Elements of Cellular

Radio System Design 3

3.1 Introduction

T
he main requirement of mobile communication systems is to make services available
regardless of time and place. This requirement enforces the reuse of limited radio
frequency spectrum. The term co-channel interference (CCI) comes with the concept of
frequency reuse.

In frequency reuse, several cells in the coverage area use the same set of frequencies.
The aim of any cellular system design is to efficiently utilize the available spectrum such that the
CCI does not degrade the performance of the system below the guaranteed grade of service (GOS).

In this chapter, the various elements of cellular system design are discussed in detail. To this
end, the chapter is divided into seven sections. Section 3.2 deals with the concept of frequency
reuse channels. In Section 3.3, the CCI concept is introduced and the CCI reduction factor is
derived. Sections 3.4 and 3.5 are concerned with the design of a cellular system in both the
normal-case and worst-case scenarios using an omnidirectional antenna. The important concept
of cell splitting is introduced in Section 3.6, and the challenges in cell splitting are discussed
in Section 3.7. Finally, in Section 3.8 the components of a cellular system are considered for
discussion as their choice affects the system design.
A challenge faced by the designer of any cellular system is that of serving the greatest
number of customers with a specified system quality. In the process of meeting this challenge,
a number of questions must be answered, which include listing out design specifications like the
number of customers who can be served in a busy hour, the number of subscribers who can be
taken into the system, and the number of frequency channels required.

3.2 Concept of frequency reuse channels

In cellular technology, a radio channel consists of a pair of frequencies used in a two-way radio
link that is maintained between the user’s mobile phone and the base station. The major drawback
with earlier mobile communication systems was the inefficient use of allocated frequency
spectrum. Repeated reuse of radio frequencies over a given geographic area provides a number
of simultaneous conversations far in excess of the voice channels derived from simply parcelling
out the available spectrum. As shown in Figure 3.1, a particular radio channel, say f1, used in one
74 Mobile Cellular Communication

Undesired signal
C2 Desired
C1 co-channel interference
signal

f1 f1

Base D
Base
Station 1 Station 2
q = frequency reuse factor = D/R
R g = propagation path loss = C/I R
R = radius of cell

Figure 3.1 Ratio of D/R

cell, say C1, with a coverage radius R can be used in another cell with the same coverage radius
at a distance D. In a frequency reuse system, users in different cells may simultaneously use the
same frequency channel. Though the frequency reuse system leads to increase in the spectrum
efficiency, serious interference may occur if the system is not properly designed. CCI is a major
concern in the concept of frequency reuse.

CCI occurs when the same carrier frequencies reach the mobile user from the transmitters of two
base stations.

3.2.1 Frequency reuse schemes

The frequency reuse concept can be used in both the time and space domains. In the time domain,
frequency reuse results in different time slots being occupied by the same frequency. This is called
time-division multiplexing (TDM). In the space domain, frequency reuse can be divided into two
categories:

• Two different geographic areas being assigned the same frequency.

• Same frequency being repeatedly used in the same area in one system. There are many
co-channel cells in the system. The total frequency spectrum allocation is divided into K
frequency reuse patterns. This concept is explained in detail in Chapter 2.

3.2.2 Number of customers in the system

During a busy hour, the traffic conditions in the area help determine both the sizes of different
cells and the number of channels in them when we design a system. At each particular cell, the
traffic conditions drive the maximum number of calls per hour per cell. The maximum number
of calls per hour can be taken care of in each cell after the maximum number of frequency
channels per cell has been implemented in each cell.

3.3 Co-channel interference

The basic idea with cellular systems, as noted in the previous chapter, is to reuse frequency in
different cells and to increase the capacity. But two problems arise. (When channels have been
assigned to two co-channel cells that are not far enough apart geographically and their signals
 Elements of Cellular Radio System Design 75

are strong enough to overlap with each other, they cause an interference called co-channel
interference.) And the same frequency assignments cannot be made in adjacent cells because of
inter-channel interference. The assignments must be spaced far enough apart geographically to keep
interference to tolerable levels.
Unlike thermal noise, CCI can be overcome by increasing the carrier-to-interference ratio
(CIR), defined to be the ratio of the desired average signal power at receiver to the total average
interference power. This ratio is comparable with the signal-to-noise ratio and can be used as a
performance measure in non-mobile communication systems. The signal-to-interference ratio
(SIR) should then be greater than a specified threshold for proper signal operation.
For example, in the first-generation (1G) Advanced Mobile Phone System (AMPS), designed
for voice calls, the desired performance threshold, in decibels (dB), is SIRdB which is equal to
18 dB. (SIRdB = 10 log10SIR.) This number has been found to be appropriate to provide acceptable
voice quality in studies of mobile calls. For the second-generation (2G) digital AMPS (D-AMPS),
a threshold of 14 dB is deemed suitable.

3.3.1 Co-channel interference reduction factor

CCI between the cells limits frequency reuse and hence is a major problem. Our aim is to find the
minimum frequency reuse distance in order to reduce this CCI.
With the assumption that the size of all cells is the same, the cell size is determined by the
coverage area of the signal strength in each cell. CCI is independent of the transmitted power of
each cell as long as the size of the cell is fixed. It is a function of a parameter q and is defined as

D (3.1)
q= = 3N
R

where
D is the distance between the centres of cells
R is the radius of the cell
q is the reuse ratio
N is the cluster size

Distance to reuse ratio (D/R) defines the geographic distance required between the cells that are
using same frequencies in order to avoid interference between these cells.

Here, q is the CCI reduction factor and N is the cluster size. CCI decreases when q increases.
Further, the separation D in Equation (3.1) can be written as a function of KI and C/I,

D = f ( K1 , C/I ) (3.2)

where
KI is the number of co-channel interfering cells in the first tier
C/I is the received CIR at the desired receiver

C C (3.3)
=
∑
K1
I I
k =1 k
76 Mobile Cellular Communication

In a hexagonal-shaped cellular system, there are always six co-channel interfering cells in the first
tier as shown in Figure 3.2; that is, KI = 6.
Assume that the interference level is very high; the local noise can be neglected. Then, C/I
can be expressed as

C R− g (3.4)
=
∑
K1
I
k =1
Dk− g

where the actual terrain environment determines γ, the propagation path-loss slope. Usually,
γ = 4 in a mobile radio medium. Weaker interference is caused by the six co-channel interfering
cells in the second tier than those in the first tier and hence is negligible. Substituting
Equation (3.1) into Equation (3.4),

C 1 1 (3.5)
= −g
= −g
I ⎛D ⎞ ∑ (q )
K1

∑ k =1 ⎜⎝ Rk ⎟⎠
K1
k =1 k

where qk is the CCI reduction factor with kth co-channel interfering cell, and

Dk (3.6)
qk =
R

R D
1
1
1 Effective interfering
cell
1

1
1

1
First tier
1
Second tier 1
1

Third tier

Figure 3.2 Six effective interfering cells of Tier 1

 Elements of Cellular Radio System Design 77

3.3.2 Relation between co-channel reduction factor

and frequency reuse factor
From the above analysis, by substituting KI = 6 in Equation (3.5), the C/I can be obtained as

C 1 (3.7)
= −g
I ⎛D ⎞
∑ k =1 ⎜⎝ Rk ⎟⎠
6

where
2  γ  5 is the propagation path loss, and depends upon the terrain environment
Dk is the distance between the mobile and the kth interfering cell
R is the cell radius
If we assume Dk to be the same for the six interfering cells for simplification, or D = Dk, then
Equation (3.7) becomes

C 1 qg (3.8)
= −g
=
I 6( q ) 6

where q = D/R is the reuse factor.

Therefore,

1
⎡ ⎛ C ⎞ ⎤g (3.9)
q = ⎢6 ⎜ ⎟ ⎥
⎣ ⎝ I ⎠⎦

Comparing Equations (3.9) and (3.1), that is, q = 3N , we have

2
1 ⎡ ⎛ C⎞ ⎤g (3.10)
N = ⎢6 ⎜ ⎟ ⎥
3 ⎣ ⎝ I ⎠⎦

Example problem 3.1

Consider the AMPS in which the C/I of 18 dB is required for the accepted voice quality. What
should be the cluster size for the system? Assume γ = 4. What will be the cluster size of the GSM
system in which a C/I of 12 dB is required?

Solution

Using Equation (3.10), we get

2
1⎡
6 (10) ⎤⎦ 4 = 6.486 ~ 7
1.8
N AMP =
3⎣
78 Mobile Cellular Communication

and

2
1⎡
6 (10) ⎤⎦ 4 = 3.251 ~ 4
1.2
NGSM = ⎣
3

Example problem 3.2

Consider a cellular system with 395 total allocated voice channel frequencies. Calculate the
mean C/I for cell reuse factor equal to 4, 7, and 12. Assume omnidirectional antennas with six
interferers in the first tier and a slope for path loss of 40 dB/decade (γ = 4).

Solution

For a reuse factor q = 4, the number of voice channels per cell site = 395/4 = 99.
Using Equation (3.10), then mean C/I for N = 4 is

2
1 ⎡ ⎛ C⎞ ⎤4
4 = ⎢6 ⎜ ⎟ ⎥
3 ⎣ ⎝ I ⎠⎦

∴ C/I = 24(13.8 dB)

Similarly,
For N = 7
∴ C/I = 18.7 dB
And for N = 12
∴ C/I = 23.3 dB

Example problem 3.3

The base station transmitter antenna in a cell site radiates a +10 dBm RF signal and is connected
using an RF coaxial cable that induces a 3 dB loss. The cable has two connectors at its either end
that induce a loss of 2 dB each. What is the signal level at the input of the antenna system?

Solution

Given data:
The RF output signal level of cell-site transmitter antenna = +10 dBm
In terms of dB = +10 − 30 = −20 dB
Signal loss due to RF coaxial cable = 3 dB
Determination of total loss in cable connectors
Signal loss due to one connector of RF coaxial cable = 2 dB
Number of connectors on a coaxial cable = 2
Therefore, signal loss due to both connectors = 2 × 2 = 4 dB
Determination of signal loss due to cable and connectors
Total signal loss due to cable and connectors = 3 + 4 = 7 dB
Determination of signal level at the input of the antenna
 Elements of Cellular Radio System Design 79

Signal level at the input of the antenna = −20 − 7 = −27 dB

In terms of dBm = −27 + 30 = 3 dBm
Hence, signal level at the input of the antenna system = +3 dBm.

3.4 Desired C/I from normal case in an omnidirectional

antenna system
The desired CIR (C/I) is related to the co-channel reuse factor q and is given in Equation (3.9). For
an isotropic or omnidirectional antenna system, the desired C/I can be obtained with a simple
analytical solution in which two cases are considered.

3.4.1 Analytic solution

In order to obtain the desired C/I, two cases are to be considered:

• The signal and CCI received by the mobile unit

• The signal and CCI received by the cell site

Both the cases are shown in Figure 3.3.

The local noises at the mobile unit and the cell site are small and can be neglected when compared
with the interference level. The system is called a balanced system as long as the received CIRs at
both the mobile unit and the cell site are the same. In order to analyse the system requirement,
either of the cases can be chosen in a balanced system. In a simplified system,

C R− g qg (3.11)
= −g
=
I 6D 6

Thus,
C (3.12)
qg = 6
I

I1 I6

r1
r6

r2 I2 I5
r5
BS BS
r3 r4 D

I3 I4

(a) (b)

Figure 3.3 Co-channel interference from six interferers: (a) Receiving at the cell site and
(b) Receiving at the mobile unit
80 Mobile Cellular Communication

and
1
⎛ C⎞ g (3.13)
q = ⎜6 ⎟
⎝ I⎠

In Equation (3.13), the values of C/I and γ are based on the required system performance and the
terrain environment, respectively. On the basis of subjective tests, the normal cellular practice is
to specify C/I to be 18 dB or higher. In a mobile radio environment, the path-loss slope γ is equal
to about 4.

q = D / R = (6 × 63.1) 4 = 4.41
1
(3.14)

Substituting q from Equation (3.14) in

D (3.15)
q= = 3N
R

We get, N = 7.
Equation (3.15) indicates that a seven-cell reuse pattern is needed for a C/I of 18 dB. The
greater the value of q, the lower the CCI.

3.5 Cellular system design in worst-case scenario

with an omnidirectional antenna
When the mobile unit receives the weakest signal from its own cell site but strong interferences
from interfering cell sites, it is considered to be the worst-case scenario. Let us re-examine the
seven-cell reuse pattern and consider the worst case in which the mobile unit is located at the cell
boundary as shown in Figure 3.4. The distances from the six interfering cells are given in Figure 3.4.
The C/I can be given as

C R− g
= (3.16)
I 2 ( D − R) + 2 D − g + 2 ( D + R)− g
−g

Using D/R = q and the path-loss exponent γ = 4, Equation (3.16) can be written as

C 1 (3.17)
=
I 2 (q − 1)− g + 2q − g + 2 (q + 1)− g

where q = 4.6 for a normal seven-cell reuse pattern (N = 7).

Substituting q = 4.6 in Equation (3.17), we get C/I = 54.3 or 17.3 dB. If we use the shortest
distance (D− R), then
 Elements of Cellular Radio System Design 81

D
D+R

D-R

D+R
D-R
D

Figure 3.4 Worst-case scenarios for co-channel interference

C 1 1
= = = 28 (14.47 dB) (3.18)
I 6 (q − 1)−4 6 (3.6)−4

The C/I could be 14 dB or lower in real situations. This can be due to imperfect cell-site
locations, terrain configuration, and also heavy traffic. The cellular system should, hence, be
designed around the C/I of worst case and for a seven-cell reuse (N = 7) pattern. We conclude
that when considering the worst-case scenario, q = 4.6 is not enough in an omnidirectional cell
system.
Therefore, in an omnidirectional cell system, N = 9 or N = 12 would be a correct choice. Then
the values of q are

⎧ D
⎪ R = 3N
⎪
q = ⎨ 5.2 N =9 (3.19)
⎪6 N = 12
⎪
⎩

By substituting these values in Equation (3.17), we obtain

C (3.20)
= 84.5 or 19.25 dB for N = 9
I

C (3.21)
= 179.33 or 22.54 dB for N = 12
I

The N = 9 and N = 12 cell patterns, shown in Figures 3.5 and 3.6, are used when the traffic
is light. Each cell covers an adequate area with adequate numbers of channels to handle the
traffic.
82 Mobile Cellular Communication

8
7 3 Cell No
2 12
D/R = 5.2
1 6 4 6
5 9 5 9 5
10 8 10 8
4 8 11 2 11
N=9 7 3 3 7 3 7
2 6 12 1 12
1 5 4 6 4 6
8 9 5 9
4 N = 12 10 8 D/R = 6
7 11

Figure 3.5 Interference with frequency Figure 3.6 Interference with frequency
reuse pattern, N = 9 reuse pattern, N = 12

Example problem 3.4

Determine the signal-to-co-channel interference ratio C/I at the mobile receiver located at the
boundary of its omnidirectional operating cell, under the influence of interfering signals from
six co-channel interfering cells in the first tier in a cellular system designed with (a) N = 7,
(b) N = 9, and (c) N = 12.
Assume the path-loss exponent is 4. What would be the correct choice of N for a system
requiring C/I = 18 dB?

Solution

Given data:
Configuration of cellular antenna design is omnidirectional.
Number of co-channel interferers = 6
The path-loss exponent γ = 4

(a) To determine the C/I at the mobile receiver for N = 7

To determine frequency reuse ratio, q
We know that:

q = 3N

For N = 7, q = 21 = 4.6.

To determine the C/I in the worst-case analysis.

In an omnidirectional antenna design of a cellular system based on regular hexagonal pattern,
the worst case C/I, under the influence of interfering signals from six co-channel interfering cells
in the first tier, is given by:

C 1
(Omni) =
2 (q − 1) + 2q − g + 2 (q + 1)
−g −g
I
 Elements of Cellular Radio System Design 83

On substituting q = 4.6 and γ = 4, we get

C 1
(Omni) = = 54
2 (4.6 − 1) + 2 (4.6) + 2 (4.6 + 1)
−4 −g −4
I

C/I (ratio) in decibels can be computed as

We know that C/I (dB) = 10 log10 C/I (ratio)
Or, C/I (dB) = 10 log 54
C/I (dB) = 17 dB

(b) Determine the C/I at the mobile receiver for N = 9

To determine frequency reuse ratio, q

q = 3N = 5.2

To determine the C/I in the worst-case analysis

On substituting q = 5.2 and γ = 4 in the relationship,
that is, on substituting these values in Equation (3.17), we get C/I =95
C/I in dB = 10 log 84.5 = 19.7 dB

(c) Determine the C/I at the mobile receiver for N = 12

To determine frequency reuse ratio, q
q = 3N = 36 = 6

To determine the C/I in the worst-case analysis

On substituting q = 6 and γ = 4 in Equation (3.17), we get C/I =179.3
C/I in decibels = 10 log 84.5 = 22.5 dB
Correct choice of the value of N
Required value of C/I = 18 dB (given)
From the above results, C/I = 17 dB for N = 7, which is lower than the desired value of C/I =
18 dB. So the system design for N = 7 is imperfect.
It is observed that the values of C/I are 19.2 and 22.5 dB for N = 9 and for N = 12, respectively,
which are higher than the desired value of C/I = 18 dB.
Therefore, in an omnidirectional cellular system, N = 9 or N = 12 cell patterns would be a
correct choice.

3.6 Cell splitting

Theoretically, a cellular system can provide services for an unlimited number of users. However,
once a system is installed, it can only provide services to a certain fixed number of users. As soon as
the number of users increases and approaches the maximum that can be served, some technique
must be developed to accommodate the increasing number of users. There are various techniques
to enhance the capacity of a cellular system. One technique is cell splitting, a mechanism by
which cells are split into smaller cells, each having the same number of channels as the original
large cells, as shown in Figure 3.7.
84 Mobile Cellular Communication

A
A

B
C G
AA
D F
E

A
A

Figure 3.7 Seven-cell cluster layout and cell splitting

3.6.1 Types of cellular networks

On the basis of cell radius, there are four types of cellular networks:

• Macrocells
• Microcells
• Picocells
• Nanocells

[Link] Macrocellular radio networks

Macrocellular radio networks are mainly used to cover large areas with low traffic densities.
Macrocells have radii between 1 and 10 km. Large macrocells, with radii between 5 and 10 km
or even higher, are used for rural areas. Small cells, with radii between 1 and 5 km, are used if
the traffic density in large cells is so high so as to cause blocking of calls. They, thus, provide
large cells with extra capacity (cell splitting). Planning small cells is more difficult because traffic
predictions for relatively small areas cannot be easily done, and these signals undergo multipath
Rayleigh fading and lognormal shadowing.

[Link] Microcellular radio networks

Microcellular radio networks are used in areas with high traffic density, like urban areas. The cells
have radii between 200 m and 1 km. For such small cells, it is hard to predict traffic densities and
area coverage because the shape of the cell is time dynamic (i.e., the shape changes from time to
time) due to propagation characteristics. One-dimensional microcells are placed in a chainlike
manner along main highways with high traffic densities, whereas “two-dimensional” refers to the
case where an antenna transmits the main ray and two additional rays are reflected off buildings
on both sides of the street. One-dimensional microcells usually cover one or two house blocks.
In these cells, antennas are placed at street lamp heights; therefore, the surrounding buildings
block the signal propagation to adjacent co-channel cells. This improves the ability to reuse
frequencies, as co-channel interference is reduced drastically by the shadowing effect caused by
the infrastructure. The typical rms delay spread is 2 μs.

[Link] Picocellular radio networks

A picocell is the smallest of the cells in a wireless system, normally covering an office area to support
an in-building wireless system. The term picocell is most commonly used in connection with
 Elements of Cellular Radio System Design 85

third-generation (3G) personal communications systems, especially GSM. Picocells are almost
always placed indoors. Picocells are also known as indoor cells, and they have cell radii between
10 and 200 m. For indoor applications, cells have three-dimensional structures. Fixed cluster sizes,
fixed channel allocations, and prediction of traffic densities are difficult for indoor applications.
Presently, picocellular radio systems are used for wireless communication in offices. The path-loss
exponent varies from 1.2 to 6.8. Signals in picocells are always Rician faded and the Rician parameter
lies between 6.8 and 11 dB. Typical values of rms delay spread lie between 50 and 300 ms.

[Link] Nanocellular radio networks

A nanocell is a miniature, portable radio unit that establishes cell phone service wherever coverage
or capacity is needed. Installation is fast and easy. Nanocells require only a power source and
a means for connecting the base station to a laptop PC, and then the minicell is ready to
operate. Nanocells are the most cost-effective cell nodes available today. The design of a nanocell
combines a small form factor with low power consumption to enable installation in a wide range
of environments. Similar to picocells, a nanocell can be mounted on the wall, in a vehicle, in a
briefcase, or in a weatherproof outdoor enclosure. While there is no limit to the number of users
who can be allowed on a nanocell, up to seven handsets may access the network at one time in
a typical installation. For greater coverage or capacity, multiple distributed nanocell base stations
can be linked together using standard Ethernet cables. Up to four nanocell units can be combined
and configured as a single network managed by one computer. The actual range of a nanocell or
picocell will be determined by the antenna gain, terrestrial obstructions, and building materials,
among other things.

The macrocells, microcells, picocells, and nanocells differ in the radius of coverage provided by
each. This coverage is defined by the power output and antenna gain used in each system.

3.7 Challenges in cell splitting

When a new system is deployed, the user demand is less and users are assumed to be uniformly
distributed over the area to be served. The initial system layout is designed to provide uniformly
reliable coverage and uniform capacity over the entire service area. As the number of users
increases, the demand for channels may begin to exceed the capacity of some base stations. This
increased demand often first shows up in those areas of cities, where the population is dense
during the working day. Once a system has been initially deployed, a system-wide reduction in
the cluster size may not be needed, since user density does not grow at the same rate in all parts
of the system.

Cell splitting is a technique that provides the capability to add smaller cells in specific areas of
the system to support the increased demand in those areas, while minimizing the need to modify
the existing cell parameters.

Cell splitting is based on the reduction of cell radius. It allows the system to grow gradually in
response to an increasing demand of traffic. It takes place by reducing the cell radius by half and
splitting the old cell into four new smaller cells. The reuse frequency can be used more often
allowing the traffic to increase by four times in the same area where an old cell was placed. The
ideal location for new cells is the mid-point between the neighbouring existing cells.
86 Mobile Cellular Communication

Smaller cells lead to increase in number of cells, which in turn requires additional base
stations. The following are two challenges faced while increasing the system capacity by reducing
the cell radius:

• Reconfiguring the system such that the existing base station towers do not have to be moved.
• Meeting an increasing demand that may vary between different geographic areas of a system.

For example, a city may have the highest density of users and, therefore, should be supported
by cells with the smallest radius. The radii of cells in a typical system generally increase as one
moves from urban to rural areas. This is because the user density typically decreases as one moves
away from a city.

The key challenge, therefore, is to add the minimum number of smaller cells wherever increased
demand dictates the need for increased capacity.

A gradual addition of new base stations and smaller cells means that the cellular system may have
to operate with cells of more than one size. When two or more sizes of cells coexist in a mobile
network, special attention needs to be paid to ensure that the minimum frequency reuse distance
is being respected.
The use of channels in the new cell sites will not cause interference problems in the larger
system. This is because the strength of the signal is smaller and is designed to comply with the
constant D/R, but considering the radius of the small cell. The problem arises with the channels of
the co-channel cell in the larger system because the small cell is within the frequency reuse distance
of the larger cell. One way to deal with this problem is by making use of the overlaid cell concept.

3.7.1 Overlaid cell concept

In the overlaid cell concept, the cellular network is seen as a superposition of the smaller cell
pattern on top of the complete larger cell pattern.

Each cell face will divide its channels between a larger and a smaller cell group. The selected
channels in the larger group will be used in all the coverage areas of a larger cell. The selected
channels installed in the smaller cell will make the smaller cell group. The formation and use of
the channel groups is then governed by the presence or absence of real smaller cell neighbours.
For example, the use of any channel installed in a smaller cell must be restricted to the other
smaller cell overlay areas in the nearest larger co-channels. If a mobile user using a smaller cell
channel goes out of the perimeter of the small cell overlay area on the larger cell, it needs to
handoff to a channel of the larger group or to a neighbouring cell. Therefore, the presence of two
sizes of cells in a network reduces the capacity of some of the larger cells. This can force their cells
to split even if they would not consider the growth of traffic in their areas.
Figure 3.8(a) shows a cellular layout with seven cell clusters. Assume that the cells in the
centre of the diagram are congested and cell A in the centre is approaching user capacity.
Figure 3.8(b) shows an overlay of smaller cells superimposed on the original layout. The new
smaller cells have half the radius of the original cells. At half the radius, the new cells will
have one-fourth of the area and consequently will need to support one-fourth the number of
subscribers.
 Elements of Cellular Radio System Design 87

A A
A A

B
C G g b c
A a
f d
D F e
E

A A
A A

(a) (b)

Figure 3.8 (a) Seven-cell cluster layout before splitting and (b) Cell splitting: overlay of
half-radius cells

Notice that one of the new smaller cells lies in the centre of each of the larger cells. If we
assume that base stations are located at the cell centres, this allows the original base stations to be
maintained as the new cell pattern spreads outward from the centre. Of course, new base stations
will have to be added for new cells that do not lie in the centre of the larger cells.
The organization of cells into clusters is independent of the cell radius. Owing to this, the
cluster size can be the same in the small cell layout as it was in the large cell layout. Recall that
the CIR is determined by the cluster size and not by cell radius.
Consequently, if the cluster size is maintained, the CIR will be the same after cell splitting
as it was earlier. If the entire system is replaced with new half-radius cells and the cluster size is
maintained, the number of channels per cell will be exactly as it was before and the number of
subscribers per cell will have been reduced. In large cities, it is common for the cellular systems
to be configured into “microcells” whose radii are measured in hundreds of meters, rather than
in kilometres.

When the cell radius is reduced by a factor, it is also possible and desirable to reduce the power
in transmitted signals.

The minimum required power level is determined by the need to maintain an adequate
signal-to-noise ratio over a significant fraction of the cell area. This in turn requires a minimum
signal-to-noise ratio at the cell radius.
Note that we are concerned with signal-to-noise ratio and not with CIR. The former is
determined by the received power and noise level, whereas the latter depends only on the cluster
size. However, the noise level is determined by the receiver noise figure, a parameter that is
completely independent of the cell radius or layout. Therefore, we can focus on the received
signal power. The received signal power at a distance d from the transmitting antenna is given by

Pr 0 (3.22)
Pr =
(d d0 )
g
88 Mobile Cellular Communication

where Pr0 is the reference power received at some reference distance d0. Pr0 is directly proportional
to the transmitted power and g is the path-loss exponent. The following example illustrates the
idea.

Example problem 3.5

In a certain cellular system, the base stations radiate 15 W. Suppose that the cells are split and
the new cells have one-fourth the radius that the original cells had. Find the power that the
base stations in the new layout must transmit to maintain the signal-to-noise ratio at the cell
boundaries. The path-loss exponent is g = 4.

Solution

Let PrOld be the reference power measured at d0 in the original cell configuration. For g = 4, the
received power at the cell boundary d = ROld is

PrOld
Pr =
(ROld d0 ) 4

When the cells are split, the cell radius becomes RNew = ROld/4. We want to maintain the received
powers at the cell radius at Pr. We have

PrOld PrNew
Pr = =
(ROld d0 )
4
⎛ ROld 4 ⎞
4

⎜⎝ d ⎟⎠
0

where PrNew is the new reference power measured at d0. On rearrangement,

PrNew = PrOld (1 4)
4

The transmitted powers change in the same proportion as the reference powers. Therefore,

PrNew = 15 (1 / 4) = 0.0585 W
4

In decibels, the power reduction is 10log (44) = 24 dB, from

PrOld |dB = 11.7 dB to PrNew |dB = 12.3 dB.

If a cellular layout is replaced by a new layout with a smaller cell radius, the CIR will not change,
provided the cluster size does not change. Special care must be taken, however, to avoid CCI
when both large and small cell radii coexist. Therefore, the only way to avoid interference
between the large cell and small cell systems is to assign entirely different sets of channels to the
two systems.
For example, if a large cell system becomes congested in the city, channels can be taken
away from the large cell system to make up channel sets for the small cell system. The capacity
 Elements of Cellular Radio System Design 89

of the large cell system will be reduced, but the large cell system will now be used primarily in
the suburban areas where the user density is low. The small cell system also does not have a full
complement of channels. But as the cell area is small, there may not be enough users per cell to
demand a full channel set. As the small cell system continues to spread, more and more channels
can be reassigned from the large cell to the small cell system, until, ultimately, the large cell
system is completely replaced.

3.8 Consideration of the components of the cellular system

In the previous sections, the elements of cellular mobile radio system design have been discussed.
Now, we consider the various components of cellular systems, such as mobile radios, antennas,
cell site, base-station controller (BSC), and mobile telephone switching centre (MTSO). Wrong
choice of these components would affect our system design. A general view of the cellular system
is shown in Figure 3.9. The various issues that affect the choice of antennas, switching equipment,
and data links are described in this section.

3.8.1 Antennas
Various issues like antenna pattern, gain, tilting, and height affect the cellular system design. In
both the vertical and the horizontal planes, the antenna pattern can be of any shape and the
transmitted power is compensated by the antenna gain. The antenna patterns in cellular systems
differ from the patterns seen in free space. Antenna tilting can enhance the weak spots in the
cell by reducing the interference to the neighbouring cells. In addition, the area and shape of the
coverage in the system are affected by the height of the cell-site antenna. For more details, the
reader is referred to Chapter 8.

3.8.2 Switching equipment

In cellular systems, the capacity of the switching equipment is based on the capacity of the
processor associated with the switches. The service life of the switching equipment is determined
by the time it takes to reach its full capacity, not by the life cycle of the equipment. More modules
can be added to increase the capacity of the equipment if the switching equipment is designed
in modules, or as distributed switches. Digital switches may be more suitable for decentralized
systems. The future trend seems to be the utilization of system handoff. This means that a call

Cell site 1

Direct
distance
dialing MTSO/BSC Cell site 2
network

Cell site N

Data
Voice

Figure 3.9 A general view of cellular telecommunication systems

90 Mobile Cellular Communication

can be carried from one system to another without the call being dropped by successive linking
of switching equipment.

3.8.3 Data links

The data links are shown in Figure 3.9. Each data link can carry multiple channels of data from
the cell site to the MTSO. This high-speed data transmission cannot be passed through a regular
telephone line; data bank devices are needed for this purpose. They can be multiplexed with
many data channels where the frequency is much higher than 850 MHz.

3.9 Summary
• The frequency reuse concept can be used in both the time and space domains.
• Frequency reuse: Several cells in coverage area use the same set of frequencies.
• Frequency reuse in the time domain takes the form of time division multiplexing. In the space
domain, frequencies can be assigned in two different geographic areas or can be repeatedly
used in the same area in one system.
• The value of C/I is based on the required system performance and the specified value of γ is
based on the terrain environment. Normal cellular practice is to specify C/I to be 18 dB or
higher based on subjective tests.
• CCI is a function of a parameter given by q = D/R, where q is the CCI reduction factor.
• Channel interference areas in a cellular system can be detected in two ways: To find the CCI
area from a mobile receiver and to find the CCI area which affects a cell site.
• Cell splitting allows an array of cells to be replaced with a similar array of cells having a
smaller radius. The cluster size and, therefore, the level of CCI are not changed. The increase
in user density can be dramatic, but additional base stations must be added to the system. It
is important to take advantage of the geometry so that the existing base stations do not have
to be moved when cells are split.
• It is necessary to consider the various components of cellular systems, such as mobile radios,
antennas, etc., while designing a cellular mobile radio system.

Review questions
1. Explain the co-channel interference reduction factor and derive the general formula for C/I.
2. Define co-channel interference and its reduction factor.
3. Derive the co-channel interference reduction factor for seven-cell reuse pattern.
4. Derive the C/I in an omnidirectional antenna system.
5. Derive the C/I in worst-case scenario with an omnidirectional antenna.
6. What are the components in a cellular system? Explain briefly.
7. Define cell splitting. How does cell splitting affect the system design?
8. Why is cell splitting needed? Define 4:1 and 3:1 cell splitting.
9. Explain cell splitting in detail and mention the various advantages of cell splitting concept.
10. Explain overlaid cell concept in detail.
11. Determine the minimum cluster size for a cellular system designed with an acceptable value
of signal-to-co-channel interference ratio C/I = 18 dB. Assume the path-loss exponent as 4
and co-channel interference at the mobile unit from six equidistant co-channel cells in the
first tier (Answer: C/I = 63.1, q = (6 × C/I)1/4 = 4.41, N = 7).
 Elements of Cellular Radio System Design 91

12. In Problem 11, if the acceptable C/I is enhanced to 20 dB, will the cluster size determined
be adequate? If not, then what should be the cluster size? (Answer: C/I for N = 7 is 74.2, C/I
(18.73 dB), N = 7 cannot meet the desired C/I requirement, new frequency ratio q = (6 ×
C/I)1/4 = 4.41, new N for C/I = 20 dB).
13. A cell-site transmitter generates a +15 dBm RF signal and is connected to an antenna using an
RF coaxial cable that induces a 3 dB loss. The cable has two connectors at its either end that
induce a loss of 2 dB each. What is the signal level at the input of the antenna? (Answer: 8 dBm)
14. Explain about the general description of the cellular radio system. (Refer Section 3.2)
15. Draw the general view of telecommunication and explain the function of the each unit. (Refer
Section 3.8)
16. Mention the two frequency reuses schemes and explain N-cell reuse pattern in detail for four-
and seven-cell reuse with illustrative diagrams. (Refer Section 3.2 )
17. Derive the CCI reduction factor for seven-cell reuse pattern. (Refer Section 3.3)
18. Describe about desired C/I from a normal case in an omnidirectional antenna system. (Refer
Section 3.4)
19. What do you mean by desired C/I? Explain. (Refer Section 3.4.1)

Objective type questions and answers

1. The desired performance threshold for proper signal operation in the first-generation AMPS
system is
(a) 4 dB (b) 12 dB (c) 18 dB (d) 24 dB
2. Reuse pattern needed for a C/I of 18 dB is
(a) 7-cell (b) 4-cell (c) 12-cell (d) 2-cell
3. Reuse pattern needed in an omnidirectional antenna in the worst case is
(a) 4-cell (b) 7-cell (c) 15-cell (d) 9-cell
4. The cellular network seen as a superposition of the smaller cell pattern on top of the complete
larger cell pattern is
(a) underlaid-cell concept (b) overlaid-cell concept
(c) small cell superposition (d) none of the above
5. It is possible to reduce the power in transmitted signals if the cell radius is
(a) increased (b) kept constant (c) made zero (d) reduced
6. The concept of repeating the same frequency is known as
(a) cell splitting (b) frequency reuse
(c) co-channel interference (d) none of the above
7. To increase the capacity of equipment while designing switching equipments the modules can be
(a) distributed (b) reduced (c) added (d) altered
8. Which of the following design and concept sets cellular communications apart from all the
preceding mobile radio systems?
(a) TDMA (b) frequency reuse
(c) half-duplex (d) none of the above
9. The cellular technical construct that requires the geographic distance that is required between
the cells using identical frequencies to avoid interference between the radio transmission at
these cells is
(a) D/R ratio (b) call handoff
(c) cell splitting (d) none of the above
92 Mobile Cellular Communication

10. The higher the frequency of a radio signal, the greater is the
(a) absorption rate (b) free space
(c) co-channel interfernce (d) none of the above
11. Co-channel interference describes which of the following:
(a) interference between two base stations transmitting on the same frequency
(b) interference between two mobile phones
(c) inability of mobile to filter out signals of neighbouring cellular channels
(d) none of the above

Answers: 1. (c), 2. (a), 3. (d), 4. (b), 5. (d), 6. (b), 7. (c), 8. (b), 9. (a), 10. (b), 11. (c).

Open book questions

1. How does the type and installation of cell site antenna system play a critical role in
determining the impact of system interference?
2. Why do all cells not have uniform size in a practical cellular network?
3. What is meant by frequency reuse concept and what are the important points to be
considered in frequency reuse concept?
4. Describe the changes that effect in the cellular architecture aspects due to cell splitting.
5. What are the elements of cellular mobile systems?
6. How do you find the differences between the simulation results and practical results in
mobile environment?
7. During a busy hour the number of calls per hour Q i for each of 10 calls is 2,000, 1,500, 3,000,
500, 1,000, 1,200, 1,800, 2,500, 2,800, and 900. Assume that 60 per cent of car phones will
be used during the busy period and that one call is made per phone. Find out the total
number of customers in the system. (Refer Section 3.2.2)
8. Explain the designing of the omnidirectional antenna under the practical case conditions
for k = 7, k = 9, and k = 12 with all the suitable values and explaining each of them.
9. Explain the designing of the directional antenna for K = 7 with all suitable values explaining
each of them, consider a noise margin of 6 dB.

Key equations

1. Carrier-to-interference ratio at the desired receiver.

C C
=
∑
KI
I I
k =1 k

2. Assume that the interference level is very high; the local noise can be neglected. Then, C/I
can be expressed as
C R− g
=
∑ k =1 Dk−g
KI
I
 Elements of Cellular Radio System Design 93

3. Co-channel interference reduction factor

1
⎡ ⎛ C ⎞ ⎤g
q = ⎢6 ⎜ ⎟ ⎥
⎣ ⎝ I ⎠⎦

4. Cluster size of the GSM system

2
1 ⎡ ⎛ C⎞ ⎤g
N = ⎢6 ⎜ ⎟ ⎥
3 ⎣ ⎝ I ⎠⎦

5. The received signal power at a distance d from the transmitting antenna is given by

Pr 0
Pr =
( d d0 )g

Further reading
Mac Donald, V. H. “Advanced Mobile Phone Service,” special issue, Bell System Technical Journal, Vol.
58, pp. 12–41, January 1979.
A. Mehrotra, Cellular Radio Analog and Digital System. Boston: Artech House, 1994.
AT&T Technical Center, “Cellular System Design and Performance Engineering I,” CC 1400,
version 1.12, 1993.
V. H. MacDonald, “The Cellular Concept,” Bell System Technical Journal, Vol. 58, January 1979,
pp. 15–42.
Lee W. C. Y., Lee’s Essentials of Wireless Communications. New York: McGraw-Hill , 2001.