可证明安全的后量子两方口令认证密钥协商协议

口令认证密钥协商可以在客户机和服务器之间建立安全的远程通信,且可以将一个低熵口令放大为一个高熵的会话密钥。然而,随着量子计算技术的快速发展,基于大数分解和离散对数等经典数学难题的PAKA协议面临着严峻的安全挑战。因此,为了构建一个高效安全的后量子PAKA协议,依据改进的Bellare-Pointcheval-Rogaway(BPR)模型,提出了一个基于格的匿名两方PAKA协议,并且使用给出严格的形式化安全证明。性能分析结果表明,该方案与相关的PAKA协议相比,在安全性和执行效率等方面有一定优势,更适用于资源受限的物联网(Internet of Things, IoT)智能移动设备。     

A provably secure and efficient two‐party password‐based explicit authenticated key exchange protocol resistance to password guessing attacks

Password‐based two‐party authenticated key exchange (2PAKE) protocol enables two or more entities, who only share a low‐entropy password between them, to authenticate each other and establish a high‐entropy secret session key. Recently, Zheng et al. proposed a password‐based 2PAKE protocol based on bilinear pairings and claimed that their protocol is secure against the known security attacks. However, in this paper, we indicate that the protocol of Zheng et al. is insecure against the off‐line password guessing attack, which is a serious threat to such protocols. Consequently, we show that an attacker who obtained the users' password by applying the off‐line password guessing attack can easily obtain the secret session key. In addition, the protocol of Zheng et al. does not provide the forward secrecy of the session key. As a remedy, we also improve the protocol of Zheng et al. and prove the security of our enhanced protocol in the random oracle model. The simulation result shows that the execution time of our 2PAKE protocol is less compared with other existing protocols. Copyright © 2015 John Wiley & Sons, Ltd.     

Two-factor mutual authentication based on smart cards and passwords

One of the most commonly used two-factor user authentication mechanisms nowadays is based on smart-card and password. A scheme of this type is called a smart-card-based password authentication scheme. The core feature of such a scheme is to enforce two-factor authentication in the sense that the client must have the smart-card and know the password in order to gain access to the server. In this paper, we scrutinize the security requirements of this kind of schemes, and propose a new scheme and a generic construction framework for smart-card-based password authentication. We show that a secure password based key exchange protocol can be efficiently transformed to a smart-card-based password authentication scheme provided that there exist pseudorandom functions and target collision resistant hash functions. Our construction appears to be the first one with provable security. In addition, we show that two recently proposed schemes of this kind are insecure.     

新的基于椭圆曲线的三方口令认证密钥协商协议

口令认证密钥协商使得参与通信的用户用一个低熵的口令就可以实现实体认证,并能通过不安全的信道安全地生成共享的高熵会话密钥。为此,设计了一种新的基于椭圆曲线的三方口令认证密钥协商协议,新协议将参与者的口令巧妙地隐藏在传输的消息中,确保了口令的安全性。新协议的安全性基于椭圆曲线离散对数问题,服务器并不需要完全可信。安全性分析和性能分析显示,新协议以较低的代价实现了通信双方的安全通信。     

Universally composable anonymous password authenticated key exchange

Anonymous password authenticated key exchange (APAKE) is an important cryptographic primitive, through which a client holding a password can establish a session key with a server both authentically and anonymously. Although the server is guaranteed that the client in communication is from a pre-determined group, but the client’s actual identity is protected. Because of their convenience, APAKE protocols have been widely studied and applied to the privacy protection research. However, all existing APAKE protocols are handled in stand-alone models and do not adequately settle the problem of protocol composition, which is a practical issue for protocol implementation. In this paper, we overcome this issue by formulating and realizing an ideal functionality for APAKE within the well-known universal composability (UC) framework, which thus guarantees security under the protocol composition operations. Our formulation captures the essential security requirements of APAKE such as off-line dictionary attack resistance, client anonymity and explicit mutual authentication. Moreover, it addresses the arbitrary probabilistic distribution of passwords. The construction of our protocol, which utilizes SPHF-friendly commitments and CCA2-secure encryption schemes, can be instantiated and proven secure in the standard model, i.e., without random oracle heuristics.     

基于hash函数和公钥算法的一次性口令方案*

在分析现有基于挑战／应答一次性口令方案不足的基础上,设计了一种新的基于hash函数和公开密钥加密算法的一次性口令身份认证方案。该方案不仅能够提供通信双方的相互认证、避免各种攻击,而且克服了传统挑战/应答方案认证开销大的缺点,有效地保护了用户身份信息,能防止重放攻击等攻击手段。最后对方案的安全性和效率作了分析。     

一个前向安全的基于口令认证的三方密钥交换协议

目前,文献中提出的基于口令认证的密钥交换协议,很多都是针对两方的情形设计的,即通信双方为客户与服务器,它们通过一个预先共享的口令来进行认证的密钥交换.随着现代通信环境的快速变化,需要能为任意客户间构建一个端到端的安全信道,这种应用的情形与那些文献中所考虑的有很大区别.针对这种情形,文中提出了一个可证前向安全的基于口令认证的三方密钥交换协议,使通信双方在认证服务器的帮助下能相互进行认证并建立一个会话密钥.与前人提出的基于口令认证的三方密钥交换协议相比,该协议在计算代价和通信代价上都较有效,因而更适用于资源受限的环境.此协议的安全性是在口令型的选择基Gap Diffie-Hellman问题难解的假设前提下在随机谕示模型下证明的.     

EC2C-PAKA: An efficient client-to-client password-authenticated key agreement

Most password-authenticated key agreement schemes described in the literature have focused on authenticated key agreement using a shared password between a client and a server. With rapid changes in the modern communication environment such as ad hoc networks and ubiquitous computing, it is necessary to construct a secure end-to-end channel between clients. This paradigm is a quite different paradigm from the existing ones.In this paper, we study client-to-client password-authenticated key agreement (C2C-PAKA) enabling two clients in different realms to agree on a common session key using different passwords. Byun et al. first presented a C2C-PAKA protocol under the cross-realm setting. However, the scheme was not formally treated, and subsequently found to be flawed. In addition, in this scheme, there is still opportunity for improvements both in the computation and communication aspects. We provide formal treatments for the C2C-PAKA protocol by using Bellare et al.’s security model. We also suggest an efficientC2C-PAKA protocol and prove that the protocol is secure under the decisional Diffie-Hellman assumption in the ideal cipher and random oracle models.     

A Generic Framework for Anonymous Authentication in Mobile Networks

Designing an anonymous user authentication scheme in global mobility networks is a non-trivial task because wireless networks are susceptible to attacks and mobile devices powered by batteries have limited communication, processing and storage capabilities. In this paper, we present a generic construction that converts any existing secure password authen- tication scheme based on a smart card into an anonymous authentication scheme for roaming services. The security proof of our construction can be derived from the underlying password authentication scheme employing the same assumptions. Compared with the original password authentication scheme, the transformed scheme does not sacrifice the authentication effciency, and additionally, an agreed session key can be securely established between an anonymous mobile user and the foreign agent in charge of the network being visited. Furthermore, we present an instantiation of the proposed generic construction. The performance analysis shows that compared with other related anonymous authentication schemes, our instantiation is more effcient.     

具有强匿名性的网关口令认证密钥交换协议

网关口令认证密钥交换协议允许用户和网关在服务器的协助下建立起一个共享的会话密钥,其中用户和服务器之间的认证通过低熵的口令来完成.已有的网关口令认证密钥交换协议对用户的匿名性研究不足.该文基于Diffie-Hellman密钥交换提出了具有强匿名性的网关口令认证密钥交换协议,并且在随机预言模型下基于标准的DDH假设证明了协议的安全性.新协议可以抵抗不可检测在线字典攻击并且计算效率高,安全性和计算效率都优于已有的同类协议.     

安全域间信息资源访问的协议和方法

为了保护内部网络的安全，必须设置应用边界安全设备．Internet上不同的应用安全域间要实现信息资源的安全访问，首先需要认证．Kerberos是目前比较常用的认证协议，一般的应用边界安全设备（如SOcks5）中就应用了该认证协议，但应用该协议存在一定的缺陷：在应用边界安全设备链的认证过程中，资源域中的应用边界安全设备认证对象是主体域中的应用边界安全设备，而不是真正发起资源请求的客户端，因此资源域中的应用边界安全设备审计的对象是主体域中的应用边界安全设备，而不是真正的客户端．在Kerberos域间认证的基础上，给出了新的域间认证协议以及身份传递协议，使用新的协议不仅能够提供应用边界安全设备对用户访问请求的安全审计而且只需要两次域间的网络连接，这两次域间网络连接不需要主体和客体直接进行，而是通过应用边界安全设备完成的，提高了系统的通信效率，扩大了该系统的应用范围，适合于现有的企业网环境，能有效地解决企业网与企业网之间的信息安全传输．     

Provably secure three-party password authenticated key exchange protocol in the standard model

Three-party password authenticated key exchange protocol is a very practical mechanism to establish secure session key through authenticating each other with the help of a trusted server. Most three-party password authenticated key exchange protocols only guarantee security in the random oracle model. However, a random oracle based cryptographic construction may be insecure when the oracle is replaced by real function. Moreover, some previous unknown attacks appear with the advance of the adversary capability. Therefore, a suitable standard model which can imitate a wider variety of attack scenarios for 3PAKE protocol is needed. Aim at resisting dictionary attack, unknown key-share attack and password-compromise impersonation attack, an expanded standard model for 3PAKE protocol is given. Meanwhile, through applying ElGamal encryption scheme and pseudorandom function, a specific three-party password authenticated key exchange protocol is proposed. The security of the proposed protocol is proven in the new standard model. The result shows that the present protocol has stronger security by comparing with other existing protocols, which covers the following security properties: (1) semantic security, (2) key privacy, (3) client-to-server authentication, (4) mutual authentication, (5) resistance to various known attacks, and (6) forward security.     

基于RLWE的双因子三方认证密钥交换协议

为了使格上Diffie-Hellman式密钥交换协议能够实现认证性并且适用于客户-服务器-客户模式的大规模通信,提出了一个基于环上误差学习RLWE的双因子三方认证密钥交换协议。该协议将口令和生物特征作为客户的长期密钥,实现服务器对客户的显式身份认证。首先利用环上误差学习的困难问题的优势（密钥及密文尺寸短、运行效率高）来构造密码体制;其次服务器通过口令和生物特征的哈希值传递环元素,并结合丁式错误协调机制使得通信方获得随机均匀的会话密钥。最后分析表明,该协议适用于大规模通信,提高了通信量,具有更高的安全属性,可以抵抗口令泄露用户假冒攻击。     

基于MAC地址的软件动态口令实现方案

随着网络技术的发展,传统的静态密码身份认证方案已不能给电子商务活动提供足够的保护。描述了动态口令技术的原理,分析了硬件和软件实现动态口令的利弊,设计了一种基于客户端MAC地址的动态口令认证协议,并在此基础上论述了系统方案的实施流程、总体设计和认证过程。最后进行了安全性分析并提出了相应的提高安全性的措施。分析表明,该方案具有适用面广、安全性高、使用方便和系统成本低的特点。     

An efficient and practical threshold gateway-oriented password-authenticated key exchange protocol in the standard model

With the assistance of an authentication server, a gateway-oriented password-authenticated key exchange (GPAKE) protocol can establish a common session key shared between a client and a gateway. Unfortunately, a GPAKE protocol becomes totally insecure if an adversary can compromise the authentication server and steal the passwords of the clients. In order to provide resilience against adversaries who can hack into the authentication server, we propose a threshold GPAKE protocol and then present its security proof in the standard model based on the hardness of the decisional Diffie-Hellman (DDH) problem. In our proposal, the password is shared among n authentication servers and is secure unless the adversary corrupts more than t+1 servers. Our protocol requires n > 3t servers to work. Compared with existing threshold PAKE protocols, our protocol maintains both stronger security and greater efficiency.     

基于Weil对改进的Kerberos协议设计

Kerberos协议是一种具有广泛应用价值的身份认证协议,但该协议存在口令猜测及无客户端认证等安全问题,通过引入Weil对,成功对Kerberos协议进行了改进,改进后的协议具有更强的安全性和实用性。     

标准模型下高效的三方口令认证密钥交换协议

三方口令认证密钥交换协议允许两个分别与服务器共享不同口令的用户在服务器的协助下建立共享的会话密钥,从而实现了用户间端到端的安全通信.现阶段,多数的三方口令认证密钥交换协议都是在随机预言模型下可证明安全的.但在实际应用中,利用哈希函数对随机预言函数进行实例化的时候会给随机预言模型下可证明安全的协议带来安全隐患,甚至将导致协议不安全.以基于ElGamal加密的平滑投射哈希函数为工具,在共同参考串模型下设计了一种高效的三方口令认证密钥交换协议,并且在标准模型下基于DDH假设证明了协议的安全性.与已有的同类协议相比,该协议在同等的安全假设下具有更高的计算效率和通信效率,因此更适用于大规模的端到端通信环境.     

基于动态口令的身份认证方法

本文提出了一种采用安全单钥管理技术,在客户机和认证中心两端建立加密和认证协议,将用户在客户机端输入的静态口令和认证系统自动生成的静态口令,通过密钥元素代替表,代替成一次一变的认证密钥,再生成动态口令实现身份认证,从而,建立基于动态口令身份认证系统.     

理想格上的多阶段认证密钥交换协议

现有的格基认证密钥交换协议普遍基于Bellare-Rogaway等单阶段模型,忽略了实际通信场景中会话密钥建立的阶段分离。针对这一问题,基于MSKE模型提出了一种格基多阶段认证密钥交换协议。该方案使用预共享的口令进行认证,并使用Peikert误差消除机制结合服务器静态密钥实现多阶段密钥协商。分析表明,该方案整体只引入少量计算开销,实现了双向认证、二阶会话密钥完美前向保密、抗量子攻击等特性,在MSKE模型下满足KD-2FS-M安全等级,是一种简单高效的后量子多阶段密钥交换协议。     

Experiences developing architectures for realizing thin‐client diagram editing tools

Diagram‐centric applications such as software design tools, project planning tools and business process modelling tools are usually ‘thick‐client’ applications running as stand‐alone desktop applications. There are several advantages to providing such design tools as Web‐based or even PDA‐ and mobile‐phone‐based applications. These include ease of access and upgrade, provision of collaborative work support and Web‐based integration with other applications. However, building such thin‐client diagram editing tools is very challenging. We have developed several thin‐client diagram editing applications realized as a set of plug‐in extensions to a meta‐tool for visual design environment development. In this paper, we discuss key user interaction and software architecture issues, illustrate examples of interacting with our thin‐client diagram editing tools, describe our design and implementation approaches, and present the results of several different evaluations of the resultant applications. Our experiences will be useful for those interested in developing their own thin‐client diagram editing architectures and applications. Copyright © 2007 John Wiley & Sons, Ltd.