September 1998

BY MIKE KATZ

No one still seriously characterizes IP Telephony as just "free calls over the net" -- the market focus has now shifted to cost-saving toll bypass applications. However, this is not the limit of IP telephony's value proposition. The market will soon come to see IP Telephony as a means to implement business enhancing applications over managed networks.

As the previous articles in this space illustrated, you can build or purchase the infrastructure needed to create a managed IP telephony intranet. This month, we will focus on three applications that make use of this network: Enterprisewide call control; single phone line usage for voice and data by telecommuters and mobile workers; and remote calling or hop off applications.

ENTERPRISEWIDE CALL CONTROL
Enterprisewide call control is an application that has only an indirect association with IP telephony. It is in use today, but has not yet been generally applied in IP telephony networks. The following scenario may be familiar to you: ABC Corporation has deployed one PBX vendor, Megatel, throughout their geographically dispersed organization. One of the major features of the Megatel deployment is that phone numbers for all employees regardless of physical location are delivered with one flat dial plan. A flat dial plan infers that employees in Atlanta are at extensions 1xxx and can pick up the telephone and dial their Boston office by simply dialing 2xxx or their California office at 3xxx - without the need for long-distance access codes or tie line access numbers such as *8 or #8.

You may ask, "What's so great about a simplified flat dial plan?" Well, here's the reason: It's hard to remember access codes and then dial 10 to 12 numbers. You might as well use the PSTN. The fact is, it's simpler for employees to have a flat dial plan and yet it's hard for telecom managers to implement. Without purchasing all of the PBX equipment from a single vendor and paying for and running the associated proprietary tie-trunk protocols, it would be difficult (if not impossible) to implement this type of simple flat dial plan. However, the complexity and distributed nature of the implementation problem represents an ideal IP telephony application - once you've decided to implement an IP telephony capable network.

Using intelligent IP Telephony gateways to create a virtual network PBX is really what the definition ought to be for a toll bypass application. Still, most gateway solutions are not intelligent enough provide this level of functionality.

The next phase in gateway deployments will require more native intelligence in the gateway, that is, the ability to provide extensive dial plan mapping for calls entering or leaving the gateway. It will also be necessary to eliminate the differences in dial plans and stop the need for proprietary tie trunk protocols. The intelligence for the call handling will be spread across both the PBX and gateway.

The local PBX will have already performed some of the mapping, (e.g., extension 3232 is not directly connected to this PBX and has a phantom extension, much like the ones used to map calls to a voicemail system) that has been mapped to a T1 link. First, the call is passed by the PBX to the T1 (remember its mapped to the 3xxx extension group), which is directly connected to the gateway. Second, the gateway picks up the call, gets the dialed number information, and then performs the balance of the mapping, resolving the appropriate remote IP telephony gateway to send the call to, setting up the call, and perhaps even providing local ring-back indication. It sounds complicated, but this scenario underscores two very strategic and valuable tasks - it removes the need to have all PBXs from the same vendors, and makes it very simple (transparent) for your employees to call across the corporation.

PC CLIENT-BASED IP TELEPHONY
The second application for IP telephony gateways and intranets entails voice enabling the remote worker, whether they are telecommuters or so-called "road warriors." This means giving them the ability to make and receive phone calls from their IP telephony-equipped personal computer. Fortunately, there is one very large company that is interested in creating and delivering a standards-based IP Telephony client application, available for every Win95 or 98 PC. That's, of course, Microsoft with NetMeeting. This IP telephony application makes use of the PC to compress the speech and send it over either a LAN or modem connection. The compressed speech is sent from a user's PC connected to an intranet-based remote access server (RAS). In this application, the person to whom the modem-based caller is speaking is not aware that the call has been turned into packets and routed over a modem.

This type of application is very valuable and has many uses to a corporation. In California, for example, there exists a mandate for a percentage of the workforce to telecommute. In such an environment, the application represents massive cost savings. I have heard stories about employers competing for workers based on the type of phone line they are willing to place in the worker's home. Normally, these workers get two phone lines, one for regular calls and one for a data connection. The second phone line is a fixed cost item that is usually idle two-thirds of the week. If a telecommuter could share a single phone line for voice and data calls, then the second line overhead cost is eliminated and the corporation can deploy its people more cost effectively.

This also applies to the mobile worker or laptop-based road warrior. This is a person who, when travelling (often), normally connects the laptop primarily for e-mail or database access. What happens when this person needs to simultaneously talk with a customer and look up an order status or some other relevant information in a centralized corporate database? Well, the employee is likely to tell the customer that he'll get back to them (not a popular option with customers). Furthermore, doing so from the road could be doubly difficult.

Let's propose a slightly different approach. The road warrior makes a data call via RAS and then simply dials his key customer, knowing that he has access to the order databases and other support people to help answer questions and provide customer satisfaction. This is yet another way to use an intelligent IP telephony gateway to provide useful new business applications and services.

LOW-COST LONG-DISTANCE
The third application, enabling a corporation's PBXs to support remote calling capabilities provides a unique value-add to the IP telephony toll bypass application. This application is often referred to as a "hop-off." Like in the road warrior solution discussed earlier, remote PBXs are used to make local calls on behalf of remote callers. In simple terms, a PBX extension in Atlanta can make a call to a local business in Boston through the IP telephony connected PBX in Boston. The Boston PBX does not know that the call originated in Atlanta, it is merely acting on behalf of a phantom extension on the system. This application not only saves the corporation money on long-distance calls between facilities, but also on routine long-distance calls made to that part of the country. There are many security, authentication, and billing issues that need to be addressed when allowing extensions to access the trunks and features of a remote PBX, but if well managed and implemented within the corporation, the possibilities are enormous.

THE FUTURE
The future of IP telephony is not just about saving money on calls made between sites. It's about new applications that take advantage of the fact that IP telephony provides a cheaper, more efficient, and flexible transport for voice then the traditional PSTN. Cost, flexibility, and efficiency - what can we think of next?

Mike Katz is vice president of marketing and business development at NetPhone, Inc. Headquartered in Marlborough, Massachusetts, NetPhone is a leading provider of computer telephony solutions for small business environments. For more information, contact the author at [email protected].