Hoang's homepage

I am a Senior Lecturer in Computer Science and the Program Manager of the Bachelor of Computer Science Program at RMIT University, Melbourne. My main research interest includes blockchain forensics, data and information privacy, coding theory, and discrete mathematics. I am most interested in research that have nice theory behind (with neat algorithms and/or mathematical proofs) and yet implementable & usable in practice.   

Contact me at sonhoang.dau@rmit.edu.au (with CV, and transcripts, if available) for available PhD and Honour/Minor thesis projects on blockchain and crypto scams. Currently, we are working on the following topic:

-Cryptocurrency scams: the goal is to understand how the scams work, how to detect them, and scammers's tactics,  

-Privacy solutions for the Ethereum blockchain.

My Google scholar: https://scholar.google.com.sg/citations?user=dEdfBuwAAAAJ 

Recent news

• January, 2025: Our latest work on crypto scams titled  Serial Scammers and Attack of the Clone has been accepted by The Web Conference (WWW) 2025 (CORE rank A*) in Sydney! In this work, we explored the phenomenon of serial scammers, each of which carried out dozens to thousands of Rug-Pull scams on decentralized exchanges like Uniswap and Pancakeswap. We discovered distinctive scam funding patterns including chain, star, and major flow, and showed that the addresses belong to such patterns use scam contracts with very high similarities, corroborating our view that they were all controled by the same scammer. The two key messages are: 1) scammers never just carried out one scam and disappeared, in fact, they often ran a series of scams (using similar smart contracts) before cashing out, and 2) their operations are highly structured and well coordinated, and we can track them once we understand how they work. Congratulations to all the co-authors Duy Huynh, Yen Tran, Nick Huppert, Josh Cervenjak, Hoonie Sun, Xiaodong Li, and Emanuele Viterbo!

• September, 2024: We've made it to IEEE S&P'25, the No. 1 computer security conference with highest h5-index ranked by Google Scholar.  Using a novel tree coloring, we propose TreePIR - a fast private information retrieval on trees that allows a client to download a Merkle proof (widely used in web-certificate log systems and blockchains) from a server without letting it know which proof is being retrieved. Our method uses fewer and smaller databases compared to existing works. The tree structure, interestingly, also allows a fast/simple indexing, exponentially faster and cheaper than previous works.  This is the joint work with Nhat Quang Cao, Rinaldo Gagiano, Phuong Duy Huynh, Xun Yi, Hung Quang Luu, Phuc Lu Le, Yu-Chih (Jerry) Huang, Jingge Zhu, Chen Feng, Emanuele Viterbo, and Mohammad M. Jalalza. Congratulations to all! 

• August, 2024: Congratulations to the team, especially Stanislav Kruglik for his lead role, on the acceptance of our paper "Querying Twice to Achieve Information-Theoretic Verifiability in Private Information Retrieval" by IEEE Transactions on Information Forensics and Security! This is the join work with Stanislav Kruglik, Han Mao Kiah, and Huaxiong Wang from NTU, Singapore, and Liangfeng Zhang from Shanghai Tech. An earlier version of our paper can be accessed here: https://www.techrxiv.org/doi/full/10.36227/techrxiv.24290563.v1  

• July, 2024: Congratulations to Duy Huynh and the team for the acceptance of our paper "Improving the Accuracy of Transaction-Based Ponzi Detection on Ethereum" by ProvSec'24. In this work, we revisited the problem of detecting Ponzi schemes on Ethereum based on Machine Learning tools. By including the temporal dimension to the feature set, we were able to significantly increase the accuracy of the detection models compared to existing works. This is the join research with Xiaodong Li from RMIT University, and Emanuele Viterbo and Phuc Luong from Monash University.

• April, 2024: Four papers accepted at the IEEE International Symposium on Information Theory 2024 in Greece! Congratulations to all collaborators from RMIT, NTU, NUS, and Monash University! The papers are available on Arxiv: Repairing a Single Erasure in Reed-Solomon Codes with Side Information, Private Repair of a Single Erasure in Reed-Solomon Codes, Repairing with Zero Skip Cost, A Family of Low-Complexity Binary Codes with Constant Hamming Weights

• October, 2023: Most existing private information retrieval (PIR) protocols, while guaranteeing privacy, either 1) cannot ensure that the data retrieved by the client from the storage server(s) are accurate, or 2) have to use a large number of servers with the assumption that less than half of them are malicious, or 3) must rely on an assumption that the malicious servers have limited computational powers and can't solve certain hard problems. Note that an unverifiable protocol may lead to devastating consequences, e.g. making the client download a piece of software/data with malware embedded. In our lastest manuscript titled "Querying Twice to Achieve Information-Theoretic Verifiability in Private Information Retrieval", we propose a neat approach that overcomes all of the aforementioned drawbacks. The key idea is conceptually simple: to retrieve a data item x_i privately from the database x = (x_1, x_2,..., x_n), i.e., not letting the malicious servers know i, and at the same time also verify that the provided data is correct,  the client can send one query to retrieve x_i privately, and an additional query  to retrieve v*x_i, for some random v known only to the client. Without knowing v, the malicious servers can pass the verification with a negligible probability, hence guaranteeing verifiability. In our work, we show how to integrate this idea into well-known polynomial-based PIR schemes such as Goldberg's and Woodruff-Yekhanin's. This work was done in collaboration with Stas Kruglik, Han Mao Kiah, Huaxiong Wang from NTU, and Liangfeng Zhang from Shanghai Tech.  

• September, 2023: If you have learned any programming language, surely you have made elementary bugs such as "divide by zero" or an "if" condition that never gets satisfied... It turns out that these types of logical bugs can be exploited to make millions of dollars on decentralized exchanges. Read our latest manuscript From Programming Bugs to Multimillion-Dollar Scams: An Analysis of Trapdoor Tokens on Decentralized Exchanges to see how scammers did just that. The work is led by Duy Huynh, our PhD student at RMIT.

• September, 2023: Our paper "On the Formation of Min-Weight Codewords of Polar/PAC Codes and Its Applications" has been accepted for publication in the IEEE Transactions on Information Theory. In this paper we propose a different approach to construct minimum-weight codewords in polar codes and moreover, based on our new understanding of the formation of such codewords, we propose a modification to the polar codes and PAC codes that result in new codes with fewer minimum-weight codewords, achieving better performance than existing codes in certain block error rate regimes. Our code construction deviates from the traditional approach in constructing polar codes in that it also takes into account the impact of weight distribution on the error-correction performance, apart from consideration of sub-channel reliabilities.  This is a joint work with Mohammad Rowshan (UNSW) and Emanuele Viterbo (Monash).

• April, 2023: Big congratulations to all collaborators, with three papers being accepted by the IEEE International Symposium on Information Theory! Two of our works are on verifiable private information retrieval, which can be found here and here. The third one is on failure recovery for blockchain-based distributed storage systems, which is available here. 

• April, 2023: Our paper titled "Committed Private Information Retrieval" has been accepted by ESORICS'23. Congratulations to all collaborators, and especially Quang, on his first accepted paper! We explore in the paper a way for a client to query a record from a database stored in multiple servers (each stores one copy of the database) that guarantees 1) (privacy) if some servers talk to each other then they can't figure out the record, and 2) (verifiability) even in the worst case when all servers collude and try to send some wrong data, the client cannot be fooled into accepting a wrong record.   

• February, 2023: Our paper "Transition Waste Optimization for Coded Elastic Computing" was accepted by the IEEE Transactions on Information Theory. 

• December, 2022: Minh Hoang Nguyen (Henry), our Minor thesis student, got his very first paper on blockchain scams detection accepted by the 2023 Australasian Information Security Conference, as part of the Australasian Computer Science Week. Congratulations, Henry! Rug-pull malicious token detection on blockchain using supervised learning with feature engineering. 

• May, 2022: The Ethereum Foundation has approved to fund our project titled "Efficient Private Information Retrieval for Ethereum Light Clients" via its Academic Grants Round 2022. The grant was submitted with Xun Yi (RMIT), Quang Cao (RMIT), and Chen Feng (UBC). Congratulations, Quang, on his first research grant! 

• April, 2022: Our paper on "Practical Considerations in Repairing Reed-Solomon Codes" has been accepted. Congratulations to the team, especially Xinh Dinh and Nhi Nguyen (their first IEEE conference paper!). We would like to thank Nguyen Dinh Quang Minh and Dau Trung Dung for their great help with the C code. 

• April, 2022: The first draft of our paper is now available on Arxiv: Parallel Private Retrieval of Merkle Proofs via Tree Colorings

• December, 2021: Present out latest work on "Balanced Ancestral Coloring of Perfect Binary Trees and Applications in Private Retrieval of Merkle Proofs" at VNMACS-2021 and AusMS Annual Meeting. This is the joint work with colleagues and PhD students from RMIT University, Vietnam National University, Monash University, University of Melbourne, National Chiao Tung University,  and University of British Columbia, Okanagan. In this work, we develop a novel coloring of binary trees, which can be used to speed up the private retrieval of Merkle proofs significantly compared to a baseline method using a state-of-the-art PIR scheme SealPIR. The paper will be available on Arxiv soon.

• March, 2021: Our paper "Repairing Reed-Solomon Codes via Subspace Polynomials" is accepted by the IEEE Transactions on Information Theory. Congratulations, Xinh, on your first journal paper as a Master student at RMIT University. Well done!

***********************************************************************************************

• October, 2020: Our detailed tutorial on the topic of repairing Reed-Solomon codes at the 2020 Hyderabad Workshop on Information Theory. This 1.5-hour tutorial is a perfect place to start if you want to look into this research direction.

• July, 2020: Our manuscript titled "Repairing Reed-Solomon Codes via Subspace Polynomials" is available online.. Co-authors: Dinh Thi Xinh, Han Mao Kiah, Tran Thi Luong,  and Olgica Milenkovic.

• July, 2020: Our paper titled "Locally Decodable Index Codes" was accepted by the IEEE Transactions on Information Theory. Co-authors: Lakshmi Natarajan, Prasad Krishnan, and V. Lalitha.

• July, 2020: Our paper titled "Access Balancing in Storage Systems by Labeling Partial Steiner Systems" was accepted by Designs, Codes, and Cryptography. Co-authors: Yeow Meng Chee, Charles J. Colbourn, Ryan Gabrys, Alan C.H. Ling, Dylan Lusi, and Olgica Milenkovic.

• June, 2020: Lakshmi J Mohan joined our group as a postdoc to work on the practical aspects of repairing Reed-Solomon codes in Hadoop Distributed File System.  

• April, 2020: Three PhD/Master scholarships are available: Reed-Solomon Codes and Distributed Storage Systems, Distributed Computing, and Blockchain. The detailed descriptions of these projects can be found here.

• April, 2020: Two papers accepted for IEEE Symposium on Information Theory (ISIT), supposedly in Los Angeles, which will go virtual in June due to the Covid-19. Links to the full papers are here: elastic coded computing and partial Steiner sytems. 

• March, 2020: Our paper "Secure Erasure Codes With Partial Reconstructibility" was accepted by the IEEE Transactions on Information Theory. Co-authors: Wentu Song and Chau Yuen (SUTD), Alex Sprintson (Texas A&M).  

***********************************************************************************************

• December, 2019: Oracle Research generously gave us USD $25k worth of cloud credit to use for one year! 

• November, 2019: Our joint Discovery Project with Monash University (Prof. Emanuele Viterbo) on  Coding Theory & Blockchain was approved by Australian Research Council! The grant is worth AUD $420k (2020-2023).

• October, 2019: The first draft of our paper on elastic coded computing was available online. 

• August, 2019: Our paper titled "On the Triangle Clique Cover and $K_t$ Clique Cover Problems" was accepted by Discrete Mathematics. Co-authors: Gregory J. Puleo (Auburn) and Olgica Milenkovic (UIUC).

• July, 2019: Four papers were presented at the IEEE Symposium on Information Theory (ISIT) in Paris, France: (1) "On the I/O Costs in Repairing Short-Length Reed-Solomon Codes" (with Weiqi Li, Zhiying Wang, Hamid Jafarkhani at UCI and Emanuele Viterbo at Monash); (2) "Set-Codes with Small Intersections and Small Discrepancies" (with Ryan Gabrys at PAWAR, Charlie Colbourn at Arizona State University, and Olgica Milenkovic at UIUC); (3) "Locality in Index Coding for Large Min-Rank" (with Laskhmi Natarajan at IIT, Hyderabah, Prasad Krishnan, V. Lalitha at IIIT H); (4) "Iterative Decoding of Reed-Solomon Codes based on Non-binary Matrices" with Viduranga Wijekoon and Emanuele Viterbo at Monash University). 

• June, 2019: Chen Feng from UBC visited us and gave a nice talk on coding theory and blockchain!

• April, 2019: I gave talks and lectures at the Mini-Workshop on Coding Theory and Applications at Vietnam Institute for Advanced Studies in Mathematics (https://viasm.edu.vn/en/hdkh/cta), VNU University of Engineering and Technology, and VNU University of Science. 

• February, 2019: I presented a talk at the Information Theory and Applications Workshop  (ITA) orgainised by UC San Diego. 

• January, 2019: I joined RMIT University!