The Enigma Cipher
mike3
Hi.
What was the weakness of the Enigma Cipher?
--- Begin EESCODE Signature ---
AeDANddvYhVJyKzGqlaxIOZjpmYaoxwSdSmI904vQybcDPK6AFGFGxGsICczlTbJ3AAA==
--- End EESCODE Signature ---
George Ou
news:1d54b7e4.04012...@posting.google.com...
Too many shared keys. Many sessions use a single key. The Germans were
suppose to change them more often, but many times they didn't. Another
example of the human link. If your crypto scheme is too hard for humans to
use, then that is a weakness in the crypto scheme.
By today's standards, it would probably be very easy to brute force a
4-wheeled Enigma session on any modern computer. They used massive spinning
wheels (called Bombes I think) during WWII to brute force Enigma. A secret
project (only second to the Manhattan project in budget) run by the NSA at
NCR had so much success with their Bombe that even the English started using
the American model. By late 1943, the German's 4-wheeled Enigma code was
all but worthless. U-boats were being attacked the minute they left their
protective docks. The American Bombe along with the advent of Air craft
carrying radar and U-boat hunting packs of ships meant your life expectancy
on a German U-boat in 1944 was about a little over one mission.
George
John Savard
part:
>What was the weakness of the Enigma Cipher?
1) No letter could represent itself. So it was easy to align probable
plaintext with messages.
2) The fast moving rotors were on the outside of the machine rather
than on the inside. This meant that their regularly-changing results
related to a constant alphabet.
3) For a period of time, the method used to encrypt the starting
positions of the rotors helped to give away the secret key - which
rotors where used in which position.
My web page provides much more detail on the technical questions
associated with the Enigma rotor machine.
John Savard
http://home.ecn.ab.ca/~jsavard/index.html
Joe Peschel
news:nemPb.13915$DJ7....@newssvr25.news.prodigy.com:
> Too many shared keys. Many sessions use a single key. The Germans
> were suppose to change them more often, but many times they didn't.
> Another example of the human link. If your crypto scheme is too hard
> for humans to use, then that is a weakness in the crypto scheme.
>
See John's response and his web page to learn about fundamental weaknesses
of the Enigma.
> By today's standards, it would probably be very easy to brute force a
> 4-wheeled Enigma session on any modern computer.
No, it isn't.
They used massive
> spinning wheels (called Bombes I think) during WWII to brute force
> Enigma.
Bombes did not decrypt Enigma by brute-force.
>A secret project (only second to the Manhattan project in
> budget) run by the NSA
The NSA din't exist in 1943.
>at NCR had so much success with their Bombe
> that even the English started using the American model. By late 1943,
> the German's 4-wheeled Enigma code was all but worthless. U-boats
> were being attacked the minute they left their protective docks. The
> American Bombe along with the advent of Air craft carrying radar and
> U-boat hunting packs of ships meant your life expectancy on a German
> U-boat in 1944 was about a little over one mission.
>
I think it would be helpful for you to read Kahn's books and Weierud's web
site.
J
--
__________________________________________
When will Bush come to his senses?
Joe Peschel
D.O.E. SysWorks
http://members.aol.com/jpeschel/index.htm
__________________________________________
Douglas A. Gwyn
> What was the weakness of the Enigma Cipher?
That many people using it didn't believe it could be broken.
If you mean, how was it broken, there are entire books on the
subject. You might also try Web sites such as
http://www.bletchleypark.org.uk/
David Eather
A bit of a re-write of history going on in your post. The US built enigma
"bombs" for the British because they had the industrial capacity. The
design was British. The British did almost all the "cracking" of enigma
messages of all types (all of this is well documented). It is not hard to
work out that the logistics of cracking a message in the US and then
transmitting it securely to Britain or the Atlantic in a secure and time
sensitive way would be a nightmare and endanger the war's second (or third,
fourth, or fifth) biggest secret. US ("only" as opposed to "joint") crypto
efforts were mainly concentrated in the pacific with good success.
As historical interest (also well documented) - At one time the British
seriously considered having to cut the US off from the naval enigma decrypts
for just the reason you say - the US attacked the U-boats almost as soon as
they left port. Of particular concern were *sequential* attacks on a class
of re-supply U-boat ordered to rendezvous at various co-ordinates in the
middle of the ocean - these attacks were done without any deception and the
sequential nature of the attacks threatened to give away the "ultra" secret.
Secrets must always be guarded by a bodyguard of lies. Read widely, question
everything. Always keep in mind every country will put there own spin on
events - and some more than others.
David Eather
"George Ou" <2038geor...@2342netzero.com2897> wrote in message
news:nemPb.13915$DJ7....@newssvr25.news.prodigy.com...
Mok-Kong Shen
David Eather wrote:
>
[snip]
> Secrets must always be guarded by a bodyguard of lies. Read widely, question
> everything. Always keep in mind every country will put there own spin on
> events - and some more than others.
That's important for history teachers to put into
the minds of the pupils, I believe.
M. K. Shen
CryptWolf
"Joe Peschel" <jpes...@no.spam.org> wrote in message
>news:Xns9476E96EEA429...@216.168.3.44...
> "George Ou" <2038geor...@2342netzero.com2897> wrote in
> news:nemPb.13915$DJ7....@newssvr25.news.prodigy.com:
>
> > Too many shared keys. Many sessions use a single key. The Germans
> > were suppose to change them more often, but many times they didn't.
> > Another example of the human link. If your crypto scheme is too hard
> > for humans to use, then that is a weakness in the crypto scheme.
> >
>
> See John's response and his web page to learn about fundamental weaknesses
> of the Enigma.
>
> > By today's standards, it would probably be very easy to brute force a
> > 4-wheeled Enigma session on any modern computer.
>
> No, it isn't.
Actually, easier and faster than you think.
> They used massive
> > spinning wheels (called Bombes I think) during WWII to brute force
> > Enigma.
>
> Bombes did not decrypt Enigma by brute-force.
Technically they only brute force searched for indicators.
These were then tested as possible keys. A Bombe was quite
capable of searching the entire key space though it took close
to a day if left to run that long.
The same search on a modern computer would be a matter of
seconds though perhaps not technically a brute force search.
The entire keyspace would be searched for possible decrypt
keys. A 3 rotor Engima has a fairly small key space (26^3)
ignoring the plugboard. Some settings modify the cipher less
than might be expected.
CryptWolf
Paul Rubin
> A 3 rotor Engima has a fairly small key space (26^3)
> ignoring the plugboard.
What about the rotor wiring?
Joe Peschel
news:35fcc94d85bf3c0e...@news.teranews.com:
>
> "Joe Peschel" <jpes...@no.spam.org> wrote in message
>>news:Xns9476E96EEA429...@216.168.3.44...
>>
>> > By today's standards, it would probably be very easy to brute force
>> > a 4-wheeled Enigma session on any modern computer.
>>
>> No, it isn't.
>
> Actually, easier and faster than you think.
>
It is? Ok, how easy and fast do I think it is? :-)
d...@florence.edu
Although the German enigma cipher could be broken in a fairly short
time period it really only was designed to be a field cipher
wasn't it?
Radio transmissions could be jammed and spoofed so the primary purpose
of the enigma must have been to authenticate messages. If a German
radio operator received an encrypted message with the correct day key
it was almost a sure thing that they were communicating with the
intended party rather than a impersonator.
Douglas A. Gwyn
> Although the German enigma cipher could be broken in a fairly short
> time period it really only was designed to be a field cipher
> wasn't it?
No, its manufacturer advertised it as secure for any purpose.
> Radio transmissions could be jammed and spoofed so the primary purpose
> of the enigma must have been to authenticate messages. If a German
> radio operator received an encrypted message with the correct day key
> it was almost a sure thing that they were communicating with the
> intended party rather than a impersonator.
No.
John Savard
<DAG...@null.net> wrote, in part:
>d...@Florence.edu wrote:
>> Radio transmissions could be jammed and spoofed so the primary purpose
>> of the enigma must have been to authenticate messages. If a German
>> radio operator received an encrypted message with the correct day key
>> it was almost a sure thing that they were communicating with the
>> intended party rather than a impersonator.
>No.
No to sentence 1.
Yes to sentence 2.
To compose and transmit a forged message in the Enigma cipher would
immediately compromise the fact that the British had cracked the
cipher. The probability that the benefits of transmitting a single
message, to temporarily deceive one German unit, would exceed the
future benefits of all future decryptions of Engima messages would be,
I assert, rather low.
John Savard
http://home.ecn.ab.ca/~jsavard/index.html
Jim Gillogly
> "Joe Peschel" <jpes...@no.spam.org> wrote in message
>>news:Xns9476E96EEA429...@216.168.3.44...
>> "George Ou" <2038geor...@2342netzero.com2897> wrote in
>> > By today's standards, it would probably be very easy to brute force a
>> > 4-wheeled Enigma session on any modern computer.
>>
>> No, it isn't.
>
> Actually, easier and faster than you think.
I'm with Joe. There's too much there for brute force. If you'd like
to prove differently, I'd love to see it. Here are three probable
M4 ciphers that so far as I know have never been broken. I've tried
them with no luck (so far).
http://members.fortunecity.com/jpeschel/erskin.htm
Please show your work!
--
Jim Gillogly
Joe Peschel
news:pan.2004.01.22....@acm.org:
> I'm with Joe. There's too much there for brute force. If you'd like
> to prove differently, I'd love to see it.
Me, too. Of the huge theoretical number (about 2x10^145) of possible 4-
rotor Enigma configurations, apparently allied cryptanalysts faced a
cryptovariable space of about 10^23, a damn big number.
d...@florence.edu
Later in the war the germans changed the physical design of the enigma
slightly adding a 5th rotor for the naval machine. Along with the
mechanical changes there was a small change in operator procedure.
Although these changes would not give much added resistance to
cracking they would clearly identify messages as received from the
german submarines rather then their land forces. Also I suppose
adding a rotor would keep any captured 4 rotor enigma from being used
to send an impersonated message.
The largest enigmas with 12 rotors were used on the teletype machines
of the german high command. the complexity of rhese machines does
seem a bit excessive for just segregating message traffic but the
Germans are a tidy sort aren't they?
Certainly would create some revisionist history if it could be proven
that the enigma was not intended to protect messages long term
wouldn't it?
Michael Amling
>
> What was the weakness of the Enigma Cipher?
>
There were two main weaknesses: Insufficient key space, and separable
key space.
The three rotor enigma (commercial version) has 26**3=17576~~2**14
possible keys. That's too few. It allows a brute force attack.
The three-of-five rotor enigma has 5*4*3*(26**3)=1054560~~2**20
possible keys, which is still too few.
The three-of-eight plus one rotor (the Naval four-rotor Enigma) has
8*7*6*(26**4)=153543936~~2**27 possible keys, which is getting up there.
But, incredibly, the rotor settings for the four-rotor enigma put the
three ordinary rotors into the same configuration as that day's
three-rotor settings. That allowed the problem of determining the
4-rotor key to be separated into two much easier problems: Finding the
three-rotor settings, and determining the setting of the fourth rotor
with the first three known.
Using ten steckers multiplies the number of keys by
26!/(10!*6!*2**10)=150738274937250~~2**47, which would have precluded
brute force (in the 1940s), if it weren't for the fact that the key
space is separable. With a suitable plaintext/ciphertext pair, the
bombes could conduct a search of the rotor key space, temporarily
ignoring the steckers, and with the rotor configuration known, the
steckers could be solved for as a monoalphabetic substitution.
SIGABA had, by contrast, 10!*2**10=111476736000~~2**36 possible keys
(not counting the index rotors, which were used to distinguish classes
of messages), which, combined with the fact that the key space is AFAIK
not separable, was enough to prevent brute force from succeeding until
computers caught up in the late 1950s.
SIGABA also did not have the Enigma property, helpful to the
cryptanalysts, that no letter encrypts to itself. I expect that SIGABA
keys were not selected, as Enigma keys were, with the arbitrary rule
that no rotor could be used in the same position on two consecutive days.
--Mike Amling
Michael Amling
The rotor wiring was not part of the key. Hence by Kerckhoffs's
principle, Enigma users should have (correctly) assumed that the wiring
would not be kept secret.
--Mike Amling
Joe Peschel
news:FgSPb.3305$BA2....@newssvr26.news.prodigy.com:
> mike3 wrote:
>>
>> What was the weakness of the Enigma Cipher?
>>
>
> There were two main weaknesses: Insufficient key space,
The theoretical cryptovariable space was huge. See my other post.
John Savard
>Certainly would create some revisionist history if it could be proven
>that the enigma was not intended to protect messages long term
>wouldn't it?
The Enigma was definitely intended to protect messages from being
read, at least in the short term. It was one of the key elements of
the Blitzkrieg strategy: allowing units to coordinate attacks that are
unpredictable to the enemy.
Authentication is a side-effect of most traditional forms of
encryption, even if today's XOR stream ciphers don't offer it. I don't
think it makes sense to claim that secrecy was not the primary goal of
Enigma encryption; but that authentication, as a benefit, was also
appreciated is true enough.
The Enigma was designed to keep messages as secure as was reasonably
practical, and improvements were continuously being made, despite the
fact that the Germans believed it was not broken. Two systems of
teletypewriter encryption were used for the highest security messages,
the Schlusselzusatz (Tunny) and the Geheimscrhreiber (Sturgeon), and
the latter was not a priority of the British because the messages sent
over it were also sent by Enigma.
So I don't think the historical evidence agrees with the idea that the
Enigma was regarded as a low-grade secrecy system either, let alone
the idea that it was primarily intended for authentication.
John Savard
http://home.ecn.ab.ca/~jsavard/index.html
d...@florence.edu
I guess it was just those grade school lockers they made me keep my
history books in that gave me the idea it was a low-grade security
system..
CryptWolf
"Paul Rubin" <http://phr...@NOSPAM.invalid> wrote in message
>news:7xzncgl...@ruckus.brouhaha.com...
This answers several questions I hope...
The rotor wiring was not part of the key. The Enigma operator
selected the rotors from a key list for that day just like the
rest of the system was keyed from a key list. Inflating the
statistics to include the rotor wiring is misleading.
I generally ignore the plugboard because the Germans didn't
distribute the latest version of the machine to all operators.
In other words, to maintain compatibility, many messages
were encrypted without taking advantage of the plugboard.
There was a similar problem with the 4 rotor Naval version.
It also had a compatibility mode that allowed it to work with
the regular 3 rotor versions.
For a 3 rotor Enigma and using all the possible rotors used
during WW2, You can brute force search the entire key
space in about 2 minutes on a 486-66. Keep in mind that
this search includes the few Naval rotors which would
not have been available with the regular 3 rotor version.
It actually takes about 2 minutes to search a full Naval
version without changing rotors.
While searching, you can safely ignore the ring positions.
The alphabet rings could be moved but this doesn't affect
the encryption a bit except where the rings affected the
step position. Then you only get a partial decrypt and would
have to manually search for the correct step position.
A brute force search could fail if the step position falls within
the text being tested. Something BP didn't need to worry
about as much because probables were manually tested
anyway.
And before anyone makes any bets... Some around here
may even have a copy of Reynard, Secret Code Breaker II.
I wrote the "Bombe" program that was included with the book.
An alpha version was capable of the numbers stated here though
it was slowed for use with the book. Also, the program with
the book used a 50 letter text sample for crib and statistical
searches. Using more modern hardware, I could extend
the attack to even search step positions and still get some
reasonable brute force search times. Coupled with one of the
better dictionary substitution cipher solvers, I think even the
plugboard would only be a minor matter of a few extra seconds.
So, yes, if we had a time machine, a modern box would
munch WW2 Enigma messages at an unbelievable rate.
CryptWolf
Paul Rubin
>
> For a 3 rotor Enigma and using all the possible rotors used
> during WW2, You can brute force search the entire key
> space in about 2 minutes on a 486-66. Keep in mind that
> this search includes the few Naval rotors which would
> not have been available with the regular 3 rotor version.
> It actually takes about 2 minutes to search a full Naval
> version without changing rotors.
I thought the naval Enigmas (at least some of them) had 4 rotors.
Also, getting hold of the rotors was the subject of a number of
big British operations (Kahn, Seizing the Enigma). I think you have
to consider the rotor wiring to be part of the key, just not a part
that changed daily.
> So, yes, if we had a time machine, a modern box would
> munch WW2 Enigma messages at an unbelievable rate.
But there are still some unsolved Enigma messages out there. How can
that be, if they're so easy to solve?
David Eather
errors) - it was intended and used as a high grade security system - a
fortunate error for the allied powers.
<d...@Florence.edu> wrote in message
news:6g9010dlrjs15b9iu...@4ax.com...
Joe Peschel
news:4010...@dnews.tpgi.com.au:
> enigma was a low grade security system (due to small defects and
> operational errors) - it was intended and used as a high grade
> security system - a fortunate error for the allied powers.
>
No, it wasn't a low-grade security system. In fact, some Enigma messages
remain encrypted.
Joe Peschel
news:c942e1b0e1133be9...@news.teranews.com:
>
> "Paul Rubin" <http://phr...@NOSPAM.invalid> wrote in message
>>news:7xzncgl...@ruckus.brouhaha.com...
>> "CryptWolf" <RWilliam...@no.spam.sceinet.no.spam.com> writes:
>> > A 3 rotor Engima has a fairly small key space (26^3)
>> > ignoring the plugboard.
>>
>> What about the rotor wiring?
>
> This answers several questions I hope...
>
> The rotor wiring was not part of the key. The Enigma operator
> selected the rotors from a key list for that day just like the
> rest of the system was keyed from a key list. Inflating the
> statistics to include the rotor wiring is misleading.
>
> I generally ignore the plugboard because the Germans didn't
> distribute the latest version of the machine to all operators.
> In other words, to maintain compatibility, many messages
> were encrypted without taking advantage of the plugboard.
> There was a similar problem with the 4 rotor Naval version.
> It also had a compatibility mode that allowed it to work with
> the regular 3 rotor versions.
>
Ignoring the plugboard is a mistake if you are trying to realistically
portray the cryptovariable space of the Enigma.
...snipped.
>
> And before anyone makes any bets... Some around here
> may even have a copy of Reynard, Secret Code Breaker II.
As Jim mentioned, there are three unsolved Enigma messages on my web site.
Good luck solving them.
> I wrote the "Bombe" program that was included with the book.
> An alpha version was capable of the numbers stated here though
> it was slowed for use with the book. Also, the program with
> the book used a 50 letter text sample for crib and statistical
> searches. Using more modern hardware, I could extend
> the attack to even search step positions and still get some
> reasonable brute force search times. Coupled with one of the
> better dictionary substitution cipher solvers, I think even the
> plugboard would only be a minor matter of a few extra seconds.
>
I don't think so. Show us otherwise.
> So, yes, if we had a time machine, a modern box would
> munch WW2 Enigma messages at an unbelievable rate.
>
Nope.
See my post and Miller's tome on the math of the Enigma.
Dan Girard
I've been trying to solve these for some time, using the
crib-and-bombe method; but haven't come up with the right crib yet.
By the way, there's another M4 message, with a good crib, which I got
from Ralph Erskine. I posted it here, with his permission, on
12-03-2003 ( Message-ID:
<7f5e8257.03120...@posting.google.com> ). Has anyone tried
to break it?
Dan Girard
Joe Peschel
news:7f5e8257.04012...@posting.google.com:
That's the one you solved with a bombe simulator, right?
I haven't tried it. Maybe Jim has.
I wonder why we don't add that ciphertext message, and while we're at it,
the three other Enigma messages to the Sandbox.
Richard Heathfield
> I wonder why we don't add that ciphertext message, and while we're at it,
> the three other Enigma messages to the Sandbox.
Perhaps we don't do it because it might win. :-)
(Actually, it's quite a cool idea.)
--
Richard Heathfield : bin...@eton.powernet.co.uk
"Usenet is a strange place." - Dennis M Ritchie, 29 July 1999.
C FAQ: http://www.eskimo.com/~scs/C-faq/top.html
K&R answers, C books, etc: http://users.powernet.co.uk/eton
Douglas A. Gwyn
> Although these changes would not give much added resistance to
> cracking they would clearly identify messages as received from the
> german submarines rather then their land forces. Also I suppose
> adding a rotor would keep any captured 4 rotor enigma from being used
> to send an impersonated message.
I don't know why you keep making (bad) guesses about this
when there are reputable histories ytou could read.
In fact, addition of a rotor posed a significant problem
for the codebreakers. And there was no special concern
about the enemy capturing Enigmas and using them to send
inauthentic traffic.
> The largest enigmas with 12 rotors were used on the teletype machines
> of the german high command.
No Enigma machine had 12 rotors. Perhaps you're referring
to the Geheimschreiber, which was based on a different
principle.
> the complexity of rhese machines does
> seem a bit excessive for just segregating message traffic but the
> Germans are a tidy sort aren't they?
I think you're continuing to be misled by the wrong idea
that somehow them main purpose was authentication. It
wasn't. The Geheimschreiber was used instead of Enigma
because it was believed to be more secure, and since it
didn't have to be field-portable the size and weight of
the equipment used for the more complex system wasn't a
problem.
> Certainly would create some revisionist history if it could be proven
> that the enigma was not intended to protect messages long term
> wouldn't it?
If pigs could fly then certainly some books would want
rewriting. So what? Pigs don't fly.
As with any cryptosystem, Enigma messages were meant to
be kept secret "long enough" that their information would
lose its value to the interceptor by the time that they
were cracked. The Germans who selected Enigma had the
(mistaken) impression that they could not be cracked with
any technology available at the time. They were *almost*
right about that, since cracking the Enigma took a heroic
effort.
d...@florence.edu
wrote:
Some of the messages from weather ships may be difficult to develop
cribs for. Useful WX info is typically choppy short numeric data.
Might be fun to try the near-far approach for the toughest examples.
Off the top of my head I can't think of a way to do that quickly on a
desktop computer though.
David Hamer
> Later in the war the germans changed the physical design of the enigma
> slightly adding a 5th rotor for the naval machine. Along with the
> mechanical changes there was a small change in operator procedure.
No Enigma had five wheels....moreover the only 'true'
four wheel variant was the one used by Abwehr - Enigma
'G' [aka 11-15-17] - in which all four wheels moved
during encryption/decryption. In the case of the navy
M4 this 'additional wheel' [Zusatzwalze] was a component
of a variable reflector. The Commercial Enigma - and all
of its variants also had a 'fourth wheel' that was in
fact a 'settable' reflector. In neither of these last
examples did the fourth wheel move during encryption or
decryption nor was it interchangeable with the regular
Enigma wheels.
David Hamer
dha...@cryptogram.org
Michael Amling
Is the wiring of I-VIII, "Beta, "Gamma", and two thin reflectors, "B"
and "C" posted somewhere?
--Mike Amling
David Hamer
> I generally ignore the plugboard because the Germans didn't
> distribute the latest version of the machine to all operators.
> In other words, to maintain compatibility, many messages
> were encrypted without taking advantage of the plugboard.
> There was a similar problem with the 4 rotor Naval version.
> It also had a compatibility mode that allowed it to work with
> the regular 3 rotor versions.
The first part of this is rubbish if you think about
it..!
Not using the plugboard in a 3-wheel Army/GAF Enigma
or with a naval M4 does not create compatibility with
a non-plugboard machine [e.g. one of the many variants
of Commercial Enigma K]...there is the small matter
of the entirely different wheel wirings/notches/etc...
Certainly the M4 could be set to be compatible with
the 3-wheel navy variant - M3 [the 'fourth' wheel at
position 'A']. This was of course a necessary compromise
since even after 1 February 1942 when M4 was introduced
to the Atlantic U-boat fleet a lot of naval facilities
and vessels continued to use the three-wheel variant.
David Hamer
dha...@cryptogram.org
David Hamer
Yes - <http://www.eclipse.net/~dhamer/Enigma1.htm>
...there is a link at the bottom of the page
'Enigma - download internal wiring information'
David Hamer
dha...@bletchleypark.org.uk
CryptWolf
I'll see if I can write this up in a paper or something that looks like
a paper. Some of my references aren't all that much crypto related.
I'll need them to provide some sort of time line and be sure I'm
remembering all this stuff right.
When you think about it, the WW2 Bombe was capable of searching
the entire key space in about a day depending on which source you use.
This was for a mechanical machine. How fast can a modern PC do
the same job? Now just make it a bit smarter. ;)
CryptWolf
Dan Girard
>
> Some of the messages from weather ships may be difficult to develop
> cribs for. Useful WX info is typically choppy short numeric data.
>
I was under the impression that weather reports were first encoded
with the Short Weather Cipher before being superenciphered with the
Enigma. These messages don't seem to have that format.
I've been operating on the theory, based on where and when they were
intercepted, that they most likely were sent by U-boats, of a
particular wolfpack (the only one operating in the area at that time).
> Might be fun to try the near-far approach for the toughest examples.
> Off the top of my head I can't think of a way to do that quickly on a
> desktop computer though.
I'm not familiar with the "near-far approach". What is it?
Dan Girard
d...@florence.edu
wrote:
>d...@Florence.edu wrote in message news:<2gj2105uvkuosgrs5...@4ax.com>...
>>
>> Some of the messages from weather ships may be difficult to develop
>> cribs for. Useful WX info is typically choppy short numeric data.
>>
>
>I was under the impression that weather reports were first encoded
>with the Short Weather Cipher before being superenciphered with the
>Enigma. These messages don't seem to have that format.
If the original plaintext were a grammatically correct verbose
transcript it should have a strong statistical signature. As you
pointed out a SWC with numerics should also have a signature. For
example although the numeric data might be random weather data the SWC
would use a subset of the keys to encode the numerics. Of course this
does not preclude some bored Nazi radio operators from going against
SOP. If the plaintext were whitened and encrypted several times with
different keys the signal to noise ratio of the original plaintext
might be too low too detect with the technology available in WW2.
>
>I've been operating on the theory, based on where and when they were
>intercepted, that they most likely were sent by U-boats, of a
>particular wolfpack (the only one operating in the area at that time).
>
Partial radio intercepts were common in WW2, the radio signals of
U-boats 1000's of miles away from the receiver sometimes would vary
with the skip. Perhaps some of the message was lost, useful uniform
formatting identifiers etc.
>> Might be fun to try the near-far approach for the toughest examples.
>> Off the top of my head I can't think of a way to do that quickly on a
>> desktop computer though.
>
>I'm not familiar with the "near-far approach". What is it?
>
As Doug Gwyn pointed out in another post there are many books that
explain the cracking of the enigma. I briefly read some accounts of
the goings on at Bletchley Park in particular the code breaking
machines located near Faulkner green.
From what I could gather the "Heath Robinson" code breaking machine
was a specialized code breaking machine tuned to the enigma hardware
parameters. Two closed loop paper tapes were used one named "near" and
t'other named "far".
The ciphertext to be cracked was punch positioned onto the near tape
and a fixed pattern was punched onto the far tape. Both tapes were run
together at high speed. The near tape was run a little slower than
the far tape stepping through the possible key combinations whith the
pattern of the far tape. Since the far tape was also in motion you
were testing combinations in parallel speeding up the cracking
considerably.
Apparently when the near tape statistics changed slightly from those
of the pseudo random far tape it signaled the proper key had been
found.
If a cracker today were to try and build a massively parallel code
cracking machine they might be avle to do better than to faithfully
reproduce a dual paper tape machine like the Heath Robinson. Perhaps
you could build something out of more modern parts such as surplus
AOM's and polarizing materials? Since these uncracked Enigma
ciphertexts got past the Bletchley Park crackers you would need a more
sensitive machine anyways.
>Dan Girard
David Hamer
> As Doug Gwyn pointed out in another post there are many books that
> explain the cracking of the enigma. I briefly read some accounts of
> the goings on at Bletchley Park in particular the code breaking
> machines located near Faulkner green.
The operative word appears to be 'briefly'...
> From what I could gather the "Heath Robinson" code breaking machine
> was a specialized code breaking machine tuned to the enigma hardware
> parameters. Two closed loop paper tapes were used one named "near" and
> t'other named "far".
Heath Robinson [a reconstruction is underway courtesy Tony Sale]
was the precursor to Colossus I and Colossus II and it looks as
if it has to be said yet again...these machines were not used to
attack Enigma. Colossus was used against the teleprinter machines
[codenamed FISH] - primarily the Lorenz SZ40/42 [Schluesselzusatz],
codenamed TUNNY by Bletchley Park.
Currently at BP there are on display examples of the SZ40/42 and
another 'FISH' machine STURGEON [Siemens T52e Geheimschreiber]
- together with, of course, Tony Sale's famous working rebuild
of Colossus and the 'in-progress' rebuild of the TUNNY analog.
David Hamer
dha...@bletchleypark.org.uk
wms
purchased the Enigma-E that they offer on their web site store? (99
Pounds Sterling, about $140 or so.) Per the paragraph below, it is
apparently compatible with the M3 and M4. Is it available anyplace on
this side of the Atlantic?
The Enigma-E is a self-build kit that enables you to build your own
electronic variant of the famous Enigma coding machine that was used by
the German army during WWII. It works just like a real Enigma and is
compatible with an M3 and M4 Enigma as well as the standard service
machines. A message encrypted on, say, a real Enigma M4 can be read on
the Enigma-E and vice versa.
Thanks,
Dave
"Douglas A. Gwyn" wrote:
>
> mike3 wrote:
> > What was the weakness of the Enigma Cipher?
>
> That many people using it didn't believe it could be broken.
>
> If you mean, how was it broken, there are entire books on the
> subject. You might also try Web sites such as
> http://www.bletchleypark.org.uk/
Marc Heusser
> Speaking of the crew at "bletchleypark.org", has anyone looked into or
> purchased the Enigma-E that they offer on their web site store? (99
> Pounds Sterling, about $140 or so.) Per the paragraph below, it is
> apparently compatible with the M3 and M4. Is it available anyplace on
> this side of the Atlantic?
Isn't it possible to get one from there sent over the pond?
I get parts across the other way round quite frequently.
BTW: There's software simulators on http://www.xat.nl/enigma/
(also info on the hardware one)
HTH
Marc
--
Marc Heusser
(remove the obvious: CHEERS and MERICAL...until end to reply via email)
Douglas A. Gwyn
> Isn't it possible to get one from there sent over the pond?
Yes, in fact the Web site is set up to make that easy.
jamotto
plugboard/steckerboard, four rotating wheels(the scrambler) and the
stationary reflector. However, to get the slowest rotor to turn
1/26th of a revolution if my math is right you would have to type 26^3
letters.
David Hamer
Two misconceptions here:
The fourth wheel or Zusatzwalze [ZW] of M4 does not move during
encryption/decryption - it cannot...it has no turnover notch.
Furthermore M4 has the same three stepping levers found in the
3-wheel machines. These act only upon the three wheels to the
right.
There are two variants of Zusatzwalze [ZW] - Beta and Gamma -
and these are not interchangeable with the regular wheels
[I-VIII]. There are two corresponding 'thin' reflectors - B
and C - each of which, when combined with the ZW, forms a
26-position, settable reflector. Each pair can be set to a
'null' position at which the combination [Beta/B or Gamma/C]
emulates either a regular B or C reflector.
For details see the paper by Ralph Erskine and Frode Weierud:
"Naval Enigma: M4 and its Rotors", Cryptologia XI(4), 1987.
The second misconception is that it takes 26^3 [17576]
keystrokes to complete a full cycle of a 3-wheel Enigma with
a single turnover notch per wheel. In fact this number is
16900 due to an anomaly which results in the 'double stepping'
of the middle wheel. See my paper: "Enigma: Actions Involved
in the 'Double Stepping' of the Middle Rotor", Cryptologia
XXI(1), 1997. If two notch-wheels [VI-VIII] are used in the
'middle' and 'fast'positions the cycle length reduces to 4056.
Cycle length C = 26.(26/m)-1.26/f where m and f are the number
of notches on the middle and 'fast' [RH] wheels respectively.
The only true four-wheel Enigma was the one used by Abwehr -
the 11-15-17 machine - which used a stepping mechanism based
upon geared wheels rather than the levers and ratchets found
in other Enigmas. See my paper: "G-312: An Abwehr Enigma",
Cryptologia XXIV(1), 2000.
--
David Hamer
dha...@bletchleypark.org.uk
jamotto
David Hamer
I erred slightly in my comments yesterday when I said [about
the movement of the Zusatzwalze] "...it cannot...it has no
turnover notch." What I intended to say was the the ZW cannot
turn during encipherment/decipherment since has no ratchet and
no corresponding stepping lever. Of course it does not have a
notch either but that is of no consequence in these discussions.
FYI some [but not all] of my articles on Enigma may be found at:
<http://www.eclipse.net/~dhamer/download.htm> but for the others
you'll have to seek out the appropriate journal(s).