基于联盟链的能源交易数据隐私保护方案

Privacy-preserving Scheme of Energy Trading Data Based on Consortium Blockchain

区块链技术可以有效地解决分布式能源交易系统中的信任缺失、恶意篡改和虚假交易等问题,但区块链开放、透明的特性使得基于区块链的能源交易系统极易受到攻击,导致用户隐私泄露。为此,提出了一种基于差分隐私算法和账户映射技术的隐私保护方案BLDP-AM(Blockchain Local Differential Privacy-Account Mapping),用于保护交易数据的隐私。该方案重新设计了本地差分隐私算法的数据扰动机制使之适用于区块链技术,并基于该扰动机制构造了BLDP(Blockchain Local Differential Privacy)算法来保护交易数据的隐私。同时,为了保证交易正确性以及隐藏交易曲线特征,该方案首先通过账户映射(Account Mapping,AM)技术实现用户与多个账户关联,然后采用指数平滑预测(Exponential Smoothing Prediction,ESP)算法计算各账户的交易预测值,最后使用BLDP算法扰动交易预测值来获得真实交易值并进行交易。通过隐私分析证明了该方案在保护数据隐私方面的可行性,且实验分析表明该方案具有较好的性能。

Blockchain technology could effectively solve the problems of lack of trust,malicious tampering and false transactions.However,the open and transparent characteristics of the blockchain make the distributed energy trading model based on the blockchain extremely vulnerable to be attacked,leading to the disclosure of user’s privacy.Therefore,a privacy-preserving scheme BLDP-AM based on differential privacy algorithm and account mapping technology is proposed to protect the privacy information of trading data.Our scheme redesigns the data perturbation mechanism of the local differential privacy algorithm to make it applicable to blockchain technology,and constructs the BLDP algorithm based on this perturbation mechanism to protect the privacy of transaction data.At the same time,in order to ensure the correctness of trading and hide the characteristics of the trading curve,our scheme first associates users with multiple accounts through account mapping technology,then uses the exponential smoo-thing prediction algorithm to calculate the trading prediction value of each account,and finally uses the BLDP algorithm to perturb the trading prediction value to obtain the real trading value and conduct trading.Our scheme not only guarantee the correctness of transactions but also achieve the purpose of protecting the privacy of trading data.The privacy analysis proves the feasibility of the scheme in protecting user privacy,and the experimental analysis shows that the scheme has better performance.