A distributed investment encryption scheme: Investcoin

0Citations
Citations of this article
10Readers
Mendeley users who have this article in their library.
Get full text

Abstract

This work presents a new framework for Privacy-Preserving Investment systems in a distributed model. In this model, independent investors can transfer funds to independent projects, in the same way as it works on crowdfunding platforms. The framework protects the investors’ single payments from being detected (by any other party), only the sums of each investor’s payments are revealed (e.g. to the system). Likewise, the projects’ single incoming payments are concealed and only the final sums of the incoming payments for every project are revealed. In this way, no other party than the investor (not even the system administration) can detect how much she paid to any single project. Though it is still possible to confidentially exchange any part of an investment between any pair of investors, such that market liquidity is unaffected by the system. On top, our framework allows a privacy-preserving return of a multiple of all the held investments (e.g. interest payments or dividends) to the indivdual investors while still revealing nothing else than the sum of all returns for every investor. We introduce Investcoin as practicable instantiation for this framework. It is a proper combination of three cryptographic protocols, namely a Private Stream Aggregation scheme, a Commitment scheme and a Range test. The security of the three protocols is based on the Decisional Diffie-Hellman (DDH) assumption. Thus, by a composition theorem, the security of Investcoin is also based on the DDH assumption. Furthermore, we provide a simple decentralised key generation protocol for Investcoin that supports dynamic join, leave and fault-tolarance of investors and moreover achieves some security guarantees against malicious investors.

Cite

CITATION STYLE

APA

Valovich, F. (2017). A distributed investment encryption scheme: Investcoin. In Lecture Notes in Computer Science (including subseries Lecture Notes in Artificial Intelligence and Lecture Notes in Bioinformatics) (Vol. 10332 LNCS, pp. 136–154). Springer Verlag. https://doi.org/10.1007/978-3-319-60080-2_10

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