|
|
|
|
|
|
|
Switzerland Zellweger Hand
Polyalphabetic substitution cipher
- under construction
FG was a polyalphabetic
manual cipher device for tactical messages —
intended for use during a war at the front lines — manufactured around 1946 1
by Zellweger
in Uster (Switzerland) and used by the
Swiss Army
[2].
The device is based on the Jefferson disk,
that was invented in 1795 by Thomas Jefferson [4].
It is believed that FG is the abbreviation of Front-Chiffriergerät [2]
(Front Cipher Device) or Feld-Gerät (Field Device) [6].
In 1949, it was succeeded by the FG-46 system.
|
The device consists of a metal cylinder that is 170 mm long and has a
diameter of 38 mm. It weights approx. 325 grams and has two cut-away
windows through which 25 movable discs, mounted onto a common spindle,
are visible.
Each disc contains a (different) scrambled Latin alphabet,
and can be set to any of 26 positions. The discs are identified by a
single letter
of the alphabet that is engraved at its side (A-Z). 2 The discs are
mounted onto the spindle in groups of five, with blank metal discs
used as separators, allowing coding of five 5-letter groups at once.
|
|
|
The device has a knurled knob at either end. It can be
separated by
pulling the leftmost knop away from the cylinder. This allows the alphabet
discs to be removed from the spindle
and re-installed in any of
15,511,210,043,330,985,984,000,000 possible orders (25!) 3 . The order in
which the discs are installed from left-to-right, is known as the secret KEY.
Up to 25 letters of the message can be encoded or decoded in each pass,
making it a very time-consuming operation.
The FG-device is based on the so-called
Jefferson Disk, that was invented
in 1795 by the 3rd US President Thomas Jefferson. It was independently
inventend a century later by Étienne Bazeries, and became known as
the Bazeries Cylinder.
In the US, the device later evolved into the
M-94 (Navy: CSP-488)
that was used from 1922 to 1945 [5].
The FG is in fact an improved M-94.
It is remarkable that it was introduced in the Swiss Army by the time it
was abandonned by the US.
The unit shown here was developed and introduced around 1946. It is
believed that only a small qantity was manufactured and used [1].
The device was succeeded by a compatible but flattened version — known as
FG-46 (later renamed FG-59) — that
entered service in 1949 and was used until at least 1973 [3].
The FG devices were used alongside the
Swiss Enigma K
and the Swiss Nema
(also made by Zellweger).
They were discarded in 1978 and have since been declassified.
|
 |
-
According to Rudolf Ritter in [2], the FG-device was built in 1944,
but we believe that he errs, as the
company name
— Zellweger AG, Apparate- und Maschinenfabriken Uster — was
first used in 1946.
-
Only 25 letters of the alphabet are used for identification of
the discs. The letter 'I' is omitted.
-
25!, or 25 factorial, is the mathematical notation for
25 x 24 x 23 x 22 ... x 2 x 1. This means that there are more
than 15 septillion possible keys.
|
The diagram below provides an overview of the various features of the
FG-device. At the center is the metal cylinder with the two cut-out windows.
At the left is the spindle with the 25 cipher discs, one of which (Z)
has been removed here. Each group of five discs is separated by a metal
spacer. Once the order of the discs has been fixed, the
spindle with the discs should be inserted into the large open end of the
cylinder. The disc setting device is stored inside the hollow spindle.
When the device is not in use, it is put in the storage position 1
and stored inside the leather bag.
|
-
In the storage position, the discs are inserted in the order of the
alphabet (A-Z), with the letter 'A' of each disc lined up with the
lower edge of the window.
|
First, the device has to be set to the daily (secret) KEY that has been
distributed in advance. The KEY consists of a string of 25 unique letters
of the Latin alphabet (A-Z minus the letter 'I'). In the example below,
we have used the default storage position, which is the order
of the alphabet:
A B C D E F G H J K L M N O P Q R S T U V W X Y Z
The core is now removed from the cylinder,
after which the 25 alphabet discs are
removed from the spindle.
Next they are mounted to the spindle in the
order dictated by the daily key (shown above).
The core is then refitted inside the cylinder.
Assume we want to send the message:
ATTACK FRIDAY MORNING AT DAWN
We now rotate the discs so that the message appears at the lower edge of
one of the windows of the cylinder. This can be
done with the special tool
that is stored inside the core of the spindle, and can be
pulled-out
from the right of the device. It can also be done manually by first moving
all discs to the far right, then setting the leftmost discs and shifting
it to the left. Next, set the second disc and move it to the left.
Continue until the entire message is visible at the lower edge:
Now turn the device over by rotating it 180 degrees, so that the other
window is at the front, and read the ciphertext message from
the lower edge of that window. At the given KEY, this results in:
Write the ciphertext message out on a message sheet in 5-letter groups.
If the message is longer than 25 letters, repeat the above procedure until
the entire message is processed. Once the message is encrypted completely,
pass the message sheet to the radio or telegraph operator and send it
to the recipient. The method is reciprocal, so the recipient follows the
same procedure.
EJJRI WGVJM HCPBE YLYAC OGZIJ
In the above procedure, there is a fixed relationship between the
plaintext letter and the ciphertext letter of each disc.
They are always exactly 180° apart (i.e. opposite each other),
in order to make the method reciprocal. This is considered a weakness of
the system. A side effect of this principle is that a letter will never
be encoded into itself, just like on the
Enigma cipher machine.
|
The security of the cipher can be improved by rotating the spindle one
or more positions — using the knob at the left — before reading-off the
ciphertext, but that means that the rotational offset must be part
of the key, so that it can be corrected at the receiving end. If, in the
example above, the spindle is rotated clockwise by one position
(i.e. offset = 1), the ciphertext will become:
QWFND CXWZI SSJXH SAOIY HMFYE
At the receiving end, the operator sets up the above ciphertext at the
lower edge of the window and then turns the spindle counterclockwise
(in this case by 1 position), before reading-off the plaintext at the other
side of the cylinder. Theoretically, the above procedure can be improved
further, by rotating the spindle by one or more positions before writing
down each ciphertext letter, but as far as we know this was not part of the
operational procedure of the Swiss Army.
|
|
The FG device described here, offers only limited security and can be broken
with pencil-and-paper methods if sufficient traffic has been intercepted.
As the device was only used for tactical field messages — which will lose
their value after several hours — it was thought to provide reasonable
security though. The device further has the following disadvantages:
|
- Cumbersome
The device works fine for short messages, but encrypting and decrypting
long messages is very time consuming [2]. As the alphabet on each disc is in
a scrambled order, it takes time to find the required letter. This is a
typical property of mixed-alphabet systems.
- Message not cleared
Once a message has been coded and sent, the last 25 letters of the message
remain 'stored' in the device. If an enemy captures the device, he might
be able to read part of the message by trying all 26 positions of the spindle.
This problem was solved in the later FG-46 device,
by returning the alphabets to their initial position when the device
is closed.
- KEY not cleared
The most serious limitation however, is that the KEY (i.e. the order of
the discs) is left behind in the device between sessions. There is no method
to purge, or zeroize, the KEY. A captured device allows the enemy to
decipher all traffic that has been sent on it.
When the device was not used for an extended period of time, it was
advised to place the discs in the storage position (i.e. the order of
the alphabet).
- Fixed offset
In the original procedure, there was a fixed offset between the plaintext
and ciphertext letter of each disc. The two were at opposite sides, which
means that the offset was 13 positions. Although this makes the cipher
reciprocal, it introduces a weakness. With the American
M-94, on which the
FG is based, an arbitrary offset was used instead.
|
Some of the limitations were overcome in newer versions like
FG-46 and FG-59 that were
introduced a few years later.
Other problems could be solved simply by improving the operational procedures.
But some limitations remained, as they are inherent to this type of cipher.
|
|
The table below shows the order of the alphabet on each disc, starting
with the letter 'A' and following the disc clockwise when viewing the
spindle from the left side of the device.
|
|
Disc
|
Alphabet | Reverse
|
|
|
|
A
|
AUMCILFJXDHRQEPSVYOKNGZBWT
|
|
B
|
AGZKBNXIOWJCUFQLVRHEYSMTDP
|
|
C
|
AYGKHTEDQPZLUICMBFJWVNORXS
|
|
D
|
AKQLFODBJGZHNRCUWETXIMBPSY
|
|
E
|
ABTNJQPFMOYSCLEVHXRUGZWKDI
|
|
|
|
F
|
AMVZSUOGPXICWDTHYEQRFBNLJK
|
|
G
|
AWSBRNQZVIFKMJLOUTCPEHXGDY
|
|
H
|
AGHOUQCFRMKYBXJPZEINWVLSDT
|
|
J
|
ARZJEHCFYBPTDSXOIGQKUNMLWV
|
|
K
|
AWRUPEFGTLDCZQNJXHOBYKIMVS
|
|
|
|
L
|
APTIEXUOCKVFSHYNLWJQBDMRZG
|
|
M
|
AVQLFIZXKJHBYERTUWPOGDNMSC
|
|
N
|
AJPQULWGKVROBEHMFSIDCTZYXN
|
|
O
|
ALPNJMFHUTIZOYWQGSDRVCEKXB
|
|
P
|
ATOCNHEQDNFWCMJYUVSRKIPLGZ
|
|
|
|
Q
|
AJGSYRMTQXEDUPCHKNFLIOVWZB
|
|
R
|
ALWQETDMZRHGCJIPSXUFCONBKY
|
|
S
|
ADXLHCVZGNUEMQJTWPBRIOYSKF
|
|
T
|
ACNWLOZVQFJYEGBSXDMTRPHKUI
|
|
U
|
AEZORMJNDUPHYCWILBTVSFKGXQ
|
|
|
|
V
|
AJCZMPVNBTGUYWXIDQKFRHOESL
|
|
W
|
AVIHPLXQZNYMGTCKEBOFUJRSWD
|
|
X
|
AMBRCVDSJTWUFZGELHKXQINYPO
|
|
Y
|
AJEYIGFOPRZLCSMBDWVTQXNUKH
|
|
Z
|
AHEJGZLUVISCRQDBNPTOYMXKFW
|
Total number of combinations
|
|
25!
|
15,511,210,043,330,985,984,000,000
|
|
|
- Günter Hütter, Front-Chiffriergerät (FG)
Photographed with kind permission. August 2017.
- Rudolf J. Ritter, Das Fernmeldematerial der Scheizerischen Armee seit 1875, 10. Folge
Communications equipment of the Swiss Army since 1875, Volume 10 (German).
Bern (Switzerland), Autumn 2002. pp. 29-30.
- HAMFU, Front-Chiffriergerät 46 (FG 46)
Retrieved September 2017.
- Wikipedia, Jefferson disk
Retrieved August 2012.
- Wikipedia, M-94
Retrieved September 2017.
- Walter Schmid, Personal correspondence
October 2017.
|
|
|
|
Any links shown in red are currently unavailable.
If you like the information on this website, why not make a donation?
© Crypto Museum. Created: Friday 01 September 2017. Last changed: Monday, 09 August 2021 - 06:44 CET.
|
|
|
|
|
|
