A Basic Introduction to Cryptanalysis

When you look at a cryptogram for the first time, they nearly always look rather daunting after all how are you expected to decipher all that code!  However there are certain techniques that can make the task much less daunting and in some senses it is actually quite fun to work them out.

One of the most basic procedures is that of frequency analysis.  In fact without this technique you won’t get very far in understanding any of the procedures behind substitution cryptanalysis. So let’s try and briefly explain what’s behind this technique in an effort to expand our mathematical education.

Cryptanalysis relies on the fact that all letters of a specific language have specific characteristics or personalities.  To the ordinary observer all the letters might look fairly similar, but to the analyst they will know the specific traits and characters of each individual letter.

 

GJXV EHRT NUCOT WMOHY UWGK HWIE

GJXV EHRT NUCOT WMOHY UWGK HWIE

GJXV EHRT NUCOT WMOHY UWGK HWIE

GJXV EHRT NUCOT WMOHY UWGK HWIE

GJXV EHRT NUCOT WMOHY UWGK HWIE

GJXV EHRT NUCOT WMOHY UWGK HWIE

GJXV EHRT NUCOT WMOHY UWGK HWIE

GJXV EHRT NUCOT WMOHY UWGK HWIE

Don’t try and decipher this example though – there’s nothing there.  The cryptanalyst would initially begin by counting each letters frequency and it’s contacts (the contacts are the letters which are adjacent. They would then construct a frequency table based on the text.  This can then be compared with a standard frequency table based on a similar number of words.  This table will list the likely frequency of each individual letter – i.e how often it would be expected to appear.
Unfortunately it is rarely as simple to line up your standard frequency table and the one you created to solve the cryptogram.  They are very unlikely to be identical even for the reason that they will be based on a different number of characters.  However it is surprising how little relative frequencies shift from one piece of text to another.  You will almost always find e,t,a,o,n,r, i, s and h in the high frequency areas whereas d,l,u, c and m will normally be found in the medium frequency group.
It’s useful to see how the basics work in encryption, especially if you use such technology to protect yourself online.  For example I can use a technology to spoof my IP called a Virtual private network which creates an encrypted tunnel between the client, effectively hiding the IP address of my client.
Using frequency you can narrow down to distinct groups each individual letter, often with very accurate results.  However you need more than this to focus more specifically on a possible solution and this is where ’contacts’ are important.  Every letter has a cluster of associations that are likely to occur.  In fact an experienced cryptanalyst can spot these associations almost without thinking when presented with a frequency distribution and a tally chart.
facebooktwittergoogle_plusredditpinterestlinkedinmailby feather

Tags: ,