This preview shows page 1. Sign up to view the full content.
Unformatted text preview: n of EarthWeb is prohibited. Read EarthWeb's privacy statement. To access the contents, click the chapter and section titles. Applied Cryptography, Second Edition: Protocols, Algorthms, and Source Code in C (cloth)
Go!
Keyword
Brief Full Advanced Search Search Tips (Publisher: John Wiley & Sons, Inc.) Author(s): Bruce Schneier ISBN: 0471128457 Publication Date: 01/01/96 Search this book:
Go! Previous Table of Contents Next
 Ideally, you only want to collect one random bit per keystroke. Collecting more may skew the results, depending on how good a typist is sitting at the keyboard. This technique is limited, though. While it’s easy to have someone type 100 words or so when it is time to generate a key, it isn’t reasonable to ask the typist to type a 100,000word essay to generate a keystream for a onetime pad. Biases and Correlations
A major problem with all these systems is that there could be nonrandomness in the generated sequence. The underlying physical processes might be random, but many kinds of measuring instruments are between the digital part of the computer and the physical process. Those instruments could easily introduce problems. A way to eliminate bias, or skew, is to XOR several bits together. If a random bit is biased toward 0 by a factor e, then the probability of 0 can be written as: P(0) = .5 + e XORing two of these bits together yields: P(0) = (.5 + e)2 + (.5  e)2 = .5 + 2e2 By the same calculation, XORing 4 bits together yields: P(0) = .5 + 8e4 XORing m bits will exponentially converge to an equal probability of 0 and 1. If you know the maximum bias you are willing to accept for your application, you can calculate how many bits you need to XOR together to get random bits below that bias. An even better method is to look at the bits in pairs. If the 2 bits are the same, discard them and look at the next pair. If the 2 bits are different, take the first bit as the output of the generator. This eliminates bias completely. Other techniques for reducing bias use trans...
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
 ALIULGER
 Cryptography

Click to edit the document details