SS7 signals significance: Growing dependence on signaling networks calls for new vigilance

We're well versed in the competitive age mantra-introduce new services, satisfy demanding customers, deal with more complex traffic and manage it all like a circus juggler.

How can carriers cope? One answer is good SS7 network management.

Reliability and robustness-the confidence that even sophisticated services and capabilities will be available-are the telecom industry's greatest assets.

But the growth and competition in today's telecom industry make reliability and robustness harder to achieve. Carriers must make routing and service information consistent across an increasingly diverse network. They must quickly activate new services and calling features.

These features must operate smoothly with the networks and services already in place. Signaling network capacity becomes a complex problem that carriers must manage before problems start if the network is to cope with more traffic and translations.

Consider the configuration challenge that local number portability presents.

Local exchange carriers must be prepared because in just a few months customers will be able to change local providers as easily as they change long-distance carriers, and they will be taking their phone numbers with them.

It sounds simple enough, except that the physical network still needs to know where to find a customer on the logical network. That knowledge comes from a signal control point (SCP) and uses a signal transfer point (STP)-the SS7 network's own switching system-to do the appropriate routing.

A typical large local service provider may need to handle more than 16 million new STP address translations in the first year of local portability-a 50% increase, Bellcore estimates.

That kind of growth will almost certainly require adding STPs that must be configured and provided with links to the rest of the network (see sidebar on page 42).

And because the network equipment market has become just as diverse as the market for telecommunications services, those STPs likely will be manufactured and programmed by more than one company.

If the STPs in an SS7 network don't process the configuration information in the same manner, they just won't work. A centralized configuration system to administer the network is needed (Figure 1).

Configuration management needs to validate the databases that network elements use to constantly verify that addresses are translated accurately. Next generation configuration managers continuously audit network elements to ensure that each pair of systems working together uses the same information. If they do not, an automated command corrects the problem.

These systems also support gateway screening that defines which SS7 messages are allowed into the network. The new systems should be able to automatically download new gateway screening rules to the STPs and test the rules before they are implemented in the network. This helps ensure that routes are properly defined and translations are correct, avoiding routing failures and errors.

Data for the taking To keep an SS7 network's performance at the levels that support customer confidence in their intelligent network services, the network operator must know ahead of time when the equipment's capacity will be exhausted, must monitor the network's quality of service and must effectively manage the network's traffic.

The most efficient new systems rely on collecting, archiving, analyzing and reporting traffic data-volume measurements from links and from various components of SS7 nodes.

Such a system provides an engineering report that might show occupancy, overflow percentages, holding times or attempted traffic exceeding thresholds that the operating company sets. If the problems are traced to congestion in the SS7 network, the network operator can plan to expand the capacity of the overloaded nodes or links.

These new performance managers operate in a client/server environment, and they are highly scalable and economically expandable to grow with the network of either a large provider or a smaller competitor. Their graphical user interfaces present network traffic data in easy-to-understand formats that help operators make the quick decisions needed to manage a complex network.

Bellcore's DCOS-2000 system has a binary interface that can collect large volumes of data but not interfere with the performance it is monitoring. Older products must directly translate performance data from ASCII as it is collected, slowing the network element's performance.

Carriers must know what's going on in the network. Fortunately, with an SS7 network, carriers can tap into a rich source of data to do everything from surveillance and fraud management to customer-specific reports that expedite marketing for new lines and services.

Every signaling message contains the initiating customer's number-if a call needs to be handled by the intelligent network-and the destination number. It also may contain information about how the customer uses the service, and if a problem develops, the information in an SS7 message will reveal the reason.

One such tool runs on its own server and gathers data through monitors on SS7 links-between an STP and an SCP database, between an STP and a service switching point (SSP), or any of the other links in the network-to unobtrusively capture data from SS7 messages in real time (Figure 2).

Such a system includes link monitors that collect data at intervals as short as five seconds. The monitoring analysis system can convert this data into reports or alarms in a variety of formats to support network-wide SS7 surveillance, customer-specific reporting, SS7 billing or fraud reduction.

The heart of any problem Surveillance at this level can keep network operators apprised of each link's state and traffic load, and can provide early warnings of conditions such as a link carrying too high a load, an out-of-service link, an isolated node, or a mass-calling event such as a radio contest or concert ticket sale.

Not even the customers who helped create the problem by calling the same number at the same time can be expected to understand.

The best bet to maintain customer confidence is to demonstrate reliability by taking advantage of the signaling network's possibilities for surveillance and quick reaction.

These systems also can filter data that SS7 link monitors collect by link, link set, channel, direction, time interval, message priority, originating or destination network element, trunk circuit, and message class. The operator then can reconstruct an event to prevent a problem from recurring.

It is also essential to effectively monitor and analyze problems at the Telecommunications Management Network (TMN) layer across direct-switched, SS7 and transport network domains. Although the three networks frequently are thought of as separate, and although they function separately in many respects, they usually share at least some network elements and physical links. If one of those shared facilities has a problem, getting the message from more than one source can add to the confusion.

Taking the view of the network management layer allows for the possibility of correlating fault reports, whether automatically or by a technician, which can lead network operators more quickly to a failure's cause and its correction.

Consider a simple fiber cut. One fiber might host a signaling link-labeled A for access-between an STP and an SSP, as well as host systems carrying voice signals for both the direct-switched and the intelligent networks. When the fiber is cut, alarms in all three systems will indicate that their links have gone down, but the alarms that the various switching systems generate aren't equipped to know whether the failure is in another switching system or in the fiber link between them (Figure 3).

These monitoring systems also do not recognize that other networks are experiencing the same link outage-not unless a network management layer system is in place for event correlation. Each network could be sending technicians to investigate problems in any of the switching systems, as well as in the fiber connection.

But with event correlation, software expert systems use experts' knowledge about how to compare alarm information from all three networks. This kind of system can even generate trouble tickets and route them to technicians according to location or function when special expertise is needed.

We clearly can see the day when almost every call made in the U.S. will rely in some way on specialized control communicated over SS7 networks. That call for celebration is also a call for robust, full-functionality network management systems to support network growth, take advantage of the service and market opportunities and ensure the kind of network integrity and availability that keeps customers coming back for more.

No longer bound by previous regulations that required Bell regional holding companies to deploy a complete set of SS7 facilities in each LATA, BellSouth is migrating to an SS7 hub architecture.

Six gateway signal transfer points (STPs), each serving one to four LATA STPs, are interconnected in a fully meshed manner (see figure).

With this architecture, other carriers can connect to BellSouth's SS7 network at a single interconnection point and gain access to the entire network. This reduces the carriers' connection costs while streamlining BellSouth's network.

BellSouth's number of STPs will decrease from 40 pairs to 19 pairs. Because the new STPs are more powerful than the ones they are replacing, capacity will not decrease, carrier representatives say.

The signaling hub design also will help in load balancing. If one gateway gets close to capacity, some of its functionality can be moved to another gateway. Another benefit of the new architecture is that other carriers can be billed for actual rather than estimated use.

DSC is supplying the STPs for BellSouth's signaling hub network. The carrier has completed three of four phases of an 18-month plan for implementing the new architecture, company officials say.

A new generation of network management systems is taking a layered approach to systems architecture. For networks, Telecommunications Management Network addresses various management layers (see figure).

Configuration management. This layer activates the translations, based on network topology, that provide for SS7 services and route the network messages that make them happen.

Performance management. This layer collects traffic data for managing network traffic, forecasting and managing equipment capacity and surveying the network in near real time.

Fault management. This layer looks for alarm conditions throughout the network, correlates network events for root cause fault analysis and tests to support these activities.

Accounting management. This layer collects billing data from SS7 messages, including the complex tasks of interconnecting wireless and wireline, local exchange and interexchange networks.

Security management. This layer provides secure access to network elements and operations support systems (OSSs) and detects fraudulent calls.

TMN enhances management information exchange by calling for interconnection among OSSs and telecom equipment. The TMN architecture has a preferred platform of client/server-based workstations and servers that uses standard interfaces, including protocols and messages. But it recognizes that some embedded OSSs and network elements will fail to meet those requirements, so it allows for interfacing with the embedded products.

TMN also supports management information display and access using Open System Interconnection-based interfaces where appropriate, such as the Q-3 interface.

It calls for standard interfaces among different building blocks from one network management system to another or between the network management system and the network element. Finally, it uses an object- oriented approach for information storage, the system design, and handling managed objects.

Every component in the network, whether it be a switch, a piece of fiber equipment or a signal transfer point (STP), is considered to be a managed object.

TMN defines five management functional areas.

Fault management involves processes such as alarm surveillance, fault localization and testing.

Configuration management provides resource and service provisioning, status and control, and service activation in the different network components.

Accounting management involves the collection of billing data and message accounting. This can be complex because of interconnection issues surrounding wireless and wireline networks, as well as local and long-distance.

Performance management includes traffic data collection and network traffic management. This type of data is used for near real-time surveillance of the network, forecasting equipment exhaust and monitoring quality of service. It also encompasses the rerouting of calls within a network to alleviate traffic congestion.

Security management provides secure access to network elements and OSSs and fraudulent call detection.

The TMN architecture also is represented by hierarchical management layers. The following functions are performed within these layers:

* The business management layer provides a total network view and strategic planning.

* The service management layer handles interfacing with customers and other providers, and it is responsible for quality of service.

* The network management layer covers a wide geographical area, and it controls and coordinates the network view of all network elements within its scope. It also provisions network capabilities to support service to customers and interacts with the service management layer for performance and wage information.

* The element management layer controls a subset of network elements and provides a gateway to the network management layer while maintaining data about its network elements. The element layer deals with functions in the network elements.