SS7 Network Architecture

SS7 can employ different types of signaling network structures. The choice between these different structures can be influenced by factors such as administrative aspects and the structure of the telecommunication network to be served by the signaling system.

The worldwide signaling network has two functionally independent levels:

• International

• National

This structure makes possible a clear division of responsibility for signaling network management. It also lets numbering plans of SS7 nodes belonging to the international network and the different national networks be independent of one another.

SS7 network nodes are called signaling points (SPs). Each SP is addressed by an integer called a point code (PC). The international network uses a 14-bit PC. The national networks also use a 14-bit PC—except North America and China, which use an incompatible 24-bit PC, and Japan, which uses a 16-bit PC. The national PC is unique only within a particular operator's national network. International PCs are unique only within the international network. Other operator networks (if they exist) within a country also could have the same PC and also might share the same PC as that used on the international network. Therefore, additional routing information is provided so that the PC can be interpreted correctly—that is, as an international network, as its own national network, or as another operator's national network. The structure of point codes is described in Chapter 7, "Message Transfer Part 3 (MTP3)."

Signaling Links and Linksets

SPs are connected to each other by signaling links over which signaling takes place. The bandwidth of a signaling link is normally 64 kilobits per second (kbps). Because of legacy reasons, however, some links in North America might have an effective rate of 56 kbps. In recent years, high-speed links have been introduced that use an entire 1.544 Mbps T1 carrier for signaling. Links are typically engineered to carry only 25 to 40 percent of their capacity so that in case of a failure, one link can carry the load of two.

To provide more bandwidth and/or for redundancy, up to 16 links between two SPs can be used. Links between two SPs are logically grouped for administrative and load-sharing reasons. A logical group of links between two SP is called a linkset. Figure 4-2 shows four links in a linkset.

A number of linksets that may be used to reach a particular destination can be grouped logically to form a combined linkset. For each combined linkset that an individual linkset is a member of, it may be assigned different priority levels relative to other linksets in each combined linkset.

A group of links within a linkset that have the same characteristics (data rate, terrestrial/satellite, and so on) are called a link group. Normally the links in a linkset have the same characteristics, so the term link group can be synonymous with linkset.

Routes and Routesets

SS7 routes are statically provisioned at each SP. There are no mechanisms for route discovery. A route is defined as a preprovisioned path between source and destination for a particular relation. Figure 4-3 shows a route from SP A to SP C.

All the preprovisioned routes to a particular SP destination are called the routeset. Figure 4-4 shows a routeset for SSP C consisting of two routes.

The following section discusses the SP types.

Node Types

There are three different types of SP (that is, SS7 node):

• Signal Transfer Point

• Service Switching Point

• Service Control Point

Figure 4-5 graphically represents these nodes.

The SPs differ in the functions that they perform, as described in the following sections.

Signal Transfer Point

A Signal Transfer Point (STP) is responsible for the transfer of SS7 messages between other SS7 nodes, acting somewhat like a router in an IP network.

An STP is neither the ultimate source nor the destination for most signaling messages. Generally, messages are received on one signaling link and are transferred out another. The only messages that are not simply transferred are related to network management and global title translation. These two functions are discussed more in Chapters 7 and 9. STPs route each incoming message to an outgoing signaling link based on routing information contained in the SS7 message. Specifically, this is the information found in the MTP3 routing label, as described in Chapter 7.

Additionally, standalone STPs often can screen SS7 messages, acting as a firewall. Such usage is described in Chapter 15, "SS7/C7 Security and Monitoring."

An STP can exist in one of two forms:

• Standalone STP

• Integrated STP (SP with STP)

Standalone STPs are normally deployed in "mated" pairs for the purposes of redundancy. Under normal operation, the mated pair shares the load. If one of the STPs fails or isolation occurs because of signaling link failure, the other STP takes the full load until the problem with its mate has been rectified.

Integrated STPs combine the functionality of an SSP and an STP. They are both the source and destination for MTP user traffic. They also can transfer incoming messages to other nodes.

Service Switching Point

A Service Switching Point (SSP) is a voice switch that incorporates SS7 functionality. It processes voice-band traffic (voice, fax, modem, and so forth) and performs SS7 signaling. All switches with SS7 functionality are considered SSPs regardless of whether they are local switches (known in North America as an end office) or tandem switches.

An SSP can originate and terminate messages, but it cannot transfer them. If a message is received with a point code that does not match the point code of the receiving SSP, the message is discarded.

Service Control Point

A Service Control Point (SCP) acts as an interface between telecommunications databases and the SS7 network. Telephone companies and other telecommunication service providers employ a number of databases that can be queried for service data for the provision of services. Typically the request (commonly called a query) originates at an SSP. A popular example is freephone calling (known as toll-free in North America). The SCP provides the routing number (translates the toll-free number to a routable number) to the SSP to allow the call to be completed. For more information, see Chapter 11, "Intelligent Networks (IN)."

SCPs form the means to provide the core functionality of cellular networks, which is subscriber mobility. Certain cellular databases (called registers) are used to keep track of the subscriber's location so that incoming calls may be delivered. Other telecommunication databases include those used for calling card validation (access card, credit card), calling name display (CNAM), and LNP.

SCPs used for large revenue-generating services are usually deployed in pairs and are geographically separated for redundancy. Unless there is a failure, the load is typically shared between two mated SCPs. If failure occurs in one of the SCPs, the other one should be able to take the load of both until normal operation resumes.

Queries/responses are normally routed through the mated pair of STPs that services that particular SCP, particularly in North America.

See Chapters 10, "Transaction Capabilities Application Part (TCAP)," and 11, "Intelligent Networks (IN)," for more information on the use of SCPs within both fixed-line and cellular networks. See Chapters 12, "Cellular Networks," and 13, "GSM and ANSI-41 Mobile Application Part (MAP)," for specific information on the use of SCPs within cellular networks.

The following section introduces the concept of link types.

Link Types

Signaling links can be referenced differently depending on where they are in the network. Although different references can be used, you should understand that the link's physical characteristics remain the same. The references to link types A through E are applicable only where standalone STPs are present, so the references are more applicable to the North American market.

Six different link references exist:

• Access links (A links)

• Crossover links (C links)

• Bridge links (B links)

• Diagonal links (D links)

• Extended links (E links)

• Fully associated links (F links)

The following sections cover each link reference in more detail.

Access Links (A Links)

Access links (A links), shown in Figure 4-6, provide access to the network. They connect "outer" SPs (SSPs or SCPs) to the STP backbone. A links connect SSPs and SCPs to their serving STP or STP mated pair.

Cross Links (C Links)

Cross links (C links), shown in Figure 4-7, are used to connect two STPs to form a mated pair—that is, a pair linked such that if one fails, the other takes the load of both.

C links are used to carry MTP user traffic only when no other route is available to reach an intended destination. Under normal conditions, they are used only to carry network management messages.

Bridge Links (B Links)

Bridge links (B links) are used to connect mated pairs of STPs to each other across different regions within a network at the same hierarchical level. These links help form the backbone of the SS7 network. B links are normally deployed in link quad configuration between mated pairs for redundancy.

Figure 4-8 shows two sets of mated pairs of B links.

Diagonal Links (D Links)

Diagonal links (D links), shown in Figure 4-9, are the same as B links in that they connect mated STP pairs.

Figure 4-9. D Links

[View full size image]

The difference is that they connect mated STP pairs that belong to different hierarchical levels or to different networks altogether. For example, they may connect an interexchange carrier (IXC) STP pair to a local exchange carrier (LEC) STP pair or a cellular regional STP pair to a cellular metro STP pair.

As mentioned, B and D links differ in that D links refer specifically to links that are used either between different networks and/or hierarchical levels, as shown in Figure 4-10.

Figure 4-10. Existence of an STP Backbone and STP Hierarchy

[View full size image]

Extended Links (E Links)

Extended links (E links), shown in Figure 4-11, connect SSPs and SCPs to an STP pair, as with A links, except that the pair they connect to is not the normal home pair. Instead, E links connect to a nonhome STP pair. They are also called alternate access (AA) links. E links are used to provide additional reliability or, in some cases, to offload signaling traffic from the home STP pair in high-traffic corridors. For example, an SSP serving national government agencies or emergency services might use E links to provide additional alternate routing because of the criticality of service.

Fully-Associated Links (F Links)

Fully-associated links (F links), shown in Figure 4-12, are used to connect network SSPs and/or SCPs directly to each other without using STPs. The most common application of this type of link is in metropolitan areas. F links can establish direct connectivity between all switches in the area for trunk signaling and Custom Local Area Signaling Service (CLASS), or to their corresponding SCPs.

Figure 4-13 shows an SS7 network segment. In reality, there would be several factors more SSPs than STPs.

Signaling Modes

The signaling relationship that exists between two communicating SS7 nodes is called the signaling mode. The two modes of signaling are associated signaling and quasi-associated signaling. When the destination of an SS7 message is directly connected by a linkset, the associated signaling mode is being used. In other words, the source and destination nodes are directly connected by a single linkset. When the message must pass over two or more linksets and through an intermediate node, the quasi-associated mode of signaling is being used.

It's easier to understand the signaling mode if you examine the relationship of the point codes between the source and destination node. When using the associated mode of signaling, the Destination Point Code (DPC) of a message being sent matches the PC of the node at the far end of the linkset, usually referred to as the far-end PC or adjacent PC. When quasi-associated signaling is used, the DPC does not match the PC at the far end of the connected linkset. Quasi-associated signaling requires the use of an STP as the intermediate node because an SSP cannot transfer messages.

In Figure 4-14, the signaling relationships between each of the nodes are as follows:

• SSP A to SSP B uses quasi-associated signaling.

• SSP B to SSP C uses associated signaling.

• STP 1 and STP 2 use associated signaling to SSP A, SSP B, and each other.

As you can see from Figure 4-14, associated signaling is used between nodes that are directly connected by a single linkset, and quasi-associated signaling is used when an intermediate node is used. Notice that SSP C is only connected to SSP B using an F link. It is not connected to any other SS7 nodes in the figure.

When discussing the signaling mode in relation to the voice trunks shown between the SSPs, the signaling and voice trunks follow the same path when associated signaling is used. They take separate paths when quasi-associated signaling is used. You can see from Figure 4-14 that the signaling between SSP B and SSP C follows the same path (associated mode) as the voice trunks, while the signaling between SSP A and SSP B does not follow the same path as the voice trunks.

Signaling Network Structure

Standalone STPs are prevalent in North America because they are used in this region to form the backbone of the SS7 network. Attached to this backbone are the SSPs and SCPs. Each SSP and SCP is assigned a "home pair" of STPs that it is directly connected to. The network of STPs can be considered an overlay onto the telecommunications network—a packet-switched data communications network that acts as the nervous system of the telecommunications network. Figure 4-15 shows a typical example of how SSPs are interconnected with the STP network in North America.

STPs are not as common outside North America. Standalone STPs typically are used only between network operators and/or for applications involving the transfer of noncircuit-related signaling. In these regions, most SSPs have direct signaling link connections to other SSPs to which they have direct trunk connections. Figure 4-16 shows an example of this type of network with most SSPs directly connected by signaling links.

SSPs often have indirect physical connections to STPs, made through other SSPs in the network. These are usually implemented as nailed-up connections, such as through a Digital Access Cross-Connect System or other means of establishing a semipermanent connection. Logically, these SSPs are directly connected to the STP. The signaling link occupies a digital time slot on the same physical medium as the circuit-switched traffic. The SSPs that provide physical interconnection between other SSPs and an STP do not "transfer" messages as an STP function. They only provide physical connectivity of the signaling links between T1/E1 carriers to reach the STP. Figure 4-17 shows an example of a network with no STP connection, direct connections, and nondirect connections. SSP 1 is directly connected to an STP pair. SSP 4 uses direct signaling links to SSP 2 and SSP 3, where it also has direct trunks. It has no STP connection at all. SSP 2 and SSP 3 are connected to the STP pair via nailed-up connections at SSP 1.

Figure 4-17. Example of Direct and Indirect SSP Interconnections to STPs

[View full size image]

Normally within networks that do not use STPs, circuit-related (call-related) signaling takes the same path through the network as user traffic because there is no physical need to take a different route. This mode of operation is called associated signaling and is prevalent outside North America. Referring back to Figure 4-14, both the user traffic and the signaling take the same path between SSP B and SSP C.

Because standalone STPs are used to form the SS7 backbone within North America, and standalone STPs do not support user traffic switching, the SSP's signaling mode is usually quasi-associated, as illustrated between SSP A and SSP B in Figure 4-14.

In certain circumstances, the SSP uses associated signaling within North America. A great deal of signaling traffic might exist between two SSPs, so it might make more sense to place a signaling link directly between them rather than to force all signaling through an STP.