Search

The thesis addresses the advantages that data visualization and semantics can bring to blockchain decentralized applications. It provides a visual tool for browsing and debugging on events over Ethereum blockchain and proposes a method to integrate Ethereum smart contract with semantics. By implementing this method, users are able to conduct knowledge reasoning, tracing and querying with the help of semantic-reinforced visualization by building up a semantic resource layer upon the basic Ethereum blockchain infrastructure. Such proof-of-concept method is examined on a case study of network function virtualization, and the utilization of semantic-reinforced visualization is verified in practical scenarios.

Network Functions Virtualization (NFV) has brought the telecommunications industry multiple benefits; however, it has also introduced many new security issues. This thesis tackles security issues related to NFV trust and defines trust as confidence in the integrity of the software and hardware in a system. Existing NFV trust solutions have added trust to the NFV infrastructure with boot time measurements, placement of Virtualized Network Functions (VNFs) on trusted infrastructure and integrity checks of a small set of VNF operations. This thesis implements the introduced trust elements from existing solutions and proposes several extensions. These extensions enable trust in the NFV management software with run time measurements, introduces a new method for building VNF trust, extends the number of trusted VNF operations and increases the user auditability of trust decisions. The proposed extensions are designed, implemented and evaluated in a trusted NFV cloud environment. Although the proposed extensions create a more trusted cloud, they come at a steep performance cost to VNF operations. However, the most impacted VNF operations only affect the cloud provider and not the telecommunications consumer. This thesis offers a valuable contribution to NFV clouds where increased trust is more important than maximized performance or where VNF operations are rarely performed.

Telecommunications systems are critical systems with high quality of service constraints. In Network Function Virtualization (NFV), commonly known as the Telco Cloud, network functions are distributed as virtual machines that run on generic servers in a datacenter. These network functions control critical elements; therefore, they should be run on trusted hardware. Trusted computing concepts can be used to guarantee the trustworthiness of the underlying hardware platform running critical workload. These concepts include the Trusted Platform Module and Remote Attestation. This work identifies limitations in existing solutions and uses those as motivation for designing and implementing a finer-grained definition of trust. This thesis designs and develops a remote attestation solution, which includes a policy and rule based mechanism for determining platform trust in a trusted cloud. Additionally, it develops a fine-grained concept of trust in a cloud environment based on NFV. Finally, this thesis utilizes the remote attestation solution to develop a forensics system based on root cause analysis, which allows the investigation of attestation failures and their mitigation.

Telecommunications systems being inherently distributed and collaborative in nature present a plurality of attack surfaces to malicious entities and hence vulnerable to many potential attacks even indirectly demanding a need in prioritising security. The choice of security implementations depends upon the currently understood threats, future possible threat vectors, and the dependencies between systems. Executing these choices while contemplating the financial aspects is exceptionally difficult. It is thus critical to have a perceptible decision support framework for better security decision-making. This thesis studies the strategic nature of the interaction between the Telecoms operators and attackers utilising game theory to understand their strategic decision-making characteristics strengthening security decisions. To understand the security investment decision-making criteria of operators, this thesis utilises static security investment games. Through these games, we study the effects of security investment decision of an operator on other operators' behaviour. We determine conditions supporting the security investment decisions and propose strategic recommendations supplementing the dependency conditions. We then study attackers' behaviour considering them with strategic incentives in contrary to their strictly-bounded rationality in traditional game-theoretic modelling approaches. We utilise a behavioural approach and design a decision-flow model capturing the choices of attackers in the attack process. An outcome of this work is a generalised attack framework. Moreover, using this framework, we derive attack strategies optimising attackers' effort. Through this work, we are probing the foundations for drawing inferences about attackers' strategic characteristics from a cybersecurity perspective.

Habitat assessments often require observers to estimate tree hollows in situ, which can be costly, destructive and prone to bias. An alternative is to count the number of trees above a specific size. The size at which a tree develops hollows differs substantially among tree species. To assist with setting standards for habitat assessment we defined a large tree as the size at which a species has a 50% probability of supporting a 2-cm diameter hollow. We estimated this size for 68 species using a meta-analysis based on 18 data sources. We found that large tree size ranged from 21 to 106 cm diameter at breast height (DBH). Each species was attributed to vegetation types (formations and classes) to explore variation in large tree sizes. Despite considerable variation within vegetation classes and formations, our results suggest that a large tree size of approximately 50 cm DBH may be appropriate for most vegetation types, with lower estimates in semi-arid vegetation (~30 cm) and higher estimates in wet sclerophyll forests (~80 cm). Our estimates provide empirical support for defining large trees at species vegetation class and formation levels within New South Wales, and highlights the need for more empirical data.