IP has taken the industry by storm, uprooting conventional wisdom about how to build network infrastructures; how to create, offer and manage services; and about what constitutes a voice service, the core application on which the entire industry has been built. And this is just the beginning: The questions the industry is asking itself now, and the steps it is taking to understand and work with IP, will have far-reaching effects many years into the future. Furthermore, IP has the potential to create new opportunities, to allow new heroes to emerge who will chip away at the strongholds of the traditional industry giants. IP is touching and changing everything we have come to know and accept about the industry.

The industry we are talking about is, of course, wireless.

The long-discussed and finally commercially ready IP Multimedia Subsystem (IMS) standard is actually a wireless industry development that could not only usher in a new era of IP-based services for wireless companies and their customers, but also help guide the telecom world at large toward a much fuller realization of the wireline and wireless convergence that has just incrementally begun to show itself.

But so much for big-bang theories. Sometimes, the best way to gauge the effect that a new technology is having on the industry is to observe the extent to which people are willing to go to sound like they understand it when they obviously don't. At the 3GSM World Congress event in Cannes earlier this year, a couple of people wandered into the Tekelec booth saying they wanted to “buy IMS,” according to Mark Whittier, vice president of corporate marketing at Tekelec.

“You can't buy IMS. It's an architectural concept,” Whittier said. “People have adopted the acronym and begun to use it in a broader context. There is a huge range of meaning to which people use the term. It was born of the wireless industry and fixed-mobile convergence, but now it's almost a name for convergence of anything. It's become the name for whatever's next in convergence.”

The forefront of convergence is not exactly a bad place to be, but in order to actually evolve a network to adhere to IMS constructs, things need to be a little more specific. The IMS standard specifications have been developed by the 3G Partnership Project in two separate phases in 3GPP Release 5 and 3GPP Release 6, the latter of which is finished but has yet to be formally released. Work on these specifications proceeded with a basic goal of creating an architecture for wireless networks that would allow peer-to-peer multimedia communication.

What the IMS standard actually calls for is an IP-based core that integrates with the existing TDM network and creates three functional layers — a service control layer, and application layer and a connection layer.

The primary function of the service control layer is to create multimedia call sessions and manage them, and this requires the implementation of two IP-based elements, a session initiation protocol (SIP) server and a home subscriber server.

The SIP server communicates with the user device and uses SIP to initiate a message to an application server in which a particular application is based. “In a traditional telephony network, a lot of call servers communicate with one another,” said Scott Wharton, vice president of marketing for BroadSoft. “But in an IMS architecture, they need to speak to the call session control function server (CSCF) [another name for the SIP server] instead. With the implementation of a CSCF, there is a need to strip the other server components out of the architecture.”

Meanwhile, the home subscriber server (HSS) operates much in the same way that a home location register (HLR) does in a more traditional cellular network infrastructure. It acts as a data store of customer profiles, including IP addresses and roaming information.

In the application layer, you will find the primary application server as well as IP media servers, also called the media resource function server and processor that allow for the delivery of packet-based multimedia services such as push-to-talk.

In a general IMS architecture, there are also media gateways to support VoIP service, but for wireless network operators, that is not yet part of the recipe. “In a fixed network, you would need some kind of gateway element, but you don't need that on the mobile side because there is no VoIP right now,” said Oscar Gestblom, strategic marketing manager for IMS at Ericsson.

The connection layer in an IMS architecture is the physical network layer of core switches and access technologies, which in wireless terms could be UMTS, GPRS, GSM, any of the various generations of CDMA, or even Wi-Fi, WiMAX or Bluetooth. Implementing an IP core as detailed in the IMS standard allows the access network and service delivery layer to operate independently of one another, with numerous benefits.

“Service providers already have an opportunity to bundle services today, with each service traversing its own infrastructure,” said Gestblom. “But with IMS, you can have a common system. You can de-couple service from infrastructure, making it easier to introduce new services and providing the flexibility to try new services more quickly.”

The service creation, delivery and management process in wireless networks has long been in need of a dramatic house-cleaning. Cindy Christy, president of Lucent Technologies' Network Solutions Group, said the way applications are developed and delivered along infrastructural lines today isn't sensible.

“Everything is done in a very monolithic fashion,” she said. “That's the way things have been, and it doesn't make sense. You want to have the ability to blend applications across different types of networks, to provide the benefit to the end user of accessing those applications from anywhere.”

Ericsson's Gestblom agrees, saying the current model for developing applications puts them into “stovepipes.” While many very useful wireless applications, including push-to-talk (P2T), have been successfully developed and delivered to market under that stovepipe model, using a common enabling protocol like SIP will allow small application developers, major vendors, carriers or anyone else who is developing an application for the wireless network to re-use software tools and work in a commonly-accepted language, rather than having to make a strategic decision about whose software language, toolkits and operating systems they will use.

“One of the golden dreams of IMS is that any guy in a garage who knows SIP could develop any application that could easily be introduced on a major wireless network,” said Marc Denton, director of wireless policy management products at Bridgewater Systems.

Well, almost. Denton added that carriers won't be changing their attitudes any time soon about which applications and application developers they most urgently want to work with. “The reality is that carriers will still be looking for the applications that will have the most impact on the mass market, and they will only want to work with the companies that can provide the applications that will be successful on the scale.”

Eric Updyke, vice president of marketing and strategic planning at Nokia Networks, said that carriers potentially could broaden their reliance on third-party application developers and could tweak applications in different ways to maintain some competitive differentiation. “With IMS, the basic building blocks would be there, and the way they put them together is how they like to be creative,” he said.

Updyke added that the delivery and management of those applications also will benefit from an IMS house-cleaning. “IMS should be able to clean up a back office in which application servers now operate autonomously of each other,” he said. “Carriers will be able to centrally control and manage registration and authentication of users as they are using those applications and crossing into different network domains. That whole process will become much more efficient.”

Much of this revolution in the applications hierarchy is made possible by SIP, which hasn't been much of a force thus far in the wireless industry because VoIP hasn't yet become a top-of-mind technical or market issue. SIP comes from the Internet and IT world, and for many years has struggled to gain a foothold of understanding in the telecom industry, but that situation is changing. Now, in the wireline industry, service providers — even the old-line telcos such as SBC, BellSouth and Verizon — are increasingly becoming comfortable with SIP and using it to a greater extent because VoIP has begun to have such a large impact in that part of the telecom industry.

“Service providers are starting to get more comfortable with SIP, especially on the wireline side,” said Lucent's Christy. “It has been around for a while now, and the industry is starting to learn how to use it in different ways.

In the wireless industry, there may be a bit more SIP education that's needed, especially among the wireless carriers that are not affiliated with the telcos that have been working further ahead with VoIP and SIP. The carriers that do have those affiliations, such as Cingular Wireless, which is owned by SBC and BellSouth, might be better off. Kelly Williams, executive director of core strategies at Cingular, said the building blocks of IMS are reminiscent of another technology concept currently more pervasive in the wireless industry — the intelligent network. “IMS reminds me of the various flavors of the IN. It's an enabler that allows us to get products and services out to market more quickly.”

In that regard, despite the overwhelming interest in looking for applications that have the greatest mass-market potential, IMS will allow carriers to take a few more chances with what services they are willing to release and when. “IMS is something that allows carriers to fail frequently and inexpensively,” said Jeff Key, vice president of strategy and business development in Tekelec's signaling group. “They'll ultimately be able to offer the largest smorgasbord of applications and be able to decide what works and what doesn't. Right now, that process of getting an application ready for launch can take a long time and is often not a pleasing experience.”

In addition to helping carriers move toward a new way of working with new applications, IMS also will help them do more with existing applications, and help end users get more out of them. Push-to-talk and instant messaging are related applications that are often put forward as examples of applications that can realize new potential when enhanced within the frame of an IMS architecture. Tekelec's Key said, “Existing applications can become more extensible and married with other applications through an IMS architecture.”

IMS is at the heart of the push-to-talk over cellular (POC) specification being developed in the GSM world to allow interoperability and application flexibility between different push-to-talk networks.

“Push-to-talk is an application that was developed without IMS and it works just fine, but POC will allow you to do more with it,” Nokia's Updyke said.

IMS, according to Lucent's Christy, will enable P2T applications with presence capability and more active real-time database capabilities. “It improves on the level of synchronization that you have today,” she said. “With IMS in place, you can create an active phonebook and buddy list of contacts with real-time information about the locations and time availabilities of the people on your contact list. You will have a phone book that is not just a passive feature as it is today.” In this scenario, the HSS would house the presence and availability information as part of a fuller customer profile.

Instant messaging is another application that is available today in both online and mobile formats, but IMS will put it into a new context, allowing an IM, SMS or MMS to be created in a mobile environment and sent to a destination on a different kind of network domain, such as Wi-Fi.

“The beauty of IMS,” Lucent's Christy said, “is that every carrier can create a new and different value proposition for its applications despite having very similar networks.”

The potential to revitalize network cores and access realms, while creating a new foundation for applications, seems like a pretty powerful selling point for wireless carriers to begin phasing in IMS architectural concepts. However, the business interests of wireless ultimately hold sway over the innovative interests, and even though IMS can introduce cost-efficiencies and eventually improve the ability of specific applications to create more revenue, carriers also have to weigh IMS as an additional network investment.

For this reason and others, the pace of wireless carrier adoption of IMS is still unclear. Some people think the newness of IP in the wireless industry could cause a period of only gradual acceptance, while other people believe the esoteric aspects of IMS architectural concepts conflict with other competitive issues that are on the minds of carriers.

Daniel Collins, vice president responsible for fixed-mobile convergence at Alcatel, said that carriers, especially those in the U.S., are still busy building out 3G coverage. “To take advantage of IMS, you really need a broadband network, and the wireless industry doesn't have enough of that coverage yet,” he said. “When you get enough broadband access infrastructure in place, you can then focus on the quality of service.”

However, Cingular's Williams said his company is moving ahead with IMS, even as it builds out its next 3G network technology, HSDPA. “I don't think any carrier would say right now that they are not considering deploying IMS technology.” (Though, Cingular and Sprint are the only two major wireless carriers in the U.S. that have publicly announced they are working on IMS.)

He added that Cingular has kept its IMS developments and purchase processes separate from its HSDPA project. Cingular is conducting IMS-related trials in both residential and enterprise deployment scenarios, though Williams declined to be more specific about the nature of the trials. He also said Cingular already knows it is planning to launch a “specific service” facilitated by IMS technology, but isn't ready to further detail its plans.

“There is not one specific application that is a driver for the overall architecture for us,” Williams said. “We actually have talked about that at Cingular as we've tried to figure out what part of the business we should assign the cost to — I mean, it's like ‘Who's going to pay for this?’ In that regard, it is very much like SS7 was in the traditional telephony industry. It's an enabling technology that bridges different systems.”

Williams said the flexibility of IMS “lends itself to a broad, rapid implementation.”

Aside from Cingular and Sprint, and international carriers such as Manx Telecom that have shown some early interest and aggressiveness to deploy IMS architectures, the wireless industry in general will likely lag behind wireline operators — such as telcos and cable TV companies — in IMS deployment.

Alcatel's Collins said that's the case because the wireline operators are under increasingly more competitive pressure to have a wireless offering. “Fixed/mobile substitution is one reason why wireline carriers are more interested in convergent solutions,” he said. “They also want to offer a complete and portable bundle wherever the customer is.”

BroadSoft's Wharton said wireless carriers can afford to take their time adopting IMS architectures exactly for that reason: “Time is on their side,” he said. “Everything is moving their way. Meanwhile, the wireline guys that don't adapt to support mobile service could go the way of the horse and buggy.”

Cingular's attitude about IMS differs in other respects from that of much of the wireless industry, because the company is owned by two major telcos that have an urgency to make wireless part of a flexible service portfolio. “We want to craft something that's complimentary to what they want to offer,” he said.

Many people in the industry are also concerned about the availability of IMS handsets slowing down other aspects of the market. Williams acknowledged it will be some time before there is broad availability of handsets equipped with SIP protocol stacks.

“But what is a SIP phone?” Williams said. “There are phones today to which you can download an SIP stack. I can do that today on my SX-66 [the Siemens phone with the PocketPC interface and keyboard],” adding, “though I guess the point is that not everyone has an SX-66.”

Lucent's Christy believes that in order to spur the market, vendors only need to go back to asking the most basic of questions. “When you're on the cusp of a new technology generation, you have to ask what it is about this technology that would enable carriers to drive additional revenue,” she said. “How do you expand the telecom wallet?”