Year: 2010
Type: dissertation
Abstract: Protecting network-accessible resources from unauthorized use requires understanding \nof network vulnerability. Modeling network attacks as chains of exploitable \nvulnerabilities, referred to as vulnerability exploits, can help security administrators \nlocate security flaws and implement appropriate preventative measures. \nHowever,mostmodel generation approaches do not scalewell to very large networks \ndue to the combinatorial explosion of the size of the search space. Much research \nhas focused on how to automatically and efficiently generate such models. \nTo realize the full benefits of network securitymodels, effectivemodel analysis is \ncrucial. Since security models are often too large and complex for pinpointing critical \nattacks manually, there is a need for automated approaches to security model \nanalysis. Moreover, network security has to deal with the trade off between accessibility \nand safety protection. Thus, selecting suitable counter-measures to secure \na network may involve conflicting judgments and subjective preferences. There \nhas been little research in this area and most current approaches lack the ability to \ndirectly assist in making informed decisions. \nThis dissertation presents a principled and practical approach to developing \nautomated preventative network security systems that show possible attacks, in \nterms of chains of vulnerability exploits, and offer mechanisms to assist in selecting \nappropriate counter-measures. To address the problem of scalability, this \ndissertation proposes host-centric model checking, an analytical framework that provides \nefficient generation of a new network security model, namely a host-centric \nattack graph, using a model checking technique along with a monotony assumption (i.e., valid preconditions of a vulnerability exploit are never invalidated). Using \nmodel abstraction and efficient reachability computation by a model checker \ntogether with the assumption, the proposed approach can be shown, both theoretically \nand empirically, to reduce the time complexity of model generation to a \nquadratic polynomial in the number of hosts. At the time of this research, compared \nwith all other existing model generation approaches, this is the best computational \nbound obtained. \nTo address the issue of security model analysis, this dissertation proposes two \nautomated approaches to intelligent analysis of a security attack model, namely \nexploit-based analysis and preference-based analysis. The former is quantitative and \nthe latter is qualitative analysis. Exploit-based analysis proposes a heuristic algorithm \nthat uses knowledge about the exploitability of network vulnerability and \na link analysis technique that employs a Markov property to rank each node in \nthe security attack model in order of its likelihood of being successfully attacked. \nPreference-based analysis provides a semi-automated analysis by means of logical \nreasoning and an explicitmodel of conditional preferences. The analysis result can \nassist a security administrator in making informed decisions about cost-effective \ncounter-measures based on the administrator’s preference criteria. To the best of \nour knowledge, preference-based analysis is the first of its kind for approaches to \nnetwork security model analysis. This research also proposes an efficient method \nfor computing attack path expressions and includes a study of a simple but realistic \ncase scenario. The proposed analysis approaches can be applied to any form \nand any topological structure of a security model. These approaches are general \nand theoretically grounded.
Author: Phongphun Kijsanayothin
Language: English
Subfield: Information Systems
Field: Computer Science
Domain: Physical Sciences
Open Access status: green
OpenAlex ID: W2788680671