities, like Denial of Service attacks; many reports of attacks can be found in both academic outcomes and daily news. In order to mitigate this wide range of issues, many research efforts have been carried out in the past decades; unfortunately, the complex architecture and the scale of the Internet make hard the evaluation and the adoption of such proposals. In order to improve the security of the Internet, the research community can benefit from sharing real network data. Unfortunately, privacy and security concerns inhibit the release of these data: its suffices to imagine the big amount of private information (e.g., political preferences or religious belief) it is possible to get while reading the Internet packets exchanged between users and web services. This scenario motivates my research, and represents the context of this dissertation which contributes to the analysis of the security issues of the Internet infrastructures and describes relevant security proposals. In particular, the main outcomes described in this dissertation are: • the definition of a secure routing protocol for the Internet network able to provide cryptographic guarantees against false route announcement and invalid path attack; • the definition of a new obfuscation technique that allow the research community to publicly release their real network flows with formal guarantees of security and privacy; • the evidence of a new kind of leakage of sensitive informations obtained hacking the models used by sundry Machine Learning Algorithms.File | Dimensione | Formato | |
---|---|---|---|
vitali-thesis.pdf
accesso aperto
Note: Tesi dottorato
Tipologia:
Tesi di dottorato
Licenza:
Creative commons
Dimensione
2.47 MB
Formato
Adobe PDF
|
2.47 MB | Adobe PDF |
I documenti in IRIS sono protetti da copyright e tutti i diritti sono riservati, salvo diversa indicazione.