Browse Prior Art Database

Identifying the Cipher Symbols of a Cryptogram From a Partially Incorrect Decryption

IP.com Disclosure Number: IPCOM000061554D
Original Publication Date: 1986-Aug-01
Included in the Prior Art Database: 2005-Mar-09
Document File: 3 page(s) / 54K

Publishing Venue

IBM

Related People

Casey, RG: AUTHOR

Abstract

A method is described for identifying the cipher symbols of a cryptogram from a partially incorrect decryption made up of sets of deciphers of the cipher symbols grouped into words. The method comprises the steps of (1) obtaining a list of candidate correct words with spelling similar to a word in the decryption by performing a dictionary search for each word in the decryption; (2) based upon the probability of errors in the decrypting process and the probability of correctness of a decipher, iteratively eliminating a subset of words from the list of candidates until only one decipher exists for each cipher symbol; and (3) repeating steps (1) and (2) until an accepted threshold of correctness is reached.

This text was extracted from a PDF file.
At least one non-text object (such as an image or picture) has been suppressed.
This is the abbreviated version, containing approximately 53% of the total text.

Page 1 of 3

Identifying the Cipher Symbols of a Cryptogram From a Partially Incorrect Decryption

A method is described for identifying the cipher symbols of a cryptogram from a partially incorrect decryption made up of sets of deciphers of the cipher symbols grouped into words. The method comprises the steps of (1) obtaining a list of candidate correct words with spelling similar to a word in the decryption by performing a dictionary search for each word in the decryption; (2) based upon the probability of errors in the decrypting process and the probability of correctness of a decipher, iteratively eliminating a subset of words from the list of candidates until only one decipher exists for each cipher symbol; and (3) repeating steps (1) and (2) until an accepted threshold of correctness is reached. By way of example, consider the following text and the cipher symbols associated with the letters in the text, such as might be generated by an image recognition system:

(Image Omitted)

Assuming that due to incorrect interpretation of cipher symbols 2, 6, 7, 11, 12, and 13, the following partially incorrect decryption is obtained: SPDID SOMY RIDE LUBOMS IN SKOLEM DREYS SO LPDY SMPK IM DERUBSIOM Each word of the decryption (hereinafter called master) is first submitted to a dictionary search program whereby, as shown in Table 1, a list of candidate correct words (hereinafter called clones) with similar spelling is returned if a master is not in the dictionary. A subset of clones for a master is then eliminated according to rules which reflect the probability of errors in the decrypting process. For example, if the occurrence of two consecutive errors in the decryption process is improbable, then a clone which differs from its master in two consecutive letters (e.g., spDId and spIEd) is eliminated, resulting in...