Title	Description	Author	Status	Purpose
Forest: Smart Contract Implementation for Enhanced Traceability in Central Bank Digital Currency Systems	An implementation of advanced data structure over token smart contract.	Sirawit Techavanitch (sirawit_tec@utcc.ac.th)	Draft	Dissertation

Forest: A Smart Contract Implementation for Enhanced Traceability in Central Bank Digital Currency Systems

Abstract

// TODO
Keyword: Anti-Money Laundering, Blockchain, Counter Financial Terrorism, Central Bank Digital Currencies, Smart Contract

Introduction

The present-day Central Bank Digital Currency concept aims to utilize the advantages of Blockchain Technology or Distributed Ledger Technology that provide immutability, transparency, and security, and adopts smart contracts, which plays a key feature in creating programmable money. However, technology itself gives an advantage and eliminates the problem ideally of compliance with the regulator and AML/CFT standard, but it does not seem practical to be done in the real world and is not efficiently responsible for the financial crime or incidents that occur in the open network of economic.

Related Works

// Opinion: AI and Deep learning recognize and analyze the pattern, but it'd be nice if the data structure also provided a more efficient and fast response to the incident.

• ERC20 fungible token intended to be currency-like, but the data structure is account-based, making it hard to separate money when it's mixed with the total balance.
• ERC721 not suitable due to metadata not being stored on-chain, it can be modified to support, but it's not intended to be exchangeable.
• ERC1155 metadata problem is the same as ERC721, however, ERC1155 can utilize tokenId as root, it can freeze the balance for each account, but it ends up with you can't to separate the money when it's stored in the total balance.
• ERC1400 have characteristic base from ERC20 but extended functionality for freeze account or freeze balance.
• ERC3643 have characteristic same as ERC1400 but extended functionality for store the document and other feature.
• MerkleTree not suitable for the payment due to its need to maintain the root hash and generate proof every time.
• UTXO maintain the amount of money or group of money in each individual transaction. To spend the transaction, the caller needs to be the owner of the transaction that needs to be spent.
• eUTXO extended version of UTXO, purpose of eUTXO is adding/carrying additional data as extraData or payload in the transaction.

Methodology

Introduce implementation call Forest used the way to modified the state to keep tracking subtree avoid to creating transaction output for change back the to spender like in UTXO.

Conclusion and Evaluation

Features	ERC20	UTXO	eUTXO	Forest
freeze the sender account.	✓	✓	✓	✓
freeze the recipient account.	✓	✓	✓	✓
freeze the certain amount token.	✗	✓	✓	✓
freeze the specifics tokenId or txId.	✗	✓	✓	✓
freeze the specifics tokenId or TxId that relevant to the root.	✗	✗	✓	✓
keep tracking child/subtree.	✗	✗	✗	✓

• For ERC20 provide events and keep tracking each Transfer,
  but the problem is the ERC20 model can't separate clean money from dirty money,
  due to the ERC20 not have tokenId to keep tracking each token when it's move.
• For ERC721 its use non-fungible, each token is unique and not intend to keep tracking amount or value.
• For ERC1400, includes features like partitioned balances,
  allowing tokens to be split into subsets based on conditions or rules (e.g., restrictions, investor categories).
  It also supports document management, enabling the attachment of legal documents to tokens, and integrates with compliance mechanisms,
  ensuring that transfers comply with rules like KYC/AML checks. However,
  ERC1400 tokens still face challenges in tracking the history or provenance of individual tokens beyond their partitioned states.
• For ERC1155 improves upon ERC20 and ERC721 by offering more flexibility and reducing operational overhead,
  it does not inherently provide a way to distinguish between clean and dirty tokens,
  as it lacks the ability to track individual token histories like UTXO models do.
• For UTXO and eUTXO facing challenge to combine multiple UnspentTransaction and spent as one,
  in case, user want to spend value that greater that selected UnspentTransaction.
  Possible solution: prepare and batching as an array,
  UTXO and eUTXO maintain the amount of money or group of money in each individual transaction.
  Each UnspentTransaction is not underlying any account, so to spend the transaction, the caller needs to be the owner of the transaction that needs to be spent.
• For Forest use to modify an existing state rather than create a new output transaction, like in UTXO or eUTXO do,
  it allows spending the transaction multiple times till it's met 0, The Forest model enables tracking of child/subtree structures,
  providing a hierarchical view of token flows and relationships,
  which is not possible in traditional token standards like ERC20, ERC721, ERC1155, ERC1400, and ERC3643.

For Further Work

Currently Forest not 100% compatible with existing ERC20 standard. To complete and fully supported ERC20 interface Forest require to have automatically select and spent the transaction.
However it's doesn't need to be store in sorted list. there is possible solution to be done.

• First in smart contract, It's can be done with FIFO or First-In-First-Out style which is can be done in smart contract but should be avoiding the gasUsed.
• Second off-load heavy computation from the smart contract to a custom pre-compiled contract.
• Third create a stateful pre-compiled contract that fully functional same as the smart contract specification.

Learn more about pre-compiled contract

The Forest project addresses significant challenge, particularly state bloat and dust transaction.
Merging multiple small transactions into a single larger transaction helps reduce the number of entries in the blockchain’s state, thereby minimizing state growth.
However, this process can complicate traceability, as it obscures the individual histories of the merged transactions.
Additionally, once a transaction's output has been spent, it can be marked as spent or removed from the dataset,
which further complicates the ability to trace the flow of funds. While these strategies are effective in managing state size,
they can hinder users' ability to follow transaction trails, making it difficult to verify the origins of funds.

Glossary of Terms

AML Anti-Money Laundering
B2G Back to Genesis
CBDC Central Bank Digital Currency
CFT Counter Financial Terrorism
C2C Customer to Customer
DLT Distributed Ledger Technology
eUTXO Extended Transaction Output
KYC Know Your Customer
UTXO Unspent Transaction Output

Reference

Conceptual Feasibility Fraud Handling from rCBDC BOT report page 33.

Comparison of As is and To be (Conceptual) from rCBDC BOT report page 33.