Telephone Transmission
Telephone Transmission
Telephone Transmission
TELEPHONE TRANSMISSION
The input–output device that provides the human interface with the telephone net-
work is the telephone instrument. It converts sound energy into electrical energy
and vice versa. The degree of efficiency and fidelity with which it performs these
functions has a vital effect upon the quality of telephone service provided. The tele-
phone set consists of transmitter, receiver, electric network for equalization, side tone
circuitry, and devices for signaling and supervision.
The transmitter (mouth piece) converts acoustic energy into electric energy by
means of a carbon granules transmitter. The transmitter requires a d.c. potential
usually on the order of 3 − 5 volts across its electrodes. It is called a talk battery,
and it is supplied over the line (central battery) from the switch. Current from the
battery flows through the carbon granules when the telephone is lifted its earphone
(off-hook). When sound hits on the diaphragm of the transmitter. Variations of
air pressure are transferred to the carbon, and the resistance of the electrical path
through the carbon changes in proportion to the pressure. Frequency response of
carbon transmitter peaks between 800 and 1000 Hz.
The propagation constant γ and characteristic impedance Z0 of a cable are given by,
p
γ = (R + jωL)(G + jωC) = α + jβ (6.1)
where,
α: is attenuation in nepers per unit length.
β: is phase shift in radian per unit length.
s
R + jωL
Z0 = (6.2)
G + jωC
√
α= RG + constant, i.e. independent of frequency
√
r
L
β = ω LC, Z0 = pure resistance
C
ω 1
vp = = √ = phase velocity = constant independent of frequency
β LC
dω 1
vg = =√ = vp = group velocity
dβ LC
For telephone cables where operating frequencies ranges from 300 Hz to 3.4 kHz,
we have ω << LR and G << ωC. Thus,
s
R R −jπ/4
Z0 = = e (6.3)
jωC ωC
and
p √
γ= jωRC = ωRCejπ/4 = α + jβ (6.4)
TELEPHONE CABLES 121
Hence,
r r
ωRC ωRC
α= and β = (6.5)
2 2
We observe that α and β are functions of frequency, the phase velocity is given
by,
r
ω 2ω
vp = = (6.6)
β RC
Twisted pair are commonly used in telephone transmission. A cable pair is made
by twisting together. Two insulated conductors usually of high purity copper. The
sizes of the cables are 6 to 900 pairs. The commonly used wire gauges and their
parameters is shown in Table 6.1.
AWG 16 is rarely used. Its diameter is 1.291 mm and R = 13.18 Ω/km. The pri-
mary constant of twisted pair cable is subject to manufacturing deviation, and change
with the physical environment such as temperature, moisture, and mechanical stress.
L is in the range of 0.625 mH/km and C is in the range of 0.041 to 0.052 µF/km.
1. Loading
From the equation of the attenuation of cable, and typical values for the primary
constants shows that substantial reduction of attenuation can be obtained by
increasing the value of L. Minimum attenuation requires a value of L nearly
122 TELEPHONE TRANSMISSION
100 times the value obtained in ordinary twisted pair. The realization of such
a value on a uniformly distributed basis is impractical. Instead, the desired
effect is obtained by “lumped” loading that is, by inserting series inductance
periodically along the pair.
Loading arrangement, are specified by a code letter designating the distance
between loading coils and by numbers which indicate the inductance value and
wire gauge. For example, the designation of one loading arrangement is 19H88.
The number 19 specifies 19 gauge wire, H indicates the spacing between coils,
the number 88 refers to the inductance in millie henries of the coils. The code
for load coil spacing are given in the following Table 6.2.
2. Bridged Taps
Bridged taps are frequently used in cables serving customer locations. This
consists of another pair of wires which are connected in shunt to the main cable
pair at any point along its length. This pair may or may not be used at some
future time. In any case only of the pairs going away from the bridging point is
likely to be used at any given time. Figure 6.2 illustrates this bridged taps.
The telephone channel may be described technically using the following parameters,
Nominal bandwidth
Phase shift
Level
Other parameters such as return loss, singing, stability, echo, reference equivalent,
· · · deals more with the voice channel in a network, which is outside the scope of this
section.
1. Bandwidth
The nominal bandwidth of a voice channel is often 4 kHz. The actual usable
bandwidth is from 300 Hz to 3400 Hz. Thus filters, attenuators, amplifiers
used to pass this voice signal should have a 3 dB bandwidth between 300 Hz
to 3400 Hz.
2. Attenuation distortion
Output signal from the channel is distorted in some manner such that it is not
an exact replica of the input. One form of distortion is called attenuation (am-
plitude) distortion. That is the amplitude response of the channel is not constant
within the pass band of the channel. To avoid attenuation distortion, an ampli-
tude equalizer is used such that all frequencies within the passband should be
124 TELEPHONE TRANSMISSION
subjected to the same loss (or gain). On typical wire systems higher frequencies
in the passband are attenuated more than lower ones.
In carrier equipment the filters used tend to attenuate frequencies around band
center the least, and attenuation increase as the band edges are approached as
shown in Fig. 6.3.
The crosshatched area in Fig. 6.3 expresses the specified limits of attenuation
distortion, and the solid line shows measured distortion on typical carrier equip-
ment for the channel band. Attenuation distortion across the voice channel is
measured compared to a reference frequency. ITU specifies the reference fre-
quency at 800 Hz. However, 1000 Hz is used in USA.
4. Level
Level is an important system parameter. If levels are maintained at too high
a point, amplifiers become over loaded, with resulting increases in intermod-
ulation products or crosstalk. If levels are too low, customer satisfaction may
suffer. The signal level is given in dBm or dBW , the signal level is measured
at the test level point a OTLP (zero test level point) is a location in a circuit
or system at which a specified test tone level is expected during alignment. A
OTLP is a point at which the test tone level should be 0 dBm.
TECHNICAL PARAMETERS OF A TELEPHONE CHANNEL 125
Thermal noise
Intermodulation noise
Cross talk
Impulse noise
Thermal Noise
Thermal noise is noise based on thermal agitation. It is a white noise, and
it is directly proportional to bandwidth and temperature, and noise figure of
the system. Thus,
Pn = kT BF (6.7)
where,
k: is Boltzman constant = 1.38 × l0−23
T : is temperature in kelven
B: is bandwidth in Hz
F : is noise figure
Intermodulation Noise
Intermodulation noise is the result of the presence of intermodulation prod-
ucts. Passing two signals with frequencies f1 and f2 through a nonlinear
device or medium, intermediation products which are spurious frequencies
will be produced. These frequencies may be present either inside or outside
the band of interest. The resulting products are,
– Second order products f1 ± f2
– Third order products f1 ± 2f2 ; 2f1 ± f2
– Fourth order products 2f1 ± 2f2 ; 3f1 ± f2
Intermodulation noise may result from a number of causes:
– Improper level setting. Too high level input to a device drives it
into its nonlinear operating region
– Improper alignment causing a device to function non linearly.
– Non linear envelope delay.
Cross talk
Cross talk refers to unwanted coupling between signal paths. There are three
causes of cross talk
– Electrical coupling between transmission media, e.g. between wire
pairs.
– Poor control of frequency response (defective filters or poor filter
design).
126 TELEPHONE TRANSMISSION
The subscriber loop connects a subscriber telephone subset with a local switch or
exchange as shown in Fig. 6.4.
maximum loop length, the radius of an equivalent area that one switch could serve
would be extended an additional 6 km out to a total of nearly 10 km.
The two basic criteria that must be considered when designing subscriber loops,
and which limit their length are,
signaling limits (obtained from resistance designs).
Attenuation limits (obtained from transmission design, taking into considera-
tion loop loss in dB at 800 Hz for Europe and many other parts of the world, or
1000 Hz for USA).
1. Resistance design
EXAMPLE 6.1
Calculate the signaling loop limit for 26 gauge and 19 gauge loops, if
switch limit is 1250 Ω.
SOLUTION
For 26 gauge d = 0.4049 mm. Thus,
43.92
Rdc = = 267.8 Ω/km
(0.4049)2
Thus,
1250
D= = 4.67 km
267.8
For 19 gauge d = 0.9116 mm. Thus,
43.92
Rdc = = 52.85 Ω/km
(0.9116)2
Thus,
1250
D= = 23.65 km
52.85
We see that signaling loop limit for 19 gauge is about 5 times longer than
for 26 gauge.
2. Transmission design
The attenuation limit is taken from the national transmission plan. The attenu-
ation of a wire pair used on a subscriber loop varies with frequency, resistance,
inductance, capacitance, and leakage conductance. Resistance of the line will
depend on temperature, so for open wire lines, attenuation may vary ±12% be-
tween winter and summer conditions. For burred cable loss variation due to
temperature are much less.
Maximum permissible loop length for transmission design is given by,
Attenuation limit in dB
DT = (6.12)
Loop attenuation loss (dB/km)
EXAMPLE 6.2
SOLUTION
For 26 gauge loop attenuation is 1.61 dB/km (from previous table). Thus,
6
DT = = 3.73 km
1.61
TELEPHONE NETWORK 129
The maximum loop length is taken to be the minimum of signaling loop limit and
transmission loop limit,
D = min{Ds , DT } km (6.13)
From the above two examples, the maximum loop length is given by,
This area is called single exchange area. The covered area is limited by the maxi-
mum loop length. To distribute the telephone network all over the area, the local ex-
change is connected to distribution tanks distributed uniformly around the exchange,
130 TELEPHONE TRANSMISSION
by primary cables. These primary cables are multi pair cables, and are connected
from these distribution tanks to distribution boards, and smaller size cables are con-
nected from these distribution boards to smaller distribution boards of sizes around
10 lines which then distributed to the nearby subscribers locations as shown in Fig.
6.8.
between the switches. The circuit shown in Fig. 6.14 may be called a small trans-
mission plan. We can define a transmission plan as a method of assigning losses end
to end on a telephone circuit. The reference equivalent is a handy device to rate such
a plan regarding subscriber satisfaction.
When studying transmission plans or developing them, we usually consider that
all sections of a circuit in a plan are symmetrical. To maintain this symmetry re-
garding reference equivalent of telephone subsets we use the term (T + R)/2, i.e.
(T RE + RRE)/2 as a symmetric circuit as given in Fig. 6.15.
Table 6.4 gives reference equivalents for subscriber sets in some countries
Thus ITU recommendation G121, states that the reference equivalent from the
subscriber set to an international connection should not exceed 20.8 dB (TRE),
and to the subscriber set at the other end from the same point of reference (RRE)
should not exceed 12.2 dB (Note: the intervening losses already are included in
these figures). By adding 12.2 and 20.8, we find 33 dB to be the overall refer-
134 TELEPHONE TRANSMISSION
1. Two–wire transmission
A telephone conversation requires transmission in both directions when both
directions are carried on the same wire pair; we call it two wire transmission.
The telephones in our home and office are connected to a local exchange by
means of two wire circuit.
2. Four–wire transmission
Carrier and radio systems require that oppositely directed portions of a single
conversation occur over separate transmission channels or paths (or using mutu-
ally exclusive time slots). Thus we have two wires for the transmit path and two
wires for the receive path or a total of four for a full duplex (two way) telephone
conversation.
For almost all operational telephone systems, the end instrument is con-
nected to its intervening network on a two wire bases.
Nearly all long distance telephone connection transverse four wire links.
From the near and user, and the far end user, the connection to the long
distance network is two wire as shown in Fig. 6.16.
3. VF repeaters
Voice frequency VF repeaters in telephony imply the use of unidirectional am-
plifiers at voice frequency on VF trunks. On a two wire trunk two amplifiers
must be used on each pair with a hybrid in and a hybrid out, as shown in Fig.
6.17
The gain of VF repeater can be run up as high as 20 or 25 dB, and previously
they were used on 80 km, 19 gauge loaded cable in the long distance network.
136 TELEPHONE TRANSMISSION
Today they are seldom found in long distance circuits but do have application on
local trunk circuit where the gain requirement is considerably less. ln practice a
repeater is installed at each end of the trunk circuit to simplify maintenance and
power feeding. Gain may be as high as 6 − 8 dB.
links are limited to assure maintenance of limits of noise, stability and reference
equivalent. Also, in a national chain the number of links are also limited. If the
average distance to a subscriber is 1000 km from the international interface,
at most three national four-wire circuits can be connected on a four wire bases
between each other.
But in countries that have average distances to subscribers in excess of 1000 km
a fourth or possibly a fifth national circuit may be added on a four wire basis.
Figure 6.18 below illustrate the case.
2. Noise
CCITT recommendation G103 treats noise. It equates noise to length of a cir-
cuit. This is equated at the rate of 4 pW p/km. Absolute noise maximum on
the receive side of an international connection should not exceed 50, 000 pwp
referred to a zero relative level point of the first circuit in the chain (Rec. G143).
This maximum noise level is permissible if there are six international circuits in
tandem.
should not exceed 1 dB. The difference between the mean value and the nomi-
nal value of the transmission loss for each circuit should not exceed 0.5 dB.
5. Attenuation
The worst condition for attenuation distortion is shown in Fig. 6.19. It assumes
that the nominal 4 kHz voice channel is used straight through. Note that the
slope from 300 to 600 Hz, and from 2400 to 3400 is approximately 6.5 dB. The
attenuation within the band 600 Hz to 2400 Hz may vary ±2.2 dB maximum
with respect to that at 800 Hz reference signal. Within the band 400 to 600 Hz
and 2400 to 3000 Hz the attenuation variation as maximum is between −2.2 dB
and +4.3 dB with respect to the reference signal at 800 , or in general the
attenuation variation between 300 Hz to 3400 Hz is between −2.2 to 8.7 dB
as maximum.
Figure 6.19 Permissible attenuation variation with respect to its value measured at 800 Hz,
for 4-wire circuits.
5. Reference equivalent
According to CCITT recommendation Q111, ”for 97% of the connections made
in a country of size 1000 km average distance to international exchange the
nominal reference equivalent between a subscriber and the 4-wire terminals
of the international circuit (international interface) should not exceed 20.8 dB
sending and 12.2 dB receiving; this gives an ORE of 33 dB. It is recommended
that telephone companies and administrations attempt to lower this figure as
much as possible to improve subscriber satisfaction. A good target would be an
ORE from a low of 6 dB to a maximum of 20 dB.
ITU (CCITT) INTERFACE 139
5. Propagation time
The propagation time at 800 Hz of the national sending or receiving system
should not exceed 50 msec (CClTT recommendation G114).
5. Echo suppression
Echo suppressors should be used on all connections where the round trip delay
(propagation time) exceeds 50 msec.