Basic Telephony 5

Objectives
After studying this chapter, you should be able to:
( Describe the topology of the switched telephone network.
( Describe the various signals present on a local-loop telephone line and
explain the function of each.
( Describe and compare in-band and out-of-band signaling systems for
telephony.
( Explain the advantages of common-channel signaling.
( Describe Signaling System Seven and explain its use in keeping track
of calls.
( Explain the use of time-division multiplexing in telephony and perform bit
rate calculations with TDM signals.
144 ! CHAPTER 5

' 5.1 Introduction

The public switched telephone system (PSTN) is undoubtedly the larg-
est, and probably the most important communication system in the world.
The reasons for this are contained in those first two words. It is public in the
sense that anyone can connect to it. Because it is switched, it is possible, in
theory at least, for anyone to communicate with anyone else. This makes the
telephone system very different from broadcasting systems and from private
communication networks.
The addition of personal wireless communication to the PSTN has made
it even more ubiquitous. Radio systems that were separate entities are in-
creasingly becoming extensions of the telephone system. Consider, for
instance, the difference between citizens’ band radio and cellular radio. The
former is a separate system, useful only when those who wish to communi-
cate have compatible radios and are within radio range. The latter can con-
nect its user to any telephone in the world, regardless of distance.
The telephone network employs many of the most interesting develop-
ments in communication practice, such as cellular radio, fiber optics, and
digital signal transmission, but it remains in many ways consistent with its
origins in the nineteenth century. This will become obvious when we look
at telephone signaling systems and the voltage and current levels found
on subscriber lines. Compatibility has been maintained in most areas of
the system, so simple dial-type telephones can coexist with modern data-
communication equipment. Though originally intended only for voice
communication, the switched telephone network has been adapted to serve
many other needs, including data communication, facsimile, and even
video.
This chapter introduces the telephone system and describes the ways in
which it can connect with wireless systems. A basic knowledge of ordinary
voice telephony (plain old telephone service or POTS in telephone jargon) will
be very useful as we consider more advanced uses.

' 5.2 Network Topology

Switched networks can be categorized as circuit-switched or packet-
switched. In a circuit-switched network there is a dedicated physical path
from transmitter to receiver for the duration of the communication. The
PSTN is a circuit-switched network. Packet-switched networks route short
bursts of data, called packets, from point to point as needed. A virtual connec-
tion may exist, but it is merely a record of the addresses on the network
between which communication takes place. Successive packets may take
 BASIC TELEPHONY ! 145

different paths through the network. The internet is a packet-switched net-

work.
Both types of networks have their advantages. Circuit switching is often
more reliable, especially when it is important that messages arrive quickly
and in the same order in which they were sent, as is the case with telephony.
On the other hand, packet switching can make more efficient use of network
resources. In a circuit-switched network the circuit is often idle. For in-
stance, during a phone call circuits are active in both directions, but most of
the time only one person is talking. It is possible to use packet-switched net-
works for telephony (internet phone is an example), and it looks as if this is
the direction networks are moving.
Wireless networks can be either circuit-switched (cellular phones) or
packet-switched (wireless local-area networks), and we shall examine both
types later in this book.

PSTN Structure Figure 5.1 shows the basic structure, or topology, of a local calling area
(known as a Local Access and Transport Area or LATA) in a typical switched
telephone system.
Each subscriber is normally connected via a separate twisted-pair line,
called a local loop, to a central office, also called an end office, where cir-
cuit switching is done. Actually the term office can be deceiving: in urban ar-
eas, it is quite possible for there to be more than one central office in the
same building.
The central office represents one exchange: that is, in a typical seven-
digit telephone number, all the lines connected to a single central office
begin with the same three digits. Thus there can be ten thousand telephones

FIGURE 5.1
Local access and
transport area
146 ! CHAPTER 5

connected to a central office. Subscribers connected to the same central

office can communicate with each other by means of the central office
switch, which can connect any line to any other line. Modern switches are
digital, so the analog local-loop signals are digitized as they enter the central
office.
The central offices themselves are connected together by trunk lines;
any subscriber can contact any other subscriber within a local calling area.
There are not enough trunks or enough switching facilities for every sub-
scriber to use the system at once, so there is the possibility of overload. This
can make it impossible for a subscriber to place a call, an occurrence that is
known as call blocking. The likely number of simultaneous conversations
is predicted by statistical methods, and should usually be exceeded only
during emergencies.
When there are no available trunks between two central offices, some-
times a connection can be made through a tandem office, which connects
central offices without having any direct connections to individual tele-
phones.
Long-distance calls are completed using a mesh of long-distance switch-
ing centers. The network usually lets the system find a direct route from one
area of the country to the other, and there is never a need for more than
one intermediate switch. Figure 5.2 shows this type of system.
The telephone system in the United States and Canada was formerly a
monopoly, and it still is in much of the world. This is changing. Competi-
tion for long-distance calls is well established, and competition for the local
subscriber loop is just beginning. Each competing long-distance carrier has
its own connection to the local access and transport area. To keep the dia-
gram simple, only two long-distance carriers are shown in Figure 5.2. Each
carrier has its own connection, called a point of presence (POP), to the local
telephone system. Wireless telephone systems are connected to the PSTN in
a similar fashion.

FIGURE 5.2
Long-distance
network
 BASIC TELEPHONY ! 147

Until now, local-loop competitors have leased their physical connec-

tions from the original local monopoly. As local-loop competitors using
their own cable (for example, cable television companies) enter the market,
they too will require points of presence.
New switching equipment and trunk lines are digital, with time-division
multiplexing used to combine many signals on one line. Fiber optics tech-
nology is increasingly used for trunk lines; other common media include ter-
restrial microwave links, geostationary satellites, and coaxial cable. Short
distances between central offices may be covered using multipair cable,
which has many twisted pairs in one protective sheath. So far, most local
loops are still analog using twisted-pair copper wire, but that seems likely
to change in the future. Eventually the system will be digital from one end to
the other, and most of it will employ fiber-optic cable. However, telephone
equipment is built for maximum reliability and lasts a long time. It is also
expensive, and telephone companies quite naturally expect to use it until
it wears out. It will be a few more years before we have a complete digital
network.

' 5.3 The Local Loop and Its Signals

Normally each individual subscriber telephone is connected to the central
office by a single twisted pair of wires. The wires are twisted to help cancel
their magnetic fields and reduce interference, called crosstalk, between cir-
cuits in the same cable. It is common practice to run a four-conductor cable
to each residence, but only two of these wires (usually red and green) are
used for a single line. The others (black and yellow) allow for the installation
of a second line without running more cable.
Recently there has been a trend toward running multiplexed digital
signals to junction boxes in neighborhoods, switching at that point to ana-
log signals on individual pairs, in an effort to reduce the total amount of
copper cable. There has been some installation of optical fiber, though the
last section of loop to the customer normally remains copper. In the future,
fiber-in-the-loop (FITL) may be used. It will cost more because of the neces-
sity of converting back and forth between electrical and optical signals at
each subscriber location, but the bandwidth will be vastly increased. This
will allow a great number of additional services, such as cable television and
high-speed internet access, to be carried on the same fiber.
The local loop performs several functions. It carries voice signals both
ways. It must also carry signaling information both ways: dialing pulses or
tones to the central office from the customer and dial tones, ringing, busy
148 ! CHAPTER 5

signals, and prerecorded messages from the network to the subscriber. Wire-
less telephones must perform all of these functions, although usually not in
exactly the same way.
In addition to the previously mentioned functions, the twisted pair
must transmit power from the central office to operate the telephone and
ring the bell.
When the phone is on hook (not in use), the central office maintains a
voltage of about 48 V dc across the line. Of the two wires in the twisted pair,
one, normally the green, is designated tip and the other (red), ring. The ring
is connected to the negative side of the supply. Most of the time in electronic
equipment a red wire is positive but not here! The “tip” and “ring” terminol-
ogy dates from the days of manual switchboards; it describes the connec-
tions to the plugs used in these boards. The positive (tip) side of the supply is
grounded.
The central office supply is called the battery. The voltage does, in fact,
derive from a storage battery that is constantly under charge. This allows the
telephone system to function during electrical power outages, whether they
occur at the central office or at the customer’s premises, and has resulted in a
well-deserved reputation for reliability. If wireless systems are to strive for
similar performance, they too need emergency power. This is especially im-
portant for cellular phone and PCS services, because many people subscribe
to these services in order to have emergency communication when needed.
When the phone is on hook, it represents an open circuit to the dc bat-
tery voltage. The subscriber signals the central office that he or she wishes to
make a call by lifting the receiver, placing the instrument off hook. The tele-
phone has a relatively low resistance (about 200 ohms) when off hook,
which allows a dc current to flow in the loop. The presence of this current
signals the central office to make a line available (the telephone is said to
have seized the line). When off hook, the voltage across the telephone drops
considerably, to about 5 to 10 volts, due to the resistance of the telephone
line. Resistance can also be added at the central office, if necessary, to main-
tain the loop current in the desired range of approximately 20 to 80 mA. Fig-
ure 5.3 illustrates this capability.

EXAMPLE 5.1 Y
A local loop has a resistance of 1 kΩ, and the telephone connected to it has
an off-hook resistance of 200 Ω. Calculate the loop current and the voltage
across the telephone when the phone is:
(a) on hook
(b) off hook
 BASIC TELEPHONY ! 149

FIGURE 5.3 Local loop

SOLUTION
(a) When the telephone is on hook, its dc resistance is infinite so the current
is zero. Since there will be no voltage drop around the loop, except at the
phone itself, the full battery voltage will appear across the phone.
(b) When the phone is off hook, the total loop resistance is

RT = 1000 Ω + 200 Ω = 1200 Ω

Then the loop current is

48 V
I = = 40 mA
1200 Ω

The voltage across the telephone is

V = IR = 40 mA × 200 Ω = 8 V
X

Once a line has been assigned, the office signals the user to proceed by
transmitting a dial tone, which consists of 350 Hz and 440 Hz signals added
together.
Dialing can be accomplished in one of two ways. The old-fashioned ro-
tary dial functions by breaking the loop circuit at a 10 Hz rate, with the num-
ber of interruptions equal to the number dialed. That is, dialing the number
5 causes five interruptions (pulses) in the loop current. This technique is
called pulse dialing and can be emulated by some electronic telephones. The
second and much more efficient way is for the phone to transmit a combina-
tion of two tones for each number. This is officially known as dual-tone
150 ! CHAPTER 5

multi-frequency (DTMF) dialing and is commonly referred to as Touch-

Tone or just tone dialing (the term Touch-Tone is a registered trade mark of
AT&T). Table 5.1 shows the combinations of tones used for each digit. This
system improves efficiency because digits can be transmitted in much less
time than with pulse dialing.

TABLE 5.1 DTMF Frequencies

Frequencies (Hz) 1209 1336 1477 1633

697 1 2 3 A

770 4 5 6 B

852 7 8 9 C

941 * 0 # D

The letters A through D are included in the system specifications but are
not present on ordinary telephones. Some wireless equipment uses them for
special functions.

EXAMPLE 5.2 Y
What frequencies are generated by a telephone using DTMF signaling when
the number 9 is pressed?

SOLUTION
Use Table 5.1. Go across from 9 to find 852 Hz; go up to find 1477 Hz. There-
fore, the output frequencies are 852 Hz and 1477 Hz.
X

Assume for now that the called party is connected to the same central of-
fice as the calling party; that is, they have the same exchange, and the first
three numbers in a typical seven-digit telephone number are the same.
When the switch connects to the called party, it must send an intermittent
ringing signal to that telephone. The standard for the ringing voltage at the
central office is 100 V ac at a frequency of 20 Hz, superimposed on the 48 V
dc battery voltage. Of course, the voltage at the telephone will be less than
that, due to the resistance of the wire in the local loop. In order to respond to
 BASIC TELEPHONY ! 151

the ac ringing signal when on-hook, without allowing dc current to flow, the
telephone ac-couples the ringer to the line. In a conventional telephone
with an electromechanical ringer, the ringer consists of two coils and a
capacitor in series across the line.
While the called telephone is ringing, the central office switch sends
a pulsed ac voltage, called a ringback signal, to the calling telephone. The
ringback signal consists of 440 and 480 Hz signals added together. When
the called phone goes off-hook, the circuit is complete, the ringing volt-
ages are switched off, and conversation can begin. If the circuit correspond-
ing to the called telephone is in use, a busy signal will be returned to the
caller.
See Table 5.2 for a summary of the signals described so far.

TABLE 5.2 Local-Loop Voltages and Currents

On-hook voltage 48 V dc

Off-hook voltage (at phone) 5–10 V dc, depending on loop resistance

Off-hook current 23–80 mA dc, depending on loop resistance

Dial tone 350 and 440 Hz

Ringing voltage (at office) 100 V ac, 20 Hz, superimposed on 48 V dc

Ringing voltage (at phone) Approximately 80 V ac, superimposed on

48 V dc

Ringback voltage 440 and 480 Hz; pulsed 2 s on, 4 s off

Busy signal 480 Hz and 620 Hz; pulsed 0.5 s on, 0.5 s off

The single twisted-pair local loop is required to carry both sides of the
conversation simultaneously, providing full-duplex communication. This is
called a two-wire system. The rest of the network uses separate transmission
paths for each direction, a topology called a four-wire system. Converting be-
tween the two systems is done using a circuit called a hybrid coil, shown in
Figure 5.4. The same thing can be done electronically. Signals from the trans-
mitter will add at the line and cancel at the receiver. Similarly, signals com-
ing in on the line will cancel at the transmitter and add at the receiver.
Deliberately unbalancing the circuit allows a small portion of the transmit-
ter signal to reach the receiver, creating a sidetone that lets the user know
the line is active and hear what is being transmitted. Hybrid coils are used
both in the phone and in the central office line cards.
152 ! CHAPTER 5

FIGURE 5.4
Hybrid coil

' 5.4 Digital Telephony

The telephone system was completely analog at its beginning in the nine-
teenth century, of course. Over the past 30 years or so, it has gradually been
converted to digital technology and the process is not yet complete. The dig-
ital techniques were originally designed to work in conjunction with the
existing analog system. This, and the fact that many of the standards for dig-
ital technology are quite old, should help the reader understand some of the
rather peculiar ways things are done in digital telephony.
The basics of digital transmission of analog signals by pulse-code modu-
lation (PCM) were discussed in Chapter 3. We can summarize the results
here as they apply to the North American system. The numbers vary slightly
in some other countries but the principles are the same.
The analog voice signal is low-pass filtered at about 3.4 kHz and then
digitized, using 8-bit samples at a sampling rate of 8 kHz. The signal is com-
pressed, either before or after digitization, to improve its signal-to-noise
ratio. The bit rate for one voice signal is then

ƒ b (voice) = 8 bits/sample × 8000 samples/second = 64 kb/s (5.1)

The sample rate is determined by the maximum frequency to be trans-
mitted, which was chosen for compatibility with existing analog FDM trans-
mission (which uses SSBSC AM with a bandwidth of 4 kHz per channel,
including guardbands between channels). An upper frequency limit of
about 3.4 kHz has long been considered adequate for voice transmission.
In Chapter 3 we found that a much lower bit rate could be used for
telephone-quality voice using data compression and vocoders. These tech-
niques are not employed in the ordinary telephone system, though data
 BASIC TELEPHONY ! 153

compression is used in special situations where bandwidth is limited and

expensive, as in intercontinental undersea cables.
The connection of wireless systems to the PSTN requires conversion of
standards in both directions. In general, wireless systems use ordinary tele-
phone quality as a guide, though many of these systems fall short of what
has been considered toll quality, that is, good enough to charge long-distance
rates for. Until now, users have been so delighted to have portable tele-
phones that they have been willing to put up with lower quality. This situa-
tion is changing quickly now that wireless phones are everywhere.

Time-Division In Chapter 3 we discussed the simplest form of time-division multiplexed

Multiplexing telephone signal. The DS-1 signal frame has one sample (8 bits) from each of
24 telephone channels plus one framing bit. This gives it a bit rate of

ƒ b (DS-1) = 24 × 64 kb/s + 1 × 8 kb/s = 1.544 Mb/s. (5.2)

When this signal is transmitted over copper wire, the result is known as a
T-1 carrier. That is, the signal includes only the coding into ones and zeros,
while the carrier also includes the voltage levels used. See Figure 5.5 for a re-
view of the DS-1 frame, which we looked at in Chapter 4.

FIGURE 5.5
DS-1 signal

A DS-1 signal can equally well be used for data communication. A single
time slot may be used or any number of time slots can be combined for
higher bit rates.
The framing bits are used to enable the receiver to determine which bit
(and in what sample) is being received at a given time. In addition, the re-
ceiver must often distinguish between frames in order to decode signaling
information. In one frame out of every six, each of the least significant bits
in the 24 samples may be used for signaling information rather than as part
of the PCM signal. This information includes on/off hook status, dial tone,
dialed digits, ringback, and busy signal. This bit robbing for signaling re-
sults in a very slight degradation of voice signal quality; for instance, the
signal-to-noise ratio is reduced by about two decibels.
154 ! CHAPTER 5

The frames are divided into groups of 12 with different signaling infor-
mation in the sixth and twelfth frames, known as the A and B frames, in a
sequence. A group of twelve frames is called a superframe. As a result, the
receiver is required to count frames up to 12. To allow the receiver to accom-
plish this, the framing bit alternates between two sequences, 100011 and
011100. The underlined bits indicate the A and B signaling frames, respec-
tively. The “stolen” signaling bits can be used to indicate basic line states
such as on-hook and off-hook, ringing, and busy signals.
Unfortunately the effect of bit robbing on data transmission is much
greater than it is for voice. Occasional bit errors are acceptable in voice sig-
nals, but certainly not in data. To avoid errors when bit robbing is used, one
bit from each 8-bit sample is discarded in every frame. This reduces the data
capacity of one voice channel from 64 kb/s to 56 kb/s. Of course, channels
can be combined for higher rates, but the loss in throughput is very substan-
tial. Bit robbing can be eliminated by using common-channel signaling,
which is described in the next section.

Digital Signal The DS-1 signal and T-1 carrier described earlier represent the lowest level in
Hierarchy a hierarchy of TDM signals with higher bit rates. All of these signals contain
PCM audio signals, each sampled 8,000 times per second. As the number of
multiplexed voice signals increases, so does the bit rate. This requires that
the channel have a wider frequency response and that variations of time de-
lay with frequency be held to a low level. Twisted-pair lines, when specially
conditioned, can be used for the T1 and T2 carriers, but higher data rates
require channels with greater bandwidth, such as coaxial cable, microwave
radio, or optical fiber. See Table 5.3 for more details.

TABLE 5.3 Digital Signal Hierarchy

Carrier Signal Voice Channels Bit Rate (Mb/s)

T1 DS-1 24 1.544

T1C DS-1C 48 3.152

T2 DS-2 96 6.312

T3 DS-3 672 44.736

T4 DS-4 4032 274.176

T5 DS-5 8064 560.16

 BASIC TELEPHONY ! 155

A glance at the table shows that the math does not seem to be exact. For
instance, a DS-1C signal carries as many voice channels as two DS-1 signals,
but the bit rate is more than twice as great. The difference is:

3.152 Mb/s − 2 × 1.544 Mb/s = 64 kb/s

The extra bits have several uses. They provide synchronization and
framing for the demultiplexer. There are also extra bits called stuff bits which
are added during multiplexing to compensate for differences between the
clock rates of the tributaries and the multiplexer. If the tributary clock rate is
slow, more stuff bits will be added to build up the bit rate; if it is fast, fewer
stuff bits are needed. This bit stuffing is more formally called justification.
Figure 5.6 is an example of the creation of a DS-3 signal by multiplexing
other signals.

FIGURE 5.6 Creation of DS-3 signal

156 ! CHAPTER 5

' 5.5 Telephone-Network Signaling

We have already looked at some of the control and supervisory signals used
with the telephone system. The local-loop signals such as DTMF tones, dial
tone, busy signal, and ringback signal are examples. All of these signals use
the same channel as the voice, but not at the same time. Since they use the
same channel, they are called in-channel signals. Their frequencies are also
in the same range as voice frequencies; they can be heard by the user, so they
are also referred to as in-band signals.
We also noted that the telephone instrument communicates its off-hook
or on-hook status to the central office by the presence or absence, respec-
tively, of a dc current. This signal is in-channel because it uses the same pair
of wires as the voice, but it is out-of-band because the dc current is not in the
same frequency range as voice signals. Consequently, the central office can
receive the off-hook signal continuously, even during the call.
Traditionally, similar methods have been used within the network for
such purposes as communicating the number of the calling party to billing
equipment, determining which trunk lines are idle (that is, ready for use),
and so on. Early systems used dc currents and dial pulses, as with local loops.
Later versions used either a switched single-frequency tone at 2600 Hz
(called SF signaling) or, still later, combinations of tones similar to but not
the same as the DTMF system and known as MF, (multi-frequency) signal-
ing. For a time, that system was plagued by fraud as people ranging from am-
ateur “phone phreakers” to members of organized crime used so-called “blue
boxes” to duplicate network signaling tones and make long-distance calls
without paying for them. However, changes to the network soon eliminated
or at least greatly reduced this problem.
In-channel but out-of-band signaling is also used. A tone at 3825 Hz can
be sent along a long-distance network. It will pass through the allotted
4-kHz channel for each call, but will be filtered out before it reaches the cus-
tomer telephone. This type of signal replaces the dc loop current as a means
of indicating whether the line is in use, since long-distance circuits, unlike
local loops, do not have dc continuity.
We have seen that the digital signal hierarchy makes available a few
bits, “stolen” from the voice channel, for signaling. These can be used to
replace the in-channel tones. However, there are insufficient bits to use for
call routing, call display, billing, and other functions of a modern telephone
network.

Signaling System Recently the trend has been to use a completely separate data channel to
Seven transmit control information between switches. This common-channel sig-
naling reduces fraud, since users have no access to the control channels, and
 BASIC TELEPHONY ! 157

also allows a call to be set up completely before any voice channels are used.
The state of the whole network can be known to the control equipment, and
the most efficient routes for calls can be planned in advance. Com-
mon-channel signaling also makes such services as calling-number identifi-
cation much more practical.
The current version of common-channel signaling is signaling system
seven (SS7). It was introduced to the Bell System in the United States in 1980
and has become, with minor variations, a worldwide system. SS7 is a packet-
switched data network linking central offices to each other, to long-distance
switching centers, and to centralized databases used for such purposes as call
display, credit card validation, voice mail, 800 and 900 number routing and
(most interesting for our purposes) cellular and PCS telephone roaming in-
formation. SS7 allows much more data to be sent more quickly, and with less
interference with voice signals, than older signaling schemes involving
in-channel signals.
SS7 uses dedicated 64 kb/s data channels. Usually one digital voice chan-
nel in each direction is reassigned for this purpose; the data rate is the same
as for a voice channel to accommodate this. If necessary, an analog channel
with modems can be used. One 64 kb/s signaling channel can handle the sig-
naling requirements of many voice channels. Figure 5.7 shows how SS7 con-
nects to the rest of the network.
With SS7, calls can be set up with no need to tie up a long-distance voice
channel until the connection is made. Since analog local loops do not sup-
port common-channel signaling, it is necessary to tie up a voice connection
from the subscriber to the central office. If ISDN, described in the next sec-
tion, is used, voice and control signals can be kept completely separate.

FIGURE 5.7
Signaling system
seven (SS7)
158 ! CHAPTER 5

' 5.6 Digital Local Loops

The analog local loop dates back to the earliest days of telephony, and is cer-
tainly outdated. Subscribers now need much greater bandwidth for uses
such as high-speed internet access or interactive video than was the case a
few years ago. One way to get this bandwidth is to replace the twisted-pair
loop with coaxial cable or optical fiber. However, the capacity of a twisted
pair is much greater than is required for the single analog signal, with a
bandwidth under 4 kHz that it usually carries. Most of the cost of the local
loop is in the labor to install it, rather than the cost of the wire itself, so when
greater bandwidth is required it makes good sense to redesign the system
to increase the capacity of the existing wiring rather than replace the wire.
Here is a brief look at two ways of increasing the capacity of twisted-pair
local loops.

Integrated The integrated services digital network (ISDN) concept is designed to

Services Digital allow voice and data to be sent in the same way along the same lines. Cur-
Network (ISDN) rently, most subscribers are connected to the switched telephone network
by local loops and interface cards designed for analog signals. This is reason-
ably well-suited to voice communication, but data can be accommodated
only by the use of modems.
As the telephone network gradually goes digital, it seems logical to send
data directly over telephone lines without modems. If the local loop could
be made digital, with the codec installed in the telephone instrument, there
is no reason why the 64 kb/s data rate required for PCM voice could not also
be used for data, at the user’s discretion.
The integrated services digital network concept provides a way to stan-
dardize the above idea. The standard encompasses two types of connections
to the network. Large users connect at a primary-access point with a data rate
of 1.544 Mb/s. This, you will recall, is the same rate as for the DS-1 signal
described earlier. It includes 24 channels with a data rate of 64 kb/s each.
One of these channels is the D (data) channel and is used for common-
channel signaling, that is, for setting up and monitoring calls. The other 23
channels are called B (bearer) channels and can be used for voice or data, or
combined, to handle high-speed data or digitized video signals, for example.
Individual terminals connect to the network through a basic interface at
the basic access rate of 192 kb/s. Individual terminals in a large organization
use the basic access rate to communicate with a private branch exchange
(PBX), a small switch dedicated to that organization. Residences and small
businesses connect directly to the central office by way of a digital local
loop. Two twisted pairs can be used for this, though more use of fiber optics
is expected in the future.
 BASIC TELEPHONY ! 159

Basic-interface users have two 64 kb/s B channels for voice or data, one
16 kb/s D channel, and 48 kb/s for network overhead. The D channel is used
to set up and monitor calls and can also be employed for low-data-rate appli-
cations such as remote meter-reading. All channels are carried on one physi-
cal line, using time-division multiplexing. Two pairs are used, one for signals
in each direction.
Figure 5.8 shows typical connections to the ISDN. The primary interface
is known as a T type interface, and the basic interface has the designation S.
Terminal equipment, such as digital telephones and data terminals, de-
signed especially for use with ISDN, is referred to as TE1 (terminal equipment
type 1), and connects directly to the network at point S. The network termi-
nation equipment designated NT2 could be a PBX, a small computer net-
work called a local area network, or a central office. Terminal equipment
not especially designed for ISDN is designated TE2 (terminal equipment type
2) and would need a terminal adapter (TA) to allow it to work with the ISDN.
Examples of type two equipment would be ordinary analog telephones, ordi-
nary fax machines, and personal computers with serial ports. Each of these
would need a different type of terminal adapter.

FIGURE 5.8
ISDN access

Implementation of the ISDN has been slow, leading some telecommuni-

cations people to claim, tongue-in-cheek, that the abbreviation means “it
still does nothing.” Several reasons can be advanced for this. First, convert-
ing local loops to digital technology is expensive, and it is questionable
whether the results justify the cost for most small users. Residential tele-
phone users would not notice the difference (except that they would have
to replace all their telephones or buy terminal adapters), and analog lines
with low-cost modems or fax machines are quite satisfactory for occasional
data users. Newer techniques like Asymmetrical Digital Subscriber Line
(ADSL), which is described in the next section, and modems using ca-
ble-television cable have higher data rates and are more attractive for resi-
dential and small-office data communication. Very large data users often
need a data rate well in excess of the primary interface rate for ISDN. They
are already using other types of networks. It appears possible that the ISDN
standard is becoming obsolete before it can be fully implemented.
160 ! CHAPTER 5

With this in mind, work has already begun on a revised and improved
version of ISDN called broadband ISDN (B-ISDN). The idea is to use much
larger bandwidths and higher data rates, so that high-speed data and video
can be transmitted. B-ISDN uses data rates of 100 to 600 Mb/s.

Asymmetrical The idea behind the asymmetrical digital subscriber line (ADSL) is to use the
Digital Subscriber frequencies above the voice range for high-speed data while leaving the use
Line (ADSL) of the local loop for analog telephony intact. This allows the subscriber to
use conventional analog telephones without special adapters, while simulta-
neously sending and receiving high-speed data. The word asymmetrical in
the name refers to the fact that the system is designed for faster communica-
tion from the network to the subscriber than from the subscriber to the net-
work. Typical uses for ADSL include internet access and interactive
television; for both of these the subscriber needs to receive data at a faster
rate than it needs to be transmitted.
There are many types of ADSL using different-frequency carriers for
downstream (to the subscriber) and upstream (from the subscriber) data.
Downstream data rates vary from about 1 to 8 Mb/s, with upstream rates
from 160 to 640 kb/s. Most systems use FDM to separate upstream from
downstream data, as illustrated in Figure 5.9. Note that the downstream
signal has wider bandwidth, as would be expected considering its higher
data rate.
Most ADSL systems require the installation of a splitter at the customer
premises to separate voice and data signals, but one variety, known as DSL
Lite, requires no splitter and provides a downstream rate of up to 1.5 Mb/s.
ADSL has the advantage over ISDN in that data signals do not have to go
through the central office switch. This means that a user can be connected
to the internet on a continuous rather than a dial-up basis. It also reduces
traffic on the switched network.

FIGURE 5.9
Spectrum of a
typical ADSL system
 BASIC TELEPHONY ! 161

' Summary The main points to remember from this chapter are:

( The public switched telephone network is of great importance for wireless

communication because it links wireless users with other users—both
wireless and wired.
( Many of the specifications for telephone equipment have their basis in
nineteenth century technology and have remained almost unchanged for
reasons of compatibility.
( Common-channel signaling allows the telephone network to set up calls
economically, without tying up voice lines, and allows the system to pro-
vide a considerable amount of data about calls.
( Wireless systems that connect with the telephone system can use its com-
mon-channel signaling system to carry information about subscribers
roaming out of their local service areas.
( Modern telephone systems are digital except for the subscriber loop, and
progress is being made towards digitizing the local loop as well.
( Digital telephone signals are time-division multiplexed with a data rate of
64 kb/s per voice channel.
( Signaling can be done by robbing bits from the voice signal but com-
mon-channel signaling using Signaling System Seven is more efficient
and allows more information to be shared.
( ISDN allows the telephone system to be completely digital from end
to end.
( ADSL allows for a conventional analog voice local loop and a high-speed
data link to be combined on one twisted pair.

( Key Terms
Asymmetrical Digital Subscriber Line (ADSL) method of providing
high-speed data transmission on twisted-pair telephone loops by
using high-frequency carriers
B (bearer) channels in ISDN, channels that carry subscriber
communication
bit robbing use of bits that normally carry payload information for
other purposes, such as controlling the communication system
bit stuffing addition of bits to a bitstream to compensate for timing
variations
call blocking failure to connect a telephone call because of lack of
system capacity
162 ! CHAPTER 5

central office switch in a telephone system that connects to local

subscriber lines
circuit-switched network communication system in which a dedicated
channel is set up between parties for the duration of the
communication
common-channel signaling use of a separate signaling channel in a
telephone system, so that voice channels do not have to carry
signaling information
crosstalk interference between two signals multiplexed into the same
channel
D (data) channel in ISDN, a communication channel used for setting up
calls and not for user communication
dual-tone multi-frequency (DTMF) dialing signaling using combinations
of two audio tones transmitted on the voice channel
end office see central office
fiber-in-the-loop (FITL) use of optical fiber for telephone connections to
individual customers
hybrid coil a specialized transformer (or its electronic equivalent) that
allows telephone voice signals to travel in both directions
simultaneously on a single twisted-pair loop
in-band signals control signals sent in a voice channel at voice frequencies
in-channel signals control signals using the same channel as a voice signal
integrated services digital network (ISDN) telephone system using
digital local loops for both voice and data, with the codec in the
telephone equipment
justification addition of bits to a digital signal to compensate for
differences in clock rates; informally known as bit stuffing
local access and transport area (LATA) in a telephone system, the area
controlled by one central office switch
local area network a small data network, usually confined to a building
or cluster of buildings
local loop in a telephone system, the wiring from the central office to an
individual customer
out-of-band in telephone signaling, a control signal that is outside the
voice frequency range
packet-switched network a communication system that works using data
divided into relatively short transmissions called packets; these are
 BASIC TELEPHONY ! 163

routed through the system without requiring a long-term connection

between sender and receiver
point of presence (POP) place where one telephone network connects to
another
private branch exchange (PBX) small telephone switch located on
customer premises
public switched telephone system (PSTN) the ordinary public wireline
phone system
ringback signal in telephony, a signal generated at the central office and
sent to the originating telephone to indicate that the destination
telephone is ringing
sidetone in telephony, the presence in the receiver of sounds picked up
by the transmitter of the same telephone
signaling system seven (SS7) system used in telephony which transmits
all call setup information on a packet-data network that is separate
from the voice channels used for telephone conversations
tandem office telephone switch that connects only to other switches,
and not to individual customers
trunk lines transmission line carrying many signals, either on multiple
pairs or multiplexed together on a single twisted-pair, coaxial cable, or
optical fiber

( Questions
1. Explain briefly how the telephone network differs from a broadcasting
network.
2. Explain the difference between circuit-switched and packet-switched
networks. Is the PSTN mainly circuit-switched or packet-switched?
3. What is a central office?
4. What is the difference between a tandem office and an end office?
5. What is a trunk line?
6. What is meant by a LATA?
7. How do wireless telephone providers connect to the wired telephone
network?
8. How has the breakup of the Bell monopoly changed the North Ameri-
can telephone network?
9. What is meant by call blocking, and why does it happen?
164 ! CHAPTER 5

10. How many wires are needed (for a single line) from the individual tele-
phone set to the central office?
11. How many wires are normally contained in the cable from an individual
residence subscriber to the network? Why is this number different from
the answer to Question 10 above?
12. Explain the meaning of the terms tip and ring. Which has negative
polarity?
13. Explain how pulse dialing works.
14. What is meant by DTMF dialing, and why is it better than pulse dialing?
15. What is the function of the hybrid coil in a telephone instrument?
16. What is sidetone and why is it used in a telephone instrument?
17. Approximately how much bandwidth, at baseband, is needed for one
channel of telephone-quality audio?
18. List the steps required in originating a local call. Include the appropri-
ate voltages and frequencies that appear at the telephone instrument.
19. Describe the difference between in-band and out-of-band signaling,
and give an example of each.
20. Describe the difference between in-channel and common-channel sig-
naling. Which is the more modern system?
21. How does common-channel signaling reduce the vulnerability of the
telephone system to fraudulent use?
22. Name some types of information that are carried by Signaling System
Seven.
23. What type of data channel is used by SS7?
24. What type of modulation is used in FDM telephony?
25. What type of modulation is used in TDM telephony?
26. What is meant by bit robbing? What is its function and why is it undesir-
able for data connections?
27. What is meant by bit stuffing? When and why is it necessary?
28. Compare basic-rate ISDN and ADSL as technologies for voice telephony.
Compare number of lines and type of equipment needed.
29. Compare basic-rate ISDN and ADSL as technologies for data communi-
cation. Compare data rates and connection type.
 BASIC TELEPHONY ! 165

( Problems
1. Suppose the voltage across a telephone line, at the subscriber, drops
from 48 V to 10 V when the phone goes off hook. If the telephone in-
strument has a resistance of 200 ohms when off hook and represents an
open circuit when on hook, calculate:
(a) the current that flows when the phone is off hook
(b) the combined resistance of the local loop and the power source at
the central office
2. The local loop has a resistance of 650 ohms and the telephone instru-
ment has a ringer voltage of 80 volts when the voltage at the central of-
fice is 100 V. Calculate the impedance of the ringer in the telephone.
3. Find the DTMF frequencies for the number 8.
4. What number is represented by tones of 770 and 1209 Hz, in the DTMF
system?
5. Calculate the overhead of a DS-4 signal:
(a) in bits per second
(b) as a percentage of the total bit rate
6. By what percentage does the use of bit robbing reduce the data capacity
of a DS-1 signal?
7. What is the proportion of overhead in a basic rate ISDN signal? (Assume
the D channel is part of the overhead.) Compare with the overhead in a
DS-1 signal.
8. The fastest modem for use on an analog telephone line operates at 56
kb/s in both directions. If an ADSL system has an upstream rate of 640
kb/s and a downstream rate of 1.5 Mb/s, by what factor does it exceed
the modem data rate in each direction?
9. Compare the data rate available with basic-rate ISDN and that available
with the ADSL system described in the previous problem. By what factor
does ADSL exceed basic-rate ISDN in each direction:
(a) if a voice call is being made simultaneously with data?
(b) if no voice call is being made so that the whole available ISDN rate
can be used for data?
10. Telephone signals can be carried by radio using either analog or digital
modulation schemes. Compare the bandwidth required to carry each of
the following signals:
(a) an analog voice signal with a baseband bandwidth of 4 kHz using
SSBSC AM (often used for terrestrial microwave links where many
voice signals are transmitted together)
166 ! CHAPTER 5

(b) the same analog voice signal using FM with a frequency deviation of
12 kHz (used for cellular phones)
(c) a standard digital voice signal using QPSK and assuming a channel
with a signal-to-noise ratio of 20 dB. (Your answer to this question
may suggest why data compression and vocoders are often used
when digital signals are to be transmitted by radio.)