Abstract: The development of cloud computing, big data, Internet of things and other technologies has laid the foundation for data sharing. Currently, governments all over the world are encouraging data sharing and trading to promote the connectivity between data and stimulate the development of big data economy. However, because the data themselves contain a lot of private information and the data sharing process lacks incentives, most data owners are reluctant to participate in the sharing. Based on this, a data sharing scheme combining auction and privacy protection  emerges. But in these auction methods, the data reserve price is not taken into account, and the relevant interests of the data requester are ignored. Therefore, this paper designs a data incentive sharing scheme based on auction mechanism to meet the interests of both parties. Firstly, attribute-based access control technology is used to prevent unauthorized requestor from entering the auction process. Secondly, in combination with the auction in the real scene, information entropy is used to set a hard reserve price for data. In order to prevent the requestor from overexaggerating the value of the owner's data in the auction process and causing the winner to regret, the soft reserve price is set as the ideal maximum transaction price of the owner. Finally, the Pedersen promise is used to hide the bid of the requester, and the zero-knowledge range proof is used to verify the bid information of each requester, and the blockchain is used to realize the public accountability of the data transaction results. The experimental analysis shows that the auction scheme can satisfy the interests of both parties and realize the individual rationality, weak budget balance and incentive compatibility of all participants.

Key words: blockchain, incentives, data auctions, information entropy, privacy protection

摘要： 云计算、大数据、物联网等技术的发展为数据共享奠定了基础，当前各国政府都在鼓励数据共享和交易，以促进数据之间的互联互通，刺激大数据经济的发展。由于数据本身含有许多隐私信息以及数据共享过程缺乏激励，大多数据拥有者不愿参与共享，基于此出现了将拍卖与隐私保护结合起来的数据共享方案。但在这些拍卖方法中，没有考虑到数据底价问题，并且忽略了数据请求者的相关利益。为此，基于拍卖机制设计了一种满足双方利益的数据激励共享方案。采用基于属性的访问控制技术防止未授权的请求者进入拍卖环节；结合现实场景中的拍卖使用信息熵为数据设置硬性底价，为了防止拍卖过程中请求者过度夸大拥有者的数据价值造成获胜者后悔的情况，设置软性底价作为拥有者理想的最高成交价；应用Pedersen承诺对请求者出价进行隐藏，并使用零知识范围证明对各请求者的出价信息进行验证，利用区块链实现数据交易结果的公开问责。实验分析表明，所设计的拍卖方案满足双方利益并且可以实现所有参与者的个体理性、弱预算平衡和激励相容等特性。

关键词: 区块链, 激励机制, 数据拍卖, 信息熵, 隐私保护