Fight the monsters

As competition escalates, service providers have to fight the demons in the signaling network. Applying Internet ideas to the signaling realm can be a solution

VILLAINS COME IN ALL FORMS, SHAPES AND sizes. Some are evil by the malicious acts they perpetrate; others are simply driven by greed. Probably the most dangerous of these scoundrels, however, are those destined to wreak havoc because by design, they have no other choice. These types of villains lurk in the signaling network.

Heightened competition, number portability, interconnection and new service deployments on closed proprietary systems in a dated, inefficient architecture have conspired to threaten network revenue and operations.

Reduced quality of service (QOS) because of increased call setup times, and heightened security threats resulting from a more open network, already have left the signaling network inefficient and vulnerable. And as service providers unite their voice and data networks, things will get even more complicated because signaling traffic then will traverse both networks and technologies.

In this environment, prudent business people must protect and enhance their existing investments while positioning themselves for the next generation.

One of the most economical ways to do this is to infuse successful Internet paradigms such as caching and firewalls into the signaling environment. Using this technique, carriers can drive intelligence from the core to the edge of the signaling network to reduce required SS7 network buildout, improve call setup times and protect their networks.

Deja vu all over again The computer networking industry had faced challenges similar to those that signaling network operators experience today. When computer networking infrastructure was designed, cumbersome mainframes acted as large centralized databases, burying information in the core of data networks. These devices were closed proprietary systems that limited the computer network provider's ability to create new services.

The centralized, closed proprietary nature of the data network in those days was a barrier to Internet growth. It wasn't until the PC broke the mainframe's stranglehold by distributing intelligence to the edge of the network on open, standards-based platforms that the Internet began its exponential growth.

Similarly, the SS7 network was designed in an era of limited competition and service offerings. In this era, the network architecture stored customer service information such as call routing instructions and caller information for a few carriers, kept in large databases buried deep inside the network's core.

Today, as the signaling network evolves, thousands of wireless and wireline carriers are generating billions of customer service queries - including 800 number translations, caller name identification requests and local number portability (LNP) requests - to massive, distant databases. Over time, the number of queries will only increase.

For service providers, this centralization of service intelligence coupled with rapid industry growth has resulted in inefficient operations, increased network vulnerabilities and rising costs.

Up to the task To overcome these challenges, computer networking companies developed new technologies. PCs grew in capability and intelligence as middleware applications and data caching emerged. Each of these developments enhanced QOS offered through data networks.

Application programming interfaces (APIs) became a reality, starting the push for open development platforms. This freed Internet data network integrators from closed infrastructures and enabled rapid deployment of new, customer-specific services.

Lastly, firewalls were put in place to protect the network from fraudulent and malicious activity inherent with increasing network growth and openness. Figure 1 shows the signaling network's current architecture where database queries must reach deeply into the signaling network to access the information stored in the network's core.

The key to these solutions is that they brought much needed intelligence to the edge of the data network. To take the signaling network to the next level, the industry must adopt a similar strategy.

Going to the edge To drive intelligence to the edge of the signaling network, carriers can adopt successful Internet paradigms such as caching and firewalling into the signaling environment. Adoption of these data networking practices will enable service providers to overcome many of the challenges in the SS7 network.

In the Internet, caching was developed to deal with the exponential rise of Web-related traffic. Caching is a special, high-speed mechanism for storing frequently requested information in a nearby location. It lets data network integrators reduce costs, improve response times and optimize network usage.

By leveraging this same technique into the signaling world for number portability, carriers can reduce network buildout requirements, eliminate the need for repeated number portability queries and improve QOS with better call setup times.

Number portability is expensive, whether carriers purchase or subscribe to a number portability database. If a carrier owns its own network, each of the links between its service control point (SCP) and signaling transfer point (STP) will cost around $3500 or more. Depending on configuration, SCPs and STPs can cost between $500,000 and $1.5 million. Service providers also can add to the equation 10% to 15% of the hardware cost annually for maintenance and administration.

These are capital and expense dollars that could be spent on the deployment of new services and new technology that would have a greater impact on profitability.

Carriers that opt to contract with a network infrastructure provider for signaling services find the costs equally staggering. In this case, a service provider can expect to pay between $0.005 and $0.01 per LNP dip. While at first this may not seem like much, consider the billions of queries made to the LNP database. A medium-sized carrier with 2 million subscribers, using five calls per subscriber per day and 20 business days per month, would generate approximately 10 million dips per day and 200 million dips per month and 2.4 billion dips per year. Put in dollar terms at $0.005 per dip and assuming that 80% of the calls were to ported numbers, this would cost approximately $9.6 million per year.

With an edge solution service providers won't have to incur huge LNP expenses. Instead, they can use a number portability caching application. A caching system placed within the signaling network can be equally complementary to network infrastructure and carriers contracting for signaling services.

The offloading option Number portability providers will improve the state of the signaling network by offloading signaling traffic to increase QOS. The offloading of traffic from the core of the signaling network reduces network infrastructure buildout. Reducing the need for network buildouts will enable carriers to reserve dollars for the development and deployment of new services and functionality. This savings will allow carriers to redirect capital dollars toward projects that offer a greater return on investment than they would realize from expanding their signaling networks.

Offloading traffic from the core of the signaling network also enhances QOS by improving call setup times. With the LNP caching system, most LNP queries can be answered from the cache database. Through caching, the signaling hops between the service switching point and STP, as well as the STP and the SCP (where the number portability database resides), are eliminated. This saves milliseconds in call setup time.

Service providers that contract for signaling services also will benefit. They will no longer have to pay for most LNP queries because, in most cases, an LNP caching application will answer between 80% to 90% of its LNP requests from its cache database. Saving 90%, or $8.64 million, on an annual bill of $9.6 million will immediately affect overall company profitability and performance.

Service providers contracting for signaling services also will realize the same QOS enhancements from reduced call setup time. Figure 2 shows how the placement of a caching solution at the edge of the network can reduce the traffic in the core of the signaling network.

The SS7 network once was a relatively safe, closed network. However, with industry deregulation and voice/data network convergence, it has become increasingly vulnerable to fraudulent activities, unproven technologies, inexperienced interconnect partners and network intrusions. Data-centric and traditional voice networks are being inter-connected via IP/SS7 gateways. The potential damage of these threats is staggering - billions lost annually to fraud, compromised operations and declining QOS.

To address this problem, some vendors are developing signaling firewalls based on Internet models. These firewalls will let carriers establish a set of rules that will define their security policy. Based on these rules, they can inspect, block, reroute or pass signaling messages. Although network administrators must be vigilant, these signaling firewalls lock and secure the signaling "vault" maintaining secure, uninterrupted operation.

Standing out New services equal new revenue sources and are an essential means of differentiation. The SS7 network is the key to delivering these new services. But, its existing infrastructure consists of proprietary, dissimilar switches and control points that make rapid deployment of new service offerings ubiquitously across their network difficult. Its closed environment restricts a service provider's ability to expand its opportunities.

In today's competitive market, a new "box" for each network function no longer meets carriers' needs.

The "edge" approach demands a new class of network element - one that is scalable and can handle multiple applications, an open API for customer or third-party development, and services that can generate new revenue while reducing costs.

This type of equipment should be "in-series" or "live" on the SS7 link, allowing carriers to observe, change, redirect or store messages. It also should handle multiple applications on a single platform and be easy to integrate into the telecom network, without requiring a point code.

This will simplify installation and reduce the need for reengineering. The platform should be an open system enabling rapid service development and network convergence. It should be switch independent to ensure easy deployment and enhance current infrastructure investments. Its design should include a distributed architecture for near-infinite scalability and cost-effective manageability. And it should reside at the network's edge or central office and have five 9s reliability.

An API is software that an application program uses to request and carry out lower-level services performed by the computer or telephone system's operating system. The major advantage of an API is that it enables the development of new features and functionality, either in-house or by outsourcing. This allows quicker service development, resulting in faster time to market. Figure 3 shows the platform required to bring intelligence to the edge of the network, including an API layer and service layer.

Keep the villains at bay. The current state of the SS7 network presents many challenges: centralized intelligence, closed systems, fraudulent activities and QOS threats. These challenges restrict service providers' revenue potential, jeopardize network operations and make it difficult to converge.

Forward-thinking service providers should seek technology that will enable a new class of services at the edge of the SS7 network. By applying successful Internet paradigms to the signaling world, with an open platform, service development will be secured and constrained only by the limits of the service provider's imagination.