
April 1999
Will The Intelligent Network Still Matter In An
IP-Centric World?
BY JOHN DE KEIJZER AND DOUGLAS TAIT
Intelligent network (IN) technology overlays the public voice network with intelligence
that enables new services, and it has been with us for several decades. But in contrast
with rapidly developing Internet technologies, the historically proprietary world of IN
services has developed services slowly. And while IN has some widely used services, such
as 800 numbers, call forwarding, and last-number redial, its full potential hasn't come
close to being realized, largely because IN services involve proprietary hardware and
software platforms at the SS7 level, with proprietary services and applications written on
top of them. Nearly one year ago, Sun Microsystems (www.sun.com)
organized the Java in Advanced Intelligent Networks (JAIN) industry initiative to address
the problem of incompatibility within the IN and SS7 environments. JAIN seeks to define
common interfaces that will enable services and protocols to run anywhere in the network,
establishing a level playing field that will benefit carriers, network equipment
providers, and application developers.
Lack of standardization has made it difficult, time consuming, and expensive for
carriers to deploy new services quickly -- in their heterogeneous networks, deploying new
services means programming to many different devices. To offer ubiquity of service, a
service provider may need to program two or more switch types, to program more than one
call model, and to take into account the interoperability between the different SS7 stacks
in their network.
At its simplest level, JAIN will define open interfaces and APIs so that services
written for one area can be deployed across many. Most telecom service providers have
mixed environments, but they will be able to use JAIN-compliant services throughout their
network. Not only can this help to solve their network integration problems, but it will
also free up staff resources to create revenue-generating new services.
JAIN is conceived as a telecom industry framework designed to spur a new wave of
telecom services by blending Internet and IN technologies to drive convergence between
traditional voice networks and IP-based data networks. A key goal of JAIN is to enable
service portability, allowing service providers to provide the same telephony services
across different networks and delivery platforms. Based on Sun's JavaBeans component
architecture, the JAIN framework promises to make IN telecom application development --
including computer telephony integration -- faster, simpler, and less expensive through
platform-independent Java technology.
INITIAL FOCUS: SS7
The initial focus of JAIN was on defining JAIN interfaces at the SS7 protocol stack layer.
Key SS7 protocol stack vendors -- DGM&S Telecom (www.dgms.com),
ADC NewNet (now ADC Telecommunications -- www.adc.com)
and Ericsson InfoTech AB (www.ericsson.se/infotech)
-- were JAIN's charter members. In January, however, Sun expanded the JAIN initiative to
begin focusing on the higher level of application service creation environments --
establishing common interfaces that will enable IN applications to be "portable"
across different network infrastructures. IBM (www.ibm.com)
and Bellcore (www.bellcore.com) are the first
members of this new expert group, with Bellcore and Trillium (www.trillium.com) also contributing to the SS7
consortium.
Let's take a closer look at these two "levels" of the JAIN architecture and
how they will complement each other. At the SS7 level, the goal is to develop a JAIN API
that will provide protocol independence and platform (both hardware and software)
independence, while supporting elements such as TCAP, ISUP, and BISUP.
The next level is the service creation layer, where JAIN will allow the creation of
capability sets: JavaBeans or applets that can be reused. For example, a "collect
digits" capability, once created, can be used for dialing numbers, credit cards, or
calling cards. Finally, complete JAIN-based IN services can be created by simply stringing
together the various JavaBeans building blocks. We thus have a complete, open architecture
for IN that for the first time supports reuse of components, speeding application time
while dramatically lowering costs.
More significantly, however, JAIN provides an ideal mechanism to divorce the SS7
gateway from the service itself, allowing the service to be run on a separate host that
can be anywhere within the network. The first implementations of JAIN may in fact have the
gateway and service running on the same platform or SCP (service control point), as is now
the case for SS7 applications. But the JAIN framework will provide operators with the
unprecedented freedom to move services out into the network -- or even onto intelligent
handsets or other advanced end-user terminals. Theoretically, a user could download a Java
service from Web site for call forwarding, customize the application for personal
preferences, and then activate call forwarding from the handset. Such functionality may be
science fiction now, but it's technologically feasible.
CONSIDERING THE PBX
PBX technology is a prime potential application area for JAIN -- especially as VoIP (voice
over IP) and IP-based PBX technology come to the fore. JAIN-compliant technology will
enable these next-generation PBX systems to utilize network-resident Java-based services
that can leverage the capabilities of the existing IN for their own purposes. These
systems could thus use JAIN to provision phone calls, to perform their own routing, and to
do real-time bandwidth allocation. Caller ID, an IN-based service, could start to flow
through the JAIN interface for use in a PBX or IP-based PBX environment.
The key point here is that JAIN links these next-generation devices with the existing
IN infrastructure, ensuring that the highly reliable, established IN services that already
exist need not be reinvented as new IP-based technologies migrate onto the PSTN.
JAIN AND JTAPI: COMPLEMENTARY STANDARDS
The JAIN initiative isn't the first time Sun's Java technology has been used to establish
a telecom-industry specification: JTAPI (Java Telephony Application Program Interface) was
developed several years ago as an effort to use Java's platform-independent capabilities
to bring application portability to the CTI space.
Although they both use Java technology to bring the benefits of interoperability to
previously "closed" environments, JTAPI and JAIN are complementary and distinct
technologies. JAIN is focused on the public network domain, while JTAPI is oriented to
private network domains as typified by enterprise CTI and call center operations. While
these two worlds are edging closer together, for the moment they are still separate realms
with distinct requirements.
VoIP is one of the most exciting areas in telephony. But the real opportunity is not
simply about replacing analog circuit-switched voice with VoIP -- it's about offering new
services. Voice traffic can be shifted from analog to VoIP, but for most operators, the
cost model doesn't justify such a move if it is done solely for voice. There's no reason
for an operator to shift over to VoIP unless they get something more. And that something
more could be unified messaging, transforming voice mail into e-mail, or e-mail messages
into voice messages (via the Java speech interface, which can convert text into spoken
messages). These are some of the next-generation applications we may start seeing as
JAIN-related efficiencies and new services make their way into the network.
Because JAIN is an implementation of Java, this new environment enables providers to
leverage Java's capabilities such as some of the existing APIs. Good examples would be
Java Wallet -- enabling electronic cash for simple, secure transactions over the Web --
and JavaSpeech, as noted above.
Clearly, the JAIN framework and Java offer significant opportunities to developers of
CTI platforms and applications. As they develop and specify new products, the CTI
community should keep abreast of JAIN developments so they can take full advantage of
JAIN's benefits.
Especially with rapidly emerging technologies such as VoIP and IP-based PBX systems,
developers will need to interface to the IN network to access services and to route calls.
JAIN holds the promise of being a highly functional, versatile interface that will help
manage the transition to an open, IP-based communications infrastructure from proprietary
CTI and IN environments. In fact, JAIN will allow the VoIP revolution to occur more
quickly: much of the infrastructure for VoIP simply does not exist, and JAIN leverages the
existing IN infrastructure so that existing IN services can be accessed -- not laboriously
re-invented from scratch.
The ever-growing group of telecom companies committed to making JAIN a reality is now
at work to propose preliminary specifications for JAIN at the SS7 level. Sun and its JAIN
development allies believe that JAIN is a powerful enabling technology that will liberate
IN from its proprietary heritage. By providing the critical link between IN and the
Internet and bringing the capabilities of Java to bear at the very heart of the public
network, the JAIN movement is working to help to bring these two worlds together. The
ultimate goal: a truly converged, function-rich public network that will combine the best
of both worlds and deliver a wealth of new services.
John de Keijzer is manager, Intelligent Networks and Enhanced Network Services at
Sun Microsystems, Inc. He can be reached at [email protected]. Douglas
Tait is an Intelligent Network specialist at Sun Microsystems, Inc., and can be reached at
[email protected]. For more
information on Sun Microsystems and JAIN, please visit Sun's Web site at www.sun.com.
|