【作者】 朱珍超；

【导师】 张玉清；

【作者基本信息】 西安电子科技大学 ， 信息安全， 2011， 博士

【摘要】 随着社会的发展,人与人之间信息交换变得越发迫切与频繁。如何确保交换过程中的信息的安全催生出以此为研究目标的一门科学—密码学。然而,目前广泛应用于保密通信中的密码体制(密码算法)的安全性大多依赖于没有严格证明的数学难题,随着经典计算机计算能力的提高和量子分解算法研究领域的重大突破,依赖于数学难题的密码体制的安全性将面临着严峻的挑战。量子密码的出现使密码研究者们摆脱了这一窘境。量子密码学是密码学与量子力学相结合的产物,其安全性受到量子力学基本规律的保证。量子不可克隆原理和Heisenberg测不准原理保证了量子密码体制具有无条件安全性和对窃听的可检测性,因此,量子密码体制具有良好的性能和应用前景。本文紧跟量子密码协议研究领域的前沿热点问题,主要取得了以下成果:(1)基于两粒子纠缠态提出一个高效的量子秘密共享协议。协议中,除去用作检测窃听的粒子外,所有的EPR纠缠对均可以用作生成最终的秘密信息,因此,协议的理论效率接近100%。协议可以抵抗内部成员的攻击,诸如不可见光子检测攻击、特洛伊木马攻击以及密集编码攻击对于提出的协议都是无效的,同时,一种被称为不透明欺骗攻击的有效性也被首次分析。(2)指出两个高效的具有高容量的量子秘密共享协议中存在的安全性问题并给出相应的改进方案。改进的协议是基于对两粒子纠缠态直接编码而不是通过对某些特定的局部正交变换的编码来实现,协议是安全高效的;同时,我们将协议成功的推广到多方的情况,为此类协议的进一步的应用奠定了坚实的基础;最后,我们对此类量子秘密共享协议中存在的一个有趣但是非常重要的问题(即协议中共享的消息不是由Boss而是由代理联合生成的,这是与量子秘密共享的应用情景相违背的)展开讨论,通过对我们提出的协议做一些细微的修改即可避免此问题。(3)分析了一个基于χ类型纠缠态的三方量子秘密共享协议中存在的安全隐患并给出相应的改进方案。通过截获—重发攻击,协议中的不诚实成员Bob(注意,并不是Charlie)可以在没有Charlie的帮助下完全获得Alice所有的秘密信息。(4)指出一个基于Bell态测量的量子秘密共享协议中的不安全性并给出相应的改进方案。在原协议中,任何代理Alice(或Bob)可以在没有别的成员的帮助下非法的获得Trent的一半秘密消息;通过修改量子态和经典信息的对照表,最终使得协议安全;改进的协议拥有以下的优点:(a)成员不需要制备任何量子态,也不需要执行对粒子的任何局部正交变换操作;(b)在生成最终的秘密时,所有的成员唯一需要做的就是对纠缠对做Bell态测量;(c)除了在检测窃听阶段需要传输经典信息外,协议在执行过程中不需要传递任何经典信息;(d)所有的信息只需在量子信道中传输一次,这也就意味着由于信道中存在噪声而损失的粒子很少。(5)指出一个可以实现多方与多方之间秘密共享的量子秘密共享协议中存在的安全性隐患并给出相应的改进方案。在原协议中,群组1中的最后一个成员可以非法获得协议中的大部分秘密。(6)基于单粒子提出一个具有高量子比特利用率的量子对话协议,此协议可以抵抗拒绝服务攻击、特洛伊木马攻击和不可见光子检测攻击,协议中不存在信息泄露问题,鉴于协议中不需要使用纠缠态粒子以及多态联合测量,因此协议是很适合于实际应用的。更多还原

【Abstract】 With the development of human society,information exchange between people becomes more urgent and frequent. How to ensure the security of the information in the exchange process gave birth to a new science, cryptography. Most conventional cryptographic algorithm widely used nowadays is based on the unproven computational assumptions. However, with the development of computer science and the breakthrough of the factoring algorithm,the information security mechanism depending on these unproven computational assumptions will face great challenge. The appearance of the quantum cryptography helps cryptographers out of this dilemma. Quantum cryptography,is the synthesis of quantum mechanics with the art of code making (Cryptography), its security is guaranteed by the fundamental laws of physics. Quantum no cloning theorem and Heisenberg’s uncertainty principle ensure the unconditional Security and the ability of detecting eavesdropper, these facts guarantee that quantum cryptography has the excellent capacity and the attractive foreground. Focusing on the frontier research subjects of quantum cryptography, the main results we obtained in the dissertation are as follows:(1) We propose a protocol for multi party quantum secret sharing utilizing four non orthogonal two particle entangled states. In the protocol, all the entangled states can be used for generating the private key except those used for eavesdropping check. The theoretic efficiency for qubits of the protocol is improved from 50% to approaching 100%. The protocol is secure against to insider participant’s attack, such as invisible photon eavesdropping attack; the delay photon Trojan horse attacker and dense coding attack. The validity of a probable attack called opaque cheat attack to this kind of protocols is considered in the paper for the first time.(2) We analyze two efficient high capacity quantum secret sharing schemes and then propose a protocol for quantum secret sharing with coding of two particle quantum entanglement states directly. In this protocol, the sharing of the secret is accomplished by coding of the two particle quantum entanglement states but not the coding of some special discrete unitary operations. The security of the proposed protocol is analyzed, the scheme has a high intrinsic efficiency for qubits and a high capacity. We give the method to generalize our QSS scheme to multiparty QSS scheme. We also discuss an awkward situation that the sharing secret information is determined by the agents but not the boss existed in some schemes available and we give a variant of our scheme to avoid this problem.(3) We analyze the security of a three party quantum secret sharing of secure direct communication based onχtype entangled states, the protocol is insecure. Applying intercept and resend attack, one agent, Bob, but not Charlie, can gain Alice’s secret without the help from the other agent, Charlie. We also give our effective modification to improve the protocol.(4) We study the security of a multiparty quantum secret sharing scheme based on Bell measurement, the protocol is not secure for any dishonest agent Alice (or Bob), who can illegally elicit half of Trent’s secret message by himself. Finally a feasible improvement of this quantum secret sharing protocol is proposed. As we just change the transformation rule, the improved protocol also has the following optimal efficiency as the original protocol: (a) All the agents need not prepare any quantum state or carry out local unitary operations. (b)When generating keys, all agents only need to perform Bell measurement no matter how many participants there are in the QSS task. (c) No classical information is required to be transmitted during the protocol except for the detection process. (d) All instances are transmitted through the quantum channel only once which means fewer particles is wasted in the channel noise.(5) We study the security of a quantum secret sharing scheme between multiparty and multiparty, the protocol is not secure for the last member in the first group can illegally obtain most secret messages without introducing any error. Finally, a possible way to avoid the security flaw is suggested.(6) We put forward an efficient quantum dialogue protocol using single photons. The protocol overcomes the information leakage problem existed in most previous quantum dialogue protocols and has a high efficiency for qubits. The protocol is secure against denial of service attack, the delay photon Trojan horse attack and invisible photon eavesdropping attack. As only single photons were used in the protocol, the protocol is experimentally feasible within current technologies.更多还原