Signaling Systems

Signaling Systems

This section discusses the nervous system of the network: the signaling system. A great deal of information needs to be passed back and forth between the network elements in the completion of a call and also in the servicing of specialized features. Four main types of signals handle this passing of information:

         Supervisory signals Supervisory signals handle the on-hook/off-hook condition. For instance, when you lift a telephone handset (that is, go off-hook), a signal tells the local exchange that you want a dial tone, and if you exist in the database as an authenticated user, you are then delivered that service; when you hang up (that is, go back on-hook), you send a notice that says you want to remove the service. A network is always monitoring for these supervisory signals to determine when someone needs to activate or deactivate service.

         Address signals Address signals have to do with the number dialed, which essentially consists of country codes, city codes, area codes, prefixes, and the subscriber number. This string of digits, which we refer to as the telephone number, is, in effect, a routing instruction to the network hierarchy.

         Information signals Information signals are associated with activating and delivering various enhanced features. For instance, a call-waiting tone is an information signal, and pressing *72 on your phone might send an information signal that tells your local exchange to forward your calls.

         Alerting signals Alerting signals are the ringing tones, the busy tones, and any specific busy alerts that are used to indicate network congestion or unavailability.

Signaling takes place in two key parts of the network: in the access network, where it's called loop signaling, and in the core, where it's called interoffice signaling (see Figure 5.10).

Figure 5.10. Customer loop and interoffice signaling

With analog loop signaling, two types of starts exist:

         Ground start Ground start means that when you seize that line, it's immediately grounded so that no other call can potentially conflict with it. Ground start is used with a contentious system, perhaps a PBX at a corporate enterprise, to avoid collisions. For example, say you seize a trunk and place a call, and now you're in the ringing state. There are short periods of silence between ringing tones. The local exchange could mistake one of these periods of silence to mean that that trunk is available and try to send a call in over that same trunk that you're trying to place a call out over; this would cause a collision (referred to as glare). Consequently, when you're dealing with systems and contention for the resource, grounding the trunk up front is the most efficient procedure.

         Loop start Pay telephones and residential phones use loop start, which means that the circuit is grounded when the connection is completed.

There are various start standards for digital subscriber signaling, and they are defined in accordance with the service being provided.

Interoffice signaling has been through several generations of signaling approaches. In the first generation, called per-trunk signaling, the complete path all the way to the destination point is set up in order to just carry the signaling information in the first place (see Figure 5.11). This method uses trunks very inefficiently; trunks may be put into place to carry 20 or 30 ringing tones, but if nobody is on the other end to take that call, the network trunk is being used but not generating any revenue. Also, when a call is initiated and begins to progress, you can no longer send any other signaling information over that trunk; being able to pass a call-waiting tone, for instance, would not be feasible.

Figure 5.11. Per-trunk signaling

We have moved away from the per-trunk signaling environment to what we use today common-channel signaling (see Figure 5.12). You can think of common-channel signaling as being a separate subnetwork over which the signaling message flows between intelligent networking components that assist in the call completion and assist in the delivery of the service logic needed to deliver the requested feature. Today, we predominantly use the ITU-T standard for common-channel signaling: SS7.

Figure 5.12. Common-channel signaling

SS7 Architecture

SS7 is critical to the functioning and operation of the modern network. With SS7, a packet data network overlays and controls the operation of the underlying voice networks; signaling information is carried on an entirely different path than voice and data traffic. Signaling doesn't take a great deal of time, so we can multiplex many signaling messages over one channel, and that's why the signaling system is a packet network. The signaling system takes advantage of the efficiencies of statistical multiplexing for what is essentially bursty data. The SS7 signaling data link is a full-duplex digital transmission channel that operates at either 56Kbps or 64Kbps, depending on the standards under which the network is operating (for example, T-carrier and J-carrier operate at 56Kbps, E-carrier operates at 64Kbps).

SS7 is an entire architecture that performs out-of-band signaling (that is, signaling in which the conversation and the signaling take place over different paths) in support of the information-exchange functions that are necessary in the PSTN, such as call establishment, billing, and routing. Database access messages convey information between toll centers and centralized databases to permit real-time access to billing-related information and other services. The SS7 architecture defines the procedures for the setup, ongoing management, and clearing of a call, and it allows you to pass along customer-related information (for example, the identity of the caller, the primary carrier chosen) that helps in routing calls. The efficiency of the network also results in faster call setup times and provides for more efficient use of the circuits when carrying the voice or data traffic. In addition, SS7 supports services that require signaling during a call as it is occurring not in the same band as the conversation.

SS7 permits the telephone company to offer one database to several switches, thereby freeing up switch capacity for other functions, and this is what makes SS7 the foundation for INs and advanced intelligent networks (AINs). It is also the foundation for network interconnection and enhanced services. Without SS7, we would not be able to enjoy the level of interoperability we have today. SS7 is also a key to the development of new generations of services on the Internet, particularly those that support traditional telephony services. To be able to accommodate features such as call forwarding, call waiting, and conference calling, you must be able to tap into the service logic that delivers those features. Until quite recently, the Internet has not been able to do this, but the year 2000 saw the introduction of SS7 gateways, which allow an interface between circuit-switched networks (with their powerful SS7 infrastructure) and the emerging packet-switched networks that need to be capable of handling the more traditional type of voice communications on a more cost-effective basis.

As Figure 5.13 shows, there are the three prerequisite components in the SS7 network: service switching points (SSPs), service control points (SCPs), and signal transfer points (STPs).

Figure 5.13. An SS7 network

SSPs

SSPs are the switches that originate and terminate calls. They receive signals from the CPE and perform call processing on behalf of a user. The user, by dialing particular digits, triggers the network to request certain services. For instance, if you preface a number with a toll-free prefix, that toll-free arrangement triggers the local exchange, or SSP, to initiate a database lookup to determine the physical address of that toll-free number (that is, where it resides in the network). The SSP reaches into the network to find the database that can translate the toll-free number into a physical address in order to then complete the toll-free call. The SSP does this by interacting with a device called the SCP, which is discussed shortly.

SSPs are typically implemented at local exchanges, access tandem offices, or toll centers that contain the network-signaling protocols. The SSP serves as the source and destination point for the SS7 messages.

SCPs

The second key component of SS7 is SCP. This is the network element that interfaces with the SSP as well as the STP. Most importantly, the SCP is the network element that contains the network configuration and call-completion database; in other words, it contains the service logic to act on the types of calls and features the users are requesting. SCPs are centralized nodes that contain service logic basically software and databases for the management of the call. They provide functions such as digit translation, call routing, and verification of credit cards. The SCPs receive traffic from the SSP via the STP and return responses, based on that query, via the STP.

STPs

The STP is responsible for translating the SS7 messages and then routing those messages between the appropriate network nodes and databases. Notice in Figure 5.13 that the SCPs and the STPs are both redundant, and that the links running between them are also redundant.

SS7 and the Internet

If a network loses its signaling system, it loses the capability to complete calls, as well as to do any form of billing or passing along of management information. This makes SS7 critical. The SS7 signaling data link, as mentioned earlier in the chapter, is a full-duplex digital transmission channel that operates at either 56Kbps or 64Kbps. A variety of other SS7 links are defined as well, and each has specific uses within the signaling network:

         A (access) links An A link interconnects an STP with either an SSP or an SCP. The SSP and SCP, collectively, are referred to as the signaling endpoints. A message sent to and from the SSPs or SCPs first goes to its home STP, which, in turn, processes or routes the message.

         B (bridge) links, D (diagonal) links, and B/D links A B link connects an STP to another STP. Typically, a quad of B links interconnect peer (or primary) STPs (for example, the STPs from one network to the STPs of another network). The distinction between a B link and a D link is rather arbitrary, and such links may be referred to as B/D links.

         C (cross) links C links interconnect mated STPs.

         E (extended) links E links provide enhanced reliability by providing a set of links from the SSP to a second STP pair.

         F (fully associated) links F links are links that directly connect to signaling endpoints.