AUB ScholarWorks

Machine learning for network resilience

Show simple item record Hussein, Ali Imad 2021-09-23T09:00:35Z 2023-01 2021-09-23T09:00:35Z 2020 2020
dc.identifier.other b25875929
dc.description Dissertation. Ph.D. American University of Beirut. Department of Electrical and Computer Engineering, 2020. ED:132.
dc.description Chairman : Dr. Ali Chehab, Professor, Electrical and Computer Engineering ; Advisor : Dr. Ayman Kayssi, Professor, Electrical and Computer Engineering ; Members of Committee : Dr. Imad Elhajj, Professor, Electrical and Computer Engineering ; Dr. Muhammad Imran, Professor, Electrical and Communication Engineering, University of Glasgow, UK ; Dr. Kassem Fawaz, Professor, Electrical and Computer Engineering, University of Wisconsin, USA.
dc.description Includes bibliographical references (leaves 232-271)
dc.description.abstract Resilience is taking networks a step further beyond security. Security is one of the main concerns facing the improvement of new networking and communications systems. Another important challenge is verifying whether or not a system is working exactly as specified, hence ensuring its consistency. We argue that a resilient network is both a secure and consistent one. It is from this point that we start our thesis research. On the other hand, advances in Artificial Intelligence (AI) technology have opened up new markets and opportunities for progress in critical areas such as network resiliency, health, education, energy, economic inclusion, social welfare, and the environment. AI is expected to play an increasing role in defensive and offensive measures to provide a rapid response to react to the landscape of evolving threats. Software Defined Networking (SDN), being centralized by nature, provides a global view of the network. It is the flexibility and robustness offered by programmable networking that lead us to consider the integration of these two concepts, SDN and AI. Inspired by the fascinating tactics of the human immunity system, we aim to design a general hybrid Artificial Intelligence Resiliency System (ARS) that strikes a good balance between centralized and distributed security solutions that may be applicable to different network environments. Another objective is to investigate and leverage the state-of-the-art AI techniques to enhance network performance in general and resiliency in particular. Being able to describe a specific network as consistent is a large step towards resiliency. Next to the importance of security lies the necessity of consistency verification. Attackers are currently focusing on targeting small and crucial goals such as network configurations or ow tables. These types of attacks would defy the whole purpose of a security system when built on top of an inconsistent network. Another important goal of our work is to propose a new AI-based consistency verification system, which
dc.format.extent 1 online resource (xix, 271 leaves) : illustrations
dc.language.iso en
dc.subject.classification ED:000132
dc.subject.lcsh Computer networks -- Security measures.
dc.subject.lcsh Artificial intelligence.
dc.subject.lcsh Machine learning.
dc.title Machine learning for network resilience
dc.type Dissertation
dc.contributor.department Department of Electrical and Computer Engineering
dc.contributor.faculty Maroun Semaan Faculty of Engineering and Architecture
dc.contributor.institution American University of Beirut

Files in this item

This item appears in the following Collection(s)

Show simple item record

Search AUB ScholarWorks


My Account