An Incentive-Compatible Reward Sharing Mechanism for Mitigating Mirroring Attacks in Decentralized Data-Feed Systems

Fuente: arXiv
Salvato in:
Dettagli Bibliografici
Autori principali: Aeeneh, Sina, Zlatanov, Nikola, Yu, Jiangshan
Natura: Preprint
Pubblicazione: 2025
Soggetti:
Accesso online:
Tags: Aggiungi Tag
Nessun Tag, puoi essere il primo ad aggiungerne!!
_version_ 1866908538640531456
author Aeeneh, Sina
Zlatanov, Nikola
Yu, Jiangshan
author_facet Aeeneh, Sina
Zlatanov, Nikola
Yu, Jiangshan
contents Decentralized data-feed systems enable blockchain-based smart contracts to access off-chain information by aggregating values from multiple oracles. To improve accuracy, these systems typically use an aggregation function, such as majority voting, to consolidate the inputs they receive from oracles and make a decision. Depending on the final decision and the values reported by the oracles, the participating oracles are compensated through shared rewards. However, such incentive mechanisms are vulnerable to mirroring attacks, where a single user controls multiple oracles to bias the decision of the aggregation function and maximize rewards. This paper analyzes the impact of mirroring attacks on the reliability and dependability of majority voting-based data-feed systems. We demonstrate how existing incentive mechanisms can unintentionally encourage rational users to implement such attacks. To address this, we propose a new incentive mechanism that discourages Sybil behavior. We prove that the proposed mechanism leads to a Nash Equilibrium in which each user operates only one oracle. Finally, we discuss the practical implementation of the proposed incentive mechanism and provide numerical examples to demonstrate its effectiveness.
format Preprint
id arxiv_https___arxiv_org_abs_2509_11294
institution arXiv
publishDate 2025
record_format arxiv
spellingShingle An Incentive-Compatible Reward Sharing Mechanism for Mitigating Mirroring Attacks in Decentralized Data-Feed Systems
Aeeneh, Sina
Zlatanov, Nikola
Yu, Jiangshan
Computer Science and Game Theory
Emerging Technologies
Information Retrieval
Information Theory
Probability
Decentralized data-feed systems enable blockchain-based smart contracts to access off-chain information by aggregating values from multiple oracles. To improve accuracy, these systems typically use an aggregation function, such as majority voting, to consolidate the inputs they receive from oracles and make a decision. Depending on the final decision and the values reported by the oracles, the participating oracles are compensated through shared rewards. However, such incentive mechanisms are vulnerable to mirroring attacks, where a single user controls multiple oracles to bias the decision of the aggregation function and maximize rewards. This paper analyzes the impact of mirroring attacks on the reliability and dependability of majority voting-based data-feed systems. We demonstrate how existing incentive mechanisms can unintentionally encourage rational users to implement such attacks. To address this, we propose a new incentive mechanism that discourages Sybil behavior. We prove that the proposed mechanism leads to a Nash Equilibrium in which each user operates only one oracle. Finally, we discuss the practical implementation of the proposed incentive mechanism and provide numerical examples to demonstrate its effectiveness.
title An Incentive-Compatible Reward Sharing Mechanism for Mitigating Mirroring Attacks in Decentralized Data-Feed Systems
topic Computer Science and Game Theory
Emerging Technologies
Information Retrieval
Information Theory
Probability
url https://arxiv.org/abs/2509.11294