(IJACSA) International Journal of Advanced Computer Science and Applications,

Vol. 2, No. 6, 2011

Performance Analysis of UMTS Cellular Network

using Sectorization Based on Capacity and Coverage
A.K.M Fazlul Haque, Mir Mohammad Abu Kyum,
Md. Baitul Al Sadi, Mrinal Kar
Department of Electronics and Telecommunication
Engineering.
Daffodil International University

AbstractUniversal Mobile Telecommunications System (UMTS)

is one of the standards in 3rd generation partnership project
(3GPP). Different data rates are offered by UMTS for voice,
video conference and other services. This paper presents the
performance of UMTS cellular network using sectorization for
capacity and coverage. The major contribution is to see the
impact of sectorization on capacity and cell coverage in 3G
UMTS cellular network. Coverage and capacity are vitally
important issues in UMTS cellular Network. Capacity depends
on different parameters such as sectorization, energy per bit
noise spectral density ratio, voice activity, inter-cell interference
and intra-cell interference, soft handoff gain factor, etc and
coverage depends on frequency, chip rate, bit rate, mobile
maximum power, MS Antenna Gain, EIRP, interference Margin,
Noise figure etc. Different parameters that influence the capacity
and coverage of UMTS cellular network are simulated using
MATLAB 2009a. In this paper, the outputs of simulation for
increasing amount of sectorization showed that the number of
users gradually increased. The coverage area also gradually
increased.
Keywords-UMTS; Capacit; Coverage and data rates; sectoring.

I.

INTRODUCTION

A cellular cell can be divided into number of geographic

areas, called sectors. It may be 3 sectors, 4 sectors, 6 sectors
etc. When sectorization is done in a cell, interference is
significantly reduced resulting in better performance for
cellular network. Capacity in WCDMA standards of UMTS
refers to maximum number of users per cell, where the area
covered by RF signal from Node B or UE (User Equipment) is
called coverage area of UMTS. Capacity and coverage are two
dynamic phenomena in UMTS network. Parameters that define
capacity and coverage of UMTS are dynamic in nature, where
increasing or decreasing values of these parameters affects
capacity and coverage of UMTS cellular network. One of the
parameters is sectorization in UMTS. There are some works on
sectorization scheme [1-4].
Bo Hagerman, Davide Imbeni and Jozsef Barta considered
WCDMA 6-sector deployment case study of a real installed
UMTS-FDD network [1]. Romeo Giuliano, Franco Mazzenga,
Francesco Vatalaro described Adaptive Cell Sectorization for
UMTS Third Generation CDMA Systems [2]. Achim Wacker,
Jaana Laiho-Steffens, Kari Sipila, and Kari Heiska considered
the impact of the base station sectorisation on WCDMA radio

Md. Fokhray Hossain

Department of Computer Science and Engineering.
Daffodil International University

network performance [3]. S. Sharma, A.G. Spilling and A.R.

Nix considered Adaptive Coverage for UMTS Macro cells
based on Situation Awareness [4]. Most of the works analyzed
the performance considering sectors with static parameters but
it is needed to analyze the performance along with all dynamic
parameters.
This paper optimizes the performance of both capacity and
coverage of UMTS not only considering sectors but also with
dynamic parameters as energy per bit noise spectral density
ratio, voice activity, inter-cell interference, soft handoff factor,
and data rates.
II.

BACKGROUND

A. Capacity in WCDMA for UMTS:

As the downlink capacity of UMTS is related to transmit
power of Node B and uplink capacity is related to numbers of
users, uplink capacity is considered in this paper.
If the number of users is Ns then for a single CDMA cell,
the number of users will be [5],

W/R 1
)
Eb / N o S
Where, Ns=total number of users, W=chip rate,

N s 1 (

(i)

R= base band information bit rate, Eb/No=Energy per bit to

noise power spectral density ratio, = background thermal
noise, S=signal power=S1-P (d)-shadow fading, S1=UE power,
P (d) =Propagation loss.
For WCDMA, the chip rate is 3.84 Mcps [8], and the
channel bandwidth is 5 MHz [8]. It is also necessary to
consider the affects of multiple cells or intra-cell interference
()[12], cell sectoring(D)[6], soft handover factor(H)[11],
Array antenna gain (Ag)[10].Thus the capacity for WCDMA in
UMTS yields:
Ns 1 (

W /R

1
)
D H Ag (ii)
Eb / N o S (1 )

B. Coverage and data rates in WCDMA for UMTS:

UMTS offered different data rates for multi services. Table
1 shows different standard bit rates offered by UMTS. Higher
class of service makes cell radius small resulting in small

98 | P a g e
www.ijacsa.thesai.org

(IJACSA) International Journal of Advanced Computer Science and Applications,

Vol. 2, No. 6, 2011

coverage area. If different class of services is classified in terms

of coverage area, it will look like figure 1.
TABLE 1: DIFFERENT CLASSES OF SERVICES
Bit Rate(Kbps)
Class
12.2

Class 5

32

Class 4

64

Class 3

144

Class 2

384

Class 1

Figure 1 shows that for service class 1, maximum distance

is observed by UE (User Equipment) from Node B. Similarly
for service class 2 and service class 3, UEs maintain maximum
distance from Node B. From this figure it is clear that different
coverage areas are needed to maintain different data rates. So
coverage area needs to increase for better class of services.
This paper optimizes the coverage area for particular services
with sectors.
Figure 2 shows a UMTS cell where Node B received power
(PR) from User Equipment (UE). The Node B sensitivity is the
power level for minimum signal necessary at the input of the
Node B receiver to meet requirements in terms of Eb/No,
processing gain (Gp) and Node B interference and noise power
given as [8]
Node B sensitivity=Eb/No-Gp+NNode B interference and noise power
Where Gp=Processing gain
Chiprate
3.84Mcps
10 log
10 log

Bitrate
R

Figure 2: UMTS cell

Now the maximum allowable path loss for Node B,

LP =EIRP-Node b sensitivity + Gp - fast fading margin----(iii)
from radio propagation model, Path loss for dense urban area
[5],
L 46.3 33.9 log( f c ) 13.82 log hb 3.2
[log(11.75h UE )] 2 4.97 (44.9 6.55
log hNodeB ) log d 3
---------------------------------------------------------------------- (iv)
From equation (iii) and (iv) a relationship can be expressed
for coverage and data rates in dense urban case,

46.3 33.9 log( f c ) 13.82 log hb 3.2

[log(11.75h UE )] 2 4.97 (44.9 6.55
log h NodeB ) log d 3 EIRP - NodeB sensitivity
Chip rate
10 log
- Fast fading Margin
R

where d is the coverage radius and R is the data rates.

After calculating the cell range d, the coverage area can be

calculated. The coverage area for one cell in hexagonal
configuration can be estimated with [9]
Coverage area, S =K.d2
where S is the coverage area, d is the maximum cell range,
and K is a constant. In Table 2, some of the K values are listed.
TABLE 2: K VALUES FOR THE SITE AREA CALCULATION [9]:
Site
Omni
Two
Three
Four
configuration
or
no sectors
sectors
sectors
sector
Value of K
2.6
1.3
1.95
2.6
Figure 1: Different Classes of Services vs. Maximum Distance

99 | P a g e
www.ijacsa.thesai.org

(IJACSA) International Journal of Advanced Computer Science and Applications,

Vol. 2, No. 6, 2011

III.

SIMULATIONS AND RESULTS

The analysis has been done for capacity and coverage with
sectoring cell for dense urban using MATLAB R2009a. The
simulated values for sectorization are shown in Table 3, Table
4, Table 5, Table 6 and Table 7. The performances are also
described in Figure 3, Figure 4, Figure 5, Figure 6, and Figure
7. The algorithms of the evaluation process have also been
introduced in appendix.
Figure 3 shows that Energy per bit to noise spectral density
ratio (Eb/No) needs to maintain small value for increasing
number of simultaneous 384 Kbps users. From this figure it is
observed that for dynamic value of Eb/No with changing the
sectors, the number of simultaneous 384 Kbps data users
increased or decreased. For example, if Eb/No value is 4 db,
then for 6 sectors, the number of simultaneous users will be 88
but for 3 sectors, the number of simultaneous users will be 45.
Thus the dynamic values of Eb/No can be the increasing or
decreasing factors in UMTS and sectorization scheme can be
effective in this case.

16

4.629

8.2581

11.887

15.516

22.774

18

4.4156

7.4516

10.677

13.903

20.355

20

3.9032

6.8065

9.7097

12.613

18.419

The interference from other cell is known as inter-cell

interference (). For multi-cell configuration, the number of
outer cells can reduce cell capacity in UMTS. Figure 4 shows,
for increasing demand of users the value of in UMTS needs
to be small. Figure 4 also represents dynamic inter-cell
interference with changing of sectors, where number of
simultaneous 384 Kbps data users increases or decreases. From
Figure 4 it has been observed that for increasing value of , it is
needed to increase sectors.

Figure 4: Number of simultaneous 384 Kbps users vs.inter-cell interference in

sectors cell

Figure 3: Number of simultaneous 384 Kbps users vs. Eb/No in sectors cell
TABLE 3: SIMULATED VALUES FOR NUMBER OF SIMULTANEOUS
384 KBPS USERS VS. EB/NO IN SECTORS CELL
Energy per

Users

Users

Users

Users

Users

bit to Noise

without

with 2

with 3

with 4

with 6

spectral

sector

sectors

sectors

sectors

sectors

59.065

117.13

175.19

233.26

349.39

15.516

30.032

44.548

59.065

88.097

density
ratio(Eb/No)

8.2581

15.516

22.774

30.032

44.548

10

6.8065

12.613

18.419

24.226

35.839

14

5.1475

9.2949

13.442

17.59

25.885

The overlapped cell can lead an extra power thus

introducing soft handover factor (H) in a UMTS cell. The value
of H in UMTS can be a factor to increase the number of users.
Figure 5 shows that for increasing H and changing value of
sectorization the number of simultaneous 384 Kbps data users
increases.
For example, if H value is 2.5 db, then for 2 sectors the
number of simultaneous users will be 195 but for 4 sectors the
number of simultaneous users will be 388.
Figure 6 shows that the number of voice users depends on
the value voice activity factors ().This is true only for 12.2
Kbps voice users, not for data users, as for data services it will
always be 1.
Figure 3 also shows that, for increasing amount voice users
the value of in UMTS needs to as small as possible. Varying
and changing the sectors the number of simultaneous voice
users from figure 6 is observed.

100 | P a g e
www.ijacsa.thesai.org

(IJACSA) International Journal of Advanced Computer Science and Applications,

Vol. 2, No. 6, 2011
TABLE 4: SIMULATED VALUES FOR NUMBER OF SIMULTANEOUS
USERS VS. INTER-CELL INTERFERENCE IN SECTORS CELL
InterUsers
Users
Users
Users
Users
cell

without

interfer

sector

ence

with

with

with

with
6 sectors

sectors

sectors

sectors

0.1

135.33

419.6

730.73

1201

5401

0.5

27.866

84.721

146.95

241

1081

14.433

42.86

73.973

121

541

1.5

9.9552

28.907

49.649

81

361

1.7

8.9017

25.624

43.925

71.588

53.941

2.0

7.7164

21.93

37.486

61

271

TABLE 5: SIMULATED VALUES FOR NUMBER OF SIMULTANEOUS

USERS VS. SOFT HANDOVER FACTOR IN SECTORS CELL
Soft
Users
Users
Users
Users
Users
Hand

without

with

with

with

with

-over

sector

2 sectors

3 sectors

4 sectors

6 sectors

Factor

Figure 6: Number of simultaneous voice users vs. voice activity factor in

sectors cell.
TABLE 6: SIMULATED VALUES FOR NUMBER OF SIMULTANEOUS
USERS VS. VOICE ACTIVITY FACTOR IN SECTORS CELL
Voice

Users

Users

Users

Users

Users

activity

without

with

with

with

with

factor

sector

2 sectors

3 sectors

4 sectors

6 sectors

0.2

228.31

455.63

682.94

910.26

1364.9

0.4

114.66

228.31

341.97

455.63

682.94

0.1

4.871

8.7419

12.613

16.484

24.226

0.4

16.484

31.968

47.452

62.935

93.903

39.71

78.419

117.13

155.84

233.26

1.5

59.065

117.13

175.19

233.26

349.39

2.5

97.774

194.55

291.32

388.1

581.65

0.6

76.771

152.54

228.31

304.09

455.63

117.13

233.26

349.39

465.52

697.77

0.8

57.828

114.66

171.49

228.31

341.97

46.463

91.926

137.39

182.85

273.78

Figure 5: Number of simultaneous 384 Kbps users vs. soft handover factor in
sectors cell

101 | P a g e
www.ijacsa.thesai.org

(IJACSA) International Journal of Advanced Computer Science and Applications,

Vol. 2, No. 6, 2011
TABLE 7: SIMULATED VALUES FOR COVERAGE VS. DATA RATES
IN DENSE URBAN USING COST 231 MODEL IN SECTORS CELL
Data
rate
(Kbps)

Cell
range
in
(meter)
773.67

Cell
Area
without
sector
(meter2)
598.57

Cell
Area
with 2
sectors
(meter2)
778.14

Cell
Area
with 3
sectors
(meter2)
1167.2

Cell
Area
with 4
sectors
(meter2)
1556.3

200
400

635.43

403.77

524.9

78.7.34

1049.8

600

566.31

320.71

416.92

625.38

833.84

800

521.88

272.36

354.07

531.1

708.14

1000

489.83

239.94

311.92

467.88

623.84

1200

465.12

216.33

281.23

421.85

562.47

1400

445.19

198.2

257.66

386.49

515.31

1600

428.63

183.72

238.84

358.26

477.67

1800

414.53

171.83

223.38

335.07

446.76

2000

402.31

161.85

210.41

315.61

420.81

Finally, consider for coverage vs. data rates in dense urban

area, where operating frequency is considered 2000 MHz with
COST 231 Model as a radio propagation model. In Figure 7 the
x axis represents data rate in Kbps and y axis represents
coverage area in meter2 with cell radius in meter. Parameters
that are related to coverage setting first, then for increasing data
rates in x axis, the coverage area is observed in y axis. From
figure 1, it is known that for higher data rates, the coverage will
be smaller. It is true only when cell area is considered without
sectors. This phenomenon is revealed by figure 7. Figure 7 also
shows, for higher data rates comprehensive coverage area is
found with increasing sectors.
IV.

CONCLUSION

In this paper, the performance analyses in coverage and

capacity of UMTS cellular network using sectorization have
been simulated and evaluated for dynamic parameters. The
number of simultaneous users increases or decreases for
increasing or decreasing sectors with dynamic parameters.
Coverage has been estimated for dense urban using COST 231
model where higher data rates need higher processing gain
resulting in smaller coverage area. But increasing sectors with
same parameters makes extensive coverage for higher data
rates.
REFERENCES
Bo Hagerman, Davide Imbeni and Jozsef Barta WCDMA 6 - sector
Deployment-Case Study of a Real Installed UMTS-FDD Network
IEEE Vehicular Technology Conference, spring 2006, page(s): 703 707.
[2] S. Sharma, A.G. Spilling and A.R. Nix Adaptive Coverage for UMTS
Macro cellsbased on Situation Awareness. IEEE Vehicular Technology
Conference, spring 2001, page(s):2786 - 2790
[3]
A. Wacker, J. Laiho-Steffens, K. Sipila, K. Heiska, "The impact of
the base station sectorisation on WCDMA radio network performance",
IEEE Vehicular Technology Conference ,September 1999,page(s): 2611
- 2615 vol.5.
[4] Romeo Giuliano, Franco Mazzenga, Francesco Vatalaro, Adaptive cell
sectorization for UMTS third generation CDMA systems IEEE
Vehicular Technology Conference, May 2001, page(s): 219 - 223 vol.1.
[5]
T.S.Rappaport, Wireless Communications Principles and
Practice-Second Edition, Prentice Hall.
[6] Bernard Sklar, Digital Communications - Fundamentals and
Applications- Second Edition Prentice Hall.
[7] INGO FO, MARC SCHINNENBURG, BIANCA WOUTERS PERFORMANCE
EVALUATION OF SOFT HANDOVER IN A REALISTIC UMTS NETWORK
IEEE VEHICULAR TECHNOLOGY CONFERENCE, SPRING 2003, PAGE(S):
1979 - 1983 VOL.3.
[8] HARRI HOLMA AND ANTTI TOSKALA, WCDMA FOR UMTS-RADIO
ACCESS THIRD GENERATION MOBILE COMMUNICATIONS- JOHN WILEY
& SONS.
[9] Jaana Laiho, Achim Wacker, Tomas Novosad, Radio Network Planning
and Optimisation for UMTS-Second Edition John Wiley & Sons.
[10] A. Skopljak, Multiantenna CDMA systems and their usage in 3G
network, University of Sarajevo, 2007.
[11] Faruque, Saleh,
Cellular Mobile System Engineering,
Artech
House Publishers, 1996.
[12] Jhong Sam Lee, Leonard E. Miller, CDMA Systems Engineering
Handbook-Artech House Publishers.
[1]

Figure 7: Coverage vs. bit rates for dense urban using COST 231 model in
sectors cell

102 | P a g e
www.ijacsa.thesai.org

(IJACSA) International Journal of Advanced Computer Science and Applications,

Vol. 2, No. 6, 2011

Appendix
Algorithm for Capacity Analysis Using Sectorization:
Begin
Set energy per bit to noise spectral density ratio (Eb/No) = [1 2 3 4 5 6 7 8 9 10 16 20]
Set soft handover gain (H) factor = [0.1 1 1.5 2 3]
Set inter-cell interference () = [.1 1.2 1.55 2]
Set channel activity for data () = [1]
Set channel activity for voice () = [0.1 .3 .38 0.7 0.9]
Set thermal noise () in (20 Kelvin) dbm/Hz= [-173.93]
Set user signal power (S1) in dbm= [21]
Set shadow fading (sh_fd) in db = [8]
Set cell range in Km (Rcell) = [2]
Set chip rate (W) = [3840000]
Set base band information rate in Kbps (R) = [12.2 64 144 384 2000]
Set base antenna height in meter (hb) = [20]
Set user antenna height in meter (hUE) = [2]
Set sector (D) = [1 2 3 4 6]
Set frequency range in MHz (fc) = [2000]
Set data rate in Kbps (R) = [12.2 64 144 384 2000]
Set array antenna gain (Ag) in db= [1 2 3.5 5]
//Processing
Processing gain (PG) = 10log (W/R)
Propagation loss in dense urban (Pro_loss) =
46.3 33.9 log( f c ) 13.82 log hb 3.2[log(11.75UE )] 2 4.97 (44.9 6.55 log hb ) log d 3
Signal Power (S) = S1-Pro_loss-sh_hd
//Output
W /R
1
Number of Users N s 1 (
)
D H Ag
Eb / N o S (1 )
End
Algorithm for Coverage and Data rates Analysis Using Sectorization:
Begin
Set Transmitter=User Equipment
Set Receiver=Node B
Set mobile max power in dbm (mo_mx) = [21]
Set mobile gain in db (M_G) = [0]
Set cable and connector losses in db (ca_cn_loss) = [3]
Set thermal noise in dbm/Hz () = [-173.93]
Set node B noise figure in db (nodeB_NF) = [5]
Set target load (tar_ld) = [.4]
Set chip rate (W) = [3840000]
Set base antenna height in meter (hb) = [20]
Set user antenna height in meter (hUE) = [2]
Set energy per bit to noise spectral density ratio (Eb/No) = [5]
Set Power Control Margin or Fading Margin (MPC) = [4]
Set Value for sectors (Sec) = [1 2 3 4]
Set constant value for sectors (K) = [2.6 1.6 1.95 2.6]
Set data rate in Kbps (R) = [100 200 300 400 500 600 2000]
//Processing
Chip rate in (W) = [3840000]
Processing gain (PG) = (W/R)
Effective isotropic radiated power (EIRP) = mo_mx-ca_cn_loss+M_G
Node B noise density (nodeB_ND) = + nodeB_NF
Node B noise power (nodeB_NPW) = nodeB_ND +W_db
Interference margin (IM) =-10log (1-tar_ld)

103 | P a g e
www.ijacsa.thesai.org

(IJACSA) International Journal of Advanced Computer Science and Applications,

Vol. 2, No. 6, 2011

Node B Interference Power (nodeB_IP) = 10 log 10 (noisepower InterfrenceM arg in) / 10 10 Noisepower/ 10

10
Node B Noise and interference (nodeB_NIFPW) = 10 log 10
Node B antenna gain (NodeB_AG) in db = [18]
Receiver Sensitivity (Srx) = Eb/No-PG+ nodeB_NIFPW
Total Allowable Path loss=EIRP-Srx+ nodeB_AG-MPC=EIRP-(Eb/No-PG+ nodeB_NIFPW) +
NodeB_AG-MPC
Path loss in dense urban (Durban_Ploss) =
46.3 33.9 log( f c ) 13.82 log hb 3.2[log(11.75UE )] 2 4.97 (44.9 6.55 log hb ) log d 3
( noisepower) / 10

( Interfrencepower) / 10

142.17 36.37logd

//Output
Cell radius (d) = 10^ ((1/36.37)* (EIRP-(Eb/No-PG+ nodeB_NIFPW) + nodeB_AG-MPC-142.17))
Cell Area (A) = K*d2
End

104 | P a g e
www.ijacsa.thesai.org