Saeed Moradi, Koosha Esmaeilzadeh Khorasani, Sara Rouhani
Interoperability is essential for transforming blockchains from isolated networks into collaborative ecosystems, unlocking their full potential. While significant progress has been made in public blockchain interoperability, bridging permissioned and permissionless blockchains poses unique challenges due to differences in access control, architectures, and security requirements. This paper introduces a blockchain-agnostic framework to enable interoperability between permissioned and permissionless networks. Leveraging cryptographic techniques, the framework ensures secure data exchanges. Its lightweight architectural design simplifies implementation and maintenance, while the integration of Clover and Dandelion++ protocols enhances transaction anonymity. Performance evaluations demonstrate the framework's effectiveness in achieving secure and efficient interoperability by measuring the forwarding time, the throughput, the availability, and their collusion impact of the system across heterogeneous blockchain ecosystems.
Decentralized storage is one of the most natural applications built on blockchains and a central component of the Web3 ecosystem. Yet despite a decade of active development -- from IPFS and Filecoin to more recent entrants -- most of these storage protocols have received limited formal analysis of their incentive properties. Claims of incentive compatibility are sometimes made, but rarely proven. This gap matters: without well-designed incentives, a system may distribute storage but fail to truly decentralize it. We analyze Shelby -- a storage network protocol recently proposed by Aptos Labs and Jump Crypto -- and provide the first formal proof of its incentive properties. Our game-theoretic model shows that while off-chain audits alone collapse to universal shirking, Shelby's combination of peer audits with occasional on-chain verification yields incentive compatibility under natural parameter settings. We also examine coalition behavior and outline a simple modification that strengthens the protocol's collusion-resilience.
Ensuring ballot secrecy is critical for fair and trustworthy electronic voting systems, yet achieving strong secrecy guarantees in decentralized, large-scale elections remains challenging. This paper proposes the concept of collectively secure voting, in which voters themselves can opt in as secret holders to protect ballot secrecy. A practical blockchain-based collectively secure voting system is designed and implemented. Our design strikes a balance between strong confidentiality guarantees and real-world applicability. The proposed system combines threshold cryptography and smart contracts to ensure ballots remain confidential during voting, while all protocol steps remain transparent and verifiable. Voters can use the system without prior blockchain knowledge through an intuitive user interface that hides underlying complexity. To evaluate this approach, a user testing is conducted. Results show a high willingness to act as secret holders, reliable participation in share release, and high security confidence in the proposed system. The findings demonstrate that voters can collectively maintain secrecy and that such a practical deployment is feasible.
The scalability of blockchain systems is constrained by inefficient P2P broadcasting, as most existing optimizations focus only on the logical layer without considering physical network conditions. To address this, we propose BlockSDN, the first SDN-based integrated architecture for blockchain. BlockSDN employs a distributed control plane for a global network view, a graph engine for hierarchical clustering, and a hybrid macro-micro neighbor selection with hierarchical broadcasting. A dedicated simulation platform shows that BlockSDN reduces global block synchronization time by 65% and 55% compared to Gossip and Mercury, respectively.These results highlight the potential of SDN-enabled cross-layer coordination to significantly enhance blockchain scalability and performance.
The necessity of blockchain systems to remain decentralised limits current solutions to blockchain governance and dynamic management, forcing a trade-off between control and decentralisation. In light of the above, this work proposes a dynamic and decentralised blockchain management mechanism based on digital twins. To ensure decentralisation, the proposed mechanism utilises multiple digital twins that the system's stakeholders control. To facilitate decentralised decision-making, the twins are organised in a secondary blockchain system that orchestrates agreement on, and propagation of decisions to the managed blockchain. This enables the management of blockchain systems without centralised control. A preliminary evaluation of the performance and impact of the overheads introduced by the proposed mechanism is conducted through simulation. The results demonstrate the proposed mechanism's ability to reach consensus on decisions quickly and reconfigure the primary blockchain with minimal overhead.
Proof-of-stake blockchains require consensus protocols that support Dynamic Availability and Reconfiguration (so-called DAR setting), where the former means that the consensus protocol should remain live even if a large number of nodes temporarily crash, and the latter means it should be possible to change the set of operating nodes over time. State-of-the-art protocols for the DAR setting, such as Ethereum, Cardano's Ouroboros, or Snow White, require unrealistic additional assumptions, such as social consensus, or that key evolution is performed even while nodes are not participating. In this paper, we identify the necessary and sufficient adversarial condition under which consensus can be achieved in the DAR setting without additional assumptions. We then introduce a new and realistic additional assumption: honest nodes dispose of their cryptographic keys the moment they express intent to exit from the set of operating nodes. To add reconfiguration to any dynamically available consensus protocol, we provide a bootstrapping gadget that is particularly simple and efficient in the common optimistic case of few reconfigurations and no double-spending attempts.
Existing distributed ledger protocols either incur a high communication complexity and are thus suited to systems with a small number of processes (e.g., PBFT), or rely on committee-sampling-based approaches that only work for a very large number of processes (e.g., Algorand). Neither of these lines of work is well-suited for moderate-scale distributed ledgers ranging from a few hundred to a thousand processes, which are common in production (e.g, Redbelly, Sui). The goal of this work is to design a distributed ledger with sub-linear communication complexity per process, sub-quadratic total communication complexity, and low latency for finalizing a block into the ledger, such that it can be used for moderate-scale systems. We propose QScale, a protocol in which every process incurs only $\widetilde{O}(κ\sqrt{n})$ communication complexity per-block in expectation, $\widetilde{O}(nκ)$ total communication complexity per-block in expectation, and a best-case latency of $O(κ)$ rounds while ensuring safety and liveness with overwhelming probability, with $κ$ being a small security parameter.
Open access
2 source records
cs.DC
Complex Systems and Decision Making
Systems Engineering Methodologies and Applications
Hongbo Liu, Jiannong Cao, Bo Yang, Dongbin Bai · 10 authors
The rapid advancement of large language models (LLMs) in recent years has revolutionized the AI landscape. However, the deployment model and usage of LLM services remain highly centralized, creating significant trust issues and costs for end users and developers. To address these issues, we propose PolyLink, a blockchain-based decentralized AI platform that decentralizes LLM development and inference. Specifically, PolyLink introduces a decentralized crowdsourcing architecture that supports single-device and cross-device model deployment and inference across heterogeneous devices at the edge. Moreover, to ensure the inference integrity, we design the TIQE protocol, which combines a lightweight cross-encoder model and an LLM-as-a-Judge for a high-accuracy inference evaluation. Lastly, we integrate a comprehensive token-based incentive model with dynamic pricing and reward mechanisms for all participants. We have deployed PolyLink and conducted an extensive real-world evaluation through geo-distributed deployment across heterogeneous devices. Results indicate that the inference and verification latency is practical. Our security analysis demonstrates that the system is resistant to model degradation attacks and validator corruptions. PolyLink is now available at https://github.com/IMCL-PolyLink/PolyLink.
Sep 30, 2025·In Proceedings of the 21st Annual IFIP International Conference on Network and Parallel Computing (NPC 2025), Nha Trang, Vietnam, 14-16 November 2025
Modern blockchains need fast, reliable propagation to balance security and throughput. Virtual-coordinate methods speed dissemination but rely on slow iterative updates, leaving nodes out of sync. We present BlockSDN-VC, a transaction-broadcast protocol that centralises coordinate computation and forwarding control in an SDN controller, delivering global consistency, minimal path stretch and rapid response to churn or congestion. In geo-distributed simulations, BlockSDN-VC cuts median latency by up to 62% and accelerates convergence fourfold over state-of-the-art schemes with under 3% control-plane overhead. In a real blockchain environment, BlockSDN-VC boosts confirmed-transaction throughput by 17% under adversarial workloads, requiring no modifications to existing clients.
Collaborative and distributed learning techniques, such as Federated Learning (FL) and Split Learning (SL), hold significant promise for leveraging sensitive data in privacy-critical domains. However, FL and SL suffer from key limitations -- FL imposes substantial computational demands on clients, while SL leads to prolonged training times. To overcome these challenges, SplitFed Learning (SFL) was introduced as a hybrid approach that combines the strengths of FL and SL. Despite its advantages, SFL inherits scalability, performance, and security issues from SL. In this paper, we propose two novel frameworks: Sharded SplitFed Learning (SSFL) and Blockchain-enabled SplitFed Learning (BSFL). SSFL addresses the scalability and performance constraints of SFL by distributing the workload and communication overhead of the SL server across multiple parallel shards. Building upon SSFL, BSFL replaces the centralized server with a blockchain-based architecture that employs a committee-driven consensus mechanism to enhance fairness and security. BSFL incorporates an evaluation mechanism to exclude poisoned or tampered model updates, thereby mitigating data poisoning and model integrity attacks. Experimental evaluations against baseline SL and SFL approaches show that SSFL improves performance and scalability by 31.2% and 85.2%, respectively. Furthermore, BSFL increases resilience to data poisoning attacks by 62.7% while maintaining superior performance under normal operating conditions. To the best of our knowledge, BSFL is the first blockchain-enabled framework to implement an end-to-end decentralized SplitFed Learning system.
Several recent proposals implicitly or explicitly suggest making use of randomized transaction ordering within a block to mitigate centralization effects and to improve fairness in the Ethereum ecosystem. However, transactions and blocks are subject to gas limits and protocol rules. In a randomized transaction order, the behavior of transactions may change depending on other transactions in the same block, leading to invalid blocks and varying gas consumptions. In this paper, we quantify and characterize protocol violations, execution errors and deviations in gas consumption of blocks and transactions to examine technical deployability. For that, we permute and execute the transactions of over 335,000 Ethereum Mainnet blocks multiple times. About 22% of block permutations are invalid due to protocol violations caused by privately mined transactions or blocks close to their gas limit. Also, almost all transactions which show execution errors under permutation but not in the original order are privately mined transactions. Only 6% of transactions show deviations in gas consumption and 98% of block permutations deviate at most 10% from their original gas consumption. From a technical perspective, these results suggest that randomized transaction ordering may be feasible if transaction selection is handled carefully.
Yury Yanovich, Victoria Kovalevskaya, Maksim Egorov, Elizaveta Smirnova · 9 authors
The Open Network (TON) blockchain employs an asynchronous execution model that introduces unique security challenges for smart contracts. A primary concern is race conditions arising from unpredictable message processing order. While previous work established vulnerability patterns through static analysis of audit reports, dynamic detection of temporal dependencies through systematic testing remains an open problem. This study proposes a dynamic evaluation methodology based on controlled message orchestration to systematically expose vulnerabilities in asynchronous smart contracts. By synthesizing precise message queue manipulation with differential state analysis and probabilistic permutation testing, we establish a framework (namely, BugMagnifier) for identifying execution flaws that static methods miss. Experimental evaluation demonstrates BugMagnifier's effectiveness through extensive parametric studies on purpose-built vulnerable contracts and five real-world vulnerability cases reproduced from recent security audits. Results reveal message ratio-dependent detection complexity that aligns with theoretical predictions. This quantitative model enables predictive vulnerability assessment while shifting discovery from manual expert analysis to automated evidence generation. By providing reproducible test scenarios for temporal vulnerabilities, BugMagnifier addresses a critical gap in the TON security tooling, offering practical support for safer smart contract development in asynchronous blockchain environments.
Traditional single-proposer blockchains suffer from miner extractable value (MEV), where validators exploit their serial monopoly on transaction inclusion and ordering to extract rents from users. While there have been many developments at the application layer to reduce the impact of MEV, these approaches largely require auctions as a subcomponent. Running auctions efficiently on chain requires two key properties of the underlying consensus protocol: selective-censorship resistance and hiding. These properties guarantee that an adversary can neither selectively delay transactions nor see their contents before they are confirmed. We propose a multiple concurrent proposer (MCP) protocol offering exactly these properties.
Spam poses a growing threat to blockchain networks. Adversaries can easily create multiple accounts to flood transaction pools, inflating fees and degrading service quality. Existing defenses against spam, such as fee markets and staking requirements, primarily rely on economic deterrence, which fails to distinguish between malicious and legitimate users and often exclude low-value but honest activity. To address these shortcomings, we present StarveSpam, a decentralized reputation-based protocol that mitigates spam by operating at the transaction relay layer. StarveSpam combines local behavior tracking, peer scoring, and adaptive rate-limiting to suppress abusive actors, without requiring global consensus, protocol changes, or trusted infrastructure. We evaluate StarveSpam using real Ethereum data from a major NFT spam event and show that it outperforms existing fee-based and rule-based defenses, allowing each node to block over 95% of spam while dropping just 3% of honest traffic, and reducing the fraction of the network exposed to spam by 85% compared to existing rule-based methods. StarveSpam offers a scalable and deployable alternative to traditional spam defenses, paving the way toward more resilient and equitable blockchain infrastructure.
Blockchain Business applications and cryptocurrencies such as enable secure, decentralized value transfer, yet their pseudonymous nature creates opportunities for illicit activity, challenging regulators and exchanges in anti money laundering (AML) enforcement. Detecting fraudulent transactions in blockchain networks requires models that can capture both structural and temporal dependencies while remaining resilient to noise, imbalance, and adversarial behavior. In this work, we propose an ensemble framework that integrates Graph Convolutional Networks (GCN), Graph Attention Networks (GAT), and Graph Isomorphism Networks (GIN) to enhance blockchain fraud detection. Using the real-world Elliptic dataset, our tuned soft voting ensemble achieves high recall of illicit transactions while maintaining a false positive rate below 1%, beating individual GNN models and baseline methods. The modular architecture incorporates quantum-ready design hooks, allowing seamless future integration of quantum feature mappings and hybrid quantum classical graph neural networks. This ensures scalability, robustness, and long-term adaptability as quantum computing technologies mature. Our findings highlight ensemble GNNs as a practical and forward-looking solution for real-time cryptocurrency monitoring, providing both immediate AML utility and a pathway toward quantum-enhanced financial security analytics.
This paper introduces Lyquor, a decentralized platform that reimagines blockchain infrastructure through a service-centric model where nodes selectively host smart contracts (called Lyquids) while preserving global composability. We present three key innovations: (1) Fate-Constrained Ordering (FCO), which decouples consensus from execution to enable selective hosting without sacrificing Layer-1 grade composability; (2) Direct Memory Architecture (DMA), which eliminates state access bottlenecks by providing each contract with persistent, byte-addressable virtual memory; and (3) Universal Procedure Call (UPC), which enables fault-tolerant, programmable coordination across distributed off-chain computation. Together, these components are powered by a Rust-macroed unified programming model where on-chain and off-chain logic coexist seamlessly, supporting both traditional smart contract patterns and novel distributed applications. Lyquor addresses critical limitations in existing systems while maintaining compatibility with Ethereum APIs, offering a path toward truly scalable decentralized computation.
Sharding has emerged as a critical solution to address the scalability challenges faced by blockchain networks, enabling them to achieve higher transaction throughput, reduced latency, and optimized resource usage. This paper investigates the advancements, methodologies, and adoption potential of sharding in the context of Hedera, a distributed ledger technology known for its unique Gossip about Gossip protocol and asynchronous Byzantine Fault Tolerance (ABFT). We explore various academic and industrial sharding techniques, emphasizing their benefits and trade-offs. Building on these insights, we propose a hybrid sharding solution for Hedera that partitions the network into local and global committees, facilitating efficient cross-shard transactions and ensuring robust security through dynamic reconfiguration. Our analysis highlights significant reductions in storage and communication overhead, improved scalability, and enhanced fault tolerance, demonstrating the feasibility and advantages of integrating sharding into Hedera's architecture.
Pipelined Byzantine Fault Tolerant (BFT) consensus is fundamental to permissioned blockchains. However, many existing protocols are limited by the requirement for view-consecutive quorum certificates (QCs). This constraint impairs performance and creates liveness vulnerabilities under adverse network conditions. Achieving "certificate decoupling"-committing blocks without this requirement-is therefore a key research goal. While the recent BeeGees algorithm achieves this, our work reveals that it suffers from security and liveness issues. To address this problem, this paper makes two primary contributions. First, we formally define these flaws as the Invalid Block Problem and the Hollow Chain Problem. Second, we propose pBeeGees, a new algorithm that addresses these issues while preserving certificate decoupling with no additional computational overhead. To achieve this, pBeeGees integrates traceback and pre-commit validation to solve the Invalid Block Problem.Further, to mitigate the Hollow Chain Problem, we introduce a prudent validation mechanism, which prevents unverified branches from growing excessively. To summarize, pBeeGees is the first protocol to simultaneously achieve safety, liveness, and certificate decoupling in a pipelined BFT framework. Experiments confirm that our design significantly reduces block commit latency compared to classic algorithms, particularly under frequent stopping faults.
Users of blockchains value scalability, expecting fast confirmations and immediate transaction processing. Odontoceti, the latest in DAG-based consensus, addresses these concerns by prioritizing low latency and high throughput, making a strategic trade-off in security by operating with a 20% fault tolerance instead of the established 33% level. It is the first DAG-based protocol to achieve commitment in just two communication rounds, delivering median latency of 300 milliseconds while processing 10,000 transactions per second under realistic network conditions. Odontoceti operates with n = 5f + 1 validators and creates an uncertified DAG with a novel decision rule for committing blocks. The protocol includes an optimization that advances progress when participants are slow, benefiting crash fault scenarios which are more common in practice than Byzantine faults. Evaluation results demonstrate 20-25% latency improvements compared to an existing production protocol, validating that reducing wave length from three rounds to two rounds yields meaningful performance benefits. This paper establishes the practical viability of lower fault tolerance consensus protocols for blockchains.
Qianyu Yu, Giuliano Losa, Nibesh Shrestha, Xuechao Wang
To maximize performance, many modern blockchain systems rely on eventually-synchronous, Byzantine fault-tolerant (BFT) consensus protocols. Two protocol designs have emerged in this space: protocols that minimize latency using a leader that drives both data dissemination and consensus, and protocols that maximize throughput using a separate, asynchronous data dissemination layer. Recent protocols such as Partially-Synchronous Bullshark and Sailfish combine elements of both approaches by using a DAG to enable parallel data dissemination and a leader that paces DAG formation. This improves latency while achieving state-of-the-art throughput. Yet the latency of leader-based protocols is still better under moderate loads, which are common in practice. We present Angelfish, a hybrid protocol that adapts smoothly across this design space, from leader-based to Sailfish-like DAG-based consensus. Angelfish lets a dynamically adjusted subset of parties use best-effort broadcast to issue lightweight votes instead of reliably broadcasting costlier DAG vertices. This reduces communication, helps lagging nodes catch up, and lowers latency in practice compared to prior DAG-based protocols. Our empirical evaluation shows that Angelfish attains state-of-the-art peak throughput while significantly lowering latency under moderate throughput, delivering the best of both worlds.
We analyze 15,097 blocks proposed for inclusion in Ethereum's blockchain over an eight-minute window on December 3, 2024, during which 38 blocks were added to the chain. We classify transactions as exclusive -- appearing only in blocks from a single builder -- or private -- absent from the public mempool but included in blocks from multiple builders. We find that, depending on the methodology, exclusive transactions account for between 77.2% and 84% of the total fees paid by transactions in winning blocks. Moreover, we show that exclusivity cannot be fully attributed to persistent relationships between senders and builders: only between 7% and 8.4% of all on-chain exclusive transaction value originates from senders who route exclusively to one builder. Finally, we observe that transaction exclusivity is dynamic. Some transactions are exclusive at the start of a bidding cycle but later appear in blocks from multiple builders. Other transactions remain exclusive to a losing builder for two or three cycles before appearing in the public mempool. These transactions are therefore delayed and then exposed to potential attacks.
Mohamed Abdessamed Rezazi, Mouhamed Amine Bouchiha, A. Bendada, Yacine Ghamri-Doudane
Roaming settlement in 5G and beyond networks demands secure, efficient, and trustworthy mechanisms for billing reconciliation between mobile operators. While blockchain promises decentralization and auditability, existing solutions suffer from critical limitations-namely, data privacy risks, assumptions of mutual trust, and scalability bottlenecks. To address these challenges, we present B5GRoam, a novel on-chain and zero-trust framework for secure, privacy-preserving, and scalable roaming settlements. B5GRoam introduces a cryptographically verifiable call detail record (CDR) submission protocol, enabling smart contracts to authenticate usage claims without exposing sensitive data. To preserve privacy, we integrate non-interactive zero-knowledge proofs (zkSNARKs) that allow on-chain verification of roaming activity without revealing user or network details. To meet the high-throughput demands of 5G environments, B5GRoam leverages Layer 2 zk-Rollups, significantly reducing gas costs while maintaining the security guarantees of Layer 1. Experimental results demonstrate a throughput of over 7,200 tx/s with strong privacy and substantial cost savings. By eliminating intermediaries and enhancing verifiability, B5GRoam offers a practical and secure foundation for decentralized roaming in future mobile networks.
Zero-Knowledge Proofs (ZKP) are protocols which construct cryptographic proofs to demonstrate knowledge of a secret input in a computation without revealing any information about the secret. ZKPs enable novel applications in private and verifiable computing such as anonymized cryptocurrencies and blockchain scaling and have seen adoption in several real-world systems. Prior work has accelerated ZKPs on GPUs by leveraging the inherent parallelism in core computation kernels like Multi-Scalar Multiplication (MSM). However, we find that a systematic characterization of execution bottlenecks in ZKPs, as well as their scalability on modern GPU architectures, is missing in the literature. This paper presents ZKProphet, a comprehensive performance study of Zero-Knowledge Proofs on GPUs. Following massive speedups of MSM, we find that ZKPs are bottlenecked by kernels like Number-Theoretic Transform (NTT), as they account for up to 90% of the proof generation latency on GPUs when paired with optimized MSM implementations. Available NTT implementations under-utilize GPU compute resources and often do not employ architectural features like asynchronous compute and memory operations. We observe that the arithmetic operations underlying ZKPs execute exclusively on the GPU's 32-bit integer pipeline and exhibit limited instruction-level parallelism due to data dependencies. Their performance is thus limited by the available integer compute units. While one way to scale the performance of ZKPs is adding more compute units, we discuss how runtime parameter tuning for optimizations like precomputed inputs and alternative data representations can extract additional speedup. With this work, we provide the ZKP community a roadmap to scale performance on GPUs and construct definitive GPU-accelerated ZKPs for their application requirements and available hardware resources.
Quorum systems are a common way to formalize failure assumptions in distributed systems. Traditionally, these assumptions are shared by all involved processes. More recently, systems have emerged which allow processes some freedom in choosing their own, subjective or asymmetric, failure assumptions. For such a system to work, individual processes' assumptions must be compatible. However, this leads to a Catch-22-style scenario: How can processes collaborate to agree on compatible failure assumptions when they have no compatible failure assumptions to start with? We introduce asymmetric grid quorum systems that allow a group of processes to specify heterogeneous trust assumptions independently of each other and without coordination. They are based on qualitative attributes describing how the processes differ. Each process may select a quorum system from this class that aligns best with its subjective view. The available choices are designed to be compatible by definition, thereby breaking the cycling dependency. Asymmetric grid quorum systems have many applications that range from cloud platforms to blockchain networks.