Research On Privacy-preserving Auditing For Distributed Ledger

Due to the benefit of decentralization,immutability and cost saving,Distributed Ledger(DL),e.g.,Blockchain,is now booming and reshaping the practice in areas such as finance,government,law and etc.As data is recorded in a distributed manner,anonymity and privacy is provided to some extent.This property has,unfortunately,posed great challenge to necessary auditing and regulation.How to enhance a confidential Distributed Ledger with auditing functions has been a focus among the cryptographic community.Current works are either designed for specific purpose or hazard user privacy.Furthermore,the efficiency for large scale auditing falls short in the era of big data.The main works of the paper are as follows:(1)We propose Audi TEE,a privacy-preserving auditing scheme on DL.Leveraging TEE(Trusted Execution Environment)and Merkle tree,the scheme is universal and practical in finance,government,law and various other industries.Conquering several intrinsic challenges of TEE,such as lack in trusted memory,speed and reliable time source,in a cryptographic way,Audi TEE promises users strong confidentiality and auditors versatile auditing tasks while no secret beyond permission can be leaked.(2)Audi TEE is instantiated among a banking scenario to prove its performance.Comprehensive experiment and testing have demonstrated its efficiency and reliability and further improvements are suggested based on data analysis.(3)The security of our auditing scheme is proven in the UC(Universally Composable)framework.We show how we choose and design our ideal functionalities and how they satisfy our security goals.Furthermore,we construct the adversary S such that no environment Z running in polynomial time is able to distinguish whether it’s interacting with ideal world with S or the hybrid model with A.Therefore,our scheme has realized the ideal functionality.