Key establishment --- security models, protocols and usage
Berkant Ustaoğlu
Key establishment is the process whereby two or more parties derive a shared \nsecret, typically used for subsequent confidential communication. However, \nidentifying the exact security requirements for key establishment protocols is \na non-trivial task. This thesis compares, extends and merges existing security \ndefinitions and models for key establishment protocols. \n \nThe primary focus is on two-party key agreement schemes in the public-key \nsetting. On one hand new protocols are proposed and analyzed in the existing \nCanetti-Krawzcyk model. On the other hand the thesis develops a security model \nand novel definition that capture the essential security attributes of the \nstandardized Unified Model key agreement protocol. These analyses lead to the \ndevelopment of a new security model and related definitions that combine and \nextend the Canetti-Krawzcyk pre- and post- specified peer models in terms of \nprovided security assurances. \n \nThe thesis also provides a complete analysis of a one-pass key establishment \nscheme. There are security goals that no one-pass key establishment scheme can \nachieve, and hence the two-pass security models and definitions need to be \nadapted for one-pass protocols. The analysis provided here includes \nthe description of the required modification to the underlying security model. \nFinally, a complete security argument meeting these altered conditions is \npresented as evidence supporting the security of the one-pass scheme. \n \nLastly, validation and reusing short lived key pairs are related to \nefficiency, which is a major objective in practice. The thesis considers the \nformal implication of omitting validation steps and reusing short lived key \npairs. The conclusions reached support the generally accepted cryptographic \nconventions that incoming messages should not be blindly trusted and extra \ncare should be taken when key pairs are reused.
•Jul 30, 2008•3 citations