BABBA Magazine, Issue 7 - September 1993, pages 15-30

ISDN: What is it - Really?

(By Donovan Dillon)

Some articles about ISDN leave the reader with the feeling of "I Still Don't kNow". Here is a practical guide intended to explain ISDN and what it means to you. ISDN stands for Integrated Services Digital Network. ISDN was conceived in the late-sixties/early-seventies by the CCITT. (Consultative Committee for International Telegraph and Telephone). It gestated during the eighties and is now being realized (in North America) in the nineties. Before I get into the details, here's a summary list of the basic capabilities of ISDN:

The original CCITT literature explained ISDN as: An end-to-end digital network, evolved from the public telephone network, providing for the multiple, simultaneous transfer of voice, data and video (image) information over a limited set of standardized interfaces.

These original principles of ISDN still hold true today. For example, multimedia (voice, data, image) and open standards (limited set of standardized interfaces) are still prime movers in today's converging telecommunications, computer, entertainment, and information industries. However, it is necessary to recognize two distinct stages of ISDN evolution.

Narrowband ISDN: Today's implementation of ISDN is based on the original ISDN specifications. This (here and now) stage is commonly referred to as Narrowband ISDN. Narrowband refers to the electrical transmission of information over copper-based telephone lines at speeds of up to 1.5 Mb/s (Megabits per second), usually in increments of 64 Kb/s.

Broadband ISDN The upcoming implementation of ISDN is referred to as Broadband ISDN. Broadband ISDN will be more appropriate for today's concepts of multimedia, (multimegabit bandwidth) video and interactive services. Broadband refers to higher speed communications (typically from 45 Mb/s to 622 Mb/s) over fiber optic network access and transport lines.

This article describes salient aspects of the Narrowband ISDN. The three primary concepts embodied in the CCITT definition of ISDN are described below:


When ISDN was conceived, analog data transmissions over voice telephone lines were the norm. Analog electrical transmission is inherently susceptible to naturally occurring electromagnetic interference. Noise-induced errors were prevalent in computer-to-computer information streams because only rudimentary error correction was provided by data communications equipment. This resulted in inefficient and expensive data communications sessions.

The early ISDN architects adopted digital encoding and transmission schemes to overcome this condition. Even with today's sophisticated error-correcting modems, the local access (local loop) portion of the telephone network is still susceptible to induced noise. This causes degraded transmission efficiency and poor data quality primarily because of the underlying analog encoding and transmission scheme.

Today's sophisticated error correcting modems (with compression techniques) are limited to the 30 to 50 Kb/s range over the conventional public telephone network. Because of its inherent digital encoding and transmission scheme, ISDN provides virtually error-free transmission at higher speeds up to 1.5 Megabits/second. These speeds are achieved without compression. Still higher throughput will be obtained with ISDN-based transmission devices that incorporate compression. (Ratios could range from 2:I to 6:1.)


The Narrowband ISDN has two defined PHYSICAL interfaces - the Basic Rate Interface (BRI) and the Primary Rate Interface (PRI). The most prevalent interface is the Basic Rate Interface.

(More common)

The BRI provides a total transmission bandwidth of 160 Kb/s. The 160 Kb/s is divided between overhead, signaling/control and user (Bearer) information. Of the 160, 16 Kb/s is dedicated to overhead. This includes data organization (framing) and an embedded operations channel for network maintenance and performance management functions.

This leaves 144 Kb/s available for control and user information. The 144 Kb/s bit stream is logically partitioned into two 64 Kb/s "Bearer" (B) channels and one 16 Kb/s signaling/data (D) channel. Hence, the BRI is often referred to as "2 B + D". The D channel always carries control/signaling information but can also simultaneously support an X.25 Packet-Switched Data Bearer service with throughput of up to 9.6 Kb/s. Bearer (B) channels carry user information and can be assigned to any of the following (Bearer) services:

The Voiceband Information Bearer service provides the digital bandwidth (56-64 Kb/s) for a 3.1 KHz telephone analog signal. The VI service is used for normal voice conversations and provides transmission compatibility with Voiceband equipment such as modems.

The Circuit Switched Data Bearer service provides for either a 56 Kb/s or a 64 Kb/s unrestricted transmission channel. CSD is suitable for continuous bit rate applications such as video conferencing and bulk file transfers. The 56 Kb/s CSD service is compatible with existing non-ISDN Switched 56 Kb/s Data Services.

X.25 PSD
The X.25 Packet Switched Data Bearer service is suitable for applications with short or "bursty" transmission characteristics. These applications include telemetry and automated teller machine/credit card verification transaction processing. X.25 PSD supports data throughput rates from 75 bits per second up to 48 Kb/s.

(Less common)

The PRI provides a total bandwidth of 1.536 Mb/s. The PRI bitstream is partitioned into 23 user-assignable 64 Kb/s logical B channels and one 64 Kb/s logical D channel which only supports user-to-network signaling. The PRI is targeted at "high speed" bulk information transfer and high call volume applications. This includes such host-to-host file transfers, group video conferencing, remote computer aided design/ computer aided engineering workstation access and Private Branch eXchange (PBX) network access.


A very important aspect of the ISDN specification is its computer-oriented, message based signaling protocol commonly referred to as the "D-channel protocol".

The open-standards-based D-channel protocol is the key to the flexibility and utility of the (BRI and PRI) ISDN interfaces. Because the D-channel protocol is derived from X.25, it inherits the X.25 multiplexing technology that allows simultaneous logical "sessions" over a single physical interface. This allows multiple calls and call types (i.e., VI, CSD, and PSD) to exist simultaneously on a single ISDN line.

What's in it for US?
ISDN technology will allow you to simultaneously:
What about Modems?
Computer modems will be with us as long as analog lines are being used for data communications. (A long time.) Current modems are analog/digital translators, changing digital computer information into analog noise and back again.

The ISDN equivalent of a modem is called a Terminal Adapter, which is dramatically different from a modem. Because ISDN capable equipment delivers digital information to your telephone jack, analog/digital conversion is not necessary. The computer modem will eventually be replaced by an inverse multiplexer. This will act as a data separator to route data to your computer.

ISDN in 1993
ISDN-based services have been available in most US metropolitan areas for several years. Until November of 1992, however, an interoperable standard implementation did not exist. This means that ISDN services on AT&T equipment wouldn't work with ISDN on other vendors' equipment. To solve this problem, a group of telephone companies, end-user groups, and telecommunications equipment manufacturers worked together and developed a standard for a nationally interoperable ISDN implementation.

In November of 1992 the first version of this standard was successfully demonstrated. Since then, actions have been taken to increase ISDN availability by:

1) Tariff agreements for ISDN service
2) Nationwide interconnection agreements between long distance voice and data carriers.
3) End-user equipment manufacturers working together to provide high quality, low cost ISDN terminal equipment.

Additional work is underway with computer industry giants like Microsoft and Intel to define Applications Programming Interfaces. The net result of these and many other activities is that technical obstacles to widespread national ISDN availability are being swiftly overcome.

Coming soon to Your Neighborhood
If it is not already there, ISDN is coming soon to your area. It is projected that nearly 2/3 of all US telephone lines will have access to ISDN by the end of 1995. Relatively soon a cornucopia of complementary end-user telecommunications products utilizing ISDN will become widely and inexpensively available. When these two events occur, be prepared for the screeching sound of ISDN terminal adapters shifting down to match the "slow" high speed analog modems. (As it turned out, here in California, the screeching sound was from the captive customers of the local telco monopoly, in reaction to the mess-ups of ISDN installation orders and billing.)

Concurrent BBS Calls?

(Multiple BBS Connections using Packet Switching/Rate Adaption)

(By Matt Young)

What if you could call one BBS, begin downloading a file, and meanwhile call a second BBS and read your messages? The current way to call multiple BBSs (at once) is to have multiple modems and phones lines connected to your computer. The near future holds a new method of simultaneously handling multiple BBSs (or other phone calls), on a single line, using the technique of rate adaption. (Now, some call this kind of thing PPP or telnet.)

Rate Adaption
Rate adaption allows a user to divide their (digital only) telephone line between multiple connections. Rate adaption was pioneered by, and is still used by, the telephone companies that use Integrated Switched Digital Network (ISDN) services.

ISDN is a digital dial-up telephone service technology that does not require modems. ISDN operates 5 to 10 times faster than current modem technology. ISDN is widely available now for businesses, but not to home users. Upgrades to telephone company equipment are necessary before offering digital dial-up service to the consumer. These upgrades are not yet completed in suburban areas.

Rate adaption is one of the many ISDN standards. Rate adaption allows multiple low speed data streams between two sites over a single ISDN telephone line. For instance, an office site could connect to both an email and a database service from the headquarter's office over a single digital telephone link.

Packet Switching
Packet switching is a method of digital communication where data is formatted into blocks (packets). The key concept of packet switching is the data does not tie up the communication link. The link is only occupied for the short time needed to transfer the blocks of data. This allows multiple data channels to be multiplexed over one communication link. Rate adaption is a streamlined form of packet switching.

X.25 packet switching is an earlier specification of a rate adaption protocol that also allows for multiple connections over a single line. X.25 takes batches of user data, provides a connection identifier, checks the data, and then sends the packet of data through the network to the destination. X.25 forms temporary connections, assigning a temporary data destination address on the fly, during connection set up.

You've used it before
Many commercial BBSs are accessible over public X.25 networks such as Tymnet and SprintNet. Another packet-switched rate adaption technology is TCP/IP. TCP/IP is the basis for the Internet, the public packet network. TCP/IP works similarly to X.25, except that the destination addresses are permanently assigned. TCP/IP uses data header information with predefined data routing information.

Several Rate Adaption Techniques
The original ISDN rate adaption specification was called V.110. Most data equipment companies rejected it. V.110 defined a complex, bit-level data manipulation scheme that consumed considerable computer processor power.

The next ISDN rate adaption specification was called Frame Relay. Multiple bytes from one source are packed into a single frame and sent to one destination. Then another frame of bytes is packaged for the second source, and so on. The frames (data packets) look like streamlined X.25 packets, but with only half of the header information. Frame Relay is very popular and today users can obtain frame relay service from most public packet networks. The Frame Relay is a competitor to X.25.

The third rate adaption specification is called Time Division Multiplexing (TDM). The digital data stream is broken into bytes, and bytes are assigned to each source. The first source uses bytes 1 and 2, the second 3 and 4, and so on. This is the current form in which traditional phone calls are packed into high-bandwidth links for long distance transmissions.

Yet another method to implement rate adaption is called fast switching. Fast switching maintains two or more connections through a network. The end user's software tells the network to direct the data stream to one of the connections. The data stream follows the requested path until the software indicates that the stream should be redirected. A convenient feature of fast switching is that if only one connection is established, the data stream behaves like a 'normal' modem connection.

Fast switching depends upon having another frame-based rate adaption technique. This other technique allows requests and responses between the network connection controller and the user's software.

The specifications for ISDN include a signaling mechanism called Q931. This mechanism uses data frames to communicate dialing information between the user and the telephone company. Future modem switching mechanisms will operate like the current ISDN Q931 specification. In fact, rate adaption for modem users will borrow heavily from the time proven ISDN standards.

Sounds neat, When can I call?
It sounds great - we use our high speed modems and open connections to three or four BBS simultaneously with a single phone call. Each independent modem connection has its own window.

Are the telephone companies going to be the ones to make this work? Maybe not. Internet and Tymnet use this technology now, but their methods are not optimized for BBS users. Because of the growing importance of online operations, the industry is beginning to take a look at rate adaption standards optimized for high speed modems. You can be sure that any standards will be compatible with all current software, and all current and planned modems.

RIP Graphics: Making BBSs User-Friendly

(By Allen Woolley)

RIP, the Remote Imaging Protocol designed by the TeleGraphics Communications Corporation, is rapidly changing the face of BBSs across the world.

BBSs are hard to figure out
For over a decade, the only way to use a BBS was to type in characters in response to a menu displayed on the user's screen. For the techno-geeks like myself, this was no great challenge or impediment to our use of BBS systems.

However, for the uninitiated, the unwilling, and just about anyone else outside of the techno-fraternity (90% male according to my BBS statistics), the BBS menuing system can be confusing, intimidating, and obscure to the point of frustration. The best we (as Sysops) could do for our novice users was to use ANSI colors and simplified layouts in an attempt to make the "What do I do now?" question as self-answering as possible.

The luckiest novice callers would have a mentor/guru at their side (or at least one only a phone call away). The rest had to suffer with vague help screens full of techno-babble.

New BBS callers can be intimidated by online menus full of BBS jargon. I have watched people on my own system give up half way through the new user questionnaire when they were asked to "Please select a default download protocol".

Who knows what a default download protocol is? Some of my users didn't even understand the full implication of what "default" meant, let alone what a "protocol" is. I have, of course, changed that question since then, but you get the idea. RIP allows almost all of this "techno-speak" to be either eliminated entirely, or learned at a much slower "learn as you BBS" pace. The key is to eliminate the traditional BBS menuing system and move to a point and click interface.

The number of people who are "Command Line Impaired" is huge. However, the number of "Mouse Enabled" people is growing faster than ever before, due to the wide spread acceptance of Graphical User Interfaces on all platforms (even Unix!).

RIP Makes it easy for the Caller
RIP allows the Sysop to design a new "Point and Click" interface to their BBS system. This opens the BBS to many more people, who will now be able to navigate the BBS without having to read the usually poorly-written BBS online help files.

RIP will allow all types of BBS software packages to appear to be the same to the callers. The caller will be spared from having to remember the differing commands for each type of BBS software.

With RIP, the caller only needs to click on the New Personal Mail button, regardless of the underlying BBS system software. The same buttons will do the same thing on all RIP-compatible systems. BBS software authors may not like it, but RIP will remove the BBS software specific "flavor" from most packages, much to the caller's delight!

The state of the RIP
RIP is not yet a panacea for new callers. The "State of RIP" is certainly not "mature" by any measure, but the promise of the technology is quite evident. You can see this from the first time you log into one of the first generation RIP-compatible BBSs. Here are a couple of problems that must be solved in order to fulfill the RIP promise:

Getting Ripped
To date, only IBM PC compatible computers support RIP. Besides that, hardware requirements for RIP are reasonable. EGA color graphics capability and a mouse are required. A 2400 baud modem will do just fine, because the RIP BBS icons are stored on the caller's hard drive. At 14.4K baud, the system really shines and is more limited by the speed of the user's PC instead of the modem speed.

Locally stored icons use a small amount of disk space. If you call 10 RIP BBSs, you're going to have 10 sets of icons stored on your hard disk. RIPTerm itself installs 500K of icons - which provides most of the icons that will be used. Additional storage demands should be minimal, depending on how "RIP Happy" your Sysop gets.

The tools and software for RIP BBSs are less than three months old, but already I am seeing virtually every BBS package announcing RIP compatibility. To give you a "Taste of RIP", download RIPTM153.EXE and call any v3.9 Wildcat BBS. Many other BBS software companies are also starting to support RIP.

Pages 19 though 28 were detailed listings of Bay Area BBSs.

Page 29 was a full-page ad for Halted Specialties, Inc. (

Page 30 (back cover) was a full-page ad for TeleText Communications.

End of Issue 7. Go back, or to Issue 8, or to Mark's home page.