November 1998

BY BROUGH TURNER

Increasingly, PC-based open telecommunications technology is penetrating the very core of the public telephone network. For some years, we've seen enhanced services platforms based on PC technology. But today's deployments are more fundamental than voice mail or voice response. Today, PC technology is carrying live voice traffic in the core of the network. How did this happen? How did "unreliable PCs" first sneak into the core of the public telephone network? The answer is as gateways. Gateways showed up, first between pieces of legacy telephone equipment and the PSTN, then between the PSTN and data networks, and now between the PSTN and the emerging world of IP telephony.

I've mentioned gateways in earlier columns - in my IP telephony column (July 1998) and my SS7 column (August 1998). This article expands on the gateway phenomenon, as gateways are a continuing opportunity for computer telephony developers.

EXPANDING OPPORTUNITIES
Many early gateway opportunities were driven by global market considerations. An equipment vendor or systems integrator got an opportunity to sell equipment into a new region, but couldn't adapt existing proprietary equipment to the new market in time to hold the deal together. Here computer telephony technology is a natural. Already proven interface components are available for most countries around the world, and - with the programmability of open telecommunications - completely custom gateway configurations can often be generated in just days or a few weeks. An early gateway I was involved with adapted a U.S. voice mail system to MFC/R2 signaling so the system could be used to provide virtual telephony (private mailboxes accessed through public telephone, that is, payphones) in Mexico and South America.

Of course, lowest system cost comes when the gateway is eliminated and the basic equipment gains direct connection. But time-to-market is usually more significant than any other factor. Interestingly enough, experience in developing CT gateways has exposed many proprietary equipment providers to open telecommunications solutions. Then, when they're ready to develop their next-generation equipment, open telecommunications approaches at least get considered (and get adopted in many cases). Indeed, the U.S. voice mail system mentioned above was subsequently implemented on an open telecommunications platform. Meanwhile, several hundred million dollars of custom gateways are sold each year.

A similar effect occurs with new technology. For example, when ISDN first became prevalent in Europe, there was a crying need to adapt existing equipment (PBXs and so on) to the new signaling capabilities. Here the opportunity for CT gateways was to convert the installed base. The peak of this market lasted only three years, but several companies grew substantially on this business alone.

The most recent example is in the voice-over-IP (VoIP) market. Most early VoIP solutions have been based on open telecommunications technology of one sort or another. Only now, two years into the market, do we see a variety of proprietary boxes being announced. And even in these seemingly proprietary systems, we see strong signs of open telecommunications influence. For example, 3Com has elected to embed PCs within their proprietary system, and Cisco is focusing considerable attention on control protocols like Simplified Gateway Control Protocol (SGCP) and IP Device Control (IPDC) that will let external PCs control the guts of their media gateways. The flexibility of the PC programming environment is critical to everyone in telecommunications.

PUBLIC NETWORK APPLICATIONS
Most of the world has moved to, or is in the process of moving to, SS7 backbones for public network signaling. But there is an enormous amount of pre-SS7 infrastructure. And on the equipment side, there are an enormous variety of platforms designed for high-end applications at the edge of the network. In today's competitive communication services market, these platforms can frequently be applied to public network services, but typically they don't support SS7.

To solve these interface problems, vendors face a choice. They can develop new interfaces, validate them, and get regulatory approvals, or they can build a gateway out of previously approved and working CT components. For proprietary equipment designs, the quickest path to market is frequently to build a gateway out of CT components - typically an adapter between SS7 and something else. Sometimes the needed gateway function is just software to adapt between different SS7 variants because, while SS7 is an ITU standard, it is not exactly uniform worldwide. There are regional differences (ANSI SS7 versus ITU SS7), and then there are country-by-country variants as well.

Recent CT-based SS7 adaptations that I have seen include:

• Video conferencing equipment. (Originally developed for enterprise deployment with ISDN connections.)
• SS7 interfaces for a variety of voice response equipment.
• SS7 interfaces for media gateways. (So that voice-over-IP gateways can connect directly to central office switches.)
• The SS7 adaptation of remote access servers. (So it's possible to divert dial-up modem calls at the local central office - before they tie up the rest of the voice network.)

Of course, the opportunities are not limited to wire line applications. There are gateways for wireless applications as well.

WIRELESS APPLICATIONS
Wireless solutions are on the rise, not only in cellular and PCS applications, but for wireless local loop (WLL) and in-building wireless applications as well. And adapting new wireless equipment to existing networks poses its own unique set of challenges.

Many of the new WLL base stations conform to a widely deployed ETSI standard called V5.2. The V5.2 protocol and its predecessor, V5.1, were originally developed to communicate with subscriber line access concentrators - curb-side boxes that provide local telephony loops to a group of subscribers. (For those readers familiar with Bellcore specifications, ETSI V5.2 performs the same function as Bellcore GR-303.)

The backhaul from the curb-side box to the central office switch is digital and increasingly uses well-defined protocols (V5.2, V5.1 or GR-303) and standard trunks (E1 or T1). This means that if the central office switch comes from one vendor and the curb-side units come from other vendors, they are still able to work together.

Recently, several vendors of WLL equipment have developed systems in compliance with V5.2. Thus these wireless systems appear to the central office switch as if they were just concentrators for conventional subscriber lines - a great idea! Unfortunately, many of the central office switches deployed around the world don't yet support V5.2 and can't easily, or at least economically, be upgraded to support V5.2 any time soon. The only way to connect to diverse legacy central offices is to develop diverse legacy protocols - a perfect gateway opportunity.

To simplify this country-by-country adaptation, some newer wireless solutions are being developed directly on PC-based technology, with the intention to deploy on CompactPCI-based technology (to provide high availability). The open telecommunications advantage is that specific interfaces are either already available off-the-shelf from a variety of vendors in the CT marketplace or interface components are available that can easily be customized using CT-based software tools. For now, the quickest way to connect wireless equipment to legacy systems it to build a gateway and adapt it for each individual sale. Either way CT technology is popping up in core PSTN deployments, and it appears this trend will only increase.

ENTERPRISE APPLICATIONS
In-building wireless applications typically have to integrate with an existing PBX. Wireless equipment vendors that also have PBX product lines can easily integrate with their own PBXs, but seldom any other brand PBXs. Others must start from scratch. Once again, CT comes to the rescue. Recall that a major issue for early CT systems was how to integrate with proprietary PBXs to provide functions like auto-attendant and voice mail. Now this experience, and this technology, is being applied to provide gateways between the in-building wireless system and the wired PBX.

Of course it's not just the in-building wireless system that needs gateways to the existing PBX. Voice-over-IP and fax-over-IP systems for enterprise use involve gateways that must integrate with existing PBX systems. An emerging opportunity will be IP-based PBXs such as those from companies like Selsius Systems and NBX. IP-based PBXs require gateways to existing PBXs and the PSTN. Then one can anticipate a new wave of in-building wireless solutions using IP backhaul over Ethernet. Of course, this step will merely change the nature of the gateway. In any event, the CT market pioneered PBX interconnections and will excel at this business as long as the traditional PBX continues to exist.

CT - THE FLEXIBLE SOLUTION
Gateways built on open telecommunications components are the way to connect new technology to legacy equipment. CT-based gateways provide the fastest, cheapest, and most flexible solution.

Of course, systems designed on open telecommunication concepts typically don't require gateways. In one recent case, a vendor was selling an inverse multiplexing system for video transmission. This system was originally developed as an enterprise premises application using ISDN, but the vendor got opportunities to deploy it in the PSTN. Because the original system was based on open telecommunications components, the PSTN version did not require a gateway. It was simple to replace the ISDN interface card with a readily available SS7 interface card, and make minor changes to the call control software to take advantage of new capabilities in SS7 signaling.

In the 1980s, the computer telephony industry pioneered the ability to add new applications, such as voice mail and auto-attendant systems, to legacy PBXs. In the present, the CT industry is pioneering the addition of new applications to the legacy PSTN, directly and via the use of gateways.

In the future, more equipment will be designed on CT technology, somewhat reducing the need for gateways. But with the ever increasing pace of technology evolution, the market for gateways - between generations of equipment and generations of protocols - should last indefinitely. And, of course, gateways based on open telecommunications technology are uniquely positioned to support new and interesting applications, even on the gateway! Meanwhile, count on the flexibility of open telecommunications to allow you to quickly and easily build the gateway de jour.

Brough Turner is senior vice president of technology at Natural MicroSystems, a leading provider of hardware and software technologies for developers of high-value telecommunications solutions. For more information, call Natural MicroSystems at 508-620-9300 or visit the company's Web site at www.nmss.com. E-mail to the author (rbt@nmss.com) is also welcome.