Coverage and Capacity Calculations for 3G Mobile Network Planning

Liang Guo, Jie Zhang and Carsten Maple

Department of Computing and Information Systems
University of Luton, Luton, LU1 3JU, U.K.
E-mail: {liang.guo, jie.zhang, carsten.maple}@luton.ac.uk

Abstract Due to the high costs and the scarcity of radio

Coverage and capacity are important issues in the resources, an accurate and efficient mobile network
planning process for cellular Third Generation (3G) planning procedure is required. The objective of network
mobile networks. The planning process aims to allow the planning is to maximise the (usually conflicting goals of)
maximum number of users sending and receiving coverage, capacity and the quality of service.
adequate signal strength in a cell. In this paper we
present calculations for capacity, coverage for 3G The structure of the paper is organised as follows:
networks. Section 2 describes the background of planning for radio
networks, concentrating on second generation (2G)
Key Words: 3G, CDMA, planning, coverage, capacity.
technologies. The issues for 3G network planning are
covered in Section 3. Relationships between coverage and
capacity are given before the concluding comments of
Section 4.
1. INTRODUCTION
2. BACKGROUND
Third Generation (3G) radio networks are based on the
code division multiple access technology (CDMA) and are
The planning of 3G is very different from the Second
currently being installed in countries such as Japan and
2G which adopts a time division multiple access
South Korea. The aim of the technology is to fulfil the
technology (TDMA). TDMA works by dividing a radio
user requirement for innovative services such as enhanced
frequency into time slots and then allocating slots to
and multimedia messaging through high-speed data
multiple calls. In this way, a single frequency can support
channels. CDMA is a digital cellular technology that uses
multiple, simultaneous data channels.
spread-spectrum techniques. It does not assign a specific
frequency to each user. Rather, every channel uses the
2G planning can be divided into a coverage planning
full available spectrum. Individual conversations are
phase and a frequency planning phase which are driven
encoded with a pseudo-random digital sequence. Third
by coverage and capacity criteria respectively. In the
Generation networks may also be referred to as Universal
coverage planning phase, base stations (BSs) are placed
Mobile Communication Telecommunication System
so that the signal strength is high enough in the area to be
(UMTS), see [1], for example. There are three standards
served. In the frequency-planning phase, each BS is
accepted by International Telecommunication Union
assigned a number of channels, then network operator
(ITU). They are Wide-band CDMA (WCDMA),
takes into account the traffic load, the level of services
CDMA2000 and time-division synchronous CDMA (TD-
measures as the signal-to-interference ratio (SIR) as well.
SCDMA).

ISBN: 1-9025-6009-4 © 2003 PGNet

 However, in the 3G planning, since all carriers in the It is well known that the coverage of the cell has an
network use the same frequency range, frequency inverse relationship with the user capacity of the same
planning is not required. Furthermore, coverage and cell. An increase in the number of active users in the cell
capacity planning should be performed in tandem since causes the total interference seen at the receiver to
capacity requirement and traffic distribution influence the increase. This causes an increase in the power required to
coverage. be received from each user. This is due to the fact that
each user has to maintain a certain Signal-to-Interference
Although there are three distinct standards in 3G Ratio (SIR) at the receiver for satisfactory performance.
networks (each used in different parts of the world), For a given maximum allowable transmission power, an
WCDMA, CDMA2000 and TD-SCDMA, the general increase in the required power reception will result in a
planning process and overall objectives are the same. decrease in the maximum distance a mobile can be from
the base station thereby reducing the coverage.
3. 3G RADIO NWETWORK PLANNING
We will consider the up-link and the ideal situation of
The 3G planning process can be divided into three the CDMA technology. That is to say, we will assume
parts; the initial phase (also called system dimension); the coverage is limited by the maximum transmit power at
detailed planning phase; and the optimisation and the mobile and no blocking and outage take place in the
monitoring phase. Each of the phases requires additional cellular system since CDMA technology can provide the
considerations such as propagation measurements, traffic enough codes for the new call to the cell in the ideal
demand measurements and so on. The process of the 3G situation.
planning is illustrated in Figure1.
3.1 Initial Model for Capacity Calculation
Initial Phase
- Traffic We first attempt to calculate the maximum number of
- Area users in a particular cell. This is based on a static
- Coverage threshold capacity calculation and the following assumptions are
made:

1. Each cell is completely isolated. That is, there is no

inter-cell interference.
2. Signals from mobile stations cause no interference
Detailed Planning within the cell. That is, there is no intra-cell
-Traffic
interference.
-Configuration
3. There is perfect power control from the base station.
-Coverage threshold & capacity
That is, all signals arrive at the base station with
requirements
equal power.
4. There is no limit to the number of spreading codes
available.

The actual capacity of a CDMA cell depends on many

Optimisation & Monitoring different factors, such as power control accuracy,
-Coverage verification interference power. We begin by calculating the signal-
-Capacity availability to-interference ratio. If there are N users in a cell and the
signal is denoted by S then the interference can be
calculated as I = (N - 1) S + η, where η is the thermal
noise. Hence the SIR is given by
Figure1. 3G system planning process
S 1
SIR = = (1)
We see that the whole process of 3G planning needs to ( N − 1) S + η ( N − 1) + η / S
take into account coverage and capacity planning. In a
cellular system where all the air interface connections
operate on the same carrier, the number of simultaneous
The value of the interference is based upon the four
users is directly influences the receivers’ noise floors.
assumptions above and is not realistic. For this work we
only use this value since other factors are small in our W /R
situation. Other interference factors are known as inter- ( N ε − 1 )= Gv G s ( -η / S ) (4)
cell and intra-cell interference. The inter-cell interference
E b/ Io
is attributed to mobile stations and base stations in one
cell affecting the operations in another. This is results in Using speech statistics we can set Gv =2.63 and based
unwanted noise. Intra-cell interference can be attributed
to the effect one mobile station has on the transmitted upon the usual three sectors for a cell, G s =2.63.
signal of another within the same cell. It is s simple step W /R
to change the interference value to consider the other two is typically a value between 20 to 100.
factors though in this paper it would only serve to
E b/ Io
complicate the formulae.
Finally, for a cellular system in which all users in all
Suppose the digital demodulator for each users can cells employ the common spectral allocation of W Hz, we
operate against the noise at an energy per bit-to noise must evaluate the interference. Normally the total
interference equal the amount of interference from other
power density level is given by E b / I o , where
cells and interference from given cells.
Eb = S / R and I o = I / W .
We define:

Eb S / R W f=(interference from other cells)/(interference from

= = SIR (2) given cell)
Io I /W R
Due to the interference, the actual numbers of user will
decrease from ( Nε − 1) to ( N φ − 1) (Ref.[7]), the
Here W is the chip rate and R is rate of data
communication. equation (4) will be modified into (4.1)

Hence using (1) and (2) we can obtain W /R 1

N φ −1 = Gv G s ( -η / S ) (5)
E b/ Io 1+ f
W /R
( N − 1) = - η / S (3)
Eb / I o Normally, the total interference from users in all other
cells equals approximately three-fifths of that caused by
In particular, if the user is not speaking during part of all users in the given cell. (See [9], for example). That is
the conversation, the output of the coder is lowered to to say, 1 + f ≈ 1.6
prevent the power from being transmitted unnecessarily.
For a uniform population, this reduces the average signal However we note that equation (5) gives a poor
power of all users and consequently the interference estimate to N φ due to the many assumptions in the
received by each user. The capacity is then increased
proportional to this overall rate reduction, provide. Speech model. It does not reflect the relationship between the
statistics shows that a user in a conversation typically capacity and the coverage.
speaks close to 38% of the time. the effects of voice
activity and sectorization should be considered. This 3.2 Coverage Versus Capacity
Eb
results in an increase in the by a voice activity gain The analysis in the above we isolate the capacity
Io calculation from coverage. However, in the 3G networks,
it is important to derive the relationship between the
factor, Gv . Similarly, the sectionsation gain factor G s
coverage and capacity.
Eb
also increases the and the equation for the effective We will consider the up-link and the ideal situation of
Io the CDMA technology. That is to say, we will assume the
number of users, Nε, becomes coverage is limited by the maximum transmission power
at the mobile and no blocking nor outage takes place in
the cellular system (since CDMA technology can provide
the enough codes for the new call to the cell in the ideal
situation). P (d ) = A + B log10 f − 13.82 log10 hBTS −
Based on (5), we can understand the performance of a a (hms ) + (44.9 − 6.55 log10 hbts ) log10 d
CDMA network by developing a simple expression for + Cm
the ratio Eb / Io, as follows.
Where A, B are constants defined by the frequency and
S
f is the frequency. a (hms ) is a function that depends
Eb R
= (6) upon the height of the mobile station that is particular to
Io ( N φ − 1) S (1 + f ) / GvGs + η the environment. Cm is the area type correction factor.
W
For simply the questions, the propagation loss and the
Solving the equation (6) for S gives: distance relationship can be expressed as

Eb P (d ) = p1 + p 2 log10 d (9)
Rη
Io
S= (7) We can now build a relationship among the received
E power, number of users, the coverage area, the gain of
W − ( b R( N φ − 1)(1 + f )) / GvGs
No voice and gain of sectionsation based on equations (7),
(8), (9).
This equation can be used to demonstrate the
dependency of S on the user data rates, R, the voice and S = S1 − ( p1 + p2 log10 d ) − z
section gain Gv , G s , respectively, and the total number Eb
Rη
of active users in a cell, N φ . Io
= (10)
E
W − ( b R( N φ − 1)(1 + f )) / GvGs
We focus on the coverage by user1 when the number No
of users in the cell is N φ . Let r be the distance of user1
from the base station. The received power at the base
station from mobile user1, S, is given by 4. CONCLUSION AND FUTURE WORK

The efficient network planning is a vital aspect of 3G

S = S 1− P ( d ) − Z (8) networks. Key differences arise between 2G and 3G
networks due to the different levels of service offered. By
Where identifying and analysing the relationships between
capacity, coverage and quality of service.
S1 : The transmission power of the user.
The equations derived in this work can be used in
P(d ) : The propagation loss at distance d from the MS
cellular system planning to set hard limits on the
to BS. maximum number of users that can be admitted into the
Z : The shadow fading cell. For example, the pre-specified coverage
requirements are met. It is also useful in planning a
The propagation loss is usually calculated by using cellular network to design cell coverage and capacities to
propagation models. The propagation model that is most match specified data services, such as 9.6Kb/s, 144Kb/s
commonly used is the Okumura-Hata. The model is etc.
developed by combining propagation theory and Capacity issues for CDMA cellular systems play a vital
extensive measurement campaigns. The model considers role in network planning. We see that in CDMA systems,
several parameters such as effective antenna height, coverage and capacity are heavily coupled and can not be
terrain type, terrain height, frequency, and so on. For planned separately such as they could in 2G systems.
details of the Okumura-Hata model the reader is referred
to [2]. The model predicts
 The relationship given in (10) describing the [2] Jaana L., Achim W., Tomas N., 2002, “Radio
relationship between coverage, capacity and data rates is Network Planning and Optimisation for UMTS”, Wiley
very useful in planning the networks.
[3] Wacker A., Laiho-Steffens J., 1999, “Static Simulator
However, there are shortcomings in the analysis and for Studying WCDMA Radio Network Planning Issues”,
modelling provided herein. The modelling is largely Proc.IEEEVTC SpringConf.
based upon an ideal situation, wherein no call will be [4] Hoppe R., Buddendick G., 2001, “Dynamic Simulator
blocked. While CDMA technology can itself provide for Studying WCDMA for Radio Network Performance”,
enough codes for the mobile terminals to assure the ideal Proc. IEEE International Symposium on Personal, Indoor
situation (enough for every user, therefore there is no and Mobile Radio Communications
blocked call) these are not used in practice. In application [5] Jukka L., Matti M., 2001, “Radio Interference System
the network operators take into account the cost of Planning for GSM/GPRS/UMTS”, Kluwer Academic
equipment (which is related to the number of codes used) Publisher
and the network load. They use Erlang B to assure certain
grade of service. [6] Venugopal V., 1999, “The Coverage-Capacity
Tradeoff in Cellular CDMA Systems”, IEEE
To overcome the shortcomings of the modelling Transactions on Vehicular Technology. Vol.48, No.5
presented herein, in future work we would need to [7] Gilhousen K. S., 1991, “On the Capacity of a Cellular
consider the outage probability and the loading control in CDMA System”, IEEE Transactions on Vehicular
the planning process. An analysis considering outage Technology. Vol.40, No.2
probability and cell loading will give a more accurate [8] Leung K. K., Mastey W. A., 1994,“ Traffic Modelling
relationship between the coverage and capacity. for Wireless Communication Networks” IEEE Journal on
selected area in communication, Vol.12, No.8
REFERENCES
[9] Kyoung K., 2000, “Handbook of CDMA System
Design, Engineering, and Optimization” Prentice Hall
[1] Harri H., Antti T., 2000, “WCDMA for UMTS Radio
Access for Third Generation Mobile Communication”,
Wiley.