On distributed ledger technology for the internet of things: design and applications

Abstract

Distributed ledger technology (DLT) can used to store information in such a way that no individual or organisation can compromise its veracity, contrary to a traditional centralised ledger. This nascent technology has received a great deal of attention from both researchers and practitioners in recent years due to the vast array of open questions related to its design and the assortment novel applications it unlocks. In this thesis, we are especially interested in the design of DLTs suitable for application in the domain of the internet of things (IoT), where factors such as efficiency, performance and scalability are of paramount importance. This work confronts the challenges of designing IoT-oriented distributed ledgers through analysis of ledger properties, development of design tools and the design of a number of core protocol components. We begin by introducing a class of DLTs whose data structures consist of directed acyclic graphs (DAGs) and which possess properties that make them particularly well suited to IoT applications. With a focus on the DAG structure, we then present analysis through mathematical modelling and simulations which provides new insights to the properties of this class of ledgers and allows us to propose novel security enhancements. Next, we shift our focus away from the DAG structure itself to another open problem for DAG-based distributed ledgers, that of access control. Specifically, we present a networking approach which removes the need for an expensive and inefficient mechanism known as Proof of Work, solving an open problem for IoT-oriented distributed ledgers. We then draw upon our analysis of the DAG structure to integrate and test our new access control with other core components of the DLT. Finally, we present a mechanism for orchestrating the interaction between users of a DLT and its operators, seeking to improves the usability of DLTs for IoT applications. In the appendix, we present two projects also carried out during this PhD which showcase applications of this technology in the IoT domain.