分布式无中心授权的属性基可变门限环签名

基于属性的环签名具有表达能力强、使用灵活、便于隐匿签名者身份等优点，因此逐渐成为相关领域的研究热点.分析结果表明，现有方案大多无法同时具备无条件强匿名性和抗合谋攻击性，并且存在属性密钥托管、签名门限固定、验证效率低下等问题.针对上述不足，首先给出分布式属性基门限环签名方案的形式化定义和安全模型，然后提出一个分布式无中心授权的属性基可变门限环签名方案.该方案采用分布式密钥生成协议，分散了属性授权机构权限，解决了属性密钥托管问题；通过在属性密钥中嵌入用户身份标识，在签名中引入用户身份模糊因子的方式，使该方案同时具备无条件强匿名性和抗合谋攻击性.另外，提出了一种适用于该方案的批验证算法，能够将验证计算量由nO（·）降为O（·）+n（其中，n为待验证签名的数量），有效提高了验证效率.在随机预言机模型和CDH困难问题假设下，该方案被证明在适应性选择消息和断言攻击下是存在性不可伪造的，并且能够抵抗由拥有互补属性集合的恶意成员发动合谋攻击.

Multi-Authority Attribute-Based Alterable Threshold Ring Signature without Central Authority

Attribute based ring signature has gradually become a hot topic in the related field, owing to its priorities including strong in expressive power, flexible in use, and easy to hide the identity of signer. By analyzing existing attribute-based ring signature schemes, it can be found that the majority of earlier schemes cannot resist the collusion attack with the premise of unconditional strong anonymity, and there are many issues such as attribute key escrow, fixed threshold and inefficient verification. To address the above defects, this paper firstly introduces the formal definitions and security model for the multi-authority attribute-based threshold ring signature scheme. Then a multi-authority attribute-based variable threshold ring signature scheme is presented. This scheme uses distributed key generation protocol to constrain the rights of attribute authority, and to overcome the problem of attribute key escrow. Through embedding a random identity factor in each user''s attribute key, and introducing a random fuzzy parameter in each signature, the scheme can provides both unconditional strong anonymity and collusion resistance. In addition, a batch verification algorithm is proposed to reduce the computation complexity of verification from nO(·) to O(·)+n. Under random oracle model and computational Diffie-Hellman assumption, the proposal can be proven to be existentially unforgeable and can resist collusion attacks launched by the malicious users with the complementary attributes in chosen message and predicate attack.