Java jolts SS7

Since its introduction in 1995, the Java programming language has spurred a wave of innovation across the telecommunications industry-encompassing wireline, wireless and broadband carriers.

Service providers and telecom software vendors that supply them are making use of the language's platform independence-the ability to run on various computing platforms without alteration-to bring new functionalityto their customer service, network management and other enterprise operations.

The ability of Java-based programs to run across a network-be it the Internet or an enterprise intranet-not only leverages the ubiquity and flexibility inherent in Internet protocol (IP)-based networks but represents a significant time and cost savings for speeding new telecom applications to market.

Upon its introduction, the intelligent network signified something genuinely new. By separating service creation and control from the underlying network infrastructure, it threatened to shake the hold of carriers, switch-makers and service providers.

It offered service providers a faster, more economical route to market, with more potential for service differentiation. The intelligent network also promised to create a space for new players.

Yet its growth has been relatively slow. Customers can be forgiven for asking what has happened to some of the services that common sense suggests should be available now. If you have call waiting, call forwarding and follow me, why can't those services be transported to another phone or carrier? Why is roaming such a hit-and-miss affair? Intelligent networking technology seemed to promise such things, so why the delay?

Incompatible standards The problem can be summed up in one word: incompatibility.

Standards-setting bodies made possible new SS7-based services, but their work has a long way to go. So today, the computer that hangs off a switch may be incompatible with other switches or may not talk to the system that generates voice messages.

The telephony services that run well on a large, expensive system can't necessarily be transported to a smaller system, which can mean that some services are restricted to urban areas.

For the moment, most intelligent networking platforms remain proprietary. Although acknowledging that the standards bodies are slowly working their way through the issues, a recent report from the market research firm Ovum underlines the need for change: "Intelligent networking has the potential to integrate technological capability with customer expectations, competitive demands and regulatory necessities. To achieve this, it is vital that the intelligent networking industry moves from inflexible, proprietary platforms to more open, standard platforms that enable easy service creation and deployment."

Part of the problem is that the standards bodies have yet to address proprietary issues at the level of standard execution environments-such as the Java Virtual Machine that allows different computing platforms to run the programs-and critically important application programming interfaces (APIs) to hardware and SS7 communication protocols.

Thus, network equipment providers still build their solutions using proprietary stacks, which in turn are based on the particular hardware platform the protocol stack vendor chooses. When a carrier chooses a solution from an equipment provider, it is committed not only to a certain stack but also to the hardware-such as the execution environment-that it runs on.

In short, telcos have become independent of the switch vendor, but they remain dependent on the underlying hardware, operating system and SS7 API. This is the key reason carriers simply cannot buy off-the-shelf services from service providers, independent software vendors and network equipment providers.

The lack of standards for APIs and execution environments forces solution developers to choose among incompatible software and hardware providers, or it makes them re-implement the same software onto multiple architectures. One thing that can break this deadlock is a strong standardization of the interfaces between the SS7 stack and the applications that run on it.

Blending intelligent network with the Internet One potential solution to this incompatibility program is to find a way to marry the platform independence of Java technology with the functionality inherent in the intelligent networking architecture. Bringing the computing environment to the Advanced Intelligent Network addresses the incompatibility and portability problems that inhibit growth.

The real strength of applying the Java language to the intelligent network is that, while it can help solve longstanding telecom industry problems, it does so in a way consistent with the latest ideas about information technology architectures. It has the potential to add a new dimension to service deployment, portability and distribution (Figure 1).

This proposed architecture blends intelligent network and Internet technologies to provide state-of-the-art telecom services. Intelligent network services could be assembled on the fly in a plug-and-play fashion, drastically reducing the time and effort to develop services.

What benefits would result? Under such an architecture, it should be possible to propagate intelligence into PCs, Web phones, personal digital assistants (PDAs) and mobile handsets. A further positive result of this Internet/intelligent network integration is that standard Web user interfaces such as browsers and HTML links would find their way into the intelligent network environment. By combining intelligent networking and Internet technology, the architecture would let carriers deliver the content and services for which their customers have been waiting.

Carriers would have the ability to deploy intelligent network services with an Internet interface on any device at the edge of the network, including any Java-enabled end user device (Figure 2).

Another component is JavaBeans-development tools and middleware components that can be reused and combined in various ways to create new applications. The idea provides a bag of JavaBeans that allow intelligent network services to be assembled on the fly for wireless and wireline environments, independent of the underlying hardware and software platforms-somewhat like putting together Lego blocks.

Examples of beans that could be on the market quickly include protocols such as TCAP, ISUP, IS-41, GSM, INAP and AIN that would have Java-based applications (Figure 3).

Using this platform-neutral foundation, providers can develop intelligent networking services such as write once, run anywhere applications. Such an application can be written once and run on any protocol stack on any hardware across any interface-finally delivering on intelligent networking's promise of flexibility and interoperability.

The open marketplace Getting rid of proprietary roadblocks under this scenario likely would set the stage for an open market where equipment providers, independent software vendors, protocol stack vendors, service providers and carriers can market bags of Java-based components.

In fact, the resulting market scenario may resemble nothing so much as the ancient Greek agora-a large forum populated with open-air shops, buyers, sellers and even methods to resolve conflicts by adhering to commonly accepted rules. This environment could produce new business models needed for component-based products.

Participants would select their component vendor based on functionality and value. The market model should stimulate the re-use of existing components and the development of additional or missing functionality-maximizing efficiency and innovation. It also opens the market for innovative new players.

An enhanced intelligent network architecture using the software holds the promise of creating a level playing field for deploying new services. To get the most out of this model, all levels of the industry should be expected to participate, based on their ability to create applications and APIs-including carriers, protocol stack vendors, network equipment providers and information technology/networking systems vendors.

The stack providers-such as Ericsson Infotech, ADC NewNet, DGM&S and integrators such as APiON-will no longer be tied to a particular equipment provider customer. That potentially will increase their market share because they will compete to provide the best applications.

Equipment vendors such as Ericsson, Lucent and Alcatel not only will be able to develop write once, run anywhere component-based applications, but they can customize the applications to meet customers' needs.

For carriers and service providers, the Java-based intelligent networking framework offers different layers of benefits. On one level, the resulting portfolio of platform-independent intelligent networking applications will help carriers control costs and protect their investments. But it will give them the tools to compete by speeding time to market because of the portability, rapid application development cycles and reusability of these new intelligent networking applications.

Ultimately, against the backdrop of telecom competition, carriers that embrace intelligent networking capabilities will grow and prosper, leveraging their ability to create new services to differentiate themselves from less nimble competitors.