Talk:Voice inversion

Cryptography: Computer science Low‑importance

	This article is within the scope of WikiProject Cryptography, a collaborative effort to improve the coverage of Cryptography on Wikipedia. If you would like to participate, please visit the project page, where you can join the discussion and see a list of open tasks.CryptographyWikipedia:WikiProject CryptographyTemplate:WikiProject CryptographyCryptography articles
Low	This article has been rated as Low-importance on the importance scale.
	This article is supported by WikiProject Computer science (assessed as Low-importance).

Common carrier frequencies[edit]

The carrier frequencies mentioned in the article do not seem correct. With a carrier frequency of 2632 Hz, a frequency of 3200 Hz, near the upper limit of telephonic speech, would be aliased with 2064 Hz after inversion. Clearly, to avoid this for simple inversion, the carrier frequency would be higher than the highest frequency of the signal being inverted. However, there is nothing qualifying that these figures are only for split band inversion.

There is no citation for the listed carrier frequencies. Accordingly, I have tagged this paragraph as disputed. Lovibond (talk) 20:33, 15 October 2009 (UTC)[reply]

Remember, this stuff was basically invented for telephone service in the 1960s, when In-band signaling was universal using 2600 Hz and other nearby frequencies. So, most likely the voice was band-limited to 2000 Hz or so, before feeding to the inverter. Thus a method to avoid talk-off and other nuisances also eliminated frequency aliasing. And yeah, Sibilant must have suffered but such is the cost of security. Presumably the higher carrier frequencies were used more, after out of band supervision became universal, but of course by then single inversion had lost whatever power of secrecy it had anyway. Too bad nobody's got a cite, but I don't think this rises to the level of a dispute. Jim.henderson (talk) 18:14, 17 October 2009 (UTC)[reply]

Is your 1960s timeframe a typo? The application for the cited patent was filed in 1918, and frequency inversion was used in southern California shortly thereafter. The carrier frequencies are suspect and uncited. As it is, they seem to me to be more speculative than encyclopaedic. Lovibond (talk) 02:35, 30 October 2009 (UTC)[reply]

No, it's historical ignorance. Thanks; moving it back half a century gives inversion some degree of security, at least with the more complex inversion methods, for a decade or two. Looking for the cited patent, I am unable to find it by a quick Web search; indeed none of the Web links leads me quickly to anything relevant. Any relevant online cites would be welcome. As for the listed carrier frequencies creating alias frequencies, far as I see my error does not affect the question. When the front end of the system filters the voice to a narrower voiceband before scrambling, all is well. Some voice multiplex systems in the late 1920s, particularly British ones, operated at 2 KHz carrier spacing anyway, and WWII prompted the diplexing of many AT&T LD circuits to that spacing as well, so it's not as though nobody can talk through a narrower band. As for cited, instantly available sources, we are essentially bereft, not only for the carrier frequencies but for most everything in the article, unless someone can dig them up. Jim.henderson (talk) 00:55, 1 November 2009 (UTC)[reply]