April 1999

Is CTI Ready For Wireless Infrastructure?

BY TONY ROUG

The information revolution rolls on and the data and telecom networks converge: it is clear that the next-generation public network will be built on open, standards-based solutions that enable a new class of integrated services. The wireless industry is no different. Wireless data -- subject to the same basic forces as the enterprise and the public network -- is enabled by an open architecture. And building today's wireless solutions using existing computer telephony standards will help wireless take its place in tomorrow's communication network.

THE CONVERGING NETWORKS
Today's enterprise IT organizations are focused on the converging data and telecom networks, and the effects of this convergence will be profound. It will mean breaking the final wall of the PBX and moving from a comparatively simply computer-telephone integration model -- computers controlling the PBX through an open interface -- to building the PBX on open computers for seamless integration between these two previously disparate worlds. Of course, convergence of data and telecom is also a big topic in the public network, and the public network backbone has already been changing to support combined voice and data. These changes also have implications for the current wireless switching architecture.

The availability of a new portion of the broadcast spectrum for wireless and the dropping cost of wireless service have led to an immense build-out of wireless networks. In this new landscape, there are three ways for wireless service providers to differentiate themselves: branding, cost, and services. Two of these factors -- cost and services -- are driving the wireless industry to look at the same open enabler the enterprise organization enjoys when deploying enhanced service solutions like prepaid, messaging, and single number.

As wireless equipment providers redesign their architecture to begin supporting third-generation wireless standards, there are many goals, including convergence of the air standards -- Time Division Multiple Access (TDMA), Global System for Mobile Communications (GSM), and Code Division Multiple Access (CDMA) -- and the enabling of high-speed data over wireless. These issues will take some time to resolve. However, building a next-generation wireless system on open standards gives wireless infrastructure providers the flexibility to adapt as new standards evolve.

OPEN WIRELESS SYSTEMS
The wireless telecommunications industry has had a mixed history with open systems. An open system has traditionally meant an open air interface. Wireless operators demanded this level of openness, since it allowed providers to choose from a variety of mobile phone vendors instead of being tied to their wireless network infrastructure equipment vendor.

For wireless local loop (WLL) systems, however, there has not been as much pressure from telecom operators. As a result, there are numerous WLL products with proprietary air interfaces. The attempts of WLL equipment vendors to reach standardization have largely failed because the resulting standard would require a logical union of many proposals.

Although they share a common architecture, wireless systems use a variety of radio technologies -- such as GSM, CDMA, and IS-136 TDMA -- which are being vigorously debated in global standards bodies. For instance, the International Telecommunication Union (ITU - www.itu.org), in its IMT-2000 initiative, attempts to define a worldwide wireless mobile standard that would come online in 2001.

WIRELESS ARCHITECTURE: AN OVERVIEW
The subscriber unit in a wireless system digitizes the subscriber's analog voice using one of a variety of voice encoding algorithms. The subscriber unit for cellular, GSM, and Personal Communications Service (PCS) networks is a mobile phone. For WLL, the subscriber unit is a directional flat-box antenna bolted onto the side of the customer's dwelling and connected to an indoor telephone jack.

The base transceiver station (BTS) -commonly referred to as base stations - contains radio transmission and reception equipment and signal processing capabilities. It connects to an antenna typically mounted high on a tower. The BTS broadcasts radio signals to all the subscriber units within a geographic cell. BTSs are typically built on custom hardware, which is evolving toward smaller and cheaper BTSs known as micro and pico cells.

The base station controller (BSC) controls the communication for a number of BTSs to optimize network deployment. The interface between the BTS and BSC is T-1 or E-1, with proprietary communication protocols. In some cases, such as North American CDMA or TDMA mobile networks, the BSC functionality is incorporated within the mobile switching center (MSC). Also, for some wireless systems, such as fixed WLL, the MSC is not required and the BSC connects directly to the PSTN.

The MSC communicates with the BSC on the mobile user side and with the PSTN and other MSCs on the network side, coordinating calls to and from the subscriber unit. An MSC controls more than one BSC and typically covers a large geographical area. The MSC needs to interface with external networks to transport signaling between network elements. Thus, an MSC requires Signaling System 7 (SS7) capabilities to communicate with other switches and with intelligent network elements. Initially based on PBXs, the MSCs were quickly migrated to the equipment vendors' large circuit-switch platforms, which were originally designed about 10 or 15 years ago.

The wireless intelligent network (WIN) provides the capabilities to support wireless subscribers when they roam between base stations, and more importantly, when they roam out of their home area. WIN functions were initially an integrated part of the MSC, but service providers are forcing WIN functions to move to open computing platforms using standard protocols like the ISO IS.41 and GSM mobile application part.

THE DRIVE FOR OPEN SOLUTIONS
Today's MSCs and BSCs are proprietary architectures built on custom hardware and closed software. This model slows the implementation of new features and impedes convergence with PSTN features, resulting in scale that is out of proportion with need.

Slow Implementation Of New Features: In the PSTN, the intelligent network drives implementation of new features out of the switch. However, the WIN architecture is defined, but it has been slow to gain momentum. Adding features to the MSC and BSC is overly complex. As a result, new enhanced services in the network, like wireless voice dialing and prepaid phone, are being implemented through platforms that do not rely on WIN software on the MSC or BSC.

Difficult Convergence With PSTN: Since the proprietary wireless infrastructure is deployed and managed separately from wireline networks, new features are hard to implement. For example, something as simple in concept as having a single number for both your wireless and PSTN phone is anything but simple to implement.

Out-Of-Proportion Scale: Today's MSCs, built on the assumption that the wireless density would not significantly penetrate remote locations, were originally designed as monolithic solutions. The BSC, in fact, is a recent addition designed to cost-effectively service a remote area without running connections from the MSC to every BTS.

CTI TO THE RESCUE?
Can the same technology that simultaneously drove down cost and increased functionality cross over to solve the cost and functionality issues in the wireless public network? Can computer telephony solutions implement the functions supported by the public network MSC/BSC? Such solutions are already being deployed as a top to bottom enterprise PBXs supporting both classic PBX phones and site-wireless capabilities. So, what additional functions are needed to create today's MSC/BSC?

1. Connecting to the BTS: Signaling and voice data are not standardized for legacy technology. Signaling is typically based on SS7 or ISDN. Voice data is typically CDMA, GSM, or TDMA packets over the voice channel. CTI has been providing standard SS7 and ISDN component solutions for a number of years.

2. Connecting to the PSTN: Voice connects through T-1, E-1, and DS3 facilities. Signaling is typically through standard SS7 integrated services user part (ISUP) for mobile, or standard TR303, or V5.2 for WLL. All these capabilities are now offered as standard CTI components.

3. Switching voice between the BTS and the PSTN: This requires transcoding from the compressed format of the BTS -- CDMA, TDMA, or GSM -- to the standard format of the public network. This transcoding function is now a standard part of components delivered to support IP telephony. The same DSP coder algorithms apply to the wireless networks.

4. Translating signaling between the BTS and the PSTN: The MSC and BSC implement the call state of the wireless subscriber unit. The MSC consults the WIN for account information and interworks with the PSTN signaling. This is capability is similar to support in the PBX call model. The standard ECTF S.100 CT Server solution supports extensible call control capabilities. Call models can leverage all the different signaling functions provided by CT components from rob-bit , ISDN, SS7, and H.323.

5. Connecting to the WIN: Standard WIN capabilities are migrating out of the MSC/BSC. This interface is now based on SS7. Leveraging standard SS7 components, the call model can now use the standard WIN messages to talk with the Home Location Register or Virtual Location Register.

Robust deployable solutions are also possible through new technologies like CompactPCI and hot plug, so that these days it is hard to distinguish the difference between a proprietary central office MSC and a standard CompactPCI CTI rack. With today's technology, you can even walk up and pull out a card without powering down the entire system. In addition, solutions are now delivering a comprehensive management architecture based on a standard SNMP management information base. New solutions can either integrate with legacy wireless management solutions or plug into standard SNMP or even Web-based management.

NEXT-GENERATION WIRELESS SYSTEMS TODAY?
Open CTI should accelerate movement to next-generation wireless solutions. This certainly happened with the leading wireless services like prepaid phone, which are implemented using enhanced service platforms based on open CTI hardware. More significant is the rapidly-evolving CT Server architecture based on the widely-accepted Enterprise Computer Telephony Forum (ECTF - www.ectf.org) S.100 architecture. The major goal of the architecture is to allow implementation of applications independent of the deployment. By leveraging a standard ECTF S.100 architecture, a next-generation MSC or BSC could transparently host the rich applications in the enterprise as solutions in the public network.

Functions the MSC and BSC need to support for the future include:

High-speed data direct to the subscriber: Wireless needs to support browsing the Internet from anywhere using a laptop or mobile phone, and at data speeds equivalent to basic rate ISDN or better. This means the MSC/BSC must move to support high-bandwidth data switching. For digital wireless, the voice path is already a data stream. Unfortunately, the MSC and BSCs treat data as a separate service, routing data traffic away from the MSC. CTI naturally integrates data and telephony into a unified solution.

Direct connection to the public backbone data network: As the public network moves to a pure data architecture, voice packets from the BTS can be transported directly rather than converted into the 64K synchronous T-1/E-1 format of the public network today. In fact, IP telephony clients like Microsoft NetMeeting already support the native GSM format.

Integrated services: Subscribers want transparent services across the wireline, wireless, and enterprise networks. They also want their services to move with them as the roam from Chicago to New York or even London. The ECTF open CT Server architecture engages the same expertise that is solving these problems for the worldwide enterprise organization for use in the public network.

To be sure, there are still many issues to be solved to achieve the next-generation wireless network - especially seamless services to roaming subscribers. By beginning with an open ECTF S.100 solution, the wireless infrastructure manufacturer can separate the applications development from the network deployment. Also, the service provider is able to leverage components from both the public network and the enterprise industry.

BUILDING NEXT-GENERATION COMMUNICATIONS
Open architectures are essential in the fast-changing communications network. And to take their place in tomorrow's communications network, wireless solutions should to look to the value of the standard ECTF S.100 enterprise CT Server provides to basic public network infrastructure like wireless MSCs and BSCs. Building the next-generation public network on open solutions will enable a new class of integrated services that will drive the information revolution forward.

Tony Roug is the director of Advanced Network Services for Dialogic Corporation. Dialogic is a leader in the manufacture of open, high-performance, standards-based telecommunications and computer telephony components. For more information, please contact them at 973-993-3000, or visit their Web site at www.dialogic.com.