The ISDN Connection

Posted On 2005-11-1 by FortyPoundHead
Keywords: The ISDN Connection
Tags: Networking Tutorial 
Views: 1692

To really understand the Integrated Services Digital Network (ISDN), you need to put it into context. You need to see what's come before it to more fully comprehend what the new technology brings. This tutorial briefly reviews the existing conventional telephone system before discussing how ISDN departs from the norm.

The conventional telephone system-often referred to by the acronym POTS (meaning plain old telephone system)—is an analog system. The pattern of voltage variations on the local telephone line is a direct analog of the acoustical pressure variations that produced them.

The telephone line in your home is typically a single unshielded twisted pair (UTP) of copper wires that runs to the telephone company's central office (CO). This line is usually referred to as the subscriber loop or local loop.

The central office is a key concentration point for telecommunications. Inside the central office is a telephone switch that connects your line with that of the person you want to call, establishing the circuit.

ISDN Spelled Out

What's different about ISDN compared to POTS? First, it's digital instead of analog. Second, the "integrated services" portion of the name is a reference to one of the basic goals of ISDN-to provide integrated, multipurpose voice, data, and video communications, all over a single system. Contrast that with today's norm, where you typically need three services to provide the same communications ability: dial-up analog phone lines to carry voice, digital leased lines for data, and coaxial-cable networks (cable TV) for video. The idea that one network can handle all communications types is obviously appealing.

There are two levels of ISDN service. The Basic Rate Interface (BRI) is lower in cost than the other option, and it can run over typical residential subscriber loops without rewiring. BRI consists of three separate channels-two bearer channels (B channels) and one data (or D) channel. Each B channel has a data rate of 64Kbits/sec, while the D channel is 16Kbits/sec. The D channel is used for signaling, such as call setup and tear-down. The B channels carry user data, which might be digitized voice, binary data, or digitized video.

The Primary Rate Interface (PRI) might be used at a business headquarters site. It consists of 23 64Kbits/sec B channels, plus one 64Kbit/sec D channel. Again, the B channels carry the payload (voice, data, or other user information), while the D channel carries telco signaling. To get PRI service, you'll need a T1 line from your site to your local exchange carrier's CO.

In Europe, the commonly used digital data line that's roughly equivalent to the T1 line used in North America is known as E1, and it has a higher capacity than the T1-2.048Mbits/sec vs. 1.544Mbits/sec. European ISDN PRIs take advantage of the increased capacity and carry 30 64Kbit/sec B channels, plus one 64Kbit/sec D channel.

With POTS, it's a given that one phone line lets you make only one call at a time. If you're on the line when other people try to call you, they'll get a busy signal and will have to wait until you've finished your call before they can get through. With ISDN, it's a little different. With an ISDN BRI, you get two phone numbers-one for each B channel. Consequently, you could make two separate phone calls to different destinations, even though you've got only a single phone line.

PRI B channels are the same as BRI B channels; you can place a call from a BRI to a PRI, or vice versa, linking one of the BRI's B channels to one of the PRI's B channels. With 23 B channels, you could simultaneously have 23 separate ISDN calls going out over a single Primary Rate Interface to the telco, even though each of these calls has a separate destination.

ISDN is a digital, synchronous, full-duplex link. When voice is carried over ISDN, it's digitized at the telco standard of 8,000 eight-bit samples per second. This analog-to-digital conversion is performed by a compression/decompression (codec) device in the ISDN phone. If a conventional analog phone is used, it must be adapted to the network by an ISDN terminal adapter (TA). In this case, the codec is part of the TA.

The sampling rate and sampling precision is the reason 64Kbits/sec was chosen for bearer channel data rates. Eight bits sent 8,000 times per second equals 64,000 bits per second.

ISDN’s Big Picture

The block diagram in the figure shows how user equipment interfaces with the telco-provided ISDN BRI connection. A key feature of the BRI is that it can run over most existing subscriber loops, which are two-wire (single-pair) UTP. On the left side of the diagram (everything left of the letter U) is the customer premises, while the right side is the telco side, including the central office ISDN switch.

The telcos don't want customer equipment to connect directly to their networks, for fear of faulty or poorly designed equipment disrupting operation of the network. So the telcos interpose a device known as a network terminator (NT1) between the subscriber loop and customer equipment.

The customer's equipment is referred to as terminal equipment (TE), and there are two classifications of these: TE1 and TE2. TE1 equipment is digital equipment that's ISDN-capable. An ISDN phone is one example of a TE1 device. A TE1 can plug directly into the NT1.

A conventional analog phone or an ASCII data terminal (or the RS-232 serial port on a computer) represents equipment that's not ISDN-ready. Such terminal equipment is classified as TE2. These devices must have an ISDN terminal adapter interposed between them and the NT1. The terminal adapter makes the TE2 equipment ISDN-capable.

If a customer premises switch, also called a PBX, is used to connect an ISDN line to the terminal equipment or terminal adapter, it will reside between that equipment and the NT1. PBXs are classified as NT2 devices.

In the United States, the demarcation point between the telco's network and the customer premises is located at the customer end of the subscriber loop-the point marked U in the figure. Martha Haywood, director of engineering at Telebit (Sunnyvale, CA), a manufacturer of both ISDN- and analog-based remote LAN access equipment, reports that in Europe, the NT1 is traditionally owned by the telecommunications provider, even though it's installed on the customer's premises. Therefore, it is part of the telecommunications network. Because of this, the International Telecommunications Union (ITU) standards body recognizes the T reference point as being the demarcation between the customer's and the telecommunication provider's respective portions of the network.

A customer premises switch (NT2) will most likely be used with PRIs, rather than BRIs, but it's shown in the figure to indicate its place in the overall scheme of things. On most BRIs, TE1s and TAs will be plugged directly into the NT1. The S and T reference points are thus essentially interchangeable; the switch merely serves as a line selector.

Some manufacturers of ISDN terminal equipment and terminal adapters incorporate an NT1 right into the product. In this case, the device is said to have a U interface, since it can plug directly into the subscriber loop. Having a built-in NT1 means there's one less thing to buy, but Telebit's Haywood points out that there can be drawbacks, too. For example, the ISDN standards permit up to eight TEs to be connected to a single NT1. (This doesn't mean that all eight can be active at the same time; only one TE at a time can use a B channel. Thus, a BRI can support only two calls at a time.)

In the case of the BRI, which has two B channels, a terminal adapter and an ISDN phone could both be online at the same time, with each device using one of the B channels. But if the terminal adapter had a built-in NT1, you couldn't plug in another device-unless the designers of the terminal adapter had the foresight to put in a port for a phone-because the subscriber loop itself cannot be shared. In other words, there can be only one NT1 per BRI. Haywood says that initially, there may be strong interest for products with built-in NT1s, but as customers gain experience with ISDN, they'll begin to see the advantage of keeping the NT1 as a separate device.


There's a wealth of information available on ISDN. If you're interested in delving further into the subject, refer to the following sources:

Bell Communications Research (Bellcore), the research arm of the Regional Bell Operating Companies (RBOCs), maintains a World Wide Web site that contains a great deal of information compiled by the National ISDN Users Forum (NIUF). This information is in encapsulated PostScript (EPS) format. To access it, set your Web browser to:

Pacific Bell's Market Applications Lab (510-823-1663) can answer questions about the suitability of ISDN for your application, and what types of equipment you should consider.

Pacific Bell Applications BBS (510-277-1037) has an ISDN tutorial that can be downloaded.

For more information, try the following books on ISDN:

Digital Telephony and Network Integration, Second Edition, by Bernhard E. Keiser and Eugene Strange, Van Nostrand Reinhold, New York, ISBN 0-442- 00901-1.

ISDN, Second Edition, by Gary C. Kessler, McGraw-Hill, New York, ISBN 0- 07-034247-4.

ISDN and Broadband ISDN, Second Edition, by William Stallings, Macmillan, Englewood Cliffs, NJ, ISBN 0-02-415475-X.

Telecommunications, Second Edition, by Warren Hioki, Prentice Hall, Englewood Cliffs, NJ, ISBN 0-13-123878-7.

About the Author

FortyPoundHead has posted a total of 1974 articles.

Comments On This Post

No comments on this post yet!

Do you have a thought relating to this post? You can post your comment here. If you have an unrelated question, you can use the Q&A section to ask it.

Or you can drop a note to the administrators if you're not sure where you should post.

Your IP address is:

Before you can post, you need to prove you are human. If you log in, this test goes away.

Code Links