Sydney Blockchain Group

Expert: Jiangshan Yu
Partially sponsored by: ARC DECRA, ARC Discovery Projects, Data61 of CSIRO

Blockchains are critical to decentralised systems but face persistent challenges in achieving resilience and security. This project addresses these challenges through architectural hybridisation and protocol analysis. By integrating trusted hardware components into blockchain architecture, the project develops hybrid consensus protocols that achieve optimal resilience, tolerate a minority of malicious actors, and maintain high communication efficiency.

In addition, the project conducts protocol analysis to uncover vulnerabilities in deployed systems. This has led to the identification of critical vulnerabilities and attacks on several blockchains, with our research findings adopted by systems such as Monero and NEO to implement patches and improve their security.

Expert: Associate Professor Jiangshan Yu

Partially sponsored by: ARC DECRA, ARC Discovery Projects

 The scalability of blockchain systems remains a fundamental challenge, limiting their capacity to support the growing demands of applications. This project aims to address this issue by exploring and advancing two complementary approaches: sharding and Layer 2 protocols.

Sharding enables blockchain workloads to be partitioned across multiple shards, allowing the system to process transactions in parallel. This improves throughput and ensures that the network scales proportionally with the number of participants, preserving decentralisation and security. Layer 2 protocols, on the other hand, provide off-chain solutions to significantly improve blockchain performance.

Some of our works undertaken in this project had been implemented and deployed by blockchains such as Nervos Network, to support large scale applications.

Expert: Associate Professor Jiangshan Yu

This project aims to address critical security challenges posed by fault interdependencies in blockchains, focusing on safeguarding these systems against catastrophic failures. Fault independence is a fundamental requirement for blockchain resilience; without it, a single software vulnerability can compromise multiple machines, undermining the security and trustworthiness of the entire network.

This project aims to bridge the gap between theoretical assumptions and practical realities by establishing robust foundations and practical solutions to ensure fault independence. By addressing this fundamental issue, this project aims to enhance the resilience and reliability of blockchain systems, paving the way for more secure and trustworthy applications.

Expert: Associate Professor Jiangshan Yu

Partially sponsored by: OCSC, AgriFutures Australia, Building 4.0 CRC, RACE for 2030 CRC, Algorand Foundation

Blockchain technology holds transformative potential for high-performance digital infrastructures, providing innovative solutions to enhance efficiency, transparency, accountability, and security across critical sectors. This set of projects explores diverse applications, including government services, supply chain management, energy systems, and credentialing.

The practical relevance of this work is demonstrated through collaborations with government agencies, which have sought our expert advice on blockchain implementation strategies and anti-money laundering solutions. These efforts exemplify how blockchain can revolutionise infrastructure systems, foster innovation, and build trust in digital ecosystems.

Expert: Associate Professor Vincent Gramoli

Web3 promises to revolutionize the economy by letting users provide services to others without the need for centralised institutions. Unforunately Web3 operates on blockchain systems that are insecure and whose participants are not held accountable for their actions.

Redbelly is a blockchain system that offers security and performance for UTXO and account models. Its security stems from its deterministic consensus protocol, called Democratic BFT (DBFT), that prevents forks and its formal verification with parameterized model. Its performance comes from its superblock optimization that combines proposed blocks instead of selecting only one and discarding the other, and its lightweight validation. On the Diablo benchmarking framework, Redbelly outperforms six mainstream blockchains.

This project's goal is to create an accountable version of the Web3 by adding accountability to the Redbelly Blockchain system that features a scalable variant of the Ethereum virtual machine.

This project is partially funded by a grant from the Ethereum Foundation.
 

Experts: Associate Professor Vincent Gramoli

Partners: Rachid Guerraoui (Swiss Federal Institute of Technology, Lausanne)

Diablo is a benchmark to evaluate blockchain systems on the same ground. It was developed in a partnership between University of Sydney and the Swiss Federal Institute of Technology Lausanne (EPFL) to evaluate blockchain and distributed ledger technologies when running realistic applications.

The name Diablo stems from DIstributed Analytical BLOckchain benchmark. We are currently extending Diablo to test the security vulnerabilities of blockchain systems by using fuzzing, injecting faults and implementing malicious behaviors (51% attacks, attack of the clones, balance attack, etc.)

We believe that the result will be insightful to improve existing blockchain technologies and protect their users.

Expert: Associate Professor Qiang Tang

Sponsors: JD.com, Ethereum Foundation, Stellar Development Foundation

Asynchronous consensus is the most robust (assuming least trust on underlying network conditions) consensus protocol, thus critical for blockchains deployed over the open Internet. Unfortunately, all previous protocols suffer from high complexity and essentially none has been widely deployed. We have developed a sequence of results called Dumbo protocol family (at PODC20, CCS20,22, NDSS 22) on pushing asynchronous BFT consensus to the optimal complexity, and finally, real. We are also exmaining real world blockchain systems, their security under asynchronous network conditions, and how to upgrade them to be more resilient with minimal changes. 

Expert: Associate Professor Qiang Tang

Sponsors: Protocol Labs

Distributed key and randomness generation protocols, and threshold cryptography are fundemantally useful for distributed systems and computing, for both security and performance; however, all of existing work incur a large communication complexity at cubic to the number of the parties. This makes them unapplicable to large scale systems.

We initiated the study of distributed key and randomness generation to be with sub-cubic communication, which enabled  the first all-hands checkpointing scheme that allows all validators of a blockchain to participate (realizing Filecoin's blueprint). The initial results were published at CCS24, and PODC 24. We continously advance the state of the art and on the way of getting all those protocols towards optimal, suitable in more dynamic internet environment and new generations of threshold crypto systems.  

We also initiated the study of Verifiable Capacity Bound Function, as a space analog of Verifiable Delay Function, which has been found uniquely useful in blockchain and applications.

Experts: Professor Alan Fekete, Associate Professor Vincent Gramoli, Associate Professor Qiang Tang

Partners: Christian Cachin   

Sponsors: ARC Discovery Projects

In this project, we will explore how to enable fair access to resources managed by decentralized systems. Conventional consensus protocols only ensure ordering of requests are consistent across all parties, but not how the ordering is generated, thus cannot enforce a basic first-come-first-serve policy.

To respond to the recent attacks leveraging this order insufficiency in DeFi, various consensus protocols with order fairness have been proposed. However, they are either too strong to be practically useful, or too weak to be effective against order manipulations. We are aiming to resolve the issue by proposing suitable order fairness notions that are both strong enough to minimize order manipulation attacks, and feasible to enable practicel distributed systems that can be actually useful. 

Expert: Associate Professor Qiang Tang

Sponsors: Stellar Development Foundation

We explore how to design efficient privacy preserving cryptocurrency exchange, while preserving the compliance properties such as KYC, overdraft prevention, solvency, and others. The first such design is called Pisces at NDSS 24; Moreover, we are currently investigating how to use a private exchange to enable more advanced finance systems such as dark pool for cryptocurrency and general digital asset. 

Expert: Associate Professor Qiang Tang

Partners: Moti Yung

Sponsors: Google

In this project, we will design, analyse and build end to end encrypted (E2EE) cloud storage and online collaboration tools such that even if data breach happens, all data would still remain secure. The first such design called End-to-Same-End Encryption (E2SE) was at USENIX Security 22, that we can augment an App (include crypto wallet) with a secure cloud storage that is fully compatible with the existing cloud storage infrastructure.

We are currently collaborating with Google on further investigating the right cryptographic abstractions for an E2EE cloud, analyze the security of existing products claiming E2EE security, and design diverse set of E2EE applications including Git, Googledoc, and more.

Expert: Liyi Zhou

This project investigates Maximal Extractable Value (MEV) patterns and their impact on blockchain fairness and security. We're developing new algorithms to detect and classify MEV opportunities, while also creating mechanisms to mitigate negative effects on network participants.

Our research includes analyzing transaction ordering exploitation, sandwich attacks, and arbitrage opportunities across DeFi protocols. The goal is to propose new consensus mechanisms and protocol designs that can maintain network fairness while managing unavoidable MEV extraction.

Expert: Liyi Zhou

This research leverages symbolic execution techniques to automatically detect inconsistencies between smart contract implementations and their documentation. By formally representing both the documented behavior and actual code execution paths, we can identify potential mismatches that could lead to security vulnerabilities or unexpected behavior.

The project develops tools to automatically extract specifications from natural language documentation and verify them against smart contract bytecode, helping developers maintain accurate documentation and secure implementations.

Expert: Liyi Zhou

This project focuses on developing advanced fuzzing techniques specifically tailored for Web3 protocols and smart contracts. By combining traditional fuzzing methodologies with blockchain-specific strategies, we aim to automatically discover vulnerabilities in smart contract implementations.

This project leverages oracles to explore complex state transitions and transaction sequences, helping identify potential attack vectors in DeFi protocols, NFT marketplaces, and other Web3 applications.

Expert: Liyi Zhou

This project develops foundation models specialized in understanding and analyzing blockchain transactions across multiple chains. By training on large-scale transaction data, these models learn to identify patterns, anomalies, and relationships between different types of blockchain activities.

The research focuses on creating embeddings that capture complex transaction behaviors, enabling better scam detection, user behavior analysis, and cross-chain activity tracking. This work aims to enhance blockchain security tools with deep learning capabilities that can adapt to evolving transaction patterns.

Expert: Dr Sasha Rubin

Distributed algorithms are hard to get right. The long-term vision is that the algorithms and protocols that underlie distributed storage and communication are subjected to formal guarantees before being deployed --- just as we don’t completely trust a manual calculation until we have verified it on a calculator, so too we should not fully trust a proof that an algorithm is correct until it has been verified by a computer.

We use automatic and semi-automatic formal methods (e.g., model-checking, interactive theorem proving, satisfiability checking, invariant generation) with a focus on Asynchronous Byzantine Fault Tolerant protocols.

Experts: Dr Aravind Thyagarajan

A knowledge-payment exchange is between a buyer and a seller, where a buyer has some money, and the seller has the knowledge of some data. The buyer wishes to pay the seller and, in turn, obtain the data in an atomic manner.

However, quite often in practice, buyers use proxies to find an appropriate seller. That is, the buyer transfers the control of the payment to a proxy and the proxy is entrusted with finding an appropriate seller and later the data is transferred locally from the proxy to the buyer.

However, what if the proxy holds the buyer hostage, wherein the payment is transferred to the seller and the proxy learns the data, but refuses to transfer it to the buyer? What if the proxy uses the payment to spend on different items than what the buyer intended? Can the buyer be assured of privacy of the data, and not let the proxy learn the data for which the buyer is the one paying the price?