首页> 外文OA文献 >Analyse des protocoles cryptographiques: des modèles symboliques aux modèles calculatoires
【2h】

Analyse des protocoles cryptographiques: des modèles symboliques aux modèles calculatoires

机译:密码协议分析:从符号模型到计算模型

代理获取
本网站仅为用户提供外文OA文献查询和代理获取服务,本网站没有原文。下单后我们将采用程序或人工为您竭诚获取高质量的原文,但由于OA文献来源多样且变更频繁,仍可能出现获取不到、文献不完整或与标题不符等情况,如果获取不到我们将提供退款服务。请知悉。

摘要

Security protocols are programs that secure communications by defining exchange rules on a network. They are used in many applications such as withdrawal machines, pay-per-view, mobile phone and e-commerce. Their objective is e.g. to guarantee the confidentiality of sensitive data, to authenticate some of the participants, or to guarantee anonymity or non-repudiation. These programs are executed on easily accessible public networks like the Internet. Thus in order to check their security, it is necessary to consider all possible attacks that could be mounted. The goal of my research habilitation thesis is to show that formal methods are well adapted for a precise analysis of cryptographic protocols, through various kind of tools. In this thesis, we present procedures to detect automatically whether a protocol is secure or not. We have proposed several algorithms, depending on which security properties are considered and which cryptographic primitives are used (encryption, signature, hash, exclusive or, etc.). Moreover, we characterize conditions that allow to combine the preceding results and to design protocol in a modular way. These results are based on symbolic models, very different from those used in cryptography where the notion of security is based on the complexity theory. This notion of security allows a better identification of all the attacks that can be mounted in reality but in return, the (heavy) security proofs are done by hand and seem difficult to automate. We have identified cryptographic hypotheses that allow to take the benefit of the symbolic and cryptographic approaches. It is then possible to obtain security proof at a cryptographic level, directly from proofs established (automatically) in symbolic models.
机译:安全协议是通过定义网络上的交换规则来确保通信安全的程序。它们被用于许多应用程序中,例如提款机,按次付费,移动电话和电子商务。他们的目标是以确保敏感数据的机密性,对某些参与者进行身份验证,或确保匿名性或不可否认性。这些程序在易于访问的公共网络(如Internet)上执行。因此,为了检查其安全性,有必要考虑所有可能发生的攻击。我的研究适应性论文的目的是表明,通过各种工具,形式化方法非常适合用于加密协议的精确分析。在本文中,我们提出了自动检测协议是否安全的过程。我们已经提出了几种算法,具体取决于要考虑的安全属性和所使用的密码原语(加密,签名,哈希,互斥或等等)。此外,我们描述了可以组合前述结果并以模块化方式设计协议的条件。这些结果基于符号模型,与安全性基于复杂性理论的密码术中使用的模型完全不同。这种安全性概念可以更好地识别现实中可能发动的所有攻击,但是作为回报,(繁重的)安全性证明是手工完成的,并且似乎很难实现自动化。我们已经确定了加密假设,可以利用符号和加密方法。这样就可以直接从符号模型中(自动)建立的证明中获得加密级别的安全证明。

著录项

  • 作者

    Cortier Véronique;

  • 作者单位
  • 年度 2009
  • 总页数
  • 原文格式 PDF
  • 正文语种 fr
  • 中图分类

相似文献

  • 外文文献
  • 中文文献
  • 专利

客服邮箱:kefu@zhangqiaokeyan.com

京公网安备:11010802029741号 ICP备案号:京ICP备15016152号-6 六维联合信息科技 (北京) 有限公司©版权所有
  • 客服微信

  • 服务号