Reconstructing SIGSALY
Lessons learned in reconstructing the World War II-era SIGSALY voice encryption system.
Comments
Nameless Cow • February 15, 2019 9:02 AM
Does anyone know how the SIG**** cipher machines got their assigned designations? I mean, is the part after the SIG- prefix chosen according to some scheme?
CallMeLateForSupper • February 15, 2019 9:49 AM
The “reconstructing” article made my morning coffee sweeter. B+; filament; grid; cathode; transformer Q. (I’m distinctly heard gears gnash as my brain shifted out of solid-state-speak, in which it had siezed over the last four decades.)
I acquired a healthy respect for B+ voltages while in A.F. technical school in 1968. I managed to touch the ~200v B+ of an open-frame oscilloscope I was trouble-shooting, and the resulting involuntary muscle contractions propelled me skyward from my stool. I tasted metal for a while afterword. A classmate who touched the same voltage on his o-scope later the same day fared worse: he was slouched on his stool, legs extended under the bench and feet on a 2×4 foot rest. His spasm sent him backwards, and he landed rather heavily on the concrete floor.
David Rudling • February 15, 2019 9:51 AM
A brief history of SIGSALY is on the NSA site.
The name is not an acronym of course. The well known crypto machine SIGABA is another example of the naming scheme. Perhaps not surprisingly, and very sensibly, for names selected for crypographic devices there was no easily guessable logical scheme for naming.
Jon Paul • February 15, 2019 10:17 AM
Hello all, Thanks ro Bruce for posting this, and many thanks for the comments.
Just a few notes..
I have researched the history of digital technology since 1990s and SIGSALY since 2003.
Besides the first ADC and DAC, SIGSALY had the first unbreakable speech scrambler, first spread spectrum an was the first true digital signal processor, all this in 6 months from Bell Labs in NYC, 1942. Claude Shannon, Harry Nyquist, Homer Dudley and Alan Turing were involved to varying degrees.
My SIGSALY Quantizer goal was to make a dramatic demonstration of analog to 5 level digital conversion with voice input, that even non-technical and young viewers can understand. After I reverse engineered the topology of the SIGSALY”Stepper” or quantizer, and its use of five argon tetrodes 2051 thyratrons as comparators, the rebuild took ~ 3 years overal.
There are 2 actually articles on the IEEE Spectrum site, one for the general project and the second for the design and debugging details.
Here are 3 min video demos in English and French.
http://crypto-museum.org/QUANT/VIDEO/QUANTEN.mp4]http://crypto-museum.org/QUANT/VIDEO/QUANTEN.mp4
http://crypto-museum.org/QUANT/VIDEO/QUANTFR2.mp4]http://crypto-museum.org/QUANT/VIDEO/QUANTFR2.mp4
A complete SIGSALY reconstruction needs 72 quantizers and 384 thyratrons.
A DSP real-time simulator in software is planned.
Your comments and feedback appreciated.
Jon Paul
ARCSI
Grant • February 15, 2019 11:01 AM
Fascinating work, Jon, showing great dedication! Simply fascinating example of pushing a technology to its literal limits.
Are you aware of any recordings of what the encrypted voice audio sounded like? I’ve heard descriptions of “Donald Duck” or “the theme from the Green Hornet”.
David Rudling • February 15, 2019 3:27 PM
@Grant
There is reported to be a recording of Churchill talking to Roosevelt “on display” (whatever that means in audio terms) in the UK Houses of Parliament.
I can’t personally vouch for this and I couldn’t readily find the recording on line.
scot • February 15, 2019 4:08 PM
Does anyone how where to find a recording of the reconstructed speech? 2.5 bits at a 50 hz seems marginal for reconstructing speech, but what I really want to hear is how noise in the playback of the key impacted the decryption.
Seth • February 15, 2019 4:19 PM
The National Cryptologic Museum (next to NSA headquarters in Maryland) has an excellent display on encrypted voice. There are recordings of SIGSALY and later systems for encrypted voice communications. The improvement in quality with each generation is noticeable – the output of SIGSALY is almost unintelligible due to noise, but it improved steadily over the years. I think the display mentioned that FDR only used it on one or two ocassions due to that. Reading about the circuitry behind it makes it clear just how much of an achievement it was.
Also, there’s a great overview of how it worked at https://www.cryptomuseum.com/crypto/usa/sigsaly/index.htm
Clive Robinson • February 15, 2019 4:41 PM
@ ALL,
stored as a vinyl recording of noise
Hides away another interesting part of the system.
The vinyl 16″ 33 & 1/3rd RPM records (SIGGRUV) only gave ~12mins of conversation time and had some interesting synchronization and production issues of their own. Also besides the KeyGen issues there were quite a few KeyMan/KeyDist issues which is why they were actually rarely used and reserved for more than above secret conversations.
For secret and below a different encryption method was used. This was a mechanical system (SIGBUSE) less than effectionatly called the “thrashing machine” which could eat bits of the unwary technician. The heart of which was an M-228 rotor machine that drove a frightening array of telephone selector switches and relays.
But it became clear that SIGGRUV was not just a KeyMan issue it was becoming “a cost to far”. So the vinal records made by Bell Technical Service staff in New York got replaced at the end of 1944 by Sig Corps (805th in Pentagon) staff using a different recording system (SIGSOBS:SIGSALY) that simultaniously produced two acetate records (SIGJING).
Of interest is that at the heart of both analog noise generators was an almost identical “Stepper” the design of which can be seen in US 3373245 Newby and Vaughn patent.
The entire cost of the system from design to scrappage in 1946 was estimated to be $28million.
But was the system secure?
Well that depends on how you defone “the system”… Apparently at the London end Winston Churchill had a hidden sound proof booth hiden away (supposadly disguised with a cover story of being a toilet). Well apparently the cable from the phone in the booth to the SIGSALY some distance away was not protected in any way and ran adjacent to other phone and signal cables. So yes it could have been tapped or crosstalk carry those oh so secret ruminations to others. Apparently Churchill was also the most frequent user of the system… Obviously an eraly Techno-Geek 😉
For those with a “human interest” hankering, the story of Meg Madsen might be of interest,
https://www.pritzkermilitary.org/explore/museum/past-exhibits/shes-wow/test/meg-madsen/
Tatütata • February 15, 2019 4:50 PM
I managed to touch the ~200v B+ of an open-frame oscilloscope I was trouble-shooting,
Diagnosing a large Tektronix vintage scope teaches respect, especially when you’re trying to persuade it to stop giving off smoke signals. You know, the ones that made your light bulb dims and the utility meter panick when you turned it on.
In my hamming days of my youth my first transmitter was a WW2 anode-keyed CW transmitter, with the 550V power from the screaming dynamotor fully exposed on the cheapo telegraph key I was using. Ouch.
The Spectrum article mentions patents. I looked these up a long time ago, I believe there are about a dozen relevant patents, which were filed 1942-1945, but were only published in 1976 after the secrecy order was lifted. The general principles are illustrated, but IIRC, their problem is that they disclose very little or misleading information on the actual implementation parameters, such as the height of quantization steps.
At least the Bell Labs patent betrays the ultimate purpose of the system.
There is no such information in US2503662 for a “Electronic valve apparatus suitable for use in counting electrical impulses”, originally filed on 17 November 1944 by one Thomas Harold Flowers, then a GPO engineer involved with a little outfit in a place called “Bletchley Park”. There weren’t that many uses for pulse counters back then, even though the illustrious RadLab Radar series has a few chapters on pulse counting circuits, and a 1943 general treaty was cited by the PTO as a reference during the prosecution of the patent.
Speaking of telecom equipment, the Tizard mission disclosed to the US an Wireless Set No. 10, an 8-channel microwave mobile communication system which enormously simplified the logistics of allied forces marching into conquered territory (see a description here. AT&T designed its own version of the system, the AN/TRC-6 which was bigger, heavier, more powerful, but in essence it was the same system, and could probably have been made interoperable with the British system without too much effort long before STANAGs were invented. What I found shocking was that the Yanks proceeded to patent everything in sight. A comparison of the GB and US patents yields insight in the respective style of patent drafting. I believe that the AN/TRC-6 carried Sigsaly signals on the continental leg of circuits.
One ancillary component of Sigsaly was the local loop connecting the White House to the rather large terminal elsewhere in DC. A very high level of white noise was inserted on a very finely balanced phantom circuit, probably carried in a lead-sheathed cable. The idea was that if someone tried to access the circuit, even by inductive means, the circuit would somehow immediately go out of balance, revealing the wiretap attempt. An alarm was also provided. See US2556677, which was published in 1951.
Clive Robinson • February 15, 2019 8:11 PM
@ ,
A very high level of white noise was inserted on a very finely balanced phantom circuit
The question is are you or I going to describe “phantom circuits” for those who have never heard of them 😉
With regards,
What I found shocking was that the Yanks proceeded to patent everything in sight.
The whole US involvment in WWII was very much a “profiteering” one. Not only did Britain have to sell most of their investments at the lowest possible value, lend lease was “over priced” to put it mildly and the US demanded full rights over any British invention as a pre-requisite to it being made –again at top dollar pricing– in the US. It’s easy to see the effect the “beggering of Britain” had on the US economy, to say “They never had it so good” was an understatment.
Then after WWII every technical stepforward Britain made, had to become a US technical achivment by demands that the war debt be repaid if Britain did not capitulate immediately. The only argument the US lost on that was because of the French over Concord, where France was in a better threatening position. The whole “special relationship” nonsense is a doubly sacrificial one for the UK as not only did the US bleed the UK white three times over, it also destroyed for many years the UK relationship with Europe[1].
It’s been noted by many over the years that US patents in the previous two centuries appear to be “knock offs” of patents else where in the world or based on technology developed else where and effectively stolen by the US in exactly the same way this century they complain that other nations are doing to them…
Weather • February 15, 2019 9:15 PM
Clive and Clive
Can you explain phantom circuit, and get as technical as you can, or I might agree with shizzo
JG4 • February 15, 2019 10:41 PM
The worst shock that I ever got was 220 volts DC from one hand to the other. B+ voltage in my friend’s tube radio that we were using with some test leads to charge capacitors. I vaguely think that it was the transmitter. I managed to grab the metal parts of both alligator clips. Once. And my hands stuck to them for about a second. I was very fortunate to drop the leads before my heart was permanently stopped. I remember my hands being far apart, probably because the spasm flung them away from each other. I don’t think that I had any muscular control in my arms and hands while the current was flowing, but I think that I was able to stand up as the shock occurred. I staggered back a step or two after the clips dropped out of my hands. The adrenaline probably restarted my heart. Not long after, I had rigged up some power rectifiers to the AC line and was using the resulting DC to make sodium metal in molten lye. The problem is that the cement floor in the basement was damp, and I was barefoot. Using a metal butter knife to nudge the electrodes around. Once. The spasm flung the knife all the way across the room. I don’t think that I’ve been shocked again, because those incidents produced real learning. A good thing too, because I started working with 1000 and 10,000 volt circuits shortly after.
1&1~=Umm • February 16, 2019 4:30 AM
A little searching on the Internet, reveals pictures of Churchill’s War Rooms London, including room 63 where the SIGSALY booth was located.
https://www.timetravelturtle.com/churchill-war-rooms-tour-london/
And yes it does have a “bathroom door” type “vacant / engaged” indicator as can be seen in the photograph.
Jon PAUL • February 16, 2019 5:10 AM
Hello all, I greatly appreciate your comments and feedback!
1/ There were 33 patents granted to BTL overall, some by very well-known engineers like Homer Dudley, inventor of the VOCODER. Many were classified secret at the time and only declassified in 1976.
Claude Shannon’s fundamental paper Communication Theory of Secrecy Systems was inspired by his work at BTL on the algorithms of SIGSALY and classified during the war. It was finally declassified in October 1949 and published in BSTJ.
http://pages.cs.wisc.edu/~rist/642-spring-2014/shannon-secrecy.pdf
https://archive.org/details/bstj28-4-656
2/ The transmission rate was not 50 Hz! The voice was analyzed into 12 parameters, 10 sub-band levels and 2 pitch parameters, also one bit for voiced/unvoiced. Each of the 12 analog parameters had 25 Hz bandwidth. 12 X 25 = 300 Hz, or 10X compression of voice information. Twelve quantizers converted the 12 vocoder parameters into six level digital signals.
See for info on vocoder theory https://en.wikipedia.org/wiki/Vocoder
3/ The coding of the 12 quantized parameters employed MOD-6 addition. Each 6 level parameter was added to a 6 level band limited noise, and indeed recorded on 16″ phonograph records.
4/ The final step before SW transmission was spread spectrum with SSB and FM, which expanded the bandwidth back to 3 kHz. SIGSALY had the first working spread spectrum, used not for secrecy but to reduce transoceanic SW radio fading by 20 dB.
5/ I know of no original SIGSALY recordings. But NSA and NCM have recordings of post war versions of SIGSALY using the same techniques. Very few if any are on the Internet.
See my interview and a short sound clip on PRI 99% invisible program
https://99percentinvisible.org/episode/vox-ex-machina/
6/ To probe further, see the IEEE and NSA papers, and my AES, NAB and SMPTE papers on the Origins of Digital Media and Origins of Compression.
This link has a precise technical description, which should answer many of your questions: https://www.cryptomuseum.com/crypto/usa/sigsaly/index.htm
7/ My 3 minute demo of the quantizer in action (pardon the amateur video quality, it was my very first video, done with limited equipment)
8/ In telephony, “phantom” is a circuit that uses common mode currents, for a second channel e.g. transmitted thru the center tap of a balanced line transformer.
https://en.wikipedia.org/wiki/Phantom_circuit
Its origin was in the 1800s to add a circuit on a telegraph line. The 4 wire SIGSALY secure circuits used noise and phantom circuits. They connected a SIGSALY terminal to a remote telephone e.g. Churchill War Rooms to SIGSALY miles away in the basement of Selfridges Department store.
Again thanks to all for your fine comments!
Jon Paul
ARCSI.FR
Paris
PS: Can we please keep the fine “Schneier on Security” Blog a politics free zone and focus on the security topics? Thank you for your consideration!
Alyer Babtu • February 16, 2019 11:29 AM
Re politics in comments
The blog seems to me happily to be socio-technical, concerned with not just pure technical security but also the social context and social impact of security and computing. Inevitably politics will enter, as it is a social art as well, interacting with security and computing.
bttb • February 17, 2019 7:53 AM
@Jon Paul, Alyer Babtu
Overall this blog seems to try to keep a lightly moderated off-topic discussion of security and politics to the current Squid threads, rather than in posts of things like “Reconstructing SIGSALY”.
For example, a quote from Tamir Pardo, the former Mossad director, in last weeks Squid https://www.schneier.com/blog/archives/2019/02/friday_squid_bl_662.html#c6788886
This weeks Squid is https://www.schneier.com/blog/archives/2019/02/friday_squid_bl_663.html#comments
Bruce Schneier • February 17, 2019 12:23 PM
@Clive Robinson:
I just deleted a chunk of one of your comments, above. Please keep off-topic politics off this blog.
herman • February 19, 2019 1:14 PM
The complexity of the Sigsaly encryptor is amazing.
I like to build simple radio circuits with 1 to 5 thermionic valves. Long ago, I worked with a Radar system that amongst other things, implemented a full adder using valves.
Calibrating and setting up complex circuits like that is not fun at all – it is hard work.
Anyhoo, my workshop has a rubber floor mat and I always solder bleed resistors across the PSU capacitors to discharge them quickly after power off. Getting zapped with high voltage is something you never forget.
Jon PAUL • February 20, 2019 5:10 AM
Hi there!
Off topic: Bruce, many thanks for decimating the off topic comments.
High Voltage: I have worked on HV as a power electronics consultant, to 55kV. Lethal to make mistakes. Nikola Tesla kept one hand in his pocket and used rubber mats, etc.
To Jure Juric Many thanks to point out my mistake of IT security in the link!
http://crypto-museum.org/QUANT/ has been scrubbed and ALL files you have accessed were non-public and (c). Please be so kind to delete any info downloaded, apologize for my mistake!
The only files that should be public are French and English 2 min videos of the demonstrations!
http://crypto-museum.org/QUANT/MOV/QUANTEN.mp4
http://crypto-museum.org/QUANT/MOV/QUANTFR.mp4
Many Thanks to all for your fine comments!
Jon Paul
ARCSI.FR
Paris
Subscribe to comments on this entry
Leave a comment
Sidebar photo of Bruce Schneier by Joe MacInnis.
wiredog • February 15, 2019 8:41 AM
“the debugging was very challenging, especially with 350 V present”
I’ve been bitten by voltages like that a couple of times. It really gets your attention. Nothing like sitting back while waiting for the arm muscles to stop twitching and thinking “You know, if that had gone across my chest…”