Soviet speech scrambers

Apart from codebooks and ciphers machines the major powers of WWII also used speech privacy systems to protect their sensitive radiotelephone communications.

The American A-3 device, built by Bell Telephone Laboratories , was used for the link London-Washington. Unfortunately for the Allies it was solved by two German teams .One working for the Army Ordnance, Development and Testing Group, Signal Branch – Wa Pruef 7 and the other working for the Post Office.

The successor of A-3 was the SIGSALY device which offered absolute security. The drawback was the huge size of the equipment .According to the  NSA  website it weighed 55 tons !

The Germans used devices similar to the A-3 but did not consider them secure (for obvious reasons). Michael Pröse mentions the Siemens & Halske   ’’Kleiner Leitungsverzerrer GK III’’ [Source: Pröse dissertation ,p172] .During the war they tried to develop a secure system but did not go beyond prototypes.

The Soviet government was also concerned about the security of its voice communications. According to an article in Agentura in the late 30’s they developed a scrambling device called EU-2 which used speech inversion. A more secure system was developed by ’42, the Cobol-P. 

German Signals intelligence units in the East intercepted those radiotelephone conversations and according to Karrenberg scientists at Wa Pruef 7 were able to solve them at will  up to 1943. It seems to me that this Soviet device, which the Germans called  X, must have been the EU-2 or one of its modifications.

In 1944 however a new scrambling device was introduced which proved secure. The Germans called this X². Again by comparing with the  Agentura article it seems they were referring to the Cobol-P. By the end of the war however their research had identified the principle of the Soviet device (time division scrambling with wobble inversion) and they may have reconstructed some conversations according to Huettenhain.

Time to take a closer look at all this information.

1). From EASI vol2 – p73,

37.Early Russian ciphony was solved by analysis of spectrograms

Radio telephone conversations between Moscow, Leningrad; Irkutsk, Alma Ata and Tcheljabinks, involving Russian Army and People's Commissariats, up until 1943, were enciphered by two simple methods which were said to be easily solvable by German engineers at  the Army Ordnance, Development and Testing Group, Signal Branch.(Wa Pruef 7), according to Corporal Karrenberg, of the Signal Intelligence Agency of the Army High Command (OKH/GdNA) .These two methods of Russian enciphering were:

a. Inversion, employing superimposed modulation of several audio frequencies; and,

b. Distortion, by artificial raising of amplitudes of speech harmonics.

German scientists were able to solve these two simple enciphering methods by recording the enciphered speech, making spectrograms from the recordings, and analyzing them .Evidently the voice engineers could see the results of the inversion and distortion, on careful inspection, and could readily identify the frequencies and methods used for encipherment. They tried it: only a few times, according to Karrenberg, but were successful at will. At the beginning of 1944, however, the simple enciphering methods were dropped by the Russians, radio telephone traffic networks themselves were changed, and no further entry was gained by the Germans. Dr. Buggisch of the Signal Intelligence Agency of the Supreme Command, Armed Forces (OKW/Chi) studied spectrograms of this later unsolved Moscow-Madrid radiophone traffic at the Army Ordnance, Development and Testing Group, Signal Branch Laboratories (Wa Pruef 7) where he became convinced that Russian ciphony then involved time scrambling, with the length of the individual time segments being 10 milliseconds each, and a  synchronizing pulse occurring every .6 second. The number of "pickup heads" used by the Russians to obtain this time, scrambling was reported in one interrogation to be three and, in another to be four. German engineers were unable to learn any more than this from the spectrograms. They could reconstruct fragments of speech, they thought, but "the validity of the solution did not satisfy Dr. Huettenhain's critical sense," when shown to him. Dr. Huettenhain, who consulted with Dr. Buggisch, believed that some form of one-time strip might have been used to key the time transposition, as he could find no: period whatever in the encipherments.

2). From EASI vol4,p44

The four independent Stationary Intercept Companies assigned to work on the eastern front had the following assignments. To Feste 11 was assigned coverage of high-frequency traffic on the Red Army and the NKVD. Originally, this Feste was located at Winniza, latterly at Kiev. The other two Feste, 7 and 8, concentrated an special Russian traffic. Feste 7 was the Russian Baudot reception station at Minsk. In 1942-43 it was moved to Loetzen where it became part of Section 4 or the HLS Ost and continued to intercept Russian Baudot traffic. Feste 8 was the former Army intercept station at Koenigsberg. After 1942, this station concentrated on Russian wireless telephone traffic called by the Germans Russian X-traffic. Attempts were made to pick up this traffic by equipment developed by Army Ordnance, Signal Equipment Testing Laboratory (Waffenpruefung, abbreviated Wa Fruef 7). The channels monitored ran east of Moscow; the traffic was mainly economic. From 1942 to 1944, this traffic was successfully recorded; but after 1944 the Russians scrambled their wireless telephone traffic, and after unsuccessful efforts to intercept this scrambled type had been made, the monitoring was dropped.

3). From TICOM I-213 ''Report on interrogation of Alfred Muche'' ,p3

Russian Systems. 

17. MUCHE himself never worked on Russian systems but he stated that they were first recorded and worked on in 1944 by a party headed by LOTZE. In February 1945 MUCHE was visited in hospital by LOTZE who told him that the Russian system had been broken and that basically it was a TIGERSTEDT system ( =TDS), with wobble inversion. MUCHE could not remember, or was never told, the number of TIGERSTEDT heads involved; he thought it might have been seven. He did not know the period of the TIGERSTEDT key, or even if it had one.

4). From I-31 ‘Detailed interrogations of Dr. Hüttenhain, formerly head of research section of OKW/Chi, at Flensburg on 18-21 June 1945’, p12

5). From TICOM I-2 ‘Interrogation of Dr. Huettenhain and Dr. Fricke at Flenshurg, 21 May 1945’, p2

A. THEY HAD FINALLY A SECRET R/T OF WHICH ONLY A  SMALL PART OF A FEW MESSAGES WERE READ. THIS WAS PROBABLY USED IN DIPLOMATIC AND ECONOMIC CIRCLES. WE WERE NOT SURE BECAUSE ONLY BITS WERE READ. WE KNOW HOWEVER THAT THE RUSSIANS WERE EXPERIMENTING WITH IT.

Q. WHAT ARE THE PRINCIPLES OF THIS MACHINE?

A. SYSTEM IS KNOWN AS TIGER-STEDT AND IS THE SAME AS THAT USED BY AMERICAN MUSTANG GROUPS. SPOKEN PLAIN TEXT IS OSCILLOGRAPHED ON STEEL TAPE AND LENGTHS OF TAPE ARE CUT AND THEN REASSEMBLED.