Revisiting the incentive mechanism of Bitcoin-NG

11Citations
Citations of this article
29Readers
Mendeley users who have this article in their library.
Get full text

Abstract

Recently, due to the inherent restriction of Bitcoin design, the throughput of Bitcoin blockchain protocol fails to meet the daily needs, leaving the scalability technology in dire need to provide better efficiency. To address this issue, numerous solutions have been proposed, including blocksize expansion, off-chain transactions and block structure modification. Among them, Bitcoin-NG, a scalable blockchain protocol introduced by Eyal et al. in USENIX 2016, improves scalability while simultaneously avoiding the deterioration of other metrics in the network. Bitcoin-NG has two types of blocks: key blocks for leader election and microblocks that contain ledger entries. Eyal et al. assert that the proportion of fee allocation of transactions in microblocks is bounded by miners’ mining power ratio out of all mining power in the system. Specifically, the upper bound is determined by the incentive sub-mechanism of longest chain extension, while the lower bound determined by the incentive sub-mechanism of transaction inclusion. We revisit the incentive mechanism of Bitcoin-NG. We point out that Eyal et al. neglect on the calculation of lower bound and manifest the over-simplification in the analysis of upper bound in detail. After that, the correct incentive mechanism is derived. Finally, we put forward an optimal proportion of transaction fee distribution.

Cite

CITATION STYLE

APA

Yin, J., Wang, C., Zhang, Z., & Liu, J. (2018). Revisiting the incentive mechanism of Bitcoin-NG. In Lecture Notes in Computer Science (including subseries Lecture Notes in Artificial Intelligence and Lecture Notes in Bioinformatics) (Vol. 10946 LNCS, pp. 706–719). Springer Verlag. https://doi.org/10.1007/978-3-319-93638-3_40

Register to see more suggestions

Mendeley helps you to discover research relevant for your work.

Already have an account?

Save time finding and organizing research with Mendeley

Sign up for free