Date of Award

2016

Degree Type

Thesis

Degree Name

Doctor of Philosophy

School

School of Science

First Advisor

Associate Professor Andrew Woodward

Second Advisor

Professor Craig Valli

Third Advisor

Dr Mike Johnstone

Abstract

This research investigated search methods for enumerating networked devices on off-link 64 bit Internet Protocol version 6 (IPv6) subnetworks. IPv6 host enumeration is an emerging research area involving strategies to enable detection of networked devices on IPv6 networks. Host enumeration is an integral component in vulnerability assessments (VAs), and can be used to strengthen the security profile of a system. Recently, host enumeration has been applied to Internet-wide VAs in an effort to detect devices that are vulnerable to specific threats. These host enumeration exercises rely on the fact that the existing Internet Protocol version 4 (IPv4) can be exhaustively enumerated in less than an hour. The same is not true for IPv6, which would take over 584,940 years to enumerate a single network. As such, research is required to determine appropriate host enumeration search methods for IPv6, given that the protocol is seeing increase global usage.

For this study, a survey of Internet resources was conducted to gather information about the nature of IPv6 usage in real-world scenarios. The collected survey data revealed patterns in the usage of IPv6 that influenced search techniques. The research tested the efficacy of various searching algorithms against IPv6 datasets through the use of simulation.

Multiple algorithms were devised to test different approaches to host enumeration against 64 bit IPv6 subnetworks. Of these, a novel adaptive heuristic search algorithm, a genetic algorithm and a stripe search algorithm were chosen to conduct off-link IPv6 host enumeration. The suitability of a linear algorithm, a Monte Carlo algorithm and a pattern heuristics algorithm were also tested for their suitability in searching off-link IPv6 networks. These algorithms were applied to two test IPv6 address datasets, one comprised of unique IPv6 data observed during the survey phase, and one comprised of unique IPv6 data generated using pseudorandom number generators. Searching against the two unique datasets was performed in order to determine appropriate strategies for off-link host enumeration under circumstances where networked devices were configured with addresses that represented real-word IPv6 addresses, and where device addresses were configured through some randomisation function.

Whilst the outcomes of this research support that an exhaustive enumeration of an IPv6 network is infeasible, it has been demonstrated that devices on IPv6 networks can be enumerated. In particular, it was identified that the linear search technique and the variants tested in this study (pattern search and stripe search), remained the most consistent means of enumerating an IPv6 network. Machine learning methods were also successfully applied to the problem. It was determined that the novel adaptive heuristic search algorithm was an appropriate candidate for search operations. The adaptive heuristic search algorithm successfully enumerated over 24% of the available devices on the dataset that was crafted from surveyed IPv6 address data. Moreover, it was confirmed that stochastic address generation can reduce the effectiveness of enumeration strategies, as all of the algorithms failed to enumerate more than 1% of hosts against a pseudorandomly generated dataset.

This research highlights a requirement for effective IPv6 host enumeration algorithms, and presents and validates appropriate methods. The methods presented in this thesis can help to influence the tools and utilities that are used to conduct host enumeration exercises.

Share

 
COinS