This preview shows page 1. Sign up to view the full content.
Unformatted text preview: the algorithm in CBC or CFB mode, a fixed key, and IV; the last ciphertext block is the hash value. These methods are described in various standards using DES: both modes in , CFB in , CBC in [55, 56, 54]. This just isn’t good enough for one-way hash functions, although it will work for a MAC (see Section 18.14) . A cleverer approach uses the message block as the key, the previous hash value as the input, and the current hash value as the output. The actual hash functions proposed are even more complex. The block size is usually the key length, and the size of the hash value is the block size. Since most block algorithms are 64 bits, several schemes are designed around a hash that is twice the block size. Assuming the hash function is correct, the security of the scheme is based on the security of the underlying block function. There are exceptions, though. Differential cryptanalysis is easier against block functions in hash functions than against block functions used for encryption: The key is known, so several tricks can be applied; only one right pair is needed for success; and you can generate as much chosen plaintext as you want. Some work on these lines is [1263, 858, 1313]. What follows is a summary of the various hash functions that have appeared in the literature [925, 1465, 1262]. Statements about attacks against these schemes assume that the underlying block cipher is secure; that is, the best attack against them is brute force. One useful measure for hash functions based on block ciphers is the hash rate, or the number of n-bit messages blocks, where n is the block size of the algorithm, processed per encryption. The higher the hash rate, the faster the algorithm. (This measure was given the opposite definition in , but the definition given here is more intuitive and is more widely used. This can be confusing.) Schemes Where the Hash Length Equals the Block Size
The general scheme is as follows (see Figure 18.8): H0 = IH, where IH is a random initial value Hi = EA(B) • C where A, B, and C c...
View Full Document
This note was uploaded on 10/18/2010 for the course MATH CS 301 taught by Professor Aliulger during the Fall '10 term at Koç University.
- Fall '10