NSA on 10-17-2012 ursuantto E O 13526 MDR Case # 5477 l DWVV £iJ l GJ LrlIlUiV fil 3lI1Uifil 3 fil 00 3 i l 3 f OO i Ui l1wWl GJ£iJVV0GJ£iJl1 1st Issue 1991 i I 1 i - T F' I xd I • Y U • T tL -__ I I I GISTER • • • • • • • • • • • • BULLETIN BOARD • • • • • THE DIGITAL DRAGON • • • ICH SPRECBE DEUTCH • • • • W T N UT ERRATA • • CRYSCO • 90 LETTERS • • • • • NORMALIZATION • in re PROFORMA • CROSSWORD PUZZLE EDITORIAL • • TO CONTRIBUTE • • • ' ARC L IL • • • • • BUSMAN I S • • • • • • • • • • • • • • • • HOLIDAY • • • • • • • • • TillS 99€lH UlNT C9NTAIN8 CODEWORD MATERIA TOP SECRElNOT RELEASABLE TO • •• • • 12 32 I • • I J 13 lm m um 16 1 ••••••• ••• I I i ' L 86- 3 6 • • • 22 ---JV '23 • 24 29 '3205 I •RLW • 31 • 32 • 33 CLASSIFIED BY NSA CSSM 123-2 DECLASSIFY ON Ori iI l8til l A eAey's DeterfAiAatiol l Re tJired EO TRAETORS DOCID 4019720 CRYPTOLOG f ul lished by Pl Tf lchniques and Standards VOL XVIII o l ' PUBLISHER • ' ' 1st Issue 1991 1 _ BOARD OF EDITORS EDITOR •••••• •••• Computer Systems Cryptanalysis Cryptolinguistics Information Resources Information Science Information Security Intelligence Reporting 1 1 963-1103 • 1 ' ' ' • ' ' • J • ' ' • I ' '- ' • 1 •• • ' 963-1103 963-5238 963-4382 963-3258 963-3456 ' • i 972-2351 ' • • 1 963-5068 L g e •• 963-3057 Lingwstlcs •• • 1 963-4814 Mathematics ' • 1 963-5566 Puzzles 1 963-1601 Research and Engineering • 1 968-7315 Science and Technology 1 1963-4958 Special Research • Vera R Filby 968-5043 ' § 963-5463 Classification Officer • Bardolph Support Clov er Support • ' Macmtosh Support l1lustrator To submit articles and letters please see inside back cover For New Subscription or Change ofAddress or Name MAIL name and old and new organizations and building to Distribution CRYPI'OLOG PI NORTH or via PLATFORM cryptlg @ barlc05 via CLOVER cryptlg @ bloomfield Please do not phone about your subscription Contents ofCRYPI'OLOG may not be reproduced or disseminated outside the National Security Agency without the permission ofthe Publisher Inquiries regarding reproduction and dissemination should be directed to the Editor All opinions expressed in CRYPTOLOG are those of the authors They do not represent the official views of the National Security Agency Central Security Service Fait aFFI€fkb USB 9HI J¥ 963-3369 963-1103 968-7315 963-3360 DOCID EO 1 4 c P L 86 '36 4019720 I GISTER I I P L IG52 86-36 1st Issue 1991 CRYI'TOLOG page 1 'l'9P SBORfi'f' UMBRA P L 86-36 EO 1 4 c L 86-36 U o DOCID 4019720 'I'OP SBOIll3'1' UMBR 1st Issue 1991 • CRYPTOLOG • page 2 'l'OP SBORBr£ Ul 1Bfl A DOCID P L 86-36 EO 1 4 Ie 4019720 1st Issue 1991 CRYPrOLOG page 3 tF8F 8f 1€RE'f UMftItA DOCID 4019720 Issue 1991 page 4' DOCID 4019720 P L 86 36 1-39 Issae 1991 page 5 DOCID P L 86-36 EO 1 4 c 4019720 1st Issue 1991 CRYPI'OLOG page 6 'l'OP S136RE'I' UMBIM DOCID 4019720 'fOP Sl JCRE'f UMBRA P L 86-36 EO 1 4 c - 50 - At an informal ceremony held in PIon 5 February 1991 a small group met to commemorate the passage of 50 years since Bill Lutwiniak started to ply his arcane trade of cryptanalysis in the sanctuary of the organization-now NSAthat was founded on and is nurtured and sustained by this discipline Bill retired in 1981 and has worked as a reemployed annuitant since then He continues to make significant contributions as a consultant to PI solving systems himself and more importantly assisting our younger generation of cryptanalysts Bill served for many years as P L 86-36 chief of A5 then of PI and is a distinguished member of both Kryptos and the Crypto-Mathematics Institute At the ceremony a crossword puzzle elsewhere in this issue was presented t o 'Bill turning the tables on this well-known crueiverbalist Later this year a series of interviews with Bill jwiUbe released on conducted b videotape I 1st Issue 1991 CRYPrOLOG page 7 'fOP SI CRE'f UMBRA I aCID 4019720 L 86-36 SFJCRE'i' J •• I N 'I'-Tlae FlJtlJre tl I••telligcllcc·' Y •• II III T U Picture this A room full of analysts repreone was worn out and it was replaced with the senting practically every intelligence element other new one Got the picture located anywhere in the Washington area and U I have just completed my assignment at the beyond folks from NSA DIA AlA AFIA NOrc Joint Intelligence Center JIC Pentagon in NPIC ITAC SRD ESC but none from CIA or support of the Secretary of Defense and the KAOS Uniforms from the Army Navy Marines Chairman Joint Chiefs of Staff during the and Air Force all side by side with civilians mixed DESERT SHIELD and DESERT STORM Operaamong them tions Icame away from there with a very good understanding of how all the different intelliPC word procesgence agencies contribute to the Joint Operations sors on many desks volumes and volumes of reievent textual materials all brought together by INTelligence JOINT jricture More importantly I became aware of how this a gency is perceived by the individual who believed that it was needed others and how attitudes affect our 'Validity and just in case and more map boards with push how knowledge affects attitude 'P L 86- 3 6 pins acetate overlays boundary lines and arrows I I I than anyone could have followed intelligently anyway Secure phones and STU-Ill's next to each other Classified conversations in this hectic environment in the background of calls home Fax machines that are tempermental and open phone lines to the various commands in Saudi Arabia Clocks set for other parts of the world with all their minute hands pointing to different minutes A few small fans blowing stale air around staler bodies CNN monitors of various sizes in different parts of the room without logic to their placement With these 24-hour operations nothing was turned on and off except the copier when the new CU I was in the first group ofpaople selected for duty at the Joint Intelligence Center in August '90 That first Satu daywhen we all met was surely a hectic day And right from the start it was evident thatNSA had sent its best Byafternoon the original dozen of us were dubbed i I lWeasels IuSN was in charge of the group an enthusiastic group eager to get to the task U We all attended many meetings We were even briefed on 1st Issue 1991 CRYPrOLOG page 8 SBOIH3'£ IIMfDLE Vf A COMHffI CffMoHfflLS OffL¥O 1 4 P L c 86-36 DOCID 4019720 P L 86-36 the other wars in the area Thanks to the professional experts at NSA who had prepped us we arrived at JIC very well informed Getting the necessary working aids phone numbers and office supplies was very reminiscent of a scavenger hunt A tour of the JIC facilities still under construction and the Pentagon badging processes completed our preparations U The Sunday of Labor Day weekend 1990 was the first day on the job in the JIC I was first on the ground desk which I thought was appropriate anyway since I was the only Army person on the NSA team It also allowed me to set many of the desk policies and procedures used throughout the lifespan of the JIC There were a few growing pains as each member rotated through the desk for the first time But being true professionals we adjusted The housekeeping -arrangement of files desk areas-was better than I expected and the equipment held up remarkedly for the abuse we gave it Transportation to and from the Pentagon was provided for us and was truly a life saver George and Steve our drivers on the night shift were especially friendly courteous and safe Participants in JOINT AFIA Air Force Intelligence Agency AIA Army Intelligence Agency DIA Defense Intelligence Agency ESC Electronic Security Command ITAC Intelligence and Threat Analysis Center AIA NOIC Naval Operations Intelligence Center NPIC National Photo Interpretation Center NSA National Security Agency SRD Special Research Division ITAC U Each team in the JIC had its own personality Of course I like to think that the team I U In the first few days of the DESERT SHIELD was on was the best The only real problems I operations when the Joint Intelligence Center was a saw were personnel changes and role defininewborn organized chaos was the best descriptor tions Noone on the desk really understood People from all these different agencies were just what the NSAer's role was as a part of the team coming together sizing each other up establishing I envisioned my role as a key playedn the boundaries assuming roles -all the routines that surface when two or more clans come face to face Shy and quiet would eliminate you from the team discussions If you wanted to be a part of the team you had to push into it Once that was done however you were in for life Duties were created as This last duty is what all the other members of they were deemed necessary Division ofefforts the team thought my job was until they were were naturally established based on the most likely convinced otherwise source of the information required The NSA desk JI had t o was responsible for all the SIGINT material the insert myself into the whole teamagain 13ut imagery guys for imagery etc after awhile I was able to steer some discussion I and even successfully suggested tasking based on my SIGINT and JOINT understandings This tasking provided positive results too I never felt at any time that anyone on the team w a s o v e rl y i n te r ecst e d i n th e te ch m · c al as pe c ts P L 1st Issue 1991 CRYPTOLOG· page 9 JRE'F 8 6- 3 6 6 IIMffilA'3 VTA COMUff CIIMfttf8LS ftt l Eo 1 4 c P L 86-36 DOCID 4019720 St CIUl't' the Klieglights coming over the NTSS But it was helpful to me to have them I could speak my native SIGINTese to the NSA ground forces analysis team based on the tech details I had seen in the KLIEGLIGHTS mission as it grew on the Ground Team was to provide all-source intelligence estimates to Rear Admiral McConnell the intelligence officer answering to GEN Powell and to Secretary Cheney All source meant all source too It also meant that efforts would have to be combined A joint effort from each of the intelligences was needed Not only that based on these efforts a reasonable assessment of what happened what is happening and what will be happening by whom and when and how and etc had to be presented each morning to the Admiral Also we provided support to military operations directly in theater at Riyadh Saudi Arabia CU Wisely two only routine reporting vehicles were established rather than the many one might expect One was the Military Situation Summary This was designed to keep the field elements apprised of the current order of battle and the JIC's assessment of events relative to the order of battle The other regular publication the Defense Special Assessment published significant tactical and strategic intelligence This vehicle had a much more limited audience and was sometimes focused for the Washington area alone CU In a matter of minutes usually all the national-level intelligence in the community could be brought to bear in a cohesive complete and conclusive picture for the warfighters and decision makers upstairs There was little interaction with parent agencies No one at NSA had to say Wait till the day shop comes in and we11 get you an answer The same held true for every other agency The around-the-clock operations and the on-hand expertise from all the different agencies could usually handle the task quickly and efficiently I really believe that this type of operation was a key element in the successes in joint operations planning and in theater and because of it we contributed significantly to the field commander's nearly instant knowledge of the battlefield immediately in front of him poum From time to time my connections with the fort were tested At the beginning I had carte blanche access to the NSA ground deskjust as the Director had intended Then for some reason that was nearly choked out and without warning I discovered later that the choking of access came from a single source and that soon 1st Issue 1991 CRyPI'OLOG page 10 EO 1 4 c P L 86-36 SECRE't' IkkNBl K Vltl OOMIN'f OHMHJELS Ol'JLY D ¢P6 4019720 SIlCRIH was obviated 4t least with most of the military members on the desk my rank helped to keep lines open The sergeants did not really want to tell the warrant officer that they were D otpermitted to talk to him Good choice on their part Don't take me wrongly there were many professional analyst who understood the magnitude of the mission and readily assisted in spite of some of the inane management directives FOUO Back to JOINT- Joint Operations INTelligence JOINT provides probably the quickest all-source analysis of events to the national decision-makers And it does it very well The down side of what the JIC could do dealt mostly with the classification sensitivities especially for SIGINT Reports that were ORCON often contained the most usable intelligence for the field commanders except that the field commanders couldn't receive the reports Attempts at stripping out the relevent parts and removing ORCON caveats was almost always a time-consuming thankless chore But we did it We also got angry when the field reported an ORCON TACREP in their summary as SECRET Releasable to Multi-National Forces Yes that happened too There were moments in the heat of the battle as they say when your conscience made the ORCON release decision for you Then the paperwork followed 1st Issue 1991 CRYPTOLOG page 11 S GRE'I' IIMffiLB VTA 001wIHff' OUMfUi3L8 mfIH EO 1 4 c P L 86-36 SECRET agency represented and an appreciation for the process of producing finished intelligence at those agencies What we did was take all the work of many many people and consolidate it encapsulating for the folks upstairs the meat of the matter in a complete picture What each of us did was provide our own expertise in the intrepretation of products for those who may have incorrectly intrepretated or misread a product My many explanations of how the SIGINT processes work changed attitudes and that in tum changed the validity factor - - - - ---1 U In light ofthe huge successes of Operation Desert Storm what happened in this first-of-itskind JIC probably indicates the future ofintelligence for the battlefield and definitely for Pentagon and the White House I think many of us agree that future conflicts will necessarily be joint service ifnot combined national ventures And the one in charge of these ventures will have to be well versed in the intelligence community and its capabilities It was smart to put the JIC together and the JIC provided tons of information where and when it was needed An intel-smart General or Admiral could appreciate the JIC response to his needs and temper it appropriately in his planning The use of national resources real-time during warfare operations can often provide a broader perspective and analysis of current events than a smaller less equipped more narrowly focused in-theater service asset can provide And the JIC proved it could be done with very good accuracy and timeliness I do not believe the JIC portends the demise of our individual organizations merely it enhances the intelligence they produce and provides a complete nationally de1st Issue 1991 rived usually more sensitive intelligence information for our leaders to respond to today's fastpaced short-fused decision requirements U Finally I very much enjoyed serving in the JIC and feel I did contribute to our victories It wasn't as sandy as the gulf but I'd bet it was just as arduous I have been in the Army for over 20 years I always felt especially in the intelligence fields my mission was not to wage war but to preserve peace-prevent war I started my career in the Army in Vietnam I am ending my military career as this war ends As I look back I wonder if have done my job to keep the peace The successes of this war and the miniscule number of casualties would suggest that our intelligence was right on-that I did do my job and did it well However this euphoria is hazy What would have happened if the enemy really wanted to fight I never want my brother and my nephew who served in the Gulf to know Nor do I want my son to know in the future wars I want the next JIC to be just as succesful if not more so in providing the politicians the ammunition they need so we as a nation do not need to provide ammunition to servicemen and women BULLETIN BOARD SOUTHEAST ASIA AND VIET NAM 0 CCO 8-CCO The Center for Cryptologic History D9 is attempting to locate surviving materials pertaining to Southeast Asia and Viet Nam whether official records or private holdings The massive destruction of records in the mid-1970's has left significant gaps in the record of our efforts especially VHNC If you know of any material about our efforts in this region-records working aids etc -please notify D9 Ops 2B 963-7292 CRYPTOLOG page 12 StiSIQ T WMmbE Vb' OHl Hf'Pf' OHMfHEL8 OUffY DOClD 4019720 SECRoET P L 86-36 U The People's Republic of China PRe has undergone a revolution in its telecommunications field within the last five years This resulted from the decision to become a formidable world economic power early in the 21st century with help from industry in Europe Japan Australia and the United States But industry of the developed world found that communication via telephone or telex was hampered by outdated and overused switching and transmission equipment Nothing could be done until the many services involved in the complex chain of modern industry could talk to one other This meant that an integrated telecommunications system had to be established quickly including modem-connected computerto-computer data transfer telex and business facsimile traffic and be interfaced efficiently with the international telecommunications system through gateway facilities via communications satellite services and international undersea cables U Telecommunication was accorded a very high priority in the Seventh Five Year Plan 1986-1990 and has contined to remain so in the Eighth Five Year Plan 1991-1995 Innovation will be required to modernize to world standards U The Ministry of Posts and Telecommunications MPI' is the primary center for telecommunications standards long haul interprovincial systems communications research and equipment production But each province and each Ministry is responsible for establishing its own standards equipment purchases and telecommunications structure Except for the People's Liberation Army PLA all entities are to use European CEPT standards The PLA which does not need to meet international standards for its telecommunications needs initially adopted a 512 kbls mobile troposcattersystem TS193 and subsequently standardized with 460 8 kbls mobile microwave 204A and troposcatter GS207 III for flexible strategic communications requirements U The MPr has the telecommunications engineering and production expertise needed to sUj port its function Through the mid 1980's the Chinese used FMlFDM equipment almost exclusively however since 1988 microwave radio relay lines serving Beijing-Tianjin BeijingShijiazhuang and Beijing-Shanghai including route cities Jinan Xuzhou and Nanjing are being connected using CEPT Level 4 PCM 140 mh's equipment Current planning provides for major Level 4 microwave lines to connect Beijing with most of the provincial capitals in the populous eastern and the industrial central portions of the 1st Issue 1991 CRYPTOLOG page 13 eti IIMft VlA COMHff CHM ftffiLS OlftlI 4019720 country by 1995 Most of the provinces have installed and are now operating digital microwave systems interconnecting the key nodes within their borders East-West cross connection between Chengdl l and Shanghai will be accomplished using multimode optical fiber cable portions of which are currently under construction In addition a national digital 300-baud telex system using V 26 standard was initiated in the Spring of 1989 This telex system is replacing the older stacked Voice Frequency Telegraphy telegr am systems long the mainstay of PRC record traffiC SWITCHING CENTERS U The wealthier coastal provinces because they are more involved with foreign business mov ed rapidly to modernize their telecommunications Indigenous production of digital telecommunications was limited mainly to smaller size Levelland Level 2 equipment or to small PBX switching systems Some Level 3 microwave equipment was produced but its quality was considered poor in comparison to Japanese American and European products The microwave routes were established towers constructed and antennas bought in-country The digital telecommunications equipment was imported S-CCQ tr modern telecommunications structure - begins with a fast flexible digital telephone switching system The current standard in advance d industrial nations today is the Digital Stor d Program Control DSPC switch This form of telephone switch is electronic and controlledby software The combination provides the necessary flexibility to allow adjustments as conditions change as well as the electronic speed to handle large volumes of telephone traffic Cost of the switch is initially expensive 150-300 USD per line but over its life it is cheaper to operate than electromechanical equipment such as crossbar switches An extremely important consideration is that DSPC switching capability is critical for an Integrated Services Digital Network ISDN long the hoped-for dream of the telecommunications world As the technology to produce the large public DSPC switches did not exist in the PRC both provincial and private networks have to import them Among the equipments imported are FETEX-150 Fujitsu Japan DMS-IOIDM8-100 Northern Telecom NT Canada AXE-10 L M Ericsson LME Sweden System 1240 Alcatel France NEC Nippon Electric Corp NEAX-61 Japan EWSD Siemens W Germany 5ESS American Telephone Telegraph U In addition to seIling full switches to PRC customers Fujitsu and Alcatel established Joint Ventures to produce the components of their switching systems in the PRC Fujitsu based its FETEX-150 software support Venture in Fujian Province and Alcatel bought out the Shanghai based Bell Venture Shanghai Bell Telephone Exchange Manufacturing Corporation SBTEMC to produce System 12 System 1240 switches U Principle consumers for the DSPC switch market in the PRC include provincial and city level PTNB's the MPT the Civil Aviation Administration of China CAAC the various electric power regional authorities the railroad MOR and the People's Liberation Army PLA The PLA although not a large scale buyer is listed because it is a user on any systems established by the other functions U As of the date of this report all provincial capital cities and a number oflarge prefecture capitals have installed one or more DSPC switches into their telephone structure The initial entry was made by the Japanese companies Fujitsu and NEC They had a good sales infrastructure within China and provided their products at a lower cost per line than European American or Canadian companies Fujitsu was particularly successful greatly aided by its Joint Venture in Fujian province The Japanese however had delivery delays caused by stringent export license regulations As the money for telecommunications dried up companies who were able to swing the necessary long-term low interest rate soft loans became more competitive Even where the soft loans were available 1st Issue 1991 CRYPrOLOG page 14 S Gagr HMHlLH V-fA OSMUN OIIMHffiLS mH71 I'D c 86-36 4 01 97 2 a P L 86-36 EQ 1 4 c SKCRE'f in Japan however the long delivery time caused many Chinese customers to look elsewhere Owing to these factors the Canadian firm Northern Telecom is in a good po'sition for marketing its DMS-lOO line the desire to attract foreign investment the percent of the population in an urban or rural environment the wealth of the province and the innovativeness of the leadership As discussed above the initial step has been to establish a core system using digital Stored Program Control Switches in the provincial capitol and key large cities SUMMARY The infusion of digital technology into PRC telecommunications over the period 1986-1990 has progressed at a rate probably surpassing any other major country Progressing from having one Level three 34 MBs microwave line operational between 1981 and 1985 the PRC now hJs Level three d Level four microwave lines operating in most provinces and state of the art digital SPC switching systems in place and operational for most major cities Telegrams are now routinely passed through a nationwide digital 300-baud Telex using V 26 modem Connectivity has been established with world digital standards in the gateway cities via IntelSat and the way is being rapidly paved towards an ISDN environment within the next five to ten years if desired In terms of telecommunications the digital world is happening now in thePRC U 1 - '- Prouincial Post and Telecommunications U The manner in which the individual provincial Post and Telecommunications Authorities or Bureaus upgrade their telecommunications is dependent upon the degree of industrialization U China is truly becoming a Digital Dragon 1st Issue 1991 CRyPTOWG page 15 CItR't' 1' 1mt 'LE Vb eOMIlff CIIMeffiLS OtfLY DOCID 4019720 or U it i 't ibt toUtUttt P L 86-36 1L --_ _ 3t2 Back in the olden days of 1971 I took a Permanent Change of Station PCS assignment to Stuttgart Germany My sponsor picked me up at the airport That day she taught me the word mun If you know your dns tuti d you know that ntUn is German for nine By stepping up to the reception desk at the ' Vaihingerhof and trying to make my lips form that unfamiliar word I was able to get the key to Room #9 That room-iHmmt #9-would be my temporary home until I found an apartment As you can imagine knowledge of this one German word did not afford me an easy passage throughout Deutschland During my tour in Germany fortunately I increased my vocabulary I took German I and II offered on post through the overseasdivisio n of the University of Maryland I also took sever l German conversational courses With the kno wledge gained from these courses rw' able to rder meals at ' '· ng I restaurants reserve a room whil e traveIIi and determine ifthe room was With or without rsa bath and if it included j1' ttbst tttk breakfast etc After awhile I was also able to recognize when German was used incorrectly I went to dinner with a friend one evening During the course of the evening my friend asked the waitress Wit ist 1fb 1toilttttn The waitress graciously gave him directions to the men's room On his way back to our table my friend was grinning sheepishly because he realized that he had used the interrogative Wit instead oftno In so doing instead of asking where the lavatory was he had instead asked the waitress how her lavatory was Once on a German bus trip to Italy my fledgling German language ability came in handy The German bus driver who spoke no English and the tour guide who spoke fluent German and English got into a spat For days they would not talk to each other but instead communicated through me It was somewhat hairy at times but nonetheless a challenge I am sure that many NSA employees who have been on PCS assignments have similar humorous experiences with the language of the country Indeed I was totally unprepared linguistically for my first experience on foreign soil Now many years since my PCS I am happy to say that the Office of Personnel provides language tapes to PCSing families Often there just isn't time to attend a language class before you PCS no matter how well-intentioned you may be As anyone who has been PCS knows there is a myriad of concerns about your field tour where will you live what will the job be like what about schooling for your kids All ofthese changes are traumatic enough without compounding your discomfort by being unable to speak read or understand the language As most of our lives become busier more complicated and often plagued by traffic language tapes can fill a void By popping a cassette into the car tape deck you have an opportunity to learn a language during your daily work commute Why not have a leg up by at least learning some of the basics of the language such as yes no please thank you etc Then when you are approached by someone who asks you 1IJo ist t 1 Babnbo£ you can competently say tbts um bit tekt ann gt btaUS So if you are processing for an overseas assignment don't forget to ask the personnel in M31 Field Staffing and Personnel Administration about the language tapes The following tapes I are available from II M31 1st Issue 1991 CRYPrOLOG page t6 FOR OFFIOW UBI ONIJ¥ P L 86-36 EO 1 4 c DOCID p 408l 93120 celT standards for the V 42 have been approved and a number ofcompanies have committed to building their version of it The British firm HAYES already has a V 42 modem on the marketo A Matter of COMPRESSION I i U The impetus for the V 42 was the rapidly advancing state of the microelectronics art Industry realized that small relatively cheap modems could be built with a capability to perform complex functions requiring a high compute and large storage capacity The V 42 does its own framing and employs an effective error detection and correction by retransmission procedure U Because the V 42 can en- 1 - COMPRESSION CU The compression algorithm we are discussing is to be embodied in a new type ofmodem the V 42 sure that virtually error-free data is received a lossless compression algorithm was selected to be incorporated in the modem Unlike lossy compression algorithms which are error-tolerant successful decompression by this P L 86-36 lossless compression algorithm known as V 42 bis requires thatthe data received be exactly thatwhich was sent The V 42 bis is adaptable an advantage over most compression algorithms which are designed to compress a specific kind of data In the latter case if the data differs from the type for which the algorithm was designed performance of the algorithm suffers greatly IPll U In the V 42 codebook more generally referred to as a dictionary the entries are made dynamically representing character strings that occurred earlier in the current transmission so V 42 bis adjusts automatically to the nature ofthe data being sent If the modem is used for an Italian language transmission for example the dictionary will fill mainly with entries that represent character strings that occur with a relatively high frequency in Italian If an 1st Issue 1991 CRYPTOLOG page 17 G9NFFB DOCID 4019720 accounting report is sent most entries will represent number strings row or column labels or other items in the report format While the compression ratio achieved by V 42 bis is a function ofthe roughness of the data being transmitted it is not affected by the type of data This versatility together with its relatively low price and small size is likely to make the V 42 attractive for a wide variety of communicators when an entry is made for codeword 1024 the bits sent per codeword increases to 11 etc D During a transmission the compression algorithm looks at the consecutive characters to be communicated Each time it adds a character to the string it is currently processing it checks the dictionary to see if it contains an entry that represents the current string Ifit does the algorithm adds the next character to the string If it does D The CCIT standards for the V 42 are flexible not the algorithm outputs the codeword that enough to allow manufacturers to build models of represents the current string minus the last charthe modem that range from stripped down to acter added and creates a new dictionary entry full bells and whistles Maximum dictionary size that represents the current string including the may extend from 512 entries to an unspecified last character If the dictionary is not yet full the upper limit Maximum string size i e the greatcodeword assigned to this new entry is the next est number of characters that may be represented one available If the last entry in the current by a codeword may range from 6 to 250 Modems dictionary is 1000 the new entry will be assigned with differing capabilities will be able to communi- codeword 1001 cate with each other At the beginning of a D Say for example that the first two characters transmission the corresponding modems autoto be communicated are AR The algorithm would matically negotiate parameters Each proposes first check the dictionary to see if there is an entry a maximum dictionary and string size In each that represents the single character A There is instance the low bid wins since the modem with The initial dictionary contains an entry for each of the greater capacity can adjust to the lesser pathe characters in the transmission code The rameter proposed by the inferior modem algorithm would then add the R to its string and D At the beginning of a transmission the dictio- checks to see if the dictionary contains an entry nary contains 259 entries Codewords 0 to 2 are that represents AR It does not The algorithm reserved for control functions The remaining 256 would then output the codeword that represents codewords represent the characters of the transthe I-character string A and create an entry to mission code Each character is assigned to the represent AR Since the last entry in the initial codeword that is 3 greater than its transmission portion of the codebook is codeword 258 codeword code value The expected transmission code is 259 would be assigned to the string AR Since ASCII Since A 65 in ASCII Codeword 68 would an output would have been made to represent the represent the one character string A These A it would be deleted from the current processinitial 259 entries known as root nodes consti· ing string The current processing string would tute the fixed portion of the dictionary then be the single character AR to which the next character to be transmitted would be added D The number of bits transmitted to represent a and processing would continue codeword is the least number needed to convey the value of the last entry in the current dictionary D The dictionary grows in the form of a tree At the beginning of a transmission the last entry structure It might be thought of as a family tree is codeword 258 requiring nine bits to represent Except for the immortal root node primogenitors as a binary number As the transmission procreated by the CCIT Gods every dictionary entry ceeds the dynamic portion of the dictionary begins has a parent The entry inherits its parent's to fill When an entry is made for codeword 512 string to which is added one additional terminal the number of bits transmitted per codeword character jumps to 10 because at that point the dictionary contains an entry whose codeword value cannot be U Once the dictionary reaches its maximum size the algorithm cycles around to the beginning represented by 9 bits Later in the transmission 1st Issue 1991 CRYPrOLOG page 18 efflft'IBHf'fft Jcb DOCID 4019720 of the non-fixed portion of the dictionary which starts at codeword 259 and begins searching for leaf nodes A leaf node is a codeword that has no children i e there is no dictionary entry that represents its string plus one additional character When a leaf node is found its string is released and its codeword number is used for the next dictionary entry to be made This is a merciless algorithm that tells its creations Be fruitful and multiply If you are barren when next I pass this way I will zap you we capture the entries that were made from the point where the cut-in began We know the process We know that a dictionary entry was made when codeword 700 the initial codeword of our cut-in was sent We know the string to assign to the new entry is the string represented by its parent 700 plus the first character ofthe string Each non-root node has a parent ' The child represents the parent's' string plus 1added character or some months Dr Ralph Jollens ten and I have been developing tools to attack this cut-in problem Designing the tools has been a cooperative venture I t rn them into software routines and then Ralph has the fun oftesting them against simulated cut-ins that I create for him 1st Issue 1991 CRYPI'OLOG page 19 QMFJg EO 1 4 c P L 86-36 30 1 4 DOCID 4019720 PAL 86 36 151 Issue 1991 page 20 GQN-F-IDENIPLAJ DOCID 4019720 SECR E't'SP81tE BELL JOlt WJlOT6D TO nO BEll JlBOt T JlI C P L 1st Issue 1991 ' CRYPrOWG' page 21 SE TSPOIGl 86-36 P L 86-36 EO 1 4 c OC I D 40119·7 2tP 6 EQ 1 4 c BECKET Sr61ffi •••••••••••••••••••••••••••••••• ERRATA In 3rd Issue 1990 several lines were inadvertently omitted between pp 14 and 15 They should read I FILTERING AND CHARACTERIZATION • ••• Filtering is decision making Every act •• • within the SIGINT cycle in which items aDD 1 4 c identified and evaluated for retention or jek • 86-36 tion is an act of filtering The act can be mechanized or can be performed by humans who make the selection allocate the sensors for the collection activity and make other decisions about the SIGINT process • In 3rd Issue 1990 on pp 14-18 and 20 please change the issue to 3rd Issue 1990 CRYPTOLOG regrets the errors C6NFIBEN FIAI •••••••••••••••••••••••••••••••• 1st Issue 1991 CRYPI'OLOG page 22 SF3GRB SP6HH DOCID 4019720 P L OONPftlt lN'f'IXL 86-36 ICRYSCOM Exec P13 The Cryptanalytic Software Committee CRYSCOM was established in May 1984 as a volunteer organization to improve the use of computers throughout the entire cryptanalytic community by providing a forum where representatives of the CA computer complexes could exchange ideas In 1987 it was officially sanctioned when its charter was signed by the Office Chiefs of the major CA complexes at NSA CRYSCOM provides a formal mechanism for the collection and dissemination of information on computer EO 1 4 C P L 86-36 SUPPo1 _- - - -- -_--- --- - ---l The Cryptanalytic Software Conference CRYSCO is an annual event where current topics on cryptanalysis and computer science are presented Following are the recommendations stemming from CRYSCO-90 1 Software Availability Develop and promote a softcopy method of listing all available software on alI appropriate machines in the CRYSCOM community to include pending software also to include software which makes use of special purpose boards or devices b Encourage organizations to adhere to agreed upon standards c Promote the goal of a man page for every piece of software Status The number of software exchanges is growing and ground level satisfaction with and success using the procedures has encouraged the organizations to adhere to the agreed upon standards The goal of a man page for every piece of software· exchanged has been attained Feedback through the Software Points of Contact has been perceived as effective 3 CRYSCOM Character Message Format Resolve outstanding disagreement on header line tags Status Parties involved recognize the header line tag issue as a difference to be aware of as opposed to an outstanding disagreement The committee agreed to table this recommendation 4 X·Window Widget Sets Status Notesfiles software ad and software quest were created Some user organizations are using them while other organizations are developing their own library functions r 2 Software Exchange I a Continue efforts to establish and promote soft- ware hange·1 I Status A working gr uphas been formed and is evaluating toolltits P L 86-36 1st Issue 1991 CRYPTOLOG page 23 Ct»W19Ji llRlMJ OOMIUr pOIIMftffiLS ONLY nA ffiLEl VfA 4019720 5 Public Domain Software a Look for a method of obtaining public domain software b Establish a procedure to propagate error reports for public domain software back to originators particularly X-Windows Gnu products c Establish a procedure to propagate fixes back to the users Status R5 is setting up a mechanism for evaluating public domain software which will be available for use by community members After this evaluation process CRYSCOM intends to submit a list of specific software to T03 with a request for consideration Status CRYSCOM is working with W352 to arrange for a brown bag seminar on optimization This recommendation is one which will be an ongoing concern 10 CRYSCOM Official Papers Create a directory catalog of such papers to include CRYSCO recommendations position papers guidelines standards etc In short any voted on document Status Completed 6 System Administrators Organize a forum for the various system administrators especially those of the numerous SUN networks to discuss mutual concerns such as security issues response times network transparency etc Status Completed 7 Training P L 86-36 Lack of UNIX training is seen as a big problem in themov promote UNIX training such as through brown bag seminars Status Brown Bag seminars will be reinstituted to promote UNIX training and expertise 8 NQS Checkpointing and Restarting Promote the resolution ofNQS checkpointing and restarting problems Status A list of capabilities wanted in NQS has been supplied to Cray who will work to incorporate them in a future release 9 Software Development Tools To the Editor Though there's a lot of expressed concern about The language problem and keeping linguists happy management actions belie that concern Look where they located the Language Career Panel for one thing It's at Friendship Where are the linguists At Fort Meade Linguists have more in front of them than they can possibly do For a lot of languages there's a severe shortfall So why are they compelled to travel to Friendship in connection with career development Why does management allow this inane situation to continue Why don't the linguists rise up and protest AI Enquist Promote development acquisition and use of appropriate software development tools especially optimization tools graphics aids timing tools 'FOUO 1st Issue 1991 CRYPrOLOG page 24 G9Ni 19 Ih'rlffiLS lith eOMHR OIIA flH3L8 OnLY DOCID 4019720 SECRE't' ¥9rmafization 86-36 1st Issue 1991 CRYPTOLOG pageZ5 j eRM' EO 1 4 c UMHJbE MIA eo mlT CUP lSU ¥ 86- 3 6 DOCID EO 1 4 c P L 86-36 4019720 1st Issue 1991 CRYPl'OLOG page 26 SHCRH'f IlA H LE VIA CO mIT CIIAP' l8 og mlLJf EO 1 4 c F L 86-36 DOCID 4019720 SECRE'f 1st Issue 1991· ••••• • page 1 ' QIl T UAPffiLE YIA CO HPff CHMHffiLS OnLY EO 1 4 c F L 86-36 4019720 SECRET 1st Issue 1991 CRYPrOLOG page 28 SECRt3'f' ItAlfflU VIA COltHlft' CItAlfPffiLS ONLY P L 86-36 EO 1 4 c DOCID 4019720 SECItE'f CON4i'IBENTlAb To the Editor I react Withiritere'st the item oy Preston Cur ' riet on the subject of cryptographic Habits and Cryptanalytic DiagnOSIS an extract frODllris' 1969' papers CRYPTOLOG 3rd Issue 1990 ' I thOught you be be interested iii kri6wi'ng tHai 154- ArchIve's' lind RepositOry Service's 'is' a ftefupUrig to record historic cryptOgraphic irif6rfua iioii i e codebooks' ana technical' pa pel's fb'r- the period pre' i9'7'S' in 8 specilii electro me database reteire'd to' as the Crypfu graphic GO'des and Ciphers Coil'ect 6Ii' 'Phis' i nforma tloIi IS' unique and d6esn6t dupl'i'cate' ¢ Ainformation iIi other Agency collectrons 'f e · pi Coll ectiori etc 'TIns endeavor by TS'4' 1S qUi·te· massi 7e and wilf reqUIte consici ta15ie' time to complete 'however the iiritiaI- progress that has been'made to'dale nas proven extremely worthwhile and' naif kttracted' the interest or-manY cryptarialjstS fioni'various OPls' in' the' AgenCy ' I' am fakihg' the'liberty-offorwarding you t'wo recent comriium cati ris byl 16424 HIl'l after his researcll' int i vari Hig T54·trea ' sures and the results'oihis'efforts iri support of a' Current high interest crYPtosysteriithat he' was assigned r believe you' Will filid these of tnteres Should' you Wish a follow-up·tothet 1 ·article these comments by a'young ypteiDlath analyst' may aHirt'eother arialyst iri ·this disci ·plihe'tO-'the'wealth of ihfonniitlon' develbped by our predeces'sors arid now'l ecomingavlliikble ·iWT54 Archives an Reposit'Ory Se Wi es' and D9 ' NSA'C nter 'f r'Ctyptrilogt ili toiy o lst-I si le'1991 CRYProtOG page 29' Sf5€RE'f' 1' i lJfJ VI ' C6 H IT 'CI IMtP'£ 6S fft Y P L 86-36 DOCID L 4019720 86-36 L - IPOS SAO EO 1 4 c P L 86-36 L 4 c 86-36 86-3 o P L 86-36 1st Issue 1991 CRYPTOLOG page 30 Sl leRE'f ttAN'f LF l Vf A eOMUff' ClfMflfFJLS OffL¥ EO 1 4 c P L 86-36 -36 DOCID 4019720 This puzzle is dedicated to William Lutwiniak the founding publisher ofCRYPTOLOG himselfa noted puzzle-maker to mark a half-century's association with cryptology See page 7 Busman's Holiday 2 3 4 10 11 12 13 ACROSS 14 1 '50's chewing gum favorite 17 6 Word with fly and about 9 Enterprise- ing mate 22 14 Seen less often 15 Basketball great Blab 24 16 Give a full 10% 17 Wilder pronoun 18 Odd assortment 21 Oldie but goodie _ Then 22 Correlative alternative to 60-0 • CLOCKWISE 23 Generic Mrs lMiss 24 Marx missing from Monkey Business 26 _ you is or ain't you ain't 27 Syria's second-largest city 29 Now-stout Guinness 31 Entertaining Lola 33 Washington August forecast 54 35 Guitar-wielding Hendrix 36 Imperial contents 59 39 Man from Medina 40 Donny's little sister 63 41 A _ by any other name 42 Hero Jones at the SaO 68 43 The class of Hitchcock's creatures 44 Thomas or Bob 45 Detroit's Joe Louis and Baltimore's Baltimore 47 Pull a long face 48 Place name in the '60's news 50 Cerium's symbol 51 _ off angry a la Curtis Strange 54 A's pre-vowel substitute 58 Behold's partner 59 Anais 61 Settle the details 62 Coulomb per sec 63 Imperturbable one 65 Explosive stuff 66 Plains home 68 Down to 69 Opp of WNW 70 Angel-owning cowboy crooner DOWN 1 2 3 4 5 6 Flashback your President is not one A Dern-good actress Hockey great Fine day in 1945 Bombeck Drinks in air 7 Far and B Xiaoping 9 Brando cry in Desire 10 are not 2 11 Baseball great 12 Bonzo for one 13 Horse racing great The Iliad or The Odyssey Netminder Fast month in 27-A One who has done his time-partially Wilbur Post's pet Bakes pottery 32 Kingsley 33 Bali 34 Coffee maker 35 Contents of above in Indonesia 37 a matter of fact 38 Another Kingsley 40 Creche component 44 Ti's successor 46 Fortify bread 47 La carte 48 Saint-Saens Macabre 49 Baker who's Giving You the Best That I Got 50 $100 52 Siwwy Wabbit's foe 53 One of Doc's little buddies 55 Participate on the 2nd Tues in Nov 56 Holiday and Days 57 _boy 60 See 22-A 62 Likely to be good at 64 Stephen King's_ 67 Prefix opp to drs19 20 25 28 30 31 RLW 1st Issue 1991 CRYPTOLOG page 31 POR OPPIOfAiJ USE ONL ' DOCID 4019720 I Editorial Lessons Learned With a certain amount of euphoria some veterans ofpast crises were commenting on the most recent one in which they were but spectators They were pleased with the high interest in lessons learned It's important to record and act upon these they believe and the record should include the minutest detail of things gone wrong The causes of serious problems can often be traced to the snowballing effect of many minor errors that were disregarded From their distant perch these veterans noticed the beginnings ofchanges in some of the operational elements that were not directly involved in the crisis but that contributed people and material resources Somehow those elements found ways to function more efficiently with fewer people Maybe they too should be included in a lessons learned In a leaner operation there was no time to pay tribute to arbitrary prerogatives Surely'the newly streamlined mode deserves close study and perhaps emulation elsewhere BULLETTN BOAlUO I RECOVERING DELETED DATA ON THE SUN F'OUO R522 found a way to recover deleted data on the SUN A typing mistake had resulted in the deletion of an entire directory of files dealing with the TANCIL software The SUN people and other experts estimated that the chances of recovery ranged from poor to none F'OUO But R522 experimented with a method that proved to be successful in recovering files especially source code All you have to do is lock the file system long enough to make a copy to work with amine Using this technique R522 was able to recover most of the TANCIL software For more information R522 caul 968-8411 P L 86-36 Notice to Subscribers ----- CRY P T 0 LOG - -- --- - _--- _ Solution to Puzzle on page 31 Distribution for this issue reflects changes received by COB 13 May 1991 If you move or are reorganized please notify CRYPrOWG DISTRIBUTION PI NORTH Be sure to indicate both your old and new organization and building CRYPrOLOG page 32 6ft 6FFICtA L USE 6NiN 1st Issue 1991 DOCID 4019720 CRYPTOLOG Editorial Policy CRYPTOLOG is a forum for the informal exchange ofinformation by the analytic workforce Criteria for publication are that in the opinion ofilie reviewers readers will find the article useful or interesting that the facts are accurate that the terminology is correct and appropriate to the discipline Articles may be classified up to and including TSC Technical articles are preferred over non-technical classified over unclassified shorter articles over longer Comments and letters are solicited We invite readers to contribute conference reports and reviews of books articles software and hardware that pertain to our mission or to any of our disciplines Humor is welcome too If you are a new author please request Guidelines for CRYPTOLOG Authors How to Submit your Article Back in the days when CRYPTOLOG was prepared on the then state-of-the-art a Selectric typewriter an article might be dashed off on the back of a used lunch bag But now we're into automation We appreciate it when authors are too N B If the following instruction are a mystery to you please call upon your local ADP support for enlightenment As each organization has its own policies and as there's a myriad of terminals out there CRYPrOLOG regrets that it cannot advise you Send two legible hard copies accompanied by a floppy disk or cartridge as described below or use electronic mail In your electronic medium floppy disk cartridge or electronic mail please heed these strictures to avoid extra data prep that will delay publication • do not type your article in capital letters • do not right-justify • do not double space between lines • but do double space between paragraphs • do not indent for a new paragraph • but do paragraph classify • do not format an HD as DD or vice-versa-our equipment can't cope The electronic mail address is via PLATFORM cryptlg@ barlc05 or via CLOVER cryptlg @bloomfield CRYPTOLOG publishes using Macintosh and Xerox Star It can read output from the equipment shown below If you have something else check with the editor as new conversions are being added Be sure to label your floppy or cartridge as to the hardware density format and software you used Don't forget your name building organization and phone HARDWARE MEDIUM SOFTWARE FORMAT SUN 60 or 150 ME cartridge ascii only TAR or RAW XEROX VP 2 0 5 1 4 floppy only nla n a MACINTOSH 3 1 2 DD disk only MS WORD MacWrite TEXT WriteNow n a IBM Compatibles 3 1 2 1 2 ME disk 5 1 4 DD or HD floppy MS WORD WordPerfect WordStar ascii DCA IBM revisable DOS WANG nla nla n a 1st Issue 1991 CRYPI'OLOG page 33 PeR 8PPfCfA L USE 8NL¥ DOCID 4019720 IIIS BOCtJflftEN CONTAINS €OBI3WORB MATI3RIAI i f9P-SEEREfNOT RELEASABLE TO CONTRACTORS