Towards verifiable and privacy-preserving account model on a consortium blockchain based on zk-SNARKs

With high efficiency, low overhead, and high scalability, consortium blockchains have been gradually and deeply applied in various fields of society. Privacy preservation and information verifiability are essential for the application development of the consortium blockchains. The existing schemes of privacy preservation and information verifiability on the blockchain are mainly designed for public blockchains. Most of the existing privacy tokens are based on the less scalable Unspent Transaction Output (UTXO) model. The existing privacy tokens based on the account model either require offline information transfer or suffer from low efficiency and insufficient scalability. The consensus mechanism and operation process differences between the public and consortium blockchains make them impossible to apply in the consortium blockchains directly. While protecting the privacy of transaction information on consortium blockchains, there is no specific solution to both directly utilize the privacy tokens for transactions in the context of privacy information and verify the legitimacy of account balance changes before payments. This work proposes VmppToken, a verifiable and privacy-preserving account model on a consortium blockchain based on zero-knowledge Succinct Non-interactive ARguments of Knowledge (zk-SNARKs). By participants generating the private data relationship credentials for each token transaction using the zero-knowledge proof, it enables each endorsing peer node to verify the data relationship for the zero-knowledge proof credentials before writing the transaction into the ledger. VmppToken realizes the verifiability of transaction information while protecting the privacy of account balance and transaction amount and ensures the legitimacy of the transaction. We elaborate on the construction of VmppToken and analyze the security of the circuit constructed by zero-knowledge proof. We implement a prototype of the model on Hyperledger Fabric based on Libsnark and conduct comprehensive experiments to evaluate its efficiency and performance. Our experimental results demonstrate the effectiveness of the proposed model. The generation time of each proof is about 10-12 seconds, and the average time of credential verification is only 2.03 milliseconds. The transaction processing speed is about 660 Transactions Per Second (TPS) on a moderately equipped server.

Automated summary

With high efficiency, low overhead, and high scalability, consortium blockchains have been widely applied in various fields of society. However, there is a lack of specific solutions for privacy preservation and information verifiability in the context of consortium blockchains. This work proposes VmppToken, a verifiable and privacy-preserving account model, based on zero-knowledge proofs, to address these challenges. Experimental results demonstrate the effectiveness of VmppToken in terms of efficiency and performance.