This page focuses on some of the cipher machines that
featured on Focus
on Photo. Each photo is
accompanied by a short description of the machine and web
links direct the reader to further information. You can
click the images to enlarge them.
All images are
copyrighted, and the owners of the images preserve all
rights. No use without explicit permission of the owner!
KWR-37 Fleet
Broadcast Receiver
The KWR-37
"JASON" was the receiver part of the
KW-37 crypto system, developed by the NSA In the
1950s. The systems consisted of a KWT-37
transmitter and a KWR-37 receiver. It was used to
secure fleet broadcasts of the US Navy. The shore
station transmitted 24 hours a day a continuous
stream of encrypted random data. If a message had
to be sent to one of the ships, the encrypted
message was inserted into the continuous stream.
An enemy eavesdropper could not detect if or how
many message were sent, when they started or
ended, or how long they were. Therefore, the
KW-37 system made traffic analysis by the
adversary impossible.
The
output of the receiver was connected to a
teletype machine that immediately printed the
decrypted stream onto paper. Transmitter and
receiver maintained synchronization whole day. If
synchronization was lost, the receiving operator
could re-establish synchro by re-setting and
running very fast through all past key stream
until the KWR-37 could pick up the current stream
again.
The KWR-37 was a marvel of
miniaturization in the 1950s. It contained
approximately 500 miniature vacuum tubes for a
large number of flip-flops and logic functions,
required for the shift registers that generated
the pseudo-random enciphering stream. The key of
the KWR-37 was an IBM style punch card that was
changed daily, just before the new synchro at
midnight. The machine remained in service until
the early 1990s.
The Russian FIALKA
M-125 is the most famous Soviet cipher machine of
the Cold War era. In 1965, the Cyrillic version
was introduced in the Soviet Army, and later on,
versions with Czech, Polish and (East) German
keyboard layout entered service in the Warsaw
Pact. The FIALKA was operational until at least
the 1990s and according to some sources even
still in use today. Therefore, the FIALKA and its
specifications were top secret until the late
1990s. The image shows the M-125-3MP2 version
with Czech keyboard.
Although
based on the WW2 German Enigma machine, Russian
cryptologists were well aware of the security
flaws of the Enigma and incorporated solutions to
all of those flaws into this wonderful piece of
mechanics. The small machine, only 10.8 by 12.5
inches and 8.3 inches high, is with its
impressive mechanism the Swiss watch of the
electro-mechanical cipher machines. The Fialka
has 10 interchangeable, alternately
counter-rotating rotors with 30 wirings each.
Pins on each rotor mechanically control the
irregular and most complex stepping of the
rotors.
The removable internal wirings
core of each rotor can be rotated, extracted and
mirrored, or exchanged with other wirings.
Several different types of rotor sets were
produced. The plugboard, as used on the Enigma,
is replaced by a punched card reader, and an
electronic 3-point circuit in the reflector
solves Enigma's flaw that a letter can never be
encrypted into itself. The output is printed on a
paper ribbon or punched on a five-bit paper tape,
and the machine is also equipped with a paper
tape reader.
The ETCRRM (Electronic
Teleprinter Cryptographic Regenerative Repeater
Mixer) from the Norwegian company STK (Standard
Telefon og Kabelfabrik A/S) applied one-time
encryption on a standard commercial teleprinter.
The ETCRRM mixed the five-bit Telex output signal
with a one-time tape by XOR-ing both signals. An
identical setup on the receiver's end, with an
identical key tape stream, reversed the
enciphering process. The one-time tapes or key
tapes were truly random five-bit values.
As long as there are only two copies of
a one-time tape, these tapes are used only once
and they are destroyed after use, the messages
that are sent by this system are mathematically
unbreakable. The ETCRRM used tubes, relays and
diodes to implement the logic XOR functions. A
solenoid and ratchet system, the only mechanical
parts in the machine, advanced the key tape.
The ETCRRM was used for high
level military communications in several
countries. ETCRRM's were also used on the
Washington/Moscow Hotline. Two were installed in
Washington and two in Moscow. Although a system
with absolute security, the unclassified standard
teleprinters and ETCRRM's were sold by commercial
firms and therefore did not disclose any secret
crypto technology to the Soviets. Only the key
tapes were considered secret crypto material. The
Hotline was a full duplex teleprinter circuit,
and not a so called 'Red Phone', as it was
believed that spontaneous verbal communications
could lead to miscommunications and
misperceptions.
The German Siemens
& Halske T-52 Geheimschreiber was an online
teletype cipher machine for high-level strategic
Luftwaffe (German Airforce) messages in the
Second World War. It was one of the most secure
German cipher machines. The early T-52a and T-52b
versions were less secure. Their traffic was
broken by Swedish cryptologists in 1940 and these
messages were read throughout the war. The
British codebreakers discovered the use of the
T-52 in 1942 and codenamed its message traffic
STURGEON.
Although they
succeeded in breaking a small part of STURGEON,
they did not achieve the same success as they had
with the Enigma and Lorenz machines. The T-52
enciphering was by far the most complex one.
However, an important mistake by the Luftwaffe
was that their messages were often sent through
both T-52 and Enigma. With the Enigma broken on a
regular basis, the T-52 traffic became less
important and Bletchley Park gave priority to the
Enigma traffic.
The T-52 was basically a
combination of a Telex machine and a cipher
machine. It used ten pinwheels of different sizes
of which the output states were XORed with each
other in a complex fashion. The output of these
logic functions was mixed with the standard
five-bit Telex signal of the machine through
XORing and swapping of the bits. The early T-52a
and T-52b had several security flaws and were
less secure. The T-52c had a more complex logic
mixing of the pinwheel output and the T-52d
incorporated a highly irregular stepping of the
wheels, controlled by cams. The T-52e was a
combination of the improvements of the T-52c and
T-52d, and was a highly secure machine.
The photo shows a
Converter M-209-B, manufactured by L.C.
Smith-Corona Typewriters. In 1940, the US
military selected the Hagelin C-38 as tactical
ciphering device and designated it as M-209. The
US Navy designated it CSP-1500. It was a lunchbox
sized - highly portable - cipher machine with a
simple and compact but ingenious design. By the
end of the Second World War over 140,000 of these
small M-209 machines were produced in the US.
The M-209 is a typical pin-and-lug
cipher machine with six relatively prime sized
pin wheels, a drum with 27 sliding bars and a
letter-dial/print-drum with reciprocal alphabet.
Although the M-209 was not very secure, the easy
to use and small machine was ideal for use in
tactical circumstances where the content of the
messages was no longer of importance after a few
hours.
The American SIGABA was
probably the most secure rotor cipher machine
during WW2. The SIGABA ECM Mark II (CSP 888/889)
had three banks of five rotors each. One set of
five rotors was used to encrypt the alphabet. The
other two banks of rotors were used to scramble
the signals that control the movement of the
encryption rotors. The result is a very irregular
and complex stepping of the encryption rotors.
The SIGABA was a most secure machine, and its
message traffic has never been broken. It
remained in service until the 1950s, when it was
replaced by more modern systems such as the KL-7
and on-line ciphering machines.
During WW2 the US and Britain both
developed a compatible cipher machine system,
based on their own machines. On American side the
special SIGABA CCM (Combined Cipher Machine),
designated ASAM 5 by the Army and CSP-1700 by the
Navy, was equipped with the CSP-1600 Typex
compatible rotor cage. This machine was
interoperable with the CCM version of the British
Typex cipher machine. After the war, the CCM
remained in service between the US, Britain and
Canada, and later on within NATO.
The SIGABA was a wonderful
machine that incorporated the newest developments
in the field of cipher machines. Unfortunately,
all machines were systematically withdrawn and
destructed for reasons of security. Only a
hand-full most rare ECM Mark II's survived in
museums and the special SIGABA CCM version is an
even more endangered species. The story of the
Famous German Enigma cipher machine is now widely
known to the public, but regretfully the SIGABA
with its far better cryptographic strength as the
Enigma is only known within the world of
cryptography.
More information is available
on several good websites. On the Pampanito
website you can read all about the history of the
ECM Mk II. They also published the complete
SIGABA manual. More technical details are found
on John Savard's website. Jerry proc provides
more military information on both the SIGABA ECM
Mk II and the CCM version. Finally, you can also
read the Cryptologia article on SIGABA.
The KL-7 ADONIS or
POLLUX is an off-line rotor cipher machine that
was developed in the late 1940's by the U.S.
Armed Security Agency (ASA) as a successor of the
SIGABA, and intruduced by the National Security
Agency (NSA) in 1953. ADONIS and POLLUX were two
different encryption procedures for the KL-7. The
American ADONIS procedure applied an encrypted
message key to pre-set the initial start position
of the rotors, whereas the ‘export
version’ POLLUX procedure used far less
secure non-encrypted message keys. The KL-7 was
compromised by John Walker who sold technical
information and key lists to the Soviets. The
KL-7 was used by the US and many of its Allies
and retired in 1983.
Output
of the KL-7 was printed on a paper ribbon. The
KL-7 has eight rotors (the fourth from the left
was stationary) with 36 contacts each. During its
service time, the rotors were recalled and
re-wired regularly. The rotors are placed in a
rotor cage called KLK-7 which can be removed from
the machine. Each rotor has an exchangeable
plastic outer ring with cams. Switches,
controlled by these cams, engage electromagnets
which in turn enable the motor to step certain
rotors. This resulted in a highly irregular
stepping of the rotors.
The 26 inputs and outputs of
the rotor cage are used to encrypt the letters.
The 10 remaining inputs and outputs are looped
back through the rotors, resulting in a very
complex signal path for the 26 letters. The
machine was non-reciprocal. This was achieved by
a sliding permuter board underneath the keyboard
which swapped all input and output contacts of
the rotor cage. Details about rotor and stepping
unit wiring are still classified. Today, all
publicly availably machines, such as this machine
from the Royal Dutch Signals Museum, are
carefully sanitized and stripped of any wiring,
related to the rotors and stepping unit.
