Basic Voice Call Station/System Features | Feature/Function Enhancements

Do station users really need six variations of call forwarding? Do managers still “buzz” their secretaries? Although PBXs have many call forwarding options and still retain the manual signaling (buzz) feature, the most significant station/system feature enhancements during the past two decades have been to improve incoming call coverage, support the needs of the new mobile workforce, and simplify the administration and maintenance operations of the system manager.

An important PBX feature developed in the days before voice messaging systems invaded the workplace was programmed call coverage. Programmed call coverage was a form of enhanced call forwarding, with some important distinctions. First introduced in 1983 by AT&T on the System 85 PBX, call coverage did not receive the market attention it deserved during the 1980s and 1990s, but renewed interest in personalized call screening and routing to improve communications service levels has revitalized the feature. Call coverage capabilities on current-generation PBX systems allow station users to define where incoming calls are directed when they are unable to answer the call and program the coverage path based on who is calling (CLID, Automatic Number Identification [ANI], internal calling number, call prompt), where the call originated (internal or external to the system), how it arrived into the system (trunk group ID), or when a call is placed (time of day, day of week). Building on the concept of call forwarding, personal call coverage programming redirects calls to a defined path of answering stations and will default to the called party’s voice mailbox only as a last resort. Calls will not be redirected to the forwarding position or voice mailbox of a station user defined in the call coverage path; the originally called party’s coverage path overrides intermediary station user call forwarding commands.

Call coverage tables and station user programming was not possible before the development of digital PBXs. The new CTI-based PBX system designs allows station users to program caller-specific call coverage paths based on identified callers. The personal call coverage function in these new-generation systems is supported at the station user desktop (a PC client softphone), not at the common control call processing system. The objective of personalized call coverage features is to reduce dependency on voice mail systems because a human answering station rather than a noninteractive machine might be preferred by the caller. Voice mailboxes should be the last option in a call coverage environment, not the first or only option.

The new mobile workforce includes station users who are rarely in the office and workers who do not have permanent desk assignments because they are constantly moving or their job function is not desk based. To support these mobile workers, it is necessary to dissociate a station user’s telephone directory number from a physical telephone instrument. Hoteling, a feature designed to support workers who work at different desks throughout the enterprise, allows station users to log into the system from a telephone and reassign their directory number to their chosen telephone. In addition to their telephone number, the individual’s station user profile (service levels, call restriction levels, group assignments) is also assigned to the physical telephone location. Account codes and call records are maintained for the station users for each telephone they use. When done using the telephone, after 1 hour, or 1 week, or 1 month, the station user logs out, freeing the telephone for the next mobile worker. Hoteling is becoming very popular in sales offices. The feature can significantly reduce system costs by optimizing common equipment hardware, telephone instrument, and cabling requirements and, more importantly, minimizing real estate requirements (fewer dedicated desks/telephones, less office space).

Today’s mobile workers who are rarely at a fixed telephone location also benefit from recent feature enhancements. The find-me feature allows station users to program their telephone to direct calls to other telephone numbers outside of the PBX system. More than one external number can be programmed. For example, on a no-answer call at the station user desktop, the call can be forwarded to another telephone number after a selected number of rings; if there is no answer at the external number, another telephone number is dialed, and the call is redirected. External telephone numbers likely to be programmed include cellular telephones, home, conference facility, remote office branch, or even a hotel. A relatively recent teleworker option available on some PBXs allows station users to bridge their line appearance to a telephone external to the system. The concept of the PBX as a mobility server can significantly improve call coverage, reduce lost or abandoned calls, and increase the number of successful call attempts between caller and called parties.

Another category of mobile workers consists of station users who require a telephone away from the formal office environment. Known as teleworkers, these station users require their high-performance telephones to function away from the workplace and receive incoming calls redirected to their remote desktop. The original teleworker option was an off-premises extension (OPX) station using highly tariffed telephone trunk circuits to link remote analog station equipment to the main PBX system. Expensive and low-performance analog OPX stations have evolved into affordable and high-performance digital desktops. The same digital telephone supported behind the PBX at the office can be supported remotely with several options, including distance extender modules and analog trunk carrier facilities, ISDN BRI services and equipment, and the recently available IP workstation (hard telephone or PC client softphone).

The most important system feature enhancements during the past decades have been systems administration and maintenance tools. The early PBX management terminals required high-level programming skills and weeks of training. A typical station move, add, or change operation could require at least 15 minutes of keyboard entries. After booting up the systems management terminal the administrator was met by a blank monitor screen waiting for a programming command. There were no menus, on-screen help command, or point, click, and drag. Computer technology evolved during the 1980s and so did PBX management tools. By 1990 a systems management terminal had a basic graphical user interface (GUI), usually a menu selection list and formatted screens. By 2000 PBX management tools were accessed through a web browser via the Internet, and a sophisticated GUI simplified the administration process. Few keyboard entries are now required, and access to a common metadirectory server simplifies the initial station user directory entry.

Modular System Design

Until the early 1980s all PBX systems were based on a centralized processing, centralized switching, and centralized cabinet equipment design. Intecom, the developer of the first digital PBX telephone, also broke system design tradition when its IBX system featured distributed port cabinets linked to the main processing/switching complex via fiber optic cabling. Each of the IBX’s distributed interface modules (IMs) could be located 10,000 feet from the main equipment room to support campus configuration requirements. Each Intecom IM cabinet had its own local processing unit operating under the control of the centrally located Master Control Unit (MCU).

The distributed cabinet design was dictated by distance limitations imposed by digital signal links to the digital desktop. Intecom was forced to bring the port cabinet closer to the station user. Analog telephones could support cabling loop lengths of 1 or 2 miles, but the Intecom ITE digital telephones were limited to 1,000 feet between wall jack and port circuit card.

The next logical step in a modular system design was to remote port cabinet miles away from the PBX common control complex using telephone company trunk carrier circuits. Northern Telecom was the first to accomplish this when it designed a remote peripheral cabinet for its SL-1 PBX in 1982. Using analog trunk circuits, the remote cabinet depended on the main PBX location for all call processing and switching functions, but at least a customer could support two or more distributed locations with a single PBX system. If the trunk circuit link to the remote location failed, however, the remote location was left without communications service. A spare processing option at the remote location would solve the link failure problem, so Intecom announced such an option about one year after Northern Telecom introduced the first remote cabinet option.

By the mid-1980s several PBXs offered remote cabinet options, but only Intecom has a remote survivable processor option. Other manufacturers offered an alternative solution to the remote cabinet option and in some ways a better PBX system design. A PBX system first announced in the early 1980s, and still working today after many upgrades and enhancements, was the Ericsson MD-110 PBX. Based on its own central office switching system, Ericsson’s MD-110 was a fully distributed communications system from a call processing, switching, and cabinet architecture perspective. Each MD-110 Line Interface Module (LIM) contained a common control complex that operated independently yet in coordination with every other LIM cabinet in the system. The LIMs could be geographically dispersed on a campus location or across a telephone network (analog or digital trunks, copper or fiber optic cabling, microwave or satellite transmission). Each LIM had its own switching system backplane and communicated with other LIMs via a centralized group switch complex. PCM links between the LIMs and group switch could be duplicated, as could the group switch (Figure 1).


Figure 1: MD110 IP evolution.
In the mid-1980s Rolm introduced a PBX design similar to the Ericsson offering. The Rolm CBX II 9000 did not have a centralized group switch but it did have functionally independent control cabinet clusters. The Northern Telecom SL-100, a modified version of the manufacturer’s DMS-100 central office switching system, became a popular PBX system for very large (thousands to tens of thousands of user stations) distributed communications configurations requiring an extremely high level of reliability and redundancy. The SL-100 Remote Switch Center (RSC) option could be located hundreds of miles from the main PBX location, support thousands of stations users, and function as a standalone system, if necessary, with minimal loss of features if the control link to the main common control complex failed. The growing availability of PBXs capable of supporting multiple common control complexes and port cabinets geographically dispersed across great distances marked a distinct change from the old, monolithic design platform of PBXs before 1980.

For customers with single-location requirements and not interested in remote port cabinet options, the most important PBX cabinet innovation of the early 1980s was the introduction of the stackable cabinet design. PBX control and port cabinets were traditionally based on large, multiple carrier steel frames. Customers would be forced to buy and install an expensive large cabinet capable of supporting several equipment shelves, even if they required expansion for a few stations. The incremental cost to add a few stations was very expensive. When the NEC NEAX2400 was introduced in 1983, it was the first PBX based on a stackable cabinet design, with dedicated single-shelf cabinets for call processing functions and stackable port cabinet shelves. Up to four Port Interface Module (PIM) single-shelf cabinets could be stacked on top of each other, sharing a common switching and processing backplane. Each PIM had a dedicated Port Processor Interface and a dedicated Time Slot Interexchange (TSI). The NEAX2400 offered customers a cost-effective solution for modest growth requirements as compared with PBX systems based solely on large expansion port cabinets costing tens of thousands of dollars even if only a few expansion ports were required.

By the early 1990s almost all PBX systems targeted to customers with small and/or intermediate port requirements were based on modular, stackable port cabinet designs. Many PBX manufacturers offered a remote port cabinet option to customers desiring a single communication system for multiple-location configuration requirements. Distributed processing and switching designs were becoming commonplace. The emergence of CTI in the 1990s allowed manufacturers to offload advanced software options, particularly for call center management, onto adjunct servers dedicated for a specific application. Optional software application programs run on proprietary or customer-provided server equipment reduced the call processing load on the main control complex and offered a more flexible migration and upgrade path to enhance older PBX system platforms that still performed the basic communications functions with little problem. The early CTI hardware solutions required proprietary hardware links between PBX and server, but evolving PBX architecture design led to standardized TCP/IP links over Ethernet LANs.

The development of call processor control signaling over LAN infrastructures simplified the installation of third-party hardware and software solutions behind the core PBX system and kickstarted development activity for IP telephony and the emerging client/server IP-PBX system design. Using the LAN infrastructure (Ethernet switches, multiservice routers) for voice transport and switching between LAN-connected PC client softphones and LAN-connected servers for call processing is the ultimate modular system design because the processing and switching functions are totally distributed across the entire network.

Digital Desktop

In the late 1970s and early 1980s most PBX manufacturers developed an electronic telephone for use behind their systems. The electronic telephone’s primary benefit was support of multiple-line appearances. Instead of using KTS equipment behind a PBX to support station user requirements for multiple line appearances, an alternative option was available. The majority of station users at the time used single-line appearance analog telephones, with no feature/function buttons, no speakerphones, and no displays. Only a few lucky station users qualified for multiple-line appearance telephone instruments. Today, of course, the typical PBX telephone instrument looks slightly less complicated than the cockpit of a Boeing 777, with more buttons, bells, and whistles than one knows what to do with.

Like the basic 2500-type analog telephone, voice transmission from the electronic telephone to the port circuit card over the inside wiring was analog, but the built-in intelligence of the telephone instrument provided an array of programmable feature/line buttons and a limited function display. A signaling link between the telephone and the PBX provided the intelligence to identify which line appearance button was being used to place the call or which feature button was depressed for activation. Control signaling between the electronic telephone and the port circuit card was embedded within the instrument’s 4-KHz voice transmission channel. The low-frequency signaling stream constrained feature/function development, but was a first step in the evolution of intelligent digital desktops behind the PBX.

The evolutionary step made by electronic telephones was a break from the traditional DTMF signaling techniques for communicating with the PBX common control equipment, as was done with traditional analog telephones. Each PBX manufacturer used a proprietary signaling scheme and dedicated station line circuit cards to support electronic telephones. An industry standard for electronic telephones was not developed for a variety of reasons, although it may have led to more sophisticated desktop terminals. Maintaining a proprietary signaling link meant that electronic telephones could be sold at a high price, with a significant profit margin, if customers required multiple-line appearances. Third-party telephones could not be manufactured unless the signaling scheme specifications were published (which they weren’t).

When Intecom introduced the first digital PBX telephone, the product marketing materials emphasized its potential for integrated voice/data communications with an optional data module. Two communications channels were available to the desktop station user, one for voice and the other for data. Little mention was made of the dedicated signaling channel used to link the telephone with the PBX common control equipment. The digital signaling channel was the major breakthrough that would be the distinguishing factor between analog transmission telephones (industry standard, 2500-type, electronic) and digital transmission telephones. The out-of-band signaling channel, operating at transmission rates between 16 and 64 Kbps (based on the individual manufacturer’s design specifications), could be used for a variety of new, advanced desktop capabilities.

The primary function of the signaling channel was to alert the PBX common control equipment when the telephone handset was taken off the hook to prepare a voice call. The signaling channel was designed to transmit keypad dialing signals and feature/function activation and implementation signals. Display information, such as calling name and number, was carried over the signaling channel, including call redirection information for forwarded calls. Station users could self-program their telephone instruments with software programs residing in the main PBX control complex but accessible via the signaling channel.

The second communications channel originally developed for desktop data communications applications was rarely used because LANs became the dominant enterprise data communications network. Eventually telephone designers were able to program the PBX to support a second voice channel to the individual station user desktop in support of an adjunct voice terminal. The intelligent signaling channel can distinguish between voice calls placed to different directory numbers and support simultaneous calls to and from discrete desktop devices. Using a special analog line adapter module, a digital telephone can be used to support an adjunct analog telephone, modem, facsimile terminal, or audioconferencing station. The adapter module converts signals from the adjunct analog communications device signals into the proprietary PBX digital format. Other uses of the second communications channel include support of a second digital telephone (using a digital line adapter module) off a single PBX communications port interface and bonding of the two channels for high-speed transmission in support of data or video applications using an ISDN Basic Rate Interface (BRI) type of adapter module.

The most impressive use of the signaling channel is the support of sophisticated display information fields and associated context-sensitive softkey feature/function access. The current generation of digital telephones have large multiple-line display fields that are used to view directories and call logs, access on-line help programs, read text messages, and perform station and/or system management operations.

One of the criticisms of traditional PBXs has been the use of a proprietary control signaling to support digital desktop equipment. The recent development of LAN telephony systems using IP signaling standards someday will eliminate the proprietary signaling link between the call processing system and the telephone, but standards are still in development. Cisco’s IP telephone currently does not work on a 3Com IP telephony system, and Avaya’s and Nortel’s IP telephones interwork do not work on each manufacturer’s respective IP-PBX offering. When a high-performance industry standard IP telephone is available to work behind a multitude of IP-PBX systems, it will be possible only because of a standardized signaling link between the call processing server and the desktop, a design specification first developed 20 years ago in the first-generation digital PBXs.

Messaging | Features/Function Enhancements

VMSs were introduced to the market in the early 1980s, and originally worked as stand-alone systems behind PBX systems. Although the first VMSs were designed and marketed by third-party suppliers, several of the leading PBX manufacturers eventually entered the market with products of their own design. Rolm and AT&T were among the first PBX manufacturers to enter the voice messaging market with products designed to work behind their own communications systems, although they could also be engineered as stand-alone systems to work behind other suppliers’ PBX systems.

Northern Telecom, one of the leading PBX suppliers, came late to the VMS market during the late 1980s, but when it introduced Meridian Mail it became the first messaging system to be fully integrated within the PBX system design. Meridian Mail used the Meridian 1 processing and switching network backplane for supporting PBX station user messaging applications. The Meridian Mail Module was installed as another cabinet stack in the Meridian 1 and tightly integrated within the overall PBX system design. Instead of using analog station interfaces and a dedicated data signaling link between the PBX system and adjunct voice messaging cabinet, Meridian Mail ports appeared to the Meridian 1 switching network as just another station port, and signaling between the Meridian Mail Module and the Meridian 1 common control complex was transmitted over the internal system processor bus. AT&T followed Northern Telecom’s example and later redesigned its Audix VMS as a multiple card slot equipment module to be installed within its Definity PBX system. The Definity Audix option offered most of the features and functions available on the larger, stand-alone Audix (later Intuity Audix) system at a reduced price.

During the early 1990s VMSs were redesigned to support integrated messaging applications with e-mail servers. The concept of a UMS designed to support voice and e-mail messaging, with both message mediums sharing a common directory and storage system, was also introduced in the early 1990s. Although demand for the enhanced messaging system designs has been limited to date, there are many productivity and cost benefits attributable to using one mailbox for all types of messages and having a single interface to the mailbox from either a telephone or PC client.

Recognizing the competitive advantage of bundling messaging applications within the PBX system, several recent start-up companies with PBX client/server designs, such as Altigen and NBX (since acquired by 3Com), integrated a UMS application running off the main system server that also provided basic PBX communications features and functions. Recently, Avaya integrated the capabilities of a full-function Intuity Audix system into the main call processing board of its small Definity One PBX system and included the Intuity Integrated Messaging appli- cation on the same board. The Altigen, 3Com, and Avaya PBX systems with the bundled messaging capabilities are designed for small/intermediate customer line size requirements, and the message storage capacities and access ports are limited. PBX systems designed for large and very large customer port requirements would not be able to integrate the messaging application into the main common control complex without affecting the basic communications responsibilities of the system. Dedicated messaging application servers will likely be the optimal solution for higher-end PBX customers, even when IP-PBX client/server system designs become standard by the end of this decade.


Figure 1: Avaya speech access with unified messenger architecture.

Computer Stored Program Control

Until PBX systems incorporated computer technology into its call processing system design, features and functions were extremely limited. Station user features were restricted to those operations that could be handled by mechanical means. The general availability of computer SPC meant that features could be based on software programming tools, and feature development was limited only by a programmer’s imagination. Many PBX functions that are currently viewed as basic telephony capabilities, such as call forwarding and station activated conferencing, were first implemented through computer SPC. Network routing tables and CDR would not be available without computer programming capabilities.

The first SPC PBX system was introduced by Northern Telecom in the early 1970s. Known by a variety of names, including the Pulse and the SG-1, the Northern Telecom system was the first PBX to use a computer software program to perform basic call processing functions, such as provisioning of dial tone, and implement simple station user features, such as hold and transfer. In the United States, Northern Telecom distributed the system through the Pacific Telephone and Telegraph local telephony company, but sales of the new PBX design were limited. It was not until AT&T introduced the Dimension PBX system in 1974 that an SPC communications system was distributed on a large scale through each of the Bell System’s local operating companies. Dimension became one of the best-selling PBXs of all time, although AT&T’s market share declined throughout the life cycle of the product. After the Dimension PBX announcement, there was a flood of SPC communications systems from AT&T’s competitors. Between 1974 and 1980 SPC PBXs went from a 1 to a 95 percent market saturation level for new system installations.

The first computer-based PBXs were based on a centralized processing design. A single computer-based call processing element was used for all system call processing and switching operations. PBX manufacturers of the early digital SPC systems designed and manufactured their own processing hardware and were the designers and developers of the operating system used as a platform for software feature applications. The first-generation digital PBXs were based on call processing designs that closely resembled the minicomputers of the 1970s. Many computer manufacturers became interested in the PBX industry as a new potential market for their products, and a few actually attempted to design a telephony system. Rolm was a manufacturer of military specification computers who successfully entered the PBX market, but most failed. IBM designed, manufactured, and marketed PBXs for the European market but was unable to compete in North America. Digital Equipment Corporation (DEC) was rumored to be developing a PBX based on its VAX minicomputer design, but no product was ever officially announced.

Computer technology in the 1970s was relatively expensive as compared with current prices, and the high cost to design and manufacture a digital PBX was reflected in the enduser price at the time. Common control equipment hardware was priced several times the current cost to customers, even though the features in the 1970s were minimal compared with those of today, and the call processing power of the system was a fraction of today’s capacity limits. PBX call processing design evolved significantly during the 1980s when third-party microprocessors were generally available, and prices began their exponential decline. Dispersed and/or distributed call processing designs became the standard architecture platform for PBX systems. The single, centralized, common control element gave way to dedicated processing elements for diagnostics and maintenance operations, localized call processing and switching functions, and systems administration. Basic function electronic telephones with internal processor chips evolved into intelligent digital telephones. Adjunct applications processors provided enhanced functionality behind the core PBX system.

During the 1980s PBXs could be classified into one of three call processing system designs: centralized, distributed, and dispersed. System processor elements expanded from the common control complex to expansion port cabinets and even to individual port circuit cards. The focus of PBX system design was shifting from hardware to software. From the 1970s through the mid-1980s more research and development dollars were spent on hardware upgrades and enhancements, with a focus on digital switching and SPC functions. By the late 1980s most research dollars were being spent on software programming. The emergence in the 1990s of third-party CTI software applications programs running on adjunct servers linked to the PBX officially signaled the beginning of the end of proprietary common control and call processing designs. At the beginning of the twenty-first century, almost 90 percent of PBX research and development dollars were devoted to software applications programming. Little money is spent on core call processing hardware because third-party microprocessors, digital signal processors, and servers, instead of the original self-designed and manufactured computer system, are used.

Today’s PBX call processing design is as likely to be based on a customer-provided Windows NT server from Compaq, IBM, or Dell rather than a proprietary common control cabinet from the PBX supplier. Customers may experience lower system reliability levels using third-party servers not designed by their manufacturer for the heavy-duty real-time call processing demands of telephony communications, but the lower price alleviates the risk factor to some extent.

Digital Switching/Transmission

PBXs based on digital switching and transmission technology debuted in the mid-1970s. Between 1974 and 1976 several communications system manufacturers claimed to be the first to announce a digital PBX system, including Northern Telecom (currently known as Nortel Networks), Rolm (acquired by IBM and then sold to Siemens), and Digital Telephone Systems (later acquired by Harris Corporation and known as Harris Digital until withdrawing from the market in 2000). The stated driving factor for developing a digital PBX system was to support desktop data communications without a modem, although data communications options would not be widely available until the early 1980s. Other benefits of digital switching/transmission included improved system quality and reliability levels and lower potential manufacturing costs.

There were no established standards for designing a digital PBX system in the 1970s, and the resulting systems reflected each manufacturer’s individual design biases. The preferred method of digital transmission used by almost all PBX designers was TDM. TDM is simply described as the sharing of a common transmission bus by many peripheral endpoints. Transmission of digital signals by each endpoint is based on assigned time slots by the PBX common control system. Although TDM was used for transmission of digital signals across the internal PBX switching network, it was possible to use different encoding schemes to convert the original analog signals into a digital format. Although most of the early digital PBXs used an 8-bit word PCM formatting scheme, including Northern Telecom’s SL-1 PBX, the first-generation Rolm CBX used a 16-bit word. The typical sampling rate used to convert analog signals to digital format was 8 KHz (a sampling rate double the maximum frequency of a human voice communications signal), but the Rolm CBX used a 12-KHz sampling scheme.

Encoding schemes other than PCM could also be used. In the early 1980s the first-generation Lexar LBX system used a Delta Modulation (DM) sampling/encoding scheme. Some manufacturers evaluated using Adaptive Differential Pulse Code Modulation (ADPCM), based on a 4-bit word encoding format, but no product was ever announced. Although no written industry standard existed, by the early 1980s it became obvious that the 8 KHz sampling using 8-bit word encoding was the preferred digital PBX switching platform. It took Rolm 8 years after its original CBX system made its debute to change its digital switching platform to conform to the 8-KHz, 8-bit word format; Lexar also converted to 8-bit PCM in the late 1980s. By 1990 100 percent of all new PBXs sold in North America were based on digital switching platforms using 8-KHz, 8-bit word TDM/PCM.

The first digital PBX systems digitized the analog voice signal at the port circuit card. Analog voice transmission signals were digitized for transmission across the internal switching network, mostly through the use of a TDM transmission scheme. After being transmitted across the internal switch network, the digitized transmission signal was reconverted back to analog format at the destination port circuit card. Analog station port cards were used to transmit communications to desktop devices, such as telephones or modems, and analog trunk circuit port cards were used to connect to telephone company trunk carrier circuits.

When Intecom introduced the first digital telephone in 1980 for its IBX communications system, the digitization process was performed with a codec in the telephone. Voice signals were digitized and transmitted over the local loop wiring from the telephone to the port circuit card. The first digital telephones used a multiple-channel communications link between the codec and the port circuit card. One channel was used for digitized voice signals and another channel was used for control and signaling functions. A third channel was also available for data communications devices attached to the digital telephone via a data module. Stand-alone data modules for data-only desktops were also available (Figure 1).


Figure 1: Digital PBX data communications.
Desktop-to-desktop digital communications was a major breakthrough for PBX systems. In addition to using the telephony communications network for voice communications, customers could use the PBX system as a local area data communications network. Very expensive modems would no longer be required to convert digital data communications to analog format, and transmission rates up to 64 Kbps could be achieved. Accessing a centralized computer mainframe system would be simplified—no more modems or coaxial cable cluster controllers. LAN technology in 1980 was in its infancy and very expensive. The early Ethernet Network Interface Cards (NICs) were more than double the cost of a digital PBX datastation. A PBX system could support an entire network of data workstations across the entire enterprise when an Ethernet LAN was limited to 50 workstations with major distance limitations. Great things were predicted for the integrated voice/data PBX system because transmission and switching could be all digital. We now know that LAN technology improved; NIC prices rapidly declined; bridges, hubs, Fast Ethernet, and routers were developed; and PBXs as data networking systems never caught on. The irony of the situation is that the digital PBX transmission and switching infrastructure is evolving toward an Ethernet LAN/IP WAN design.