4
\$\begingroup\$

Today with the help of integrated circuit, data encryption is used in transferring all kind of data in wireless communication. But back to some early days when people build telegram transmitter with vacuum tubes. How were data encrypted at that time. I know if someone transmit message in the form of Morse code, he can encrypt the data in various of ways, like the German Enigma machine. But how about the voice, for example, the voice chat between a navy pilot and aircraft carrier in World War II. It is possible to encrypt the voice at that time, or were there other ways to protect the talk from eavesdropping?

\$\endgroup\$
  • 4
    \$\begingroup\$ en.wikipedia.org/wiki/Code_talker Don't underestimate the value of going old-school. \$\endgroup\$ – user3624 Aug 27 '13 at 14:46
  • 4
    \$\begingroup\$ I recommend The Code Book by Simon Singh for a very readable overview of the history of this subject. It goes from Caesar substitution cyphers through to public key encryption. Written for the general public, so very little maths in it. \$\endgroup\$ – Martin Aug 27 '13 at 15:03
  • 2
    \$\begingroup\$ In WWII some voice was encrypted by using the Navajo language. These Navajo code talkers then applied additional encription by using various words differently than their normal meanings and making some meanings context-sensitive. This encryption scheme was used extensively by the US in the Pacific, and the Japanese never cracked the code. \$\endgroup\$ – Olin Lathrop Aug 27 '13 at 15:36
  • \$\begingroup\$ I second Martin's recommendation of Simon Singh's book. The Navajo story Olin mentions is also in the book. \$\endgroup\$ – radagast Aug 27 '13 at 15:53
8
\$\begingroup\$

Back in the day, voice "scramblers", as they were called at the time, were strictly analog-domain. Typically, the voice band would be divided into a number of sub-bands by filters, and each subband would be modified (shifted in frequency and/or inverted) by means of oscillators and balanced modulators, in a manner similar to how SSB radio transmission works.

As long as the other end of the circuit had a matching configuration, clear speech would come out. But the signal in the middle would be essentially unintelligible. You would hear the loudness pattern of words and syllables, but not be able to make out the vowels and consonants.

The systems were finicky to keep calibrated; components drifting by just a few percent (or less) could make the system useless. Fortunately, the first mass-produced transistors and the first forms of digital telephony and digital signal procesing (DSP) came along shortly thereafter, and these were much more amenable to true encryption.

\$\endgroup\$
8
\$\begingroup\$

Voice chat was rarely encrypted, although there was a "scrambler" system: http://ciphermachines.com/voice

Scramblers were built around synchronised recordings or pseudo-random frequency shift patterns. A noise signal was "added" at one end and "subtracted" at the other, in the analog domain. Obviously this rests heavily on the synchonisation.

The "vocoder" was developed to give true voice encryption: http://en.wikipedia.org/wiki/SIGSALY , developing a lot of useful technology on the way. Early units were huge and not portable.

Spread-spectrum technology was theoretically available, having been invented by actress Hedy Lamarr, but I can't find references to it being used for analog voice comms. In practice there was a small amount of operational frequency hopping, where units would change frequency to force enemy radio operators to go looking for their signal again.

Ground units often used field telephones, which were wired and therefore harder to intercept.

\$\endgroup\$
2
\$\begingroup\$

The American troops used Code Talkers in both World Wars. The language was never written down and so only native speakers could communicate without the danger of translation.

see http://www.defense.gov/specials/nativeamerican01/code.html

\$\endgroup\$
-3
\$\begingroup\$

It's worth noting that in your example of real-time chatter between a pilot and a carrier in WWII, it usually wouldn't matter if the communication was in the open or not. The odds of an enemy monitoring the particular frequency in use, within the range of the plane, with a translator available, and learning something they didn't already know, in time to make good use of the information, would be slim.

The biggest piece of information would be simply that there was a unit with a radio in the area, and that information would be revealed with or without crypto. Even a simple system of code words would mask most of the remaining signal.

Imagine a carrier duel such as the Battle of Midway. One of your ships picks up a weak radio transmission from a general northwesterly direction that's simply: "Pelican 2, sighted 2 teacups at grid A7". What's "teacup" code for? Maybe a small surface ship? Did they stumble across your destroyers? The destroyers shouldn't have been that far west yet! Or maybe it means aircraft, and they saw your own scout flight. You've learned almost nothing that you wouldn't have learned from a 5 second burst of gibberish.

Certainly some other kinds of communications about future plans would be worth intercepting and translating, but a lot of communication could be done in the clear without much danger.

\$\endgroup\$
  • 3
    \$\begingroup\$ While this post is interesting reading, it does not answer the question asked. This site is not a discussion forum, and a post such as this one does not belong in the answers section of the page. Perhaps you might want to summarize it into a comment. Thanks. \$\endgroup\$ – Anindo Ghosh Aug 27 '13 at 22:37

Your Answer

By clicking “Post Your Answer”, you agree to our terms of service, privacy policy and cookie policy

Not the answer you're looking for? Browse other questions tagged or ask your own question.