An Overview of Signaling System No. 7
`
`D D I R. MODARRESSI, MEMBER, IEEE, AND RONALD A. SKOOG, MEMBER, IEEE
`Invited Paper
`
`In modern telecommunication networks, signaling constitutes
`the distinct control infrastructure that enables provision of ALL
`other services. The component of signaling systems that controls
`provision of services between the user and the network is the access
`signaling component, and the component that controls provision of
`services within the network, or between networks, is the network
`signaling component. There are international standards for both
`access signaling and network signaling protocols. From a network
`structure viewpoint, access signaling structures generally provide
`point-to-point connectivity between the user and a network node,
`while network signaling structures provide network-wide commu-
`nication capability (directly or indirectly) between the nodes of
`the public network(s). Since the network signaling system acts as
`a traffic collectorldistributor for many access signaling tributaries,
`its functions are more complex, its structure more involved, and its
`performance more stringent. This paper provides an overview of
`modern network signaling systems based on the Signaling System
`No. 7 international standard.
`
`I.
`
`INTRODUCTION
`In the context of modern telecommunications, signaling
`can be defined as the system that enables stored program
`control exchanges, network databases, and other ‘intelli-
`gent’ nodes of the network to exchange a) messages related
`to call setup, supervision, and tear-down (call/connection
`control); b) information needed for distributed application
`processing (inter-process query/response, or user-to-user
`data); and c) network management information. As such,
`signaling constitutes the control infrastructure of the mod-
`ern telecommunication network.
`Modern signaling systems are essentially data communi-
`cation systems using layered protocols. What distinguishes
`them from other data communication systems are basically
`two things: their real time performance and their reliability
`requirements. No matter how complex the set of network
`interactions are for setting up a call, the call setup time
`should still not exceed a couple of seconds. This imposes
`quite a stringent end-to-end delay requirement on the sig-
`naling system. On the other hand, because of the absolute
`reliance of the telecommunication network on its signaling
`system, requirements for signaling network reliability (mes-
`Manuscript received October 23, 1991; revised December 18, 1991.
`A. R. Modarressi is with AT&T Bell Laboratories, Columbus, OH
`43213.
`R. A. Skoog is with AT&T Bell Laboratories, Holmdel, NJ 07733.
`IEEE Log Number 9108075.
`
`sage integrity, end-to-end availability, network robustness,
`recovery from failure, etc.) are extremely demanding. For
`example, current objectives require the down-time between
`any arbitrary pair of communicating nodes in the signaling
`network not to exceed 10 midyear. This is at least two
`orders of magnitude smaller than the corresponding require-
`ment in a general-purpose data network. Requirements on
`real-time performance and reliability of signaling systems
`are likely to become even more stringent with advances in
`technology and new application needs.
`Over the last century or so, signaling has evolved with the
`technology of telephony, although the pace of this evolution
`has never been faster than in the last two decades, a period
`characterized by the marriage of computer and switching
`technologies. The advent of the Integrated Services Digital
`Network (ISDN) has further accelerated the pace of de-
`velopment and deployment of signaling systems to support
`an ever increasing set of “intelligent network” services on a
`worldwide basis. When viewed as an end-to-end capability,
`signaling in ISDN has two distinct components: signaling
`between the user and the network (access signaling), and
`signaling within the network (network signaling). The cur-
`rent set of protocol standards for access signaling is known
`as the Digital Subscriber Signaling System No. 1 (DSS1).
`The current set of protocol standards for network signaling
`is known as the Signaling System No. 7 (SS7).
`This paper provides an overview of Signaling System No.
`7. It is a somewhat abridged and updated version of a tuto-
`rial on SS7 that was published in 1990 [l]. Following this
`introduction, the salient features of SS7’s Network Services
`Part (NSP) are described in Section 11. Functionally, NSP
`corresponds to the first three layers of the Open System
`Interconnection (OSI) Reference Model. This section also
`provides a discussion of signaling network structures that,
`in conjunction with the NSP, provide ISDN nodes with a
`highly reliable and efficient means of exchanging signaling
`messages. Once this reliable signaling message transport
`capability is realized, each network node has to be equipped
`with capabilities for processing of the transported messages
`in support of a useful function like setting up of a call
`(connection). In an increasingly large number of cases, call
`setup has to be preceded by invocation of some distributed
`
`590
`
`PROCEEDINGS OF THE IEEE, VOL. 80, NO. 4, APRIL 1992
`
`0018-9219/92$03.00 0 1992 IEEE
`
`CISCO SYSTEMS, INC. Ex. 1014 Page 1
`
`

`
`networks (one of the disadvantages of CCS6). Later, it
`became clear that there were other applications that would
`need additional network services (full OS1 Network service
`capabilities) like an expanded addressing capability and
`connection-oriented message transfer. SCCP was developed
`to satisfy this need. The resulting structure, and specifically
`the splitting of the OS1 Network functions into MTP level
`3 and SCCP, has certain advantages in the sense that the
`higher overhead SCCP services can be used only when
`needed, allowing the more efficient MTP to serve the needs
`of those applications that can use a connectionless message
`transfer with limited addressing capability.
`Sections 11-A and 11-B provide an overview of MTP and
`SCCP, respectively. Section 11-C describes the signaling
`network structures that can be used to implement the
`Network Services Part.
`
`A. The Message Transfer Part (MTP)
`The overall purpose of MTP is to provide a reliable
`transfer and delivery of signaling information across the
`signaling network, and to react and take necessary actions
`in response to system and network failures to ensure that
`reliable transfer is maintained. Figure 2 illustrates the
`functions of MTP levels, and their relationship to one
`another and to the MTP users. These three levels are now
`described.
`I ) Signaling Data Link Functions (Level 1): A Signal-
`ing Data Link is a bidirectional transmission path for
`signaling, consisting of
`two data channels operating
`together in opposite directions at the same data rate. It
`fully complies with the OSI’s definition of the physical
`layer (layer 1). Transmission channels can be either digital
`or analog, terrestrial or satellite.
`For digital signaling data links, the recommended bit
`rate for the ANSI standard is 56 kb/s, and for the CCITT
`International Standard it is 64 kb/s. Lower bit rates may
`be used, but the message delay requirements of the User
`Parts must be taken into consideration. The minimum bit
`rate allowed for telephone call control applications is 4.8
`kb/s. In the future, bit rates higher than 64 kb/s may be
`required (e.g., 1.544 Mb/s in North America and 2.048 Mb/s
`elsewhere), but further study is needed before these rates
`can be standardized.
`2) Signaling Link Functions (Level 2): The Signaling Link
`functions correspond to the OSI’s data link layer (layer
`2). Together with a signaling data link, the signaling link
`functions provide a signaling link for the reliable transfer
`of signaling messages between two directly connected
`signaling points. Signaling messages are transferred over
`the signaling link in variable length messages called signal
`units. There are three types of signal units, differentiated
`by the length indicator field contained in each, and their
`formats are shown in Fig. 3. The Signaling Information
`Field (SIF) in a Message Signal Unit (MSU) must have
`a length less than or equal to 272 octets. This limitation
`is imposed to control the delay a message can impose on
`other messages due to its emission time (which is limited
`by the maximum standardized link speed of 64 kb/s).
`
`-------------
`
`MTP Level 3
`MTP Level 2
`
`Fig. 1. SS7 protocol architecture.
`
`application processes, the outcome of which determines the
`nature as well as the attributes of the subsequent call or
`connection control process. These nodal capabilities of call
`control and remote process invocation and management are
`part of the Signaling System No. 7 User Parts, which are
`described in Section 111. In Section IV, we dwell on the very
`stringent performance requirements of signaling systems.
`These requirements reflect the critical nature of signaling
`functions and their real time exigencies. Finally, in Section
`V we sketch a broad outline of the likely evolution of
`network signaling in the remaining years of this century.
`
`11. SIGNALING SYSTEM No. 7 NETWORK
`SERVICES PART (NSP)
`In this section, we describe the Signaling System No. 7
`protocols that correspond to the first three layers (Physical,
`Data Link, and Network) of the OS1 Reference Model. This
`component of the Signaling System No.7 protocol is called
`the Network Services Part (NSP), and it consists of the
`Message Transfer Part (MTP) and the Signaling Connection
`Control Part (SCCP). Figure 1 shows how these relate to
`each other and to the other components of the protocol.
`MTP consists of levels 1-3 of the Signaling System No.
`7 protocol, which are called the Signaling Data Link,
`the Signaling Link, and the Signaling Network functions,
`respectively. SCCP is an MTP user, and therefore is in
`level 4 of Signaling System No. 7 protocol stack. MTP
`provides a connectionless message transfer system that
`enables signaling information to be transferred across the
`network to its desired destination. Functions are included
`in MTP that allow system failures to occur in the net-
`work without adversely affecting the transfer of signaling
`information. SCCP provides additional functions to MTP
`for both connectionless and connection-oriented network
`services.
`MTP was developed before SCCP and it was tailored
`to the real time needs of telephony applications. Thus a
`connectionless (datagram) capability was called for which
`avoids the administration and overhead of virtual circuit
`
`MODARRESSI AND SKOOG: SS7: AN OVERVIEW
`
`591
`
`CISCO SYSTEMS, INC. Ex. 1014 Page 2
`
`

`
`User
`Messagc
`Processir
`
`i-l~-Itt
`
`Functions
`
`Common
`Transfer
`Functions
`
`Fig. 2. MTP functional diagram
`
`The SS7 link functions show a strong similarity to typ-
`ical data network bit-oriented link protocols (e.g., HDLC,
`SDLC, LAP-B), but there are some important differences.
`These differences arise from the performance needs of
`signaling (e.g., lost messages, excessive delays, out-of-
`sequence messages) that require the network to respond
`quickly to system or component failure events. The standard
`flag (01111110) is used to open and close signal units,
`and the standard CCITT 16-bit CRC checksum is used
`for error detection. However, when there is no message
`traffic, Fill-In Signal Units (FISU’s) are sent rather than
`flags, as is done in other data link protocols. The reason for
`this is to allow for a consistent error monitoring method
`(described below) so that faulty links can be quickly
`detected and removed from service even when traffic is
`low.
`a) Error correction: Two forms of error correction are
`specified in the signaling link procedures. They are the
`Basic Method and the Preventive Cyclic Retransmission
`(PCR) Method. In both methods only errored MSU’s and
`Link Status Signal Units (LSSU’s) are corrected, while
`errors in FISU’s are detected but not corrected. Both
`methods are also designed to avoid out-of-sequence and
`duplicated messages when error correction takes place. The
`PCR method is used when the propagation delay is large
`(e.g., with satellite transmission).
`The Basic Method of error correction is a non-compelled
`positive/negative acknowledgment retransmission error cor-
`rection system. It uses the “go-back-N” technique of re-
`transmission used in many other protocols. If a negative
`acknowledgment is received, the transmitting terminal stops
`sending new MSU’s, rolls back to the MSU received in
`error, and retransmits everything from that point before
`resuming transmission of new MSU’s. Positive acknowl-
`edgments are used to indicate correct reception of MSU’s,
`and as an indication that the positively acknowledged
`buffered MSU’s can be discarded at the transmitting end.
`For sequence control, each signal unit is assigned forward
`and backward sequence numbers and forward and backward
`indicator bits (see Fig. 3). The sequence numbers are seven
`bits long, which means at most 127 messages can be
`transmitted without receiving a positive acknowledgment.
`The PCR method is a non-compelled positive acknowl-
`edgment cyclic retransmission, forward error correction
`system. A copy of a transmitted MSU is retained at the
`transmitting terminal until a positive acknowledgment for
`that MSU is received. When there are no new MSU’s to be
`
`sent, all MSU’s not positively acknowledged are retrans-
`mitted cyclically. When the number of unacknowledged
`MSU’s (either the number of messages or the number of
`octets) exceeds certain thresholds, it is an indication that
`error correction is not getting done by cyclic retransmission.
`This would occur, for example, if the traffic level was
`high, which causes the retransmission rate to be low. In
`this situation a forced retransmission procedure is invoked.
`In this procedure new MSU transmission is stopped and
`all unacknowledged MSU’s are retransmitted. This forced
`retransmission continues until the unacknowledged message
`and octet counts are below specified threshold values. These
`threshold values must be chosen carefully, for if they are set
`too low, and the link utilization is large enough, the link
`will become unstable (i.e., once a forced retransmission
`starts, the link continues to cycle in and out of forced
`retransmission [2]).
`h) Error monitoring: Two types of signaling link error
`rate monitoring are provided. A signal unit error rate
`monitor is used while a signaling link is in service, and it
`provides the criteria for taking a signaling link out of service
`due to an excessively high error rate. An alignment error
`rate monitor is used while a signaling link is in the proving
`state of the initial alignment procedure, and it provides the
`criteria for rejecting a signaling link for service during the
`initial alignment due to too high an error rate.
`The signal unit error rate monitor is based on a signal unit
`(including FISU) error count, incremented and decremented
`using the “leaky bucket’’ algorithm. For each errored signal
`unit the count is increased by one, and for each 256
`signal units received (errored or not), a positive count is
`decremented by one (a zero count is left at zero). When the
`count reaches 64, an excessive error rate indication is sent
`to level 3, and the signaling link is put in the out of service
`state. When loss of alignment occurs (a loss of alignment
`occurs when more than six consecutive Is are received or
`the maximum length of a signal unit is exceeded), the error
`rate monitor changes to an octet counting mode. In this
`mode it increments the counter for every 16 octets received.
`Octet counting is stopped when the first correctly-checking
`signal unit is detected.
`The alignment error rate monitor is a linear counter that
`is operated during alignment proving periods. The counter
`is started at zero at the start of a proving period, and the
`count is incremented by one for each signal unit received in
`error (or for each 16 octets received if in the octet counting
`mode). A proving period is aborted if the threshold for the
`
`552
`
`PROCEt.I)INGS OF TIIE IktE. VOL.. 80. NO. 3. APRIL 1992
`
`CISCO SYSTEMS, INC. Ex. 1014 Page 3
`
`

`
`8n. n > 2
`
`First Bit
`Transmitted
`
`F
`
`8
`
`CK
`
`LI
`
`FSN
`
`BSN
`
`F
`
`16
`
`2
`
`6
`
`B
`
`1
`
`B
`
`1
`
`7
`
`7
`
`. First Lt
`Transmitted
`
`8
`
`Fig. 3. Signal unit formats.
`
`alignment error rate monitor count is exceeded before the
`proving period timer expires.
`c') Flow, control: The flow control procedure is initi-
`ated when congestion is detected at the receiving end of
`the signaling link. The congested receiving end notifies the
`transmitting end of its congestion with a link status signal
`unit (LSSU) indicating busy, and withholds acknowledg-
`ment of all incoming signal units. This action stops the
`transmitting end from failing the link due to a time-out
`on acknowledgment. However, if the congestion condition
`lasts too long (3-h s), the transmitting end will fail the link.
`A processor outage condition indication is sent by level
`2. called signaling indication processor outage (SIPO),
`Lvhenever an explicit indication is sent to level 2 from
`level 3 or when level 2 recognizes a failure of level 3.
`This indicates to the far end that signaling messages cannot
`be transferred to level 3 or above. The far-end level 2
`responds by sending fill-in signal units and informing its
`ievel 3 of the SIPO condition. The far-end level 3 will
`reroute traffic in accordance with the signaling network
`management procedures described as follows.
`3) Signulirig Network Fiinctions (Level 3): The signaling
`network functions correspond to the lower half of the
`OSI's Network layer, and they provide the functions and
`procedures for the transfer of messages between signaling
`points. which are the nodes of the signaling network.
`The signaling network functions can be divided into two
`basic categories: signaling message Iiawdling and signaling
`nehvork munagmzent. The breakdown of these functions
`
`and their interrelationship is illustrated in Fig. 4.
`a) signaling niessuge handling: Signaling message han-
`dling consists of message routing, discrimination, and
`distribution functions. These functions are performed at
`each signaling point in a signaling network, and they are
`based on the part of the message called the routing luhel.
`and the Service Information Octet (SIO) shown in Fig. 3.
`The routing label is illustrated in Fig. 5 and consists of
`the Destination Point Code (DPC). the Origination Point
`Code (OPC), and the Signaling Link Selection (SLS) field.
`In the international standard the DPC and OPC are 14 bits
`each, while the SLS field is 4 bits long. For ANSI, the
`OPC and DPC are each 24 bits (to accommodate larger
`networks), while the SLS field has 5 bits, and there are 3
`spare bits in the routing label. The routing label is placed
`at the beginning of the Signaling Information Field. and it
`is the common part of the label that is defined for each
`MTP user.
`When a message comes from a level 3 user. or originates
`at level 3, the choice of the particular signaling link on
`which it is to be sent is made by the message routing
`function. When a message is received from level 2, the
`discrimination function is activated, and it determines if it is
`addressed to another signaling point or to itself based on the
`DPC in the message. If the received message is addressed
`to another signaling point, and the receiving signaling
`point has the transfer capability. i.e., the Signal Transfer
`Point (STP) function, the message is sent to the message
`routing function. If the received message is addressed to the
`
`CISCO SYSTEMS, INC. Ex. 1014 Page 4
`
`

`
`L m l 4
`
`c m l 2
`
`I I I
`
`..............I......................
`
`6 .......................................
`
`I
`
`I&-
`
`I
`I
`..........................................................................
`
`:
`
`
`
`I
`I
`
`- SigMthrg Meocege Flow
`_-_---___
`ktdumons end Conrrols
`
`Fig. 4. Signaling network functions
`
`Fig. 5. Routing label structure
`
`receiving signaling point, the message distribution function
`is activated, and it delivers the message to the appropriate
`MTP user or MTP level 3 function based on the service
`indicator, a sub-field of the SI0 field. Message routing is
`based on the DPC and the SLS in almost all cases. In some
`circumstances the SIO, or parts of it (the service indicator
`and network indicator), may need to be used.
`Generally, more than one signaling link can be used to
`route a message to a particular DPC. The selection of the
`particular link to use is made using the SLS field. This is
`called load sharing. A set of links between two signaling
`points is called a link set, and load sharing can be done
`over links in the same link set or over links not belonging
`to the same link set. A load sharing collection of one or
`more link sets is called a combined link set.
`The objective of load sharing is to keep the load as
`evenly balanced as possible on the signaling links within
`a combined link set. For messages that should be kept in
`sequence, the same SLS code is used so that such messages
`take the same path. For example, for trunk signaling with
`ISUP (see Section IV-A) the same SLS code is used for all
`
`messages related to a particular trunk. In order to ensure
`proper load balance using SLS fields, it is critical that
`the SLS codes are assigned such that the load is shared
`evenly across all the SLS codes. Even then, the SLS load
`sharing method does not provide a fully balanced loading
`of signaling links in all cases. For example, if there are six
`signaling links in a combined link set, the 16 SLS codes
`would be assigned so that four signaling links would each
`carry three SLS codes and two of the signaling links would
`each carry only two SLS codes.
`h) Signaling network management: The purpose of
`the signaling network management functions is to provide
`reconfiguration of the signaling network in the case of
`signaling link or signaling point failures, and to control
`traffic in the case of congestion or blockage. The objective
`is that, when a failure occurs, the reconfigurations be
`lost, duplicated, or
`carried out so messages are not
`put out of sequence, and that message delays do not
`become excessive. As shown in Fig. 4, signaling network
`management consists of three functions: signaling traffic
`management, signaling route management, and signaling
`link management. Whenever a change in the status of a
`signaling link, signaling route or signaling point occurs,
`these three functions are activated as summarized below.
`The signmling trafJic management procedures are used
`to divert signaling traffic, without causing message loss,
`missequencing, or duplication, from unavailable signaling
`links or routes to one or more alternative signaling links
`or routes, and to reduce traffic in the case of congestion.
`When a signaling link becomes unavailable, a changeover
`procedure is used to divert signaling traffic to one or
`more alternative signaling links, as well as to retrieve
`for retransmission messages that have not been positively
`acknowledged. When a signaling link becomes available, a
`changehack procedure is used to reestablish signaling traffic
`on the signaling link made available. When signaling routes
`(succession of links from the origination to the destination
`signaling point) become unavailable or available, forced
`rerouting and controlled rerouting procedures are used,
`respectively, to divert the traffic to alternative routes or to
`the route made available. Controlled rerouting is also used
`to divert traffic to an alternate (more efficient) route when
`the original route becomes restricted (i.e., less efficient
`because of additional transfer points in the path). When a
`signaling point becomes available after having been down
`for some time, the signaling point restart procedure is used
`to update the network routing status and control when
`signaling traffic is diverted to (or through) the point made
`available.
`The signaling route management procedures are used to
`distribute information about the signaling network status
`in order to block or unblock signaling routes. The fol-
`lowing procedures are defined to take care of different
`situations. The transfer-controlled procedure is performed
`at a signaling transfer point in the case of signaling link
`congestion. In this procedure, for every message received
`having a congestion priority less than the congestion level
`of the signaling link, a control message is sent to the
`
`594
`
`PKOCEEDINGS OF THE IEEF. VOL. XO. NO. 4, APRIL I992
`
`CISCO SYSTEMS, INC. Ex. 1014 Page 5
`
`

`
`OPC of the message asking it to stop sending traffic
`that has a congestion priority less than the congestion
`level of the signaling link to the DPC of the message.
`In ANSI Standards four congestion message priorities are
`used; in international networks only one is used. The
`transfer-prohibited procedure is performed at a Signal
`Transfer Point to inform adjacent signaling points that
`they must no longer route to a DPC via that STP. This
`procedure would be invoked, for example, if the STP had
`no available routes to a particular destination. The transfer-
`restricted procedure is performed at a Signal Transfer
`Point to inform adjacent signaling points that, if possible,
`they should no longer route messages to a DPC via that
`STP. The transfer-allowed procedure is used to inform
`adjacent signaling points that routing to a DPC through
`that STP is now normal. In the ANSI standards, the
`above procedures are also specified on a cluster basis
`(a cluster being a collection of signaling points), which
`significantly reduces the number of network management
`messages and related processing required when there is a
`cluster failure or recovery event. The signaling-route-set-
`test procedure is used by the signaling points receiving
`transfer prohibited and transfer restricted messages in order
`to recover the signaling route availability information that
`may not have been received due to some failure. Finally,
`in ANSI standards the signaling-route-set-congestion-test
`procedure is used to update the congestion status associated
`with a route toward a particular destination.
`The signaling link management function is used to restore
`failed signaling links, to activate new signaling links, and
`to deactivate aligned signaling links. There is a basic
`set of signaling link management procedures, and this
`set of procedures are provided for any international or
`national signaling system. Two optional sets of signaling
`link management procedures are also provided, which allow
`for a more efficient use of signaling equipment when
`signaling terminal devices have switched access to signaling
`data links. The basic set of procedures are signaling link
`activation (used for signaling links that have never been
`put into service, or that have been taken out of service),
`signaling link restoration (used for active signaling links
`that have failed), signaling link deactivation, and signaling
`link set activation. The optional sets of procedures address
`automatic allocation of signaling terminals, and automatic
`allocation of data links and signaling terminals.
`
`B. The Signaling Connection Control Part (SCCP}
`SCCP enhances the services of the MTP to provide the
`functional equivalent of 0 3 ’ s Network layer (layer 3). The
`addressing capability of MTP is limited to delivering a
`message to a node and using a four bit service indicator
`(a sub-field of the SIO) to distribute messages within
`the node. SCCP supplements this capability by providing
`an addressing capability that uses DPC’s plus Subsystem
`Numbers (SSN’s). The SSN is local addressing information
`used by SCCP to identify each of the SCCP users at a
`node. Another addressing enhancement to MTP provided by
`SCCP is the ability to address messages with global titles,
`
`addresses (such as dialed 800 or free phone numbers) that
`are not directly usable for routing by MTP. For global titles
`a translation capability is required in SCCP to translate the
`global title to a DPC + SSN. This translation function can
`be performed at the originating point of the message, or at
`another signaling point in the network (e.g., at an STP).
`In addition to enhanced addressing capability, SCCP
`provides four classes of service, two connectionless and
`two connection-oriented. The four classes are:
`
`Class 0: Basic connectionless class;
`Class 1 : Sequenced (MTP) connectionless class;
`Class 2: Basic connection-oriented class;
`Class 3: Flow control connection-oriented class.
`
`In Class 0 service, a user-to-user information block,
`called a Network Service Data Unit (NSDU), is passed
`by higher layers to SCCP in the node of origin; it is
`transported to the SCCP function in the destination node
`in the user field of a Unitdata message; at the destination
`node it is delivered by SCCP to higher layers. The NSDU’s
`are transported independently and may be delivered out of
`sequence, so this class of service is purely connection-less.
`In Class 1, the features of Class 0 are provided with an
`additional feature that allows the higher layer to indicate
`to SCCP that a particular stream of NSDU’s should be
`delivered in sequence. SCCP does this by associating the
`stream members with a sequence control parameter and
`giving all messages in the stream the same SLS code.
`In Class 2, a bidirectional transfer of NSDU’s is per-
`formed by setting up a temporary or permanent signaling
`connection (a virtual channel through the signaling net-
`work). Messages belonging to the same signaling connec-
`tion are given the same SLS code to ensure sequencing.
`In addition, this service class provides a segmentation and
`reassembly capability. With this capability, if an NSDU is
`longer than 255 octets, it is split into multiple segments
`at the originating node, each segment is transported to the
`destination node in the user field of a Data message, and at
`the destination node SCCP reassembles the original NSDU.
`In Class 3, the capabilities of Class 2 are provided with
`the addition of flow control. Also the detection of message
`loss and missequencing is provided. In the event of lost or
`missequenced messages, the signaling connection is reset
`and notification is given to the higher layers.
`The structure of SCCP is illustrated in Fig. 6, and consists
`of four functional blocks. The SCCP connection-oriented
`control block controls the establishment and release of
`signaling connections and provides for data transfer on
`signaling connections. The SCCP connectionless control
`block provides for the connectionless transfer of data units.
`The SCCP management block provides capabilities beyond
`those of MTP to handle the congestion or failure of either
`the SCCP user or the signaling route to the SCCP user. With
`this capability, SCCP can route messages to backup systems
`in the event failures prevent routing to the primary system.
`The SCCP routing block takes messages received from
`MTP or other SCCP functional blocks and performs the
`
`MODARRESSI AND SKOOG: SS7: AN OVERVIEW
`
`SYS
`
`CISCO SYSTEMS, INC. Ex. 1014 Page 6
`
`

`
`SCCP Users
`
`-Requat
`Kcomact Response
`
`Request Type 1
`Requast Type 2
`f w v
`
`Oriented
`Control
`(SCOC)
`
`Routing
`
`SCCP
`Routing
`Control
`(SCRC)
`
`CL Message
`
`Message Received for
`Unavailable Local SS
`
`SCCP
`Management
`(SCMG)
`
`M T W l n d i c a i
`
`SCCP
`Connection
`less
`Control
`(SCLC)
`
`- -
`c- - -
`
`h
`
`Fig. 6. SCCP functions.
`
`necessary routing functions to either forward the message
`to MTP for transfer or pass the message to other SCCP
`functional blocks.
`
`C. Signaling Network Structures
`Signaling networks consist of signaling points and sig-
`naling links connecting the signaling points together. As
`alluded to earlier, a signaling point that transfers messages
`from one signaling link to another at level 3 is said to be
`a Signal Transfer Point (STP). Signaling points that are
`STP’s can also provide functions higher than level 3, such
`as SCCP and other level 4 functions like ISUP (see Section
`IV-A). When a signaling point has an STP capability and
`also provides level 4 functions like ISUP, it is commonly
`said to have an integrated STP functionality. When the
`signaling point provides only STP capability, or STP and
`SCCP capabilities, it is commonly called a standalone STP.
`Signaling links, STP’s (stand alone and integrated), and
`signaling points with level