[Milsurplus] overheard telephone conversations

[jvendely at cfl.rr.com]

---- Al Klase <ark at ar88.net> wrote: 
> OK.  It's time to read the SIGSALY article on Wikipedia: 
Yes, in the pre-war years, Bell Labs had a 5-band inversion-type speech privacy device known as the A-3.  The 5 bands could be arranged in six different combinations of inverted or noninverted speech, and the combinations were changed about every 15 seconds or so.  It therefore had relatively few permutations, and though it would have been impractical for amateurs to break, it afforded no real secrecy against a sophisticated eavesdropper. It was known to be easily breakable, and the Germans developed a real-time decoder and routinely listened to high level traffic with A-3 encoding on HF links between the U.S. and England. The limitation of the A-3 system was the impetus for development of the totally secure Sigsaly system.  

73,

John K9WT 

> https://en.wikipedia.org/wiki/SIGSALY
> 
> They mention that the existing voice security on international 
> radio-telephone circuits was "A-3."  Do any of you have a handle on A-3?
> 
> Other clues from an old phone company radio guy I know:
> 
>      The rearranged analog sub-band scheme may be a Lenkurt product.
> 
>      AT&T may have used frequency inversion on some channels, making 
> them unintelligible to the casual eavesdropper.
> 
> Interesting thread.
> 
> Al
> 
> Al Klase – N3FRQ
> Jersey City, NJ
> http://www.skywaves.ar88.net/
> 
> On 5/31/2016 2:06 AM, AKLDGUY . wrote:
> > I'm highly skeptical. The only way frequency slicing like that
> > could have been done in 1937 was by translating the entire
> > audio band up to some much higher frequency, then using
> > highly selective crystal filters to pick out each slice (say 3 of
> > them), then translating each one to a segment of another
> > baseband, then translating that baseband back to audio.
> >
> > Given the state of crystal filter design in the 1930s, I think
> > even the telecomm industry would have balked at that level
> > of complexity.
> >
> > 73 de Neil ZL1ANM
> >
> > On 5/31/16, Rob Flory <farmer.rob.flory at gmail.com> wrote:
> >>   "I don't know whether SSB was
> >> used for those 1937 radio telephone links, but if it was,
> >> an enthusiast may have figured out that a BFO would
> >> allow reception."
> >>
> >> You can be sure an enthusiast would have intercepted the signals if they
> >> were simple SSB without encoding, and no one would have counted on the
> >> thousands of enthusiasts, never mind the commercial and political
> >> interests, from figuring that out.
> >>
> >> If I recall correctly, the baseband audio was broken up into several slices
> >> that were shuffled, as in 0-600Hz was re-inserted at 1200-1800 etc. and
> >> that shuffle was changed frequently according to a pseudorandom sequence
> >> based on "one time pads" known only at TX and RX locations.
> >>
> >> The transmitting station was just a few miles from here.
> >>
> >> RF
> >>
> >>
> >>
> >>
> >> On Mon, May 30, 2016 at 9:21 AM, AKLDGUY . <neilb0627 at gmail.com> wrote:
> >>
> >>> How were they encoded?
> >>>
> >>> Be careful how you answer. You're addressing a former
> >>> telecommunications technician, 1968-1984.
> >>>
> >>> When I was transferred into the long distance transmission
> >>> division in 1975, there was a very old Western Electric
> >>> system that had been in use back in the 1930s and was still
> >>> being kept running as a standby in case of emergency.
> >>> It was junked not long after, because by that stage we had
> >>> enough diversity with multiple routes such as coaxial cable
> >>> and microwave links out of Auckland.
> >>>
> >>> It was an industry-standard 60-108 KHz carrier system of
> >>> 12 channels at 4 KHz spacing designed for open wire 600
> >>> ohm line.
> >>>
> >>> Essentially it was an upper sideband broadband system,
> >>> with crystal filters for each direction of speech, 24 filters
> >>> in all. Each filter was in a sealed (soldered) copper case
> >>> about a foot long, 4" wide, 5" high. I can't remember what
> >>> tubes it used, but they were big glass ones like in radios
> >>> of the early 30s.
> >>>
> >>> That was probably the state of telecommunications
> >>> technology in the 1930s. I don't know whether SSB was
> >>> used for those 1937 radio telephone links, but if it was,
> >>> an enthusiast may have figured out that a BFO would
> >>> allow reception.
> >>>
> >>> So I'm interested to hear what kind of encoding you
> >>> think was used.
> >>>
> >>> 73 de Neil ZL1ANM
> >>>
> >>> On 5/30/16, Rob Flory <farmer.rob.flory at gmail.com> wrote:
> >>>>   "Radio telephone communications opened in 1927 on a 24 hour
> >>>>   basis, US-UK. By 1949, there were 70 radio telephone circuits for
> >>>>   all 5 continents."
> >>>>
> >>>> They were encoded, so an enthusiast would not have been able to
> >>>> overhear
> >>>> them.
> >>>>
> >>>> RF
> >>>>
> > ______________________________________________________________
> > Milsurplus mailing list
> > Home: http://mailman.qth.net/mailman/listinfo/milsurplus
> > Help: http://mailman.qth.net/mmfaq.htm
> > Post: mailto:Milsurplus at mailman.qth.net
> >
> > This list hosted by: http://www.qsl.net
> > Please help support this email list: http://www.qsl.net/donate.html
>