TABLE OF CONTENTS Content
Password access control: UNIX passwords, basic HTTP, PAP
Challenge-response protocols: digest HTTP, CHAP
One-time passwords: Lamport scheme, S/Key
Key distribution: Needham-Schroeder, Kerberos, Merkle puzzles
Authentication chains: Merkle signature scheme, timestamps
Case study: GSM network, Bluetooth network
In this chapter we look at several examples of protocols which use conventional cryptography. As we can see, the most important problems are peer authentication and key distribution. Once these problems are solved, the conventional cryptographic primitives that we have seen in previous chapters can be used in order to build up secure communications, e.g. establish communication sessions between a client and a server which preserve confidentiality, integrity, authentication, and in which no adversary can replay, reorder, or erase messages.
The most intuitive way to perform access control is to request a password as depicted in Fig. 5.1. When a client wishes to connect to a server, they can proceed as follows.
The client first sends an access request to the server.
The server acknowledges and sends a password request to the client.
The client sends his password to the server.
The server checks the correctness of the password and either provides or denies access to the client.
Here, the password does not need to be kept in memory: the server only keeps a message digest (using a hash function) of the password. This is an interesting property for security since this protects against potential damages when...
