| layout |
|---|
page |
This document summarizes several open research problems related to Grin. Feel free to join us on Keybase to discuss them. Funding available.
Introduced in 2001 by Boneh, Lynn and Shacham, BLS signatures introduces non-interactive kernel aggregation and potentially simpler multisignature schemes.
While it is not for certain that such signatures are well suited for Grin it is still interesting to explore this design.
-
D. Boneh, B. Lynn, and H. Shacham. Short signatures from the weil pairing. In Proceedings of the 7th International Conference on the Theory and Application of Cryptology and Information Security: Advances in Cryptology, ASIACRYPT '01, pages 514–532, Berlin, Heidelberg, 2001. Springer-Verlag. https://www.iacr.org/archive/asiacrypt2001/22480516.pdf
-
D. Boneh, M. Drijvers, and G. Neven. Compact multi-signatures for smaller blockchains. Cryptology ePrint Archive, Report 2018/483, 2018. https://eprint.iacr.org/2018/483
In Grin, kernels, outputs and range proofs are stored in a type of accumulator called a Merkle Mountain Range. While efficient, there have been several improvements to this kind of data structure, for example RSA Accumulators. The goal of this research problem is to identify whether or not Grin can leverage more advanced or faster cryptographic accumulators.
- D. Boneh, B. Bünz, and B. Fisch. Batching techniques for accumulators with applications to IOPs and stateless blockchains. Cryptology ePrint Archive. 2018 https://eprint.iacr.org/2018/1188.pdf
This research subject is about investigating recent advances in zero knowledge proofs and how Grin can leverage their potential improvements. Improvements can be on:
- Alternative small proof of existence of kernel history up to horizon
- Privacy
- Soundness
- Hardness
- Efficiency (whether speed or size)
- ...
- https://eprint.iacr.org/2020/1618.pdf - Proof-Carrying Data without Succinct Arguments
- https://eprint.iacr.org/2019/1021.pdf - Halo: Recursive Proof Composition without a Trusted Setup
- https://www.benthamsgaze.org/2019/02/07/introducing-sonic-a-practical-zk-snark-with-a-nearly-trustless-setup/ - SuperSonic
- https://github.com/matter-labs/awesome-zero-knowledge-proofs - Awesome Zero Knowledge Proofs
Mimblewimble does not support any kind of script. In 2017, Andrew Poelstra introduced a way to add smart contract ability to this type of blockchain: scriptless scripts.
While a significant research effort has been done on the subject there are very few concrete implementations of such cryptographic structure.
The goal of this research problems is to investigate scriptless scripts in Grin and potentially implement basic contract features.
"To validate transactions, cryptocurrencies such as Bitcoin and Ethereum require nodes to verify that a blockchain is valid. This entails downloading and verifying all blocks, taking hours and requiring gigabytes of bandwidth and storage. Hence, clients with limited resources cannot verify transactions independently without trusting full nodes.[...] FlyClient is a novel transaction verification light client for chains of variable difficulty. FlyClient is efficient both asymptotically and practically and requires downloading only a logarithmic number of block headers while storing only a single block header between executions."
The goal of this research problems is to investigate/implement the necessaries prerequisites for the introduction of FlyClient in Grin.
- B. Bünz, L. Kiffer, L. Luu, and M. Zamani. Flyclient: Super-light clients for cryptocurrencies. IACR Cryptology ePrint Archive, 2019:226, 2019. https://eprint.iacr.org/2019/226.pdf
One of the trade-offs of the Mimblewimble type blockchain is that it requires a round trip between the payer and payee to construct a valid transaction. Meaning that the sender and receiver must be online to transact.
Several research attempts have been made in that direction. For example, relying on federated relays or Beam shared bulletin board system.
The goal is this research problem is to investigate and develop an asynchronous, robust, and privacy preserving method of building Grin transactions.
Mimblewimble/Grin leverage confidential transactions to hide the identity of the sender and recipients. As such, there are no public amounts or addresses.
However, it is possible for someone listening on the network to build a transaction graph and possibly clustering entities together.
Techniques like Dandelion++ mitigate this issue but are insufficient for a privacy coin.
A much more promising design is this Mimblewimble CoinSwap proposal
The goal of this research is to investigate ways to obfuscate the Grin transaction and implement such design.
- Ian Miers - Blockchain Privacy: Equal Parts Theory and Practice (especially Flashlight Attack part) https://www.zfnd.org/blog/blockchain-privacy/#flashlight
Erlay is a transaction dissemination protocol proposed by G. Naumenko, G. Maxwell, P. Wuille. The motivation is to improve connectivity between Bitcoin nodes and reduce the bandwidth required to operate a full node. It claims to reduce the bandwidth consumption by 40% assuming current connectivity, and keeps bandwidth use almost constant as connectivity increases. It also hardens the network against attacks that attempt to learn the origin node of a transaction.
This protocol could prove useful to Grin as well, and it is of value to the community to investigate how this best should be implemented.
https://arxiv.org/pdf/1905.10518.pdf
https://forum.grin.mw/t/erlay-bandwidth-efficient-transaction-relay-in-bitcoin/5081
Payment channel hubs offer the possibility of instant payments that improve privacy by joining transactions (like a coinjoin server), and improving scalability by pruning intermediate txs and potentially even aggregating kernels. Grin already has NRD (NoRecentDuplicate) kernels waiting to be activated, which provide the relative timelocks needed for implementing basic payment channels. The difficulty is in trying to design a PCH that is trustless (can't steal/freeze coins or rollback txs), and that doesn't learn everything about the individual transactions (amounts, input/output links, etc). These past few years have seen a lot of developments in this area (TumbleBit, Bolt, A2L, etc.) but nothing that can be directly applied to Grin in a way that doesn't requiring trusting hub operators to preserve your privacy.
- https://eprint.iacr.org/2016/701 - Bolt
- https://eprint.iacr.org/2016/575 - TumbleBit
- https://eprint.iacr.org/2019/589 - Anonymous Atomic Locks (A2L)