Copyright IC) 1978 American Telephone and Telegraph Company
`THE BELL SYSTEM TECHNICAL JOURNAL
`Vol. 57, No. 2, February 1978
`Printed in U.S.A.
`
`Common Channel Interoffice Signaling:
`
`History and Description of a New Signaling
`System
`By C. A. DAHLBOM and J. S. R Y A N
`(Manuscript submitted May 7, 1977)
`
`A new interoffice signaling system known as the Common Channel
`Interoffice Signaling System (CCIS) has been introduced into the Bell
`System's DDD toll network. It represents a major step forward in sig-
`naling systems by providing high speed data links between processors
`of stored-program-controlled switching offices to carry all signaling
`and network control information, completely independent of the
`communication paths used by customers. As CCIS implementation
`proceeds it will have an expanding and significant impact on the DDD
`network system performance due to improved speed of signaling and
`provision of signals to provide a multitude of new network and customer
`services. The history of common channel signaling is traced from early
`mechanical implementation to use of present-day technological ad-
`vances. The fundamental concepts, basic features, signal formats and
`system operation are described.
`
`I. INTRODUCTION
`Until now signaling systems have to a large extent been provided on
`a per trunk in-band basis and generally have provided adequate per-
`formance for the present-day operating environment. The limitations
`of the systems have been enumerated in the lead article, and these to-
`gether with the requirements for higher-speed signaling and vastly ex-
`panded signal capacity have led to the introduction of common channel
`signaling systems for both domestic and international telephone sys-
`tems.
`The concept of common channel sig naling is not new but only recently
`have advances in technology made it possible for large-scale imple-
`
`225
`
`AT&T, Exh. 1008, p. 1
`
`

`
`mentation of such systems. As a result, new customer services and so-
`phisticated network controls requiring additional signals become pos-
`sible, all with complete independence between the transmission channel
`used by the customer and the channel used for signaling.
`
`II. EARLY COMMON CHANNEL SIGNALING SYSTEMS
`The earliest use of common channel signaling employed mechanical
`distributors and provided for multiplexing the signaling information
`for 30 trunks on one full duplex telegraph circuit.1 Eight such distributor
`systems, catering to 240 trunks, were placed on commercial trial in 1922
`and standardized in 1924. The trunks, which were installed between New
`York and Philadelphia, continued in service through the 1940s. Instal-
`lations in other cities were limited mainly because there were few trunk
`groups of a sufficient size to economically justify the system and because
`maintenance costs were high owing. to the mechanical implementa-
`tion.
`A second form of common channel signaling employed voice-frequency
`telegraph channels to carry the signaling information required for up
`to 18 trunks over one voice frequency circuit. This was, of course, an
`improvement over a much earlier plan of using a separate de telegraph
`circuit for each voice circuit. In the 1940s development of voice-frequency
`in-band and out-of-band per trunk signaling arrangements were un-
`dertaken, and they became the predominant methods used for interoffice
`signaling on carrier-derived trunks.2
`Anticipating significant advances in the available and future tech-
`nology, interest in and studies of common channel signaling were re-
`newed in the early 1960s. The approaches taken included proposals for
`signaling on a trunk group basis between markers, in the case of elec-
`tromechanical switching systems, and between processors, in the case
`of electronic switching offices. The trunk group concept catered to from
`12 to 60 trunks. In this approach, supervisory signaling was assigned to
`the common signaling path while the address.information was trans-
`mitted over the individual trunks. An advantage for such a division of
`the total signaling information was that the continuity of the speech path
`was checked by the successful transmission of the address informa-
`tion.
`Technical and economic studies indicated that if common channel
`signaling were to be implemented, it should provide for vastly improved
`signaling speeds and signal capacities for large numbers of trunks and
`should cater to future needs, then under consideration, as well as tofu-
`ture needs still in a dreamer's mind. Since the future switching system
`hierarchy was to be electronic with the offices being processor controlled,
`the concept of using high speed data links between processors was chosen
`
`226 THE BELL SYSTEM TECHNICAL JOURNAL, FEBRUARY 1978
`
`AT&T, Exh. 1008, p. 2
`
`

`
`as the most appropriate approach. Such studies at Bell Laboratories,
`together with concurrent studies and participation in discussion of in-
`ternational signaling needs in the CCITT,* led to the specification of the
`common channel interoffice signal-ing system (ccrs) for use in the Bell
`Systems DDD network and a similar system known as the CCITT signaling
`system No. 6 for international and intercontinental signaling applica-
`tions. Such systems were to carry all supervisory and address signaling
`information as well as a wealth of signals designed to cater to special
`services and network control features on the interprocessor data link,
`completely independent of the circuits used by the customer.
`
`Ill. ccrrr STUDIES OF INTERNATIONAL COMMON CHANNEL
`SIGNALING SYSTEMS
`With the expansion of semiautomatic and automatic systems within
`national networks, it was natural that there would be a desire to inter-
`connect the national networks on a continent. The United States and
`Canadian telephone networks had evolved almost as a single unit and
`as a result had a common national-continental system. Such was not the
`case within Europe. Due to the differences in national signaling systems,
`when it was desired to establish semiautomatic service between countries
`in Europe, it was necessary to find a common interexchange language.
`Two systems, CCITT Nos. 3s and 4,3 were standardized for semiautomatic
`and automatic service, and system No. 4 is used quite extensively in
`Europe.
`Shortly after the laying of the Atlantic telephone cable in 1956, it was
`proposed that semiautomatic service be introduced between North
`America and Europe. The North American signaling system was not
`compatible with system Nos. 3 and 4 nor with other systems in use in
`European countries. In addition to the technical differences between
`signaling systems, e.g., signaling frequencies, method of sending digits,
`etc., there were certain network differences which the signaling system
`switching systems, and thus between networks,
`as the interface tween
`had to take into account.
`The United States (American Telephone and Telegraph Com-
`pany-AT&T), the United Kingdom, German, and French post offices
`joined forces to design a new T A S I4 compatible signaling system, now
`known as the Atlantic System. This system provided an intercontinental
`common language and was used in both the Atlantic and Pacific cables
`to provide semiautomatic service. As more and more countries were
`connected to the cables, it became evident that it would be desirable to
`have a worldwide standard signaling system. The problem was posed
`
`• International Telegraph and Telephone Consultative Committee.
`
`HISTORY AND DESCRIPTION
`
`227
`
`AT&T, Exh. 1008, p. 3
`
`

`
`to the CCI'IT at the Ilnd Plenary Assembly (New Delhi, 1960); and study
`of a standard system was authorized.5
`During the study the Atlantic System was examined, and after some
`modification was standardized as system No. 5.6 It is now used in all of
`the undersea cables to provide both semiautomatic and automatic ser-
`vice. System No. 5 was the only system available which was compatible
`with the longer propagation delays inherent in synchronous satellites
`when they were introduced and thus is also the system used today for
`international satellite circuits.
`System No. 5 is a per circuit in-band system; that is, the sig n als are
`carried within the voiceband of the circuit used by the customer. It
`consists of a line sig n aling part and an interregister part. The line sig-
`naling part uses two frequencies, 2400 and.2600 Hz, separately or to-
`gether in a fully compelled mode. The interregister part for sending
`address signals uses a multifrequency code pulsed at a rate of 10 digits
`per second in the forward direction.
`During the 1960-1964 studies which led to the standardization of
`system No. 5, there was a wide difference of opinion over the techniques
`to be used. Even after the agreement on the specification some had
`reservations about the adequacy of system No. 5 for the future in a
`greatly expanded, fully automatic worldwide network. As a result, an
`agreement was reached that the study of a new signaling system to be
`known as system No. 6 should be undertaken in the 1964-1968 study
`period. Some of the major reservations concerning system No. 5 were
`post-dialing delay, answer signal delay, limited number of signals, in-
`terregister signaling in forward direction only, and slow signaling.
`There were two schools of thought concerning the new system, one
`favoring a system utilizing conventional techniques, and the other a
`system utilizing a new technique, i.e., a separate signaling channel
`common to a number of speech circuits. Because preliminary studies of
`common channel sig naling then underway at Bell Labs showed promise,
`AT&T was one of the supporters of the common channel approach. A
`compromise was reached and a question was formulated which called
`for the initial study to be of a system with a common channel for line
`(supervisory) type sig nals and inband interregister sig naling for address
`type signals.6
`Guidelines for the design of the interregister and common channel
`systems were drawn up, and a preliminary division of signals between
`the two was made. The concept of nonassociated signaling was defined
`and several important parameters of the data link which formed the basis
`of the ultimate desig n were accepted, i.e., the data system to operate over
`standard 3 or 4 kHz spaced voice frequency channels, the data links to
`be nonswitched, a serial mode of data transmission to be used, error
`
`228 THE BELL SYSTEM TECHNICAL JOURNAL, FEBRUARY 1978
`
`AT&T, Exh. 1008, p. 4
`
`

`
`ITC.._
`
`PARIS
`PTT
`'
`'-..J
`ROME
`
`Fig. 1-CCTTT system No. 6 total field trial network, phases Band C.
`
`detection by redundant coding, error correction by retransmission, de-
`pendability requirements, and security arrangements.
`A meeting in Stockholm marked a major trnning point in the devel-
`opment of system No. 6. After a review, it became evident that most
`administrations had become convinced in the coUI'Se of their studies that
`a full common channel system should be specified capable of carrying
`all the necessary signals. The data rate was established at 2400 bits per
`second, definitions of the sig naling network were defined, error detection
`and correction methods were elaborated, further security methods were
`defined and guidelines for the format were established.
`A series of meetings followed in New York, Tokyo, Prague, and Flor-
`ence which led to the specification of a common channel sig naling system
`which was presented to and approved by the IV th Plenary Assembly of
`the CCI'IT in Mar del Plata during October 1968. 7 At this same meeting,
`a special group was organized to conduct field trials of the new No. 6
`signaling system.
`Eleven Administrations or Recognized Private Operating Agencies
`participated in the field trials. AT&T participated utilizing equipment
`located in Bell Laboratories in Columbus, Ohio. The other participants
`and the extent of the trials are shown in Fig. 1.
`As a result of the trials, two significant decisions made were: choice
`of the link-by-link rather than the end-to-end method of making the
`continuity check of the speech path, and the design of a new format
`
`HISTORY AND DESCRIPTION
`
`229
`
`AT&T, Exh. 1008, p. 5
`
`

`
`which improved the overall efficiency of the system.
`The results of this most extensive field trial gave every confidence that
`system No. 6 as finally specified would provide the facilities required
`in a vastly expanded worldwide automatic network, with the desired
`reliability under actual operating conditions.3 A vast potential for new
`signals is available in the format and it has already been shown that the
`problems of interworking between national systems based on different
`design philosophies can be eased by utilizing some of this potential.
`The final specifications after the completion of the field trials were
`presented to and were approved by the CCITT's Vth Plenary Assembly,
`Geneva 1972,s which also authorized further study of the structure of
`the international common channel network, digital version of CCITT
`signaling system No. 6, maintenance methods for system No. 6, and in-
`terworking between international signaling system No. 6 and national
`common channel signaling systems.8
`The specification of a digital version of system No. 6 was completed
`and Recommendations were proposed to guide the desig n of common
`channel signaling systems for national or regional use in a compatible
`fashion so that they may form a part of a future worldwide signaling
`network. These Recommendations and the specifications of the digital
`version of system No. 6 were approved by the Vlth Plenary Assembly
`of the CCITT in Geneva in October 1976.9
`The design of system No. 6 represents a first in many technical areas,
`e.g., it is the first telephone signaling system to employ a dedicated
`processor-to-processor high-speed data link. Even more important,
`however, it is the first system ever designed entirely within the CCITT.
`This, for a system this complex, is quite a remarkable achievement and,
`of course, was only possible because of the goodwill of the members and
`their determination to succeed. In addition to the hours spent in meet-
`ings, many more hours of engineering time were devoted to the study
`in various laboratories around the world.
`
`IV. INTRODUCTION Of COMMON CHANNEL SIGNALING IN THE DOD
`NETWORK
`As indicated in the.above section, the studies in Bell and similar other
`laboratories in the CCITT led to the specification of common channel
`signaling systems. The international version is known as the CCITT
`sig n aling system No. 6 and the system for domestic use in the U.S.A. is
`desig nated Common Channel Interoffice Signaling, or CCIS. An intensive
`development program was undertaken to implement the system in the
`DDD network to be used between processor controlled switching offices.
`The first offices considered were No. 4A crossbar offices equipped with
`the Electronic Translator System (ETS),10 a processor with sufficient
`capacity not only to provide the translation function but also to process
`
`230 THE BELL SYSTEM TECHNICAL JOURNAL, FEBRUARY 1978
`
`AT&T, Exh. 1008, p. 6
`
`

`
`-
`
`-
`
`/ S P E E C H
`
`CIRCUITS
`
`-
`
`-
`
`--}-------- --{--
`--
`--
`--
`--
`
`COMMON CHANNEL
`PATH
`
`/ S I G N A L I N G
`
`(a)
`
`_A
`
`/
`
`II
`
`II
`
`--
`
`c : s
`
`t
`
`--},
`/',
`
`
`
`-- ::
`;;}----------{;; .
`
`\j
`
`COMMON CHANNEL SIGNALING PATHS
`
`(b)
`Fig. 2-(a) Associated signaling. (b) Nonassociated signaling.
`
`common channel signaling information. In addition, the choice of the
`No. 4 switch would insure high penetration of CCIS in the DDD network.
`The second type of office to implement CCIS was the new No. 4 ESS11 toll
`switch then under parallel development.
`Systems engineering economic studies of several implementation
`proposals indicated that in order to accelerate the introduction of
`common channel signaling a plan that would provide the greatest con-
`nectivity at minimum costs should be followed. Obviously, the greater
`the number of trunks served by a single sig naling link, the lower the
`per-trunk costs would be and since the getting-started costs were non-
`trivial, serious consideration was given to a sig naling network plan that
`would cater to small as well as large trunk group sizes. A plan which is
`cost-effective on large trunk groups calls for an associated sig naling link.
`However, since the majority of trunk groups are not large enough to
`economically support their own associated link a plan was developed
`which employed a form of nonassociated sig naling. The concepts of these
`two plans are illustrated in a simple fashion in Figs. 2a and 2b, respec-
`
`HISTORY AND DESCRIPTION
`
`231
`
`AT&T, Exh. 1008, p. 7
`
`

`
`TO OTHER
`REGIONAL AREAS
`
`' - - - - -
`
`"A" LINKS ARE DUPLICATED
`Q - USER OFFICES
`- SPEECH CIRCUITS (ONLY A FEW SHOWN FOR ILLUSTRATION)
`- -
`Fig. 3 - U s e r offices connected to STP quad.
`
`-
`
`_ . . . . J
`
`tively. In the associated case the common channel signaling link is cot-
`erminus with the group of speech circuits between offices A and B. In
`the nonassociated plan the groups of speech circuits between offices A
`and Band A and Care large enough to economically use associated sig-
`naling, while the signaling required for the smaller group of speech cir-
`cuits between offices Band C is carried over the common signaling paths
`B to A to C with office A serving as a Signal Transfer Point (STP). This
`form of nonassociated signaling is also referred to as quasiassociated
`since there is still some degree of association between signaling and
`sp ech paths. A further extension of nonassociated signaling, known as
`disassociated signaling, is where in a completely separated sig naling
`network the signaling paths do not have a fixed association with the
`speech paths they serve.
`Economics having dictated a form of quasiassociated sig naling for the
`network, implementation planning resulted in the assignment of two
`signal transfer points in each of the ten regional areas of the DDD net-
`work. Figure 3 indicates the chosen arrangement with the STPs inter-
`connected by a quad of "B" signaling links and user switching offices
`connected to their area STPs by two "A" sig naling links. The redundancy
`furnished by the "A" and "B" links assures a high degree of signaling
`reliability. In addition "C" links are provided to permit signaling of
`update and status information between mate STPs and to carry traffic
`
`232 THE BELL SYSTEM TECHNICAL JOURNAL, FEBRUARY 1978
`
`AT&T, Exh. 1008, p. 8
`
`

`
`between STPs within the same area under failure conditions. Adaptation
`of the CC ITT No. 6 system to this quasiassociated signaling network re-
`quired some minor changes which will be discussed in the following
`system descriptions.
`
`V. DESCRIPTION OF CCIS AND CCITT NO. 6 SIGNALING SYSTEMS
`The system description that follows is generally applicable to both
`the CCIS system for the domestic DDD network and the CCITT signaling
`system No. 6 for international-intercontinental networks. Where dif-
`ferences exist they will be indicated but it should be recognized that the
`two systems are completely interworkable and such operation will be
`catered to at International Switching Centers (ISCs) served by No. 4 ESS
`offices.
`Figure 4 is a basic block diagram of the common channel interoffice
`signaling system. Table I indicates the definitions of the various com-
`ponents. The system was designed to operate between stored program
`controlled switching offices where it is not practical to specify well de-
`fined equipment interfaces because there is considerable latitude per-
`mitted in the distribution of signaling functions between the processor
`and its peripheral equipment. The major signal transfer functions can,
`however, be delineated, and the blocks shown in Fig. 4 depict functions
`rather than specific equipment arrangements.
`Each signaling link transmits synchronously a continuous stream of
`data in both directions. The data stream is divided into signal units (SU)
`of 28 bits each, of which 20 bits convey information and 8 bits are check
`bits. The signal units are in turn grouped into blocks of 12, with the
`twelfth signal unit always an acknowledgment signal unit (ACU). The
`latter unit is coded to indicate the number of the block being transmitted,
`the number of the block being acknowledged and whether or not each
`of the other 11 signal units in the block being acknowledged were received
`without detected errors. Fig u re 5 shows blocks of signal units transmitted
`in opposite directions with one of the ACUs expanded to show the bit
`make-up. In the example given the third signal unit in block i was re-
`ceived in error and the ACU in block j indicates this fact. In response to
`this ACU the receiving terminal will retransmit the message which con-
`tains the signal unit in error. Thus, CCIS achieves error control by re-
`dundant coding and error correction by retransmission.
`Common channel signaling can utilize either analog or digital trans-
`mission facilities. In the analog case, data modems are provided at each
`terminal and operate over standard analog voice -bandwidth channels.
`In the digital case, channels are derived either from bit streams of pulse
`code modulated systems, e.g., subframing bits of T l lines, or dedicated
`digital channels. In the digital case no modems are required but an ap-
`propriate digital interface must be specified. The present design of CCIS
`
`HISTORY AND DESCRIPTION
`
`233
`
`AT&T, Exh. 1008, p. 9
`
`

`
`MODEM OR
`INTERFACE
`ADAPTOR
`
`--
`
`I
`
`'
`
`+ S W I T C H I N G
`NETWORK
`
`I
`
`·
`
`
`
`
`
`SIGNALING
`TERMINAL
`
`PROCESSOR
`
`I
`
`I
`I
`I
`I
`I
`I
`
` m
`
`CJJ m F
`(/) -<
`
`+ S W I T C H I N G
`NETWORK
`
`I
`
`/
`
`I
`
`' - - - -
`
`PROCESSOR
`
`SIGNALING
`TERMINAL
`
`MODEM
`INTERFACI!
`ADAPTOR
`
`' - -
`
`'
`
`'
`
`I
`I
`I
`I
`I
`
`II
`
`
`
`iil ( ' ) :I: z
`
`c5
`) >
`r
`c...
`0 c
`:D z
`
`) > r .,, m
` c
`
`) >
`:D -<
`
`SIGNALING CHANNEL
`
`OATA CHANNEL - -
`
`r
`I
`I
`I
`I
`I
`I
`I
`L
`
`..
`
`- T R A N S F E R - - ,
`I
`CHANNEL
`I
`I
`I
`I
`I
`_J
`
`TRANSFER
`CHANNEL
`
`.!_R S R...!"I
`
` TA
`
`CHANNEL
`
`SIGNALING DATA LINK
`
`- - -
`
`SIGNALING CHANNEL
`
`SIGNALING LINK
`
`COMMON CHANNEL SIGNALING SYSTEM
`Fig. 4-Basic block diagram of the Common Channel Interoffice Signaling System.
`
`AT&T, Exh. 1008, p. 10
`
`

`
`Transfer
`channel
`
`Transfer
`link
`
`Table I - Definitions of various links and channels in cc1s
`Digital Version
`Analog Version
`(Voice-frequency channel)
`(Digital channel)
`A one-way voice-frequency
`A one-way digital transmission
`transmission path from the output
`path from the output of the
`interface adaptor to the input of
`of a data modulator to the input
`the interface adaptor, made up of
`of a data demodulator, made up
`of one or more voice-frequency
`one or more digital channels in
`channels in tandem.
`tandem.
`(Digital link)
`(Voice-frequency link)
`A two-way voice-frequency
`A two-way digital transmission
`path between two interface
`transmission path between two
`adaptors, made up of one digital
`data modems, made up of one
`channel in each direction.
`voice-frequency channel in each
`direction.
`A one-way data transmission path
`A one-way data transmission path
`between two points, made up of
`between two points, made up of
`a modulator, a voice-frequency
`a digital channel terminating on
`an interface adaptor at each end.
`channel and a demodulator.
`A two-way data transmission path between two points, made up of one
`data channel in each direction.
`A one-way signaling path from the processor of one switching machine
`to the processor of another switching machine.
`A two-way signaling path from processor to processor made up of one
`signaling channel in each direction.
`
`Data channel
`
`Data link
`
`Signaling
`channel
`Signaling
`link
`
`operates at 2400 bps in the analog application and in the future at 4000
`bps in the digital case using the signaling subframing bits. Higher
`speeds-perhaps as high as 64 kbs-are expected in future designs.
`
`< I SU I SU I SU I SU I SU I ACU I SU I SU I SU I SU I SU I SU I SU I SU I SU I SU I SU I ACU l -
`
`5
`
`4
`
`3
`
`2
`
`1
`
`12
`
`11
`
`10
`
`9
`
`B
`
`7
`
`6
`
`S
`
`4
`
`3
`
`2
`
`1
`
`12
`
`BLOCK i
`
`-
`
`l ACU I SU I SU I SU I SU I SU I SU I SU I SU I SU I SU I SU I ACU I SU I SU I SU I SU I SU I l
`
`1
`
`2
`
`3
`
`4
`
`5
`
`6
`
`7
`
`B
`
`9
`
`10-11
`
`12
`
`"-1
`
`2
`
`3
`
`4
`
`5
`
`BLOCK j
`
`UNIT NO. 3 RECEIVED IN ERROR
`/ S I G N A L
`1 0 0 0 0 0 0 0 0
`
`' - - - - - - . . . J
`' - - - - J ' - - - - . . . . 1 - - - - - - - -
`HEADING ACKNOWLEDGEMENT INDICATORS @
`CHECK BITS
`@
`0 - NO ERROR DETECTED
`1 - ERROR DETECTED
`@ SEQUENTIAL NUMBER OF BLOCK BEING ACKNOWLEDGED
`@SEQUENTIAL NUMBER OF BLOCK COMPLETED BY THIS ACU
`Fig. 5-Block structure, CCIS SUS and ACU.
`
`HISTORY AND DESCRIPTION
`
`235
`
`AT&T, Exh. 1008, p. 11
`
`

`
`SWITCHING
`NETWORK
`
`PROCESSOR
`
`---------t C O N T R O L - - - - - - ,
`
`ERROR
`
`OATA
`CHANNEL
`
`DATA
`CHANNEL
`
`- OUTPUT BUFFER
`OB
`INPUT BUFFER
`-
`IB
`SYU - SYNCHRONIZATION SIGNAL UNIT GENERATOR
`ACU - ACKNOWLEDGEMENT SIGNAL UNIT GENERATOR
`COD - CODER
`DEC - DECODER
`Fig. 6-Functional block diagram of a CCIS terminal.
`
`A functional block diagram of a CCIS terminal is shown in Fig. 6. Sig-
`nals originating in the processor are transmitted in a specified format
`in parallel form to the output buffer (OB) where they are stored according
`to their priority level. The signals are then passed to the coder (COD) in
`serial form where they are encoded by the addition of check bits and then
`delivered to the outgoing data· channel.
`I n the receiving direction, signals in serial form are passed from the
`data channel to the decoder (DEC) where each signal unit is checked for
`error on the basis of the included check bits. Information-carrying signal
`units that are error free are passed on to the input buffer (IB) after de-
`letion o f the check bits. The input buffer passes the signals in parallel
`form to the processor for action.
`Information carrying signal units with detected errors are discarded
`and this information is conveyed to the originating terminal via the ac-
`knowledg m ent sig n al unit (ACU) where action is taken to retransmit the
`message containing the failed signal units. This procedure requires, o f
`course, that all signal units be stored until they are acknowledged as
`having been received correctly. Further, sig n al messages made up of two
`or more signal units must be stored and i f any signal unit in the message
`is in error the entire sig n al message must be retransmitted. Sig n al units
`that are not carrying information, e.g., synchronizing signal units (SYU)
`can be discarded i f received in error and no request is made for their
`retransmission. A data channel failure detector complements the error
`control mechanism for longer error bursts.
`
`238
`
`THE BELL SYSTEM TECHNICAL JOURNAL, FEBRUARY 1978
`
`AT&T, Exh. 1008, p. 12
`
`

`
`HEADING
`
`SIGNAL
`INFO
`
`CCIS I : : I : : : I :
`
`: : : : : :
`
`BAND NO.
`
`TRUNK LASE LS (8192)
`
`CHECK BITS
`
`: I : : : I : : : : : : : I
`
`TRUNK NO.
`
`HEADING
`
`SIGNAL
`INFO
`
`TRUNK LABELS (2048)
`
`CHECK BITS
`
`c g;1 : : : : I : : : I : : : : : : I : : : I : : : : : : : I
`
`TRUNK NO.
`BAND NO.
`Fig. 7-Lone Signal Unit (LSU) format.
`
`The present modems used in the transmission of CCIS serial binary
`data over analog facilities employ 2400 bps differential four-phase
`modulation. The transmitted binary data is grouped into dibits for en-
`coding making the rate of carrier phase shifts or baud rate equal to 1200
`per second. The receiving demodulator uses differentially coherent de-
`tection to recover the binary data from the line signal. This type of de-
`tection is relatively insensitive to the types of distortion and interference
`found on telephone-type transmission facilities. Timing information is
`extracted from the zero crossings, on a dibit basis, of the received base-
`band data signals which provides for synchronization holdover through
`extended drop-outs and periods of high noise.
`
`VI. SIGNAL FORMATS
`As indicated earlier a signal unit is made up of 28 bits-20 bits for
`information plus 8 bits for a cyclic check code used for error detection.
`The codiiag formats for the cc1s and CCITT No. 6 systems differ because
`of the need in the CCIS-STP signaling network to identify a larger number
`of individual trunks than that provided for in the CCITT No. 6 system.
`Figure 7 compares the lone signal unit (LSU) of the two systems. In CCIS,
`13 bits are set aside for trunk identification or labels while CCITT No.
`6 uses 11 bits. Hence CCIS can identify 8192 trunks while CCITT No. 6
`identifies 2048 trunks. In both cases the bits assigned for labels are di-
`vided between band numbers and trunk numbers, i.e., 16 trunks within
`512 or 128 bands, respectively.
`A lone signal unit (LSU) is used to transmit a one-unit message such
`as a single telephone signal, a signaling system control signal or a man-
`agement signal. The type of signal is defined by the "signal information"
`bits immediately following the "heading" code. A multi-unit message
`(MUM) consists of several signal units in tandem in order to transmit a
`number of related pieces of information in an efficient manner. The first
`signal unit in an MUM is referred to as an initial signal unit (ISU) and the
`
`HISTORY AND DESCRIPTION 237
`
`AT&T, Exh. 1008, p. 13
`
`

`
`HEADING
`r - - - ,
`
`: : :
`CCIS 1, : 0 : 1 I I : : I :
`:
`:
`:
`\._J..._____, ' - - - - - - - - - -
`@ @
`BAND NO.
`@ ISU TYPE INDICATOR
`@ MULTI-UNIT MESSAGE LENGTH INDICATOR
`Fig. 8-lnitial Sig nal Unit (ISU) format.
`
`TRUNK LABELS 18192)
`
`CHECK BITS
`
`: I : : : I : : : : : : : I·
`,.____ _ _ _,
`TRUNK NO.
`
`CCIS format is shown in Fig. 8. The second and any following sig nal units
`are referred to as subsequent signal units (ssu). The format for the
`CCITI' No. 6 system ISU is the same as for the LSU. Table II indicates the
`"heading" code to identify the type of signal unit class for the CCIS sys-
`tem.
`Table III indicates the "heading" code to identify the type of sig n al
`unit class for the CCITI' No. 6 system. Being an international sig naling
`system, as opposed to a strictly domestic (regional) or national system,
`it is necessary to assign blocks of sig nals for international, regional, and
`national uses. Another difference should be noted-all heading codes
`
`use five bits except for two distinct cases, namely the use of two bits (0,0)
`to identify subsequent sig nal units (SSU) and three bits (0,1,1) to identify
`
`the acknowledgment signal unit (ACU).
`Figure 9 compares the coding of subsequent sig nal units. The heading
`codes differ and in the CCITI' No. 6 system each SSU includes information
`on the total number of ssus in the message. In the case of CCIS the in-
`formation on the number of ssus in only contained in the initial signal
`unit. CCIS has 17 bits for signal information use while CCITT No. 6 has
`16 bits. The signal information can be routing information, address
`digits, etc., as will be indicated in typical telephone signal formats that
`follow.
`In establishing a CCIS-controlled telephone connection, an initial
`address message (IAM) is transmitted from the originating terminal. This
`message, made up of several signal units in tandem, will contain trunk
`
`Table II - Heading code for cc1s system
`Signal unit class
`Signal type
`Heading code
`000
`LSU}
`001
`LSU
`010
`LSU
`100
`LSU
`111
`LSU
`
`{ Lone signal unit-telephone signals
`
`011
`101
`110
`
`ACU
`
`!SU ssu
`
`Lone Signal Units-
`Telephone sig nals
`Sig naling system control signals
`Management sig nals
`Acknowledgment signal unit
`Initial sig nal unit
`Subsequent signal unit
`
`238
`
`THE BELL SYSTEM TECHNICAL JOURNAL, FEBRUARY 1978
`
`AT&T, Exh. 1008, p. 14
`
`

`
`Table Ill - Heading code for c c , n system No. 6
`Signal unit class
`Heading code
`00
`01001
`01010
`01011
`011
`
`01000 l
` )10010
`
`10011
`10100
`10101
`10110
`10111
`
`u m 1 }
`11010
`11011
`11100
`11101
`
`11110
`11111
`
`Subsequent signal unit
`Spare (reserved for regional and/or national use)
`
`Acknowledgment signal unit
`Initial sig nal unit of an initial address message (or of a multiunit
`message)
`
`Subsequent address message (one-unit message or multiunit mes-
`sage)
`
`International telephone signals
`
`Spare (reserved for regional and/or national use)
`Sig naling-system-control sig nals (except acknowledgment of signal
`unit) and management sig nals
`Spare (reserved for regional and/or national use)
`
`identification, abbreviated or expanded routing information and address
`information. T h e most common 1AM to be used in the DDD network will
`be an 1AM with abbreviated routing information and seven or ten digits
`for the address, resulting in either thre