A Secure-by-Design Federated Microchain Fabric for Internet-of-Things (IoT) Systems

Abstract

The rapid advancements in the Internet of Things (IoT) technology and Edge-Fog-Cloud Computing paradigm make the concept of Smart Cities become realistic. Thanks to the fusion of Big Data and artificial intelligence (AI) based on heterogeneous IoT networks, seamless smart applications, and public safety services are deployed to maintain a sustainable urban environment for residents. With an ever-increasing presence of highly connected IoT devices, the centralized framework adopted by conventional IoT systems is facing challenges in terms of performance bottleneck and single-point failure. Owing to attractive features like decentralization, immutability, and auditability, Blockchain is promising to enable a tamper-proof and trust-free framework to address performance and security issues in existing IoT solutions. However, directly integrating cryptocurrency-oriented blockchain technologies into IoT systems still meets tremendous limitations, such as high computation and storage demands, low transaction throughput, poor scalability, and complicated security and privacy requirements.

In this dissertation, a secure-by-design federated microchain fabric is proposed to balance trade-offs on performance, scalability and security when applying blockchain for IoT scenarios. Following an in-depth review of background knowledge and state-of-the-art solutions to the efficient and scalable blockchain, key principles and techniques of designing blockchain for IoT networks are identified. Splitting the whole blockchain network into multiple independent small-scale microchain networks, the federated microchain fabric is promising to guarantee performance, scalability and security in heterogeneous IoT networks. Under a generic microchain framework, a periodically random-elected consensus committee delegates all participants to execute a lightweight and hybrid consensus protocol: Proof of Credit (PoC)+Voting based Chain Finality (VCF). Thus, each microchain maintains a private distributed ledger with computing efficiency, low latency, and privacy guarantees at the network of edge. Regarding a practical microchain implementation for IoT networks, I develop Econledger which utilizes a novel Proof-of-ENF (PoENF) consensus protocol to improve performance and security. In addition, I design an epoch randomness based committee configuration called ECOM, which ensures that robustness and security are not sacrificed by a small-scale microchain network including fewer validators. Moreover, I introduce Fed-DDM, which leverages a smart contract-enabled inter-ledger protocol for cross-microchain transactions, to handle trade-offs among performance, scalability, and security in the federated microchain framework. Furthermore, I develop Fairledger that adopts a novel Proof-of-Sequential-Work (PoSW) consensus protocol to enhance the fairness and security of microchain for IoT networks. Finally, I implement µDFL as a case study, which integrates a hierarchical microchain framework into a cross-devices federated learning scheme. The experimental results demonstrate our federated microchain solution is promising to provide efficiency, security, and privacy guarantees for cooperative multi-domain IoT settings.