Towards Robust Blockchain Systems: From Fair Consensus to Verifiable Data Access

The Hong Kong University of Science and Technology
Department of Computer Science and Engineering


PhD Thesis Defence


Title: "Towards Robust Blockchain Systems: From Fair Consensus to Verifiable Data
Access"

By

Mr. Weijie SUN


Abstract:

Blockchain technology has transformed distributed systems by enabling mutually 
untrusted nodes to reach agreement without a central authority. Such trustless 
decentralized paradigm relies on the robustness of system design mainly from 
two components: the consensus layer governing block production and the data 
layer governing data consumption. However, these pillars of robustness could 
be undermined in a Byzantine environment, where adversarial participants may 
strategically misbehave, leading to biased data production and compromised 
data access. This thesis systematically addresses robustness vulnerabilities 
across both layers, ensuring that blockchain systems remain fair, predictable, 
and verifiable throughout the entire data lifecycle.

First, at the consensus layer, we defend against the selfish mining attack in 
the classic Proof-of-Work (PoW) consensus that damages the system fairness. To 
address this, we first introduce the unfairness measurement based on the 
KL-divergence from the computing power distribution to the revenue 
distribution of miners. Then we propose a novel block promotion strategy 
called Tit-for-Tat (TFT) for honest miners detecting selfish behavior based on 
fork observations and then selectively delaying block promotions to suspicious 
nodes. To determine the optimal withholding time, we formulate the Delay 
Vector (DV) problem to minimize the attacker's unfair profit and propose 
efficient approximation algorithms to solve it. Extensive experiments show 
that the TFT strategy effectively improves overall system fairness.

Furthermore, we tackle the block withholding attack in the prevalent 
Proof-of-Stake (PoS) consensus that downgrades the system performance 
predictability. While Byzantine Fault Tolerant (BFT-PoS) is designed with 
stable block generation intervals, malicious participants can deliberately 
delay their block proposals to capture extra Maximal Extractable Value (MEV) 
from upcoming transactions. To handle this, we introduce InTime to 
economically motivate timely proposals. InTime features an Arrival Rate 
Incentive (ARI), which allocates transaction tips at a finer granularity based 
on their arrival rates across the network, removing the financial benefit of 
delaying a block.

To robustly collect and verify these arrival times in a malicious environment, 
we designed the Committee Time Witness (CTW) workflow and a Shift-Mean 
Estimation (SME) algorithm. Our evaluation demonstrates that InTime 
effectively reduces latency variability.

Finally, at the data layer, we address the limitation of verifiable data 
access for advanced on-chain queries. Specifically, we propose the Merkle 
Bloom Filter Tree (MBFT), an efficient framework for authenticated aggregate 
queries that combine boolean keywords and range predicates on blockchains. At 
its core is a Bloom filter- based authenticated data structure that supports 
both types of predicates, constructed per block for efficient transaction 
indexing. For temporal predicates, we optimize time window queries through 
value pruning and block consolidation. We design a novel Merge Bloom Filter 
(MBF) for space-efficient handling of dynamic sets during query 
authentication. We provide a theoretical analysis of the storage overhead 
caused by the Bloom filter's false positive rates. Our framework employs data 
sketches to support various aggregate operations. The experimental results 
demonstrate that MBFT can significantly improve the query speed.

In conclusion, this thesis enhances blockchain robustness by securing the 
fairness of PoW, the predictability of PoS, and the verifiability of complex 
data retrieval. These contributions provide a holistic framework for building 
resilient and robust blockchain systems.


Date:                   Wednesday, 11 March 2026

Time:                   2:00pm - 4:00pm

Venue:                  Room 2132C
                        Lift 22

Chairman:

Committee Members:      Prof. Lei CHEN (Supervisor)
                        Prof. Qiong LUO
                        Prof. Ke YI
                        Prof. Jiheng ZHANG (IEDA)
                        Prof. Jianliang XU (HKBU)