The ISDN Network
The ISDN network architecture is deployed throughout the world in different ways, but the underlying principles remain the same. Several new pieces of equipment are required to facilitate the digital loop between the service provider and the customer.
The building blocks of ISDN are Bearer channels (B channels) and Data channels (D channels). The B channels are used for the transmission of customer video, voice, and data, and the D channel is used for call-control signaling. There are some cases in which the D channel can be used for data applications. These channel types will be discussed in more detail a bit later.
ISDN is referred to as a circuit-switching technology. If you think back to Chapter 2, "The Signaling System 7 Network," you'll remember that Signaling System 7 (SS7) operates by routing calls on a hop-by-hop basis. Don't think about the signaling path, but think about the actual path that the bearer traffic takes. The call moves from switch to switch until it arrives at its final destination. After it is there, a circuit path is opened from one end of the call to the other. That circuit remains open until it is closed by one of the participating sides.
This is exactly how the ISDN network operates. In Figure 8-1, when Keith wants to connect to Chris he must dial a number that is associated with Chris's ISDN circuit. These numbers are commonly referred to as phone numbers, but in the ISDN world they're actually E.164 numbers, or addresses. When Keith calls 555-1212, the call is switched through the network to Chris's location and the call is connected. Remember that as the call transits through the public switched telephone network (PSTN), it's not necessarily ISDN. The circuit path between the two parties is reserved for the duration of the call and released upon disconnect.
Figure 8-1 Circuit Switching
One of the problems with network structures such as this is that they can be quite inefficient. The connection might not require all the bandwidth that is allocated and, therefore, can waste what has been given. This is a problem not only with ISDN but also with any traditional time-division multiplexing (TDM)-based technology. To correct this problem, you need to use something similar to statistical multiplexing (stat-muxing). This way, even individual DS0s can be allocated for different applications. For more information on stat-muxing, refer to Chapter 9, "Frame Relay."
Before moving on, an important point to understand about ISDN is that it is a local loop technology. This means that from your local central office (CO) to your premises it is ISDN. However, on the other side of your CO's switch, it can be just about anything, such as ATM or Synchronous Optical Network/Synchronous Digital Hierarchy (SONET/SDH). Figure 8-2 shows the local ISDN loop interconnected to the digital transport network of the service provider.
Figure 8-2 High-Level Overview of an ISDN Local Loopt
The transit of ISDN can look something similar to ISDN to SONET for transport through the network infrastructure, back into ISDN on the physical plane, and to ISDN Q.931 signaling to SS7 signaling through the network infrastructure, and finally back into ISDN Q.931 signaling on the signaling plane. You will learn more about some of the call flows a little later in the chapter.
The Local Loop
ISDN is deployed between the service provider's CO and the customer's premises. The local loop, deployed over a combination of copper and possibly optical cable, must observe the same stipulations of any digital technology. In other words, the ISDN carrier facilities are equally as susceptible to problems such as cross talk and attenuation as any analog circuit. As far as interference goes, fiber-optic trunks do not need to worry about electromagnetic interference (EMI). Optical trunks can also transport information for much greater distances before the need for signal regeneration. Service providers tend to deploy high bandwidth, fiber trunks to integrated digital loop carrier (IDLC) systems and then provide copper service from there to the customer's premises.
The local loop, if deployed solely over copper facilities, must observe a maximum loop length of about 18 kft (18,000 feet) for BRI and 4800 ft for E1 or 6000 ft for T1 PRI. After 18 kft, the signal has to be regenerated with a device called a digital regenerator. In this chapter, these are referred to as mid-span repeaters. A mid-span repeater functions differently from an analog amplification system in that the signal is completely regenerated upon receipt. Figure 8-3 shows the cable lengths associated with ISDN BRI or PRI circuit deployments in relation to mid-span repeaters.
Figure 8-3 Mid-Span Repeaters
The signal travels from the CO, through the copper pairs, until it reaches the mid-span repeater. At that point, the mid-span repeater dusts off and regenerates the digital signal by locking onto the combination of 1s that are being transmitted from the network. The ISDN loop can travel another 18 kft because the mid-span repeater now looks similar to the origination point of the signal. The mid-span repeater functions the same way in both directions and is not limited to network to user transmission.
There is a grey area with mid-span repeater deployment when circuits fall just to one side or the other of the requirements. A question that a lot of people ask is: when will a mid-span repeater be added to an ISDN circuit? If the loop length is close to 18 kft, sometimes a repeater is added and sometimes it isn't. It depends on who is installing the ISDN circuit and whether they feel that one is necessary. Unfortunately, mid-span repeaters aren't exactly cheap, so these devices aren't deployed unless absolutely necessary.
It is also possible that mid-span repeaters can be installed even if the loop length is well within the allowed limit. The other factor that can dictate the deployment of a mid-span repeater is decibel (dB) loss. Normally, even if the circuit is within 18 kft, a mid-span repeater is deployed if the dB loss on the circuit is between 38 dB and 42 dB or higher. High dB could mean that there is a problem with some of the facilities or there might be some extreme environmental conditions that cannot be controlled.
Repeaters are normally housed in an environmentally safe casing that shields it from the elements. The cases are normally about five times the size of the actual mid-span repeater card.
Repeaters aren't the only devices that are found on the ISDN local loop. Remember that IDLCs require less copper to customer locations. Connected to a CO through a high-bandwidth trunk, these devices are positioned closer to customer locations to provide more efficient use of available facilities. ISDN services have been integrated into most IDLC equipment by using either three available DS0s for BRI or a full digital signal level 1 (DS-1) circuit for PRI. These functions will be discussed in more detail in the BRI and PRI sections of this chapter.
Echo cancellation is used on BRI circuits because there is only a single pair for both transmitting and receiving information. PRI circuits do not have this problem because they run off of either coaxial cable or two copper twisted pairs, one pair for transmit and one for receive.
Echo cancellation is designed to remove the signal echo from a circuit so that your equipment knows what information is being sent to it. Too much echo without cancellation measures can cause problems with the network equipment. Refer to Figure 8-4 and the description that follows for the basic echo cancellation procedure. Figure 8-4 shows a pair of echo cancellers removing their own signal from the circuit to ascertain the proper signal to be received.
Figure 8-4 Echo Cancellation
The voltage levels in Figure 8-4 are only examples of possible values.
The echo canceller on the right sends a signal equivalent to a 3v charge, and the echo canceller on the left sends a signal equivalent to a +1v charge.
To figure out what is being sent to it, the echo canceller on the right takes the total voltage of the circuit (+1v + 3v = 2v).
With the total line voltage known, the echo canceller then applies the voltage of what it transmitted, but in the opposite polarity. In other words, 3v becomes +3v.
The final equation looks something similar to the following: 2v + +3v = +1v. It is now known that a signal of +1v was meant for its receiver.
Network Interface Device (NID)
NID is a common term that describes the interface at which an ISDN BRI circuit connects to the customer's premises in North American implementations. The NID separates the service provider's portion of the ISDN circuit from the customer's ISDN equipment and inside wiring (ISW). This device is a result of the deregulation of the phone company after the divestiture of 1984.
A typical NID is a grey box that sits on the side of the customer's house or office, and it is split into two halves. The customer's side is usually accessible by the customer, but the service provider's side is normally locked with a special type of bolt. A special tool is required to access that side of the NID and service technicians normally carry it with them. If an ISDN technician is ever sent out to your premises, the first thing they will do is try to test from the NID back to the CO. If they test successfully and you can't access the network, chances are there is something wrong with your equipment or ISW.
ISDN Reference Points
ISDN specifies a set of reference points that describe the various network functions associated with an ISDN BRI circuit. These reference points are shown in Figure 8-5. The reference points and equipment include U, S, T, TE1, TE2, and R.
Figure 8-5 ISDN Reference Pointst
The U interface or reference point, also known as the user interface, describes the access point from the service provider's network into the customer's Network Terminator Type 1 (NT1). Also referred to as the U-loop interface, it is normally represented with a RJ-45 modular plug on the NT1. This interface is used in the United States for customer access, but most of the rest of the world uses an S/T interface.
Next, there are the S and T reference points. You might see these referred to separately as is done here or as a single reference point denoted S/T. The T reference point is located between the NT1 and NT2 devices, and the S reference point is located on the far end of the NT2 device, if the S/T interface is separated. NT1 devices terminate the ISDN circuit. NT2 devices are smart ISDN devices that make multilayer decisions on the ISDN network, such as on routers or private branch exchanges (PBXs). Devices known as NT12s are devices that integrate both functions of the NT1 and NT2 into a single device.
The two pairs of wires on the other side of the NT1 are known as the S/T bus. On this bus you can have up to eight devices, but only two can operate at any one time. The S/T bus can support a maximum length of about 3000 feet, and the bus terminates at the far end to prevent circuit echo, line noise, and distortion. The S/T bus uses pins 3, 4, 5, and 6 for transmission and reception. The S/T interface uses pins 3 and 6 for transmit and pins 4 and 5 for receive if used in a terminal equipment type 1 (TE1) or terminal adapter (TA). If the S/T interface is used in a NT, the pin assignments are reversed. Pins 3 and 6 receive and pins 4 and 5 transmit.
The next two items are types of equipment rather than actual reference points in the network. TE1 is a device that is ISDN-ready. ISDN-ready means that you can plug it directly into the ISDN network and it functions properly. A common TE1 is an ISDN phone or an ISDN video conferencing device. Terminal equipment type 2 (TE2) is a device that is not directly compatible with the ISDN network. TE2s require a TA to adapt their signal to a proper ISDN network signal. Typically, devices such as personal computers require an ISDN TA to function on the network properly. The link between the TE2 and the TA is referred to as the R reference point.
Most equipment that is sold in the United States integrates the NT1 and TA functionality into the same box. For this reason, there is no S/T bus available to users unless a standalone NT1 is purchased. In that case, the stand-alone product provides the two-pair S/T bus for a device, such as an ISDN video conferencing unit. Most devices also provide plain old telephone service (POTS) interfaces for analog communication through the ISDN circuit. This feature is provided with an analog-to-digital converter.