DOCID 3112865 U Cryptologic Almanac 50 th Anniversary Series · o 1 4 c U Collection and Processing at NSA TShlgI It was mornings like this that made you really appreciate an overseas assignment I was working mids in the spring of 1965 a lone year into a fIveyear PCS Along toward dawn you would hear through the open window the sound of a cowbell Soon a sheepherder with his flock led by a bellwether would go past en route to the antenna field to graze Nothin like it at Fort Meade Sure five miles awa '--_ Iwere probably getting ready to start thekiIling again but at that moment in that place peace and the sound of the bellwether's bell prevailed eNS I By the beginning of the 1960s radio communications worldwide were overwhelmingly high-frequency manual Morse HFMM and high-frequency radioprinter HFRP and U S collection capabilities were geared toward this There was a large overseas establishmentof intercept stations which aimed at intercepting all HF communications ofinterest NSA and the SCAs Service Cryptologic Agencies had intercept stationsin countries and the Continental United States Alaska Hawaii and I I C L L8I Rhombic antennas had dominated the 1950s These were used as the standard collection antennas because the gain exceeded other antennas by up to five times but a rhombic field at a major site could occupy 100 acres Atl 'ocal fanners used the rhombic field to graze sheep Where space was a problem log periodic antennas were also used In most cases stations used a number of different types of antennas In addition to the rhombic antennas we had amonQ others a large log oeriodic I The rhombics were connected by coaxial cable to the intercept building Once inside the buildin the cables terminated in an RF distribution room I 'Operators in the early days often used Hammarlund SP-600 or RYCOM receivers The SP-600 was an analog receiver without any sort of di gital readout so the operator had to guess the frequency In the 1960s the Watkins Johnson R-390 which had digital readout with manual tuning became the standard Declassified and approved for release by NSA on 27 February 2007 pursuant to E G 12958 as amended MDR-51909 DOCID 3112865 U1 FOUO An improvement in collection was the introduction in the 1950s of the Wullenweber or Circularly Disposed Antenna Array CDAA This design pioneered by the Germans in WWII was first used in U S SIGINT operations by the U S Navy which was primarily interested in the CDAA's DF capabilities The Naval Security Group NSG began its systems R D work in 1956 and fielded its first CDAA an FRD-10 in 1962 By 1966 NSG had built 13 FRD-IOs in three foreign countries the U S and its territories C ' lSI7 In the 1950s the Air Force Security Service USAFSS began designing from the ground up a new collection system based on the Wullenweber finally included two components GLR-1 and GLR-4 After extensive modifications theGLR-1 was fielded as the FLR-l2 The FLR-9 antenna of the GLR-4 system was with a circumference of 1 200 feet the largest single antenna ever designed fo l8IldEO 1 4 earned the cognomen of elephant cage EO 1 4 P L Ctt81 - While improvements were being made in antennas there 'Yasalso improvement in the actual means of copying traffic After WWII intercept operators used standard typewriters to copy manual Morse traffic on continuous-form a er '-'-_ _--- - - -- -_ _- -- --- These typewriters were modified to type letter text only in uppercase Hard-copy was forwarded via courier while particularly important items wl reforwarded in TECSUM Technical Summary format via regular communica ior1s circuits I I 8 51 - The first attempt to improve the copy and forwarding system was called The equipment consisted of a modified Remington-Rand typewriter with special keys to indicate such traffic components as callsigns preambles and case notations The new feature was that it could output both hard-copy and a seven-level paper tape which was used to input the data into communications equipment This was replaced by the Teletype Model 35 extensively modified with the addition of the special tagging keys NSA named the device the AG-22 and changed the output to eight-level tape These tapes could then be introduced directly into the OPSCOMM system The plan was to install these at every HF field site Unfortunately the OPSCOMM system could not handle the volume of data this would produce It operated at only 60-100 words per minute and was already being swamped by other requirements such as TECSUM forwarding The solution developed in 1968 and first installed in December 1969 was STRAWHAT a 9600-baud data link system from field sites to processing centers It used a time division multiplex system capable of up to eight-level forwarding and could be patched directly from the circuit terminal to a computer I I 51181 The AG-22 STRAWHAT system however caused other problems A large field site with many manual Morse positions produced a considerable amount of eight-level tape and handling this tape at the station could be a major problem In the mid-60s NSA c d 86-36 DOCID 3112865 began working on a system for accepting manual Morse data directly onto a magnetic tape The method chosen called IATS Improved AG-22 Terminal System called for wiring up to 128 intercept positions to a Honeywell 316 computer Intercept files were packed onto a magnetic tape to be transmitted periodically on a high-speed data link to NSA Even with this there were sometimes problems During a major crisis when events were breaking rapidly the periodic transmissions of data only every four or six hours caused an unacceptable delay in processing In one case the station started converting the tapes to five-level paper tape which was then sent via the CRITICOMM circuits originally designated for the transmission of critical information or data Using this procedure the elapsed time between intercept and publication of the item to users was sometimes less than two hours On the other hand toward the end of that crisis the field site requested immediate notification when they could conclude this procedure as they were hip-deep in paper tape CAtS I Once the data were received at NSA whether by AG22 STRAWHAT or by IATS they were immediately accessible for processing Plaintext traffic EO __ _ _ _----' were passed on to the appropriate analysts via internal NSA electrical circuits Encrypted traffic was transmitted directly either to the appropriate computer for solution or to a cryptanalyst for study L - _ --- - ' As a result the U S Air Force started using airborne -in-t-er-c-ep-t-c-a-rr ' ie-d- ---m-o-st- 'ly--- in-C' -1 3 0 aircraft The AirForce had IRC-130s emphasizing voice collection and in 1963DRC-135s withlimited ELINT capability were added to the fleet During the 1960s the Strategic Air Command SAC used RB-47s with limited ELINT capability and the Navy used EC-121s and P3 OrioDs in support of naval operations 7 UHFOUO Another method for carrying intercept to the enemy was on board a ship The first Technical Research Ship TRS was the Oxford which put to sea in 1961 Oxford was a WWII Liberty ship configured withDCOMINT andr JELINT positions The TRS program was originally designed to provide additional intercept in the eastern Mediterranean but Oxford was diverted almost immediately to Cuba The next ship in the program was the Valdez a former Military Sea Transport Service MSTS vessel withL positions also launched in 1961 The Navy ships Oxford Georgetown Jamestown Belmont and Liberty were designated USS and had Navy crews the MSTS ships Valdez and Muller were designated USNS and had civilian contract crews with Navy SIGINT personnel In 1968 a third MSTS ship was added the Pueblo 1 4 c DOCID 3112865 EO 1 4 c S JSI There were no further major advances in collection or forwarding until the introduction of space satellites as collectors in 1967 These satellites were used orimarilv to collect ELINT J IThe next step in the usage of sa' 'lt eTTIll - 'lt-es- -r-em---o tl-n-g -w-OU 'lIMld -p-r-o-ve '7' to b'e-ev-e-n-m-o-re-e--'xpensive but promised great savings in the long run 5 H The remoting concept involved the establishment of unmanned intercept sites in safe areas which would then transmit the intercepted signals to remote ooerating facilities ROFs in the United States V FOVO But there are no sheep UHFOUO David P Mowry Center for Cryptologic History 972-2893s dpmowry Almanac 50th Anniversary Series Content Owner F eedhack Web Poc Feedback Last Modified by nsr Last Reviewed February 28 2003 Next Review 365 days -eEIW OM NSAlCSS MOHUAL 12 3-2 ----'-' TED 2 4 1998 OB LASSIFY ON Xl