Boris Caesar Wilhelm Hagelin was a Swedish businessman and inventor of encryption machines. Born of Swedish parents in Adshikent, Hagelin attended Lundsberg boarding school and studied mechanical engineering at the Royal Institute of Technology in Stockholm, graduating in 1914, he gained experience in engineering through work in the United States. His father Karl Wilhelm Hagelin worked for Nobel in Baku, but the family returned to Sweden after the Russian revolution. Karl Wilhelm was an investor in Arvid Gerhard Damm's company Aktiebolaget Cryptograph, established to sell rotor machines built using Damm's 1919 patent. Boris Hagelin was placed in the firm to represent the family investment. In 1925, Hagelin took over the firm reorganising it as Aktiebolaget Cryptoteknik in 1932, his machines sold rather better. At the beginning of World War II, Hagelin moved from Sweden to Switzerland, all the way across Germany and through Berlin to Genoa, carrying the design documents for the company's latest machine, re-established his company there.
That design was small and moderately secure, he convinced the US military to adopt it. Many tens of thousands of them were made, Hagelin became quite wealthy as a result. Hagelin fraudulentely sold compromised machines to a variety of customers. Historian David Kahn has suggested that Hagelin was the only cypher-machine maker who became a millionaire. U. S. Patent 1,846,105 U. S. Patent 2,089,603 U. S. Patent 2,247,170 U. S. Patent 2,802,047 U. S. Patent 2,851,794 U. S. Patent 3,083,263 U. S. Patent 3,485,948 B-21 C-36 M-209 C-52 Boris CW Hagelin, The Story of the Hagelin Cryptos, Cryptologia, 18, July 1994, pp 204–242. History of Boris Hagelin Chart of the evolution of Hagelin / CRYPTO AG machines Hagelin BC-52 Cipher Machine Simulator Hagelin US M-209 Cipher Machine Simulator
The CD-57 was a portable, mechanical cipher machine manufactured by Crypto AG, first produced in 1957. It was derived from the earlier CD-55, was designed to be compatible with the larger C-52 machines. Compact, the CD-57 measured 5 1/8in × 3 1/8in × 1 1/2in and weighed 1.5 pounds. The CD-57 used six wheels. A variant is a similar device using a one-time pad system rather than rotating wheels; the STG-61 was a licensed copy of the CD-57 by Hell. Sullivan shows. M-209 Wayne G. Baker, Solving a Hagelin, Type CD-57, Cryptologia, 2, January 1978, pp1–8. Louis Kruh, Cipher Equipment: Hagelin Pocket Cryptographer, Type CD-57, Volume 1, 1977, pp255–260. Geoff Sullivan, Cryptanalysis of Hagelin machine pin wheels, Cryptologia, 26, pp257–273, October 2002. Photographs and a simulator Photographs of the CD-57:, Jerry Proc's pages:, Information about the STG-61
In cryptography, Fialka is the name of a Cold War-era Soviet cipher machine. A rotor machine, the device uses 10 rotors, each with 30 contacts along with mechanical pins to control stepping, it makes use of a punched card mechanism. Fialka means "violet" in Russian. Information regarding the machine was quite scarce until c. 2005 because the device had been kept secret. Fialka contains a five-level paper tape reader on the right hand side at the front of the machine, a paper tape punch and tape printing mechanism on top; the punched-card input for keying the machine is located on the left hand side. The Fialka requires 24 volt DC power and comes with a separate power supply that accepts power at 100 to 250 VAC, 50–400 Hz; the machine's rotors are labelled with Cyrillic. The keyboard, at least in the examples of East German origin, had both Latin markings. There are at least two versions known to exist, the M-125-MN and the M-125-3MN; the M-125-MN had a typewheel that could handle Cyrilic letters.
The M-125-3MN had separate typewheels for Cyrilic. The M-125-3MN had three modes, single shift letters, double shift with letters and symbols, digits only, for use with code books and to superencrypt numeric ciphers; the Fialka rotor assembly has 10 rotors mounted on a 30 by 30 commutator. The commutator consists of two sets of 30 contact strips set at right angles to each other. A punched card is placed between the two sets of contacts via a door on the left hand side of the unit; each punched card has 30 holes, with one hole per row and column pair, thereby specifies a permutation of the 30 rotor contact lines. This feature is comparable to the plug board on the Enigma machine. A triangular plate was used to enter the null permutation for testing purposes. There are two types of rotors: disassemblable rotors, used with M-125-3MN. Rotorset name is "PROTON." The disassemblable rotors consisted of an insert with electrical contacts and scramble wiring, an outer ring with mechanical pins whose presence or absence controlled rotor stepping.
As part of the key setup, the stepping control pins could be rotated relative to the outer ring. The inner, electrical ring could be rotated relative to the outer ring and could be inserted in one of two ways, with side 1 or side 2 up. Unitary rotors, used with M-125-MN; these had mechanical pins. The only key adjustment was the order of the rotors on the initial rotor settings. There was one combination for the disassemblable rotors, compatible with the unitary rotor. One East German manual that has become public contains typed-in and hand written addenda that suggest the East Germans, at least stopped using the added features of the disassemblable rotors and only used them in unitary compatibility mode. Adjacent rotors step in opposite directions. A spare rotor assembly was kept in unit's top cover; the keying material for the Fialka consists of a daily key book, a message key book and a message identifier book for broadcast traffic. The daily key book contains day keys for one month. A day key consists of a punched card.
For fixed rotor systems, the key table specifies the order of the rotors on the axle and the initial rotor settings to be used to decrypt the indicator of broadcast messages. As its name implies, the day key was changed at 00:01 hours. For the disassemblable rotors, the table specifies the electrical insert for each outer rotor, which side was to be up, the orientation of the insert relative to the outer rotor. Here is a sample M-125-3NM day key table for use on the 14th of the month: ИДЖЗА ВКБГЕ 14 OCAHE PTБВЕ БДВИА ГЕЗКЖ 2II22 I22I2 КУЛКЮ ЫХВУГThe message key table contained the initial rotor settings to be used with each message. A message key was never to be used more than once; the keying material was distributed in a foil-covered package, with the daily key tables and punched cards fan-folded in a pouch with perforations between each item. The other tables were in a side pouch; the Fialka design seems to derive from the Swiss NEMA, but the NEMA only has 5 electrical rotors vs. the Fialka's 10 and NEMA lacks a punched card commutator or an equivalent, such as a plug board.
Fialka seems most comparable to the U. S. KL-7 which has eight electrical rotors and lacks a commutator, but its keyboard permutor switch eliminated the need for a reflector, which proved to be a weakness in the Enigma system. Enigma NEMA M-125 Operation manual, DV A 040/1/321, December 1978, National People's Army, German Democratic Republic, scanned PDF from Dr. Tom Perera, Fialka Cipher Machines — by Tom Perera Fialka M-125 on the Crypto Museum website Fialka Cipher Machines — by Nick Gessler RUSSIAN M-125 FIALKA — by Jerry Proc A Simulation of M125MN and M125-3MN find under Fialka: The Bigger, Russian Enigma Fialka in the German Spy Museum
The TSEC/KW-26, code named ROMULUS, was an encryption system used by the U. S. Government and by NATO countries, it was developed in the 1950s by the National Security Agency to secure fixed teleprinter circuits that operated 24 hours a day. It used vacuum tubes and magnetic core logic, replacing older systems, like SIGABA and the British 5-UCO, that used rotors and electromechanical relays. A KW-26 system contained over 800 cores and 50 vacuum-tube driver circuits, occupying more than one half of a standard 19-inch rack. Most of the space in the rack and most of the 1 kW input power were required for the special-purpose vacuum tube circuits needed to provide compatibility with multiple input and output circuit configurations; the military services' requirements for numerous modes and speeds increased costs and delayed delivery. NSA says it is doubtful that more than three or four of the possible configurations were used; the KW-26 used an NSA-developed encryption algorithm based on shift registers.
The algorithm produced a continuous stream of bits that were xored with the five bit Baudot teleprinter code to produce ciphertext on the transmitting end and plaintext on the receiving end. In NSA terminology, this stream of bits is called the key; the information needed to initialize the algorithm, what most cryptographers today would call the key, NSA calls a cryptovariable. Each KW-26 was given a new cryptovariable once a day. NSA designed a common fill device, for loading the cryptovariable, it used. The operator inserted the daily key card into the CFD and closed the door securely, locking the card in place. Decks of cards were sent by courier; the cards were accounted for. Because the KW-26 used a stream cipher, if the same key card was used twice, the encryption could be broken. To prevent re-use, the card was automatically cut in half upon reopening the CFD; as the units aged, the card reader contacts became less dependable, operators resorted to various tricks, such as hitting the card reader cover with a screwdriver, to get them to work properly.
Card readers were cleaned and the spring loading of the contacts checked as part of the routine maintenance of the device. Because the KW-26 sent a continuous stream of bits, it offered traffic-flow security. Someone intercepting the ciphertext stream had no way to judge how many real messages were being sent, making traffic analysis impossible. One problem with the KW-26 was the need to keep the transmitter units synchronized; the crystal controlled clock in the KW-26 was capable of keeping both ends of the circuit in sync for many hours when physical contact was lost between the sending and receiving units. This capability made the KW-26 ideally suited for use on unreliable HF radio circuits. However, when the units did get out of sync, a new key card had to be inserted at each end; the benefit of traffic-flow security was lost each time. In practice, operational protocol led to the cards being replaced more than was desirable to maintain maximum security of the circuit; this was so on radio circuits, where operators changed the cards many times each day in response to a loss of radio connectivity.
In any case, it was necessary to change the cards at least once per day to prevent the cypher pattern from repeating. Early KW-26 units protected the CRITICOMM network, used to protect communications circuits used to coordinate signals intelligence gathering; the initial production order for this application, awarded to Burroughs in 1957, was for 1500 units. Other services demanded KW-26's and some 14000 units were built, beginning in the early 1960s, for the U. S. Navy, Air Force, Defense Communications Agency, State Department and the CIA, it was provided to U. S. allies as well. When the USS Pueblo was captured by North Korea in 1968, KW-26's were on board. In response, the NSA had modifications made to other units in the field changing the crypto algorithm in some way by changing the shift register feedback taps. Starting in the mid-1980s, the KW-26 system was decommissioned by NSA, being replaced by the more advanced solid-state data encryptor, TSEC/KG-84. NSA encryption systems KW-26 history page NSA brochure - Securing Record Communications: The TSEC/KW-26
The STU-II is a secure telephone developed by the U. S. National Security Agency, it permitted up to six users to have secure communications, on a time-shared basis. It was made by ITT Defense Communications, New Jersey. An OEM partner was Northern Telecom. According to information on display in 2005 at the NSA's National Cryptologic Museum, the STU-II was in use from the 1980s to the present, it uses the linear predictive coding algorithm LPC-10 at 2.4 kilobits/second to digitize voice, the "Key Distribution Center" for key management. The display stated that the STU-II B is the standard narrow band secure telephone. STU-II replaced the STU-I, KY-3 and the Navajo I; the last was a secure telephone in a briefcase, of which 110 were built in the 1980s for use by senior government officials when traveling. The Navaho I used LPC-10; some 10 000 STU-II units were produced. Delusion.org - National Cryptologic Museum pictures Pictures of president Reagan using a STU-II phone STU-III SCIP
In the history of cryptography, Typex machines were British cipher machines used from 1937. It was an adaptation of the commercial German Enigma with a number of enhancements that increased its security; the cipher machine was used until the mid-1950s when other more modern military encryption systems came into use. Like Enigma, Typex was a rotor machine. Typex came in a number of variations, but all contained five rotors, as opposed to three or four in the Enigma. Like the Enigma, the signal was sent through the rotors twice, using a "reflector" at the end of the rotor stack. On a Typex rotor, each electrical contact was doubled to improve reliability. Of the five rotors the first two were stationary; these provided additional enciphering without adding complexity to the rotor turning mechanisms. Their purpose was similar to the plugboard in the Enigmas, offering additional randomization that could be changed. Unlike Enigma's plugboard, the wiring of those two rotors could not be changed day-to-day.
Plugboards were added to versions of Typex. The major improvement the Typex had over the standard Enigma was that the rotors in the machine contained multiple notches that would turn the neighbouring rotor; this eliminated an entire class of attacks on the system, whereas Enigma's fixed notches resulted in certain patterns appearing in the cyphertext that could be seen under certain circumstances. Some Typex rotors came in two parts, where a slug containing the wiring was inserted into a metal casing. Different casings contained different numbers of notches around such as 5, 7 or 9 notches; each slug could be inserted into a casing in two different ways by turning it over. In use, all the rotors of the machine would use casings with the same number of notches. Five slugs were chosen from a set of ten. On some models, operators could achieve a speed of 20 words a minute, the output ciphertext or plaintext was printed on paper tape. For some portable versions, such as the Mark III, a message was typed with the left hand while the right hand turned a handle.
By the 1920s, the British Government was seeking a replacement for its book code systems, shown to be insecure and which proved to be slow and awkward to use. In 1926, an inter-departmental committee was formed to consider whether they could be replaced with cipher machines. Over a period of several years and at large expense, the committee investigated a number of options but no proposal was decided upon. One suggestion was put forward by Wing Commander Oswyn G. W. G. Lywood to adapt the commercial Enigma by adding a printing unit but the committee decided against pursuing Lywood's proposal. In August 1934, Lywood began work on a machine authorised by the RAF. Lywood worked with J. C. Coulson, Albert P. Lemmon, Ernest W. Smith at Kidbrooke in Greenwich, with the printing unit provided by Creed & Company; the first prototype was delivered to the Air Ministry on 30 April 1935. In early 1937, around 30 Typex Mark I machines were supplied to the RAF; the machine was termed the "RAF Enigma with Type X attachments".
The design of its successor had begun by February 1937. In June 1938, Typex Mark II was demonstrated to the cipher-machine committee, who approved an order of 350 machines; the Mark II model was bulky, incorporating one for ciphertext. As a result, it was larger than the Enigma, weighing around 120 lb, measuring 30 in × 22 in × 14 in. After trials, the machine was adopted by the Army and other government departments. During World War II, a large number of Typex machines were manufactured by the tabulating machine manufacturer Powers-Samas. Typex Mark III was a more portable variant, using the same drums as the Mark II machines powered by turning a handle; the maximum operating speed is around 60 letters a minute slower than the 300 achievable with the Mark II. Typex Mark VI was another handle-operated variant, measuring 20 in ×12 in ×9 in, weighing 30 lb and consisting of over 700 components. Plugboards for the reflector were added to the machine from November 1941. For inter-Allied communications during World War II, the Combined Cipher Machine was developed, used in the Royal Navy from November 1943.
The CCM was implemented by making modifications to Typex and the United States ECM Mark II machine so that they would be compatible. Typex Mark VIII was a Mark II fitted with a morse perforator. Typex 22 and Typex 23 were late models. Mark 23 was a Mark 22 modified for use with the CCM. In New Zealand, Typex Mark II and Mark III were superseded by Mark 22 and Mark 23 on 1 January 1950; the Royal. This amalgamation allowed a single operator to use punch tape and printouts for both sending and receiving encrypted materiel. Erskine estimates that around 12,000 Typex machines were built by the end of World War II. Typex was used by the British armed forces and was used by Commonwealth countries including Canada and New Zealand. From 1943 the Americans and the British agreed upon a Combined Cipher Machine; the British Typex and American ECM Mark II could be adapted to become interoperable. While the British showed Typex to the Americans, the Americans never permitted the British to see the ECM, a more complex design.
Instead, attachments were built for both. Although a British test cryptanalytic attack made considerable progress, the results were not as significa