Editing Final Report 2011 (section)

==Cipher Investigation==
===Concept===
The Cipher Investigation component of the project provides a comprehensive cipher analysis of the Somerton Man Code through the methodical investigation of as many ciphers as was possible, to determine whether they had been used in generating the Somerton Man code. In the process, contributing to the ongoing cipher investigation associated with the case.

===Previous Work===
Two previous attempts at testing ciphers against the code were considered in 2011. These were the examination within the 2009 University of Adelaide Honours Project and the Department of Defence investigation conducted in 1978.
The cryptanalysts from the Department of Defence did not derive any definitive results, they concluded<ref name=InsideStory>''Inside Story'', presented by Stuart Littlemore, ABC TV, 1978.</ref>:
# There are insufficient symbols to provide a pattern.
# The symbols could be a complex substitute code or the meaningless response to a disturbed mind.
# It is not possible to provide a satisfactory answer.
In 2009, the project group investigated transposition ciphers, the One Time Pad, the Vigenère Cipher and the Playfair Cipher. Their conclusion was that transposition ciphers warranted no further investigation, the Vigenère and Playfair Ciphers were unlikely to be have been used but the One Time Pad was plausible<ref name=FinalReport2009>''Final Report 2009'', Bihari, Denley and Turnbull, Andrew, 2009, https://www.eleceng.adelaide.edu.au/personal/dabbott/wiki/index.php/Final_report_2009:_Who_killed_the_Somerton_man%3F</ref>

This year the One Time Pad, the Vigenère Cipher and the Playfair Cipher have been re-examined in greater depth and in acknowledgement of the 2009 results, no additional ciphers were tested that used a transposition methodology.

===Technical Challenges===
The key technical challenges faced in the Cipher Analysis portion of the project are identified below.
# Cipher Possibilities
# Sample Size
# Code Ambiguities
# Circumstantial Unknowns

[[Image:The_Code.png|thumb|170px|right|Ambiguous letters.]]
The first challenge, cipher possibilities, refers to the almost infinite number of potential ciphers that could have been used; there are many different algorithms to transform plaintext to ciphertext. For example, considering just simple mono-alphabetic substitution ciphers, there are 26! (factorial) possible ciphers; that is over 400,000,000,000,000,000,000,000,000 or 400 trillion trillion ciphers<ref name=TheCodeBook>''The Code Book'', author Simon Singh, The Fourth Estate, 2000.</ref>.

The sample size of the source material, 44 letters, also adds challenges. Frequency Analysis was introduced in the Background section as being the traditional method of solving substitution ciphers, however it relies on an accurate representation of the letter frequency distributions. This typically requires a sample of at least 100 letters.

Code ambiguity refers to some of the letters in the Somerton code being hard to identify. For example, the first letter in the first and second line in the figure below could be M’s or W’s. Code ambiguities added further challenges to the cipher analysis.

Finally, the circumstantial unknowns of the case introduce more complications. It is not known what nationality the Somerton Man was so we cannot be sure what language any hidden message would be in.

===Methodology===
Cipher analysis was done through a methodology established at the beginning of the project.
<center>[[File:Cipher Analysis methodology.png|Cipher Analysis Methodology]]</center>
<center>'''Figure 7 - Cipher analysis methodology'''</center>

Initially potential ciphers were '''identified''' by broad research aimed at isolating relevant ciphers. Relevance was determined as the cipher meeting several pre-conditions:
# The cipher needed to have been invented and in use prior to the time the unknown man was found in 1948. Thus no cipher invented after December 1 1948 was included in the 2011 investigation as they can be trivially discounted.
# It must be possible to perform the cipher methodology by hand. This acknowledges the original code written in the back of the Rubaiyat of Omar Khayyam whilst also reflecting that computers were not in common use in 1948. Mechanically derived codes such as those generated by the Enigma machine were not included.
# In accordance with the results from 2009, the ciphers were not to use a transposition-based methodology. 
Once identified, detailed '''research''' was conducted into each cipher determining methods for its use. Following the research, each cipher was '''investigated''' in its potential relationship to the Somerton Man code. The investigative methods that were used depended on the function and scope of the individual cipher. These ranged from exhaustive software testing to statistical analysis, structural analysis and frequency analysis. 

The product of the investigation stage was a '''conclusion''': whether the cipher could be ruled out as being used in encrypting the code or not. In the '''Compilation''' stage, the results of the previous three sections, namely Research, Investigation and Conclusion were compiled onto a central database; this database being a sub-page on the project’s publicly accessible wiki page known as the [[Cipher Cross-off List]].

The final stage was to '''implement''' the cipher in the cipher analysis software tool. This software tool is referred to as the CipherGUI in this report.

===Results===
The result of the cipher investigation is a comprehensive list of approximately 30 different ciphers tested complete with detailed reasoning. The following table summarises these results.

For complete descriptions of each cipher and the full results and data, a visit to the centralised official database; the [[Cipher Cross-off List]] is strongly recommended. 

<center>'''Table 3 - Cipher investigation results'''</center>
{| border="1" cellspacing="0" style="text-align:center; margin: 1em auto 1em auto"
|- style="color:white; background:#191970; font-weight:bold"
| width="180" | Cipher || width="250" | Test techniques || width="100" | Status || width="120" | Student

|- style="text-align:left" 
| style="color:white; background:#4682B4" | ADFGVX || Structural analysis || style="text-align:center" | Disproven || style="text-align:center" | Steven

|- style="text-align:left; background:#DCDCDC" 
| style="color:white; background:#4682B4" | Affine || Direct decryption || style="text-align:center" | Disproven || style="text-align:center" | Patrick

|- style="text-align:left" 
| style="color:white; background:#4682B4" | Alphabet reversal || Direct decryption || style="text-align:center" | Disproven || style="text-align:center" | Patick

|- style="text-align:left; background:#DCDCDC" 
| style="color:white; background:#4682B4" | Auto-key || Direct decryption || style="text-align:center" | Disproven || style="text-align:center" | Steven

|- style="text-align:left" 
| style="color:white; background:#4682B4" | Baconian || Structural analysis || style="text-align:center" | Disproven || style="text-align:center" | Patick

|- style="text-align:left; background:#DCDCDC" 
| style="color:white; background:#4682B4" | Beaufort || Frequency analysis || style="text-align:center" | Unlikely || style="text-align:center" | Steven

|- style="text-align:left" 
| style="color:white; background:#4682B4" | Bifid || Frequency analysis || style="text-align:center" | Disproven || style="text-align:center" | Steven

|- style="text-align:left; background:#DCDCDC" 
| style="color:white; background:#4682B4" | Book || Structural analysis || style="text-align:center" | Disproven || style="text-align:center" | Patrick

|- style="text-align:left" 
| style="color:white; background:#4682B4" | Dvorak encoding || Direct decryption || style="text-align:center" | Disproven || style="text-align:center" | Patrick

|- style="text-align:left; background:#DCDCDC" 
| style="color:white; background:#4682B4" | Flat-frequency || Statistical analysis || style="text-align:center" | Disproven || style="text-align:center" | Patrick

|- style="text-align:left" 
| style="color:white; background:#4682B4" | Four square || Structural analysis || style="text-align:center" | Unlikely || style="text-align:center" | Patrick

|- style="text-align:left; background:#DCDCDC" 
| style="color:white; background:#4682B4" | Gronsfeld || Structural analysis || style="text-align:center" | Disproven || style="text-align:center" | Steven

|- style="text-align:left" 
| style="color:white; background:#4682B4" | Hill || Frequency analysis || style="text-align:center" | Disproven || style="text-align:center" | Steven

|- style="text-align:left; background:#DCDCDC" 
| style="color:white; background:#4682B4" | Homophonic substitution || Structural analysis, Statistical analysis || style="text-align:center" | Disproven || style="text-align:center" | Patrick

|- style="text-align:left" 
| style="color:white; background:#4682B4" | Keyword || Frequency analysis || style="text-align:center" | Possible || style="text-align:center" | Patrick

|- style="text-align:left; background:#DCDCDC" 
| style="color:white; background:#4682B4" | Nihilist || Structural analysis || style="text-align:center" | Disproven || style="text-align:center" | Steven

|- style="text-align:left" 
| style="color:white; background:#4682B4" | Null || Structural analysis || style="text-align:center" | Unlikely || style="text-align:center" | Patrick

|- style="text-align:left; background:#DCDCDC" 
| style="color:white; background:#4682B4" | Number based || Structural analysis || style="text-align:center" | Disproven || style="text-align:center" | Patrick

|- style="text-align:left" 
| style="color:white; background:#4682B4" | One time pad || Statistical analysis || style="text-align:center" | Unlikely || style="text-align:center" | Patrick

|- style="text-align:left; background:#DCDCDC" 
| style="color:white; background:#4682B4" | Pigpen || Direct decryption || style="text-align:center" | Disproven || style="text-align:center" | Patrick

|- style="text-align:left" 
| style="color:white; background:#4682B4" | Playfair || Structural analysis || style="text-align:center" | Disproven || style="text-align:center" | Patrick

|- style="text-align:left; background:#DCDCDC" 
| style="color:white; background:#4682B4" | Playfair (Double) || Frequency analysis || style="text-align:center" | Unlikely || style="text-align:center" | Steven

|- style="text-align:left" 
| style="color:white; background:#4682B4" | Porta || Frequency analysis || style="text-align:center" | Unlikely || style="text-align:center" | Steven

|- style="text-align:left; background:#DCDCDC" 
| style="color:white; background:#4682B4" | Rail fence || Direct decryption || style="text-align:center" | Disproven || style="text-align:center" | Patrick

|- style="text-align:left" 
| style="color:white; background:#4682B4" | Shift || Direct decryption || style="text-align:center" | Disproven || style="text-align:center" | Steven

|- style="text-align:left; background:#DCDCDC" 
| style="color:white; background:#4682B4" | Templar || Direct decryption || style="text-align:center" | Disproven || style="text-align:center" | Steven

|- style="text-align:left" 
| style="color:white; background:#4682B4" | Trifid || Frequency analysis || style="text-align:center" | Unlikely || style="text-align:center" | Steven

|- style="text-align:left; background:#DCDCDC" 
| style="color:white; background:#4682B4" | Trithemius || Direct decryption || style="text-align:center" | Disproven || style="text-align:center" | Steven

|- style="text-align:left" 
| style="color:white; background:#4682B4" | Two square || Structural analysis || style="text-align:center" | Unlikely || style="text-align:center" | Patrick

|- style="text-align:left; background:#DCDCDC" 
| style="color:white; background:#4682B4" | VIC || Structural analysis || style="text-align:center" | Disproven || style="text-align:center" | Steven

|- style="text-align:left" 
| style="color:white; background:#4682B4" | Vigenere || Frequency analysis || style="text-align:center" | Unlikely || style="text-align:center" | Steven
|-
|}

The table shows that the majority of ciphers analysed have been ruled out, with most of those that remain declared unlikely. Only one cipher was determined to be plausible; the Keyword Cipher which yielded a profile consistent with the Somerton Man code. Once again, for more extensive details, please refer to the [[Cipher Cross-off List]].