MEASUREMENTS AND EVALUATION OF RADIO FREQUENCY

RADIATION AT CELLULAR BASE STATIONS

NG' ANG' A FRANCIS MWAURA

REG. NO. 156//7707/2002

1\ Thesis submitted in partial fulfillment of the requirements for the award of the degree
of Masters of Science in Electronics and Instrumentation in the School of Pure and
Applied Sciences of Kenyatta University

April 2015
 DECLARATION

1 declare that this thesis is my original work and has not been submitted or published

for any award of a degree in this or any other University.

Sign ------~----- Date ~ ?2._~_:::.f?.!f:_-.::.J5_~L?= _

Ng'ang'a Francis Mwaura

156/17707/2002

Recommendation by supervisors
•
We confirm that the work reported in this thesis has been submitted for review with our

approval as university supervisors.

Supervisors

Dr. M'. Munji

Department of physics

;::a~~ .~' Date~~::: t>. ':to ::: ..?:",:-?~..

Professor J.N. Nderu

Department of Electrical and electronics

 ii

DEDICATION

This work is dedicated to my sons Edwin, Jeff, Charles and wife Hannah
 iii

ACKNOWLEDGMENTS

First, and foremost, I would like to graciously thank my supervisors Dr. M. Munji of

Kenyatta University and Prof .. LN. Nderu of Jomo Kenyatta University of Agriculture

and Technology for the inspiration and guidance they have given me during my work.

Without their introduction of MSc studies as "learning a method of thought", rather

than only a pursuit of more knowledge, I would not have reached this far. I value the

freedom i have had in choosing the research topic. It has been frustrating when leading

into points of dead-end, but valuable for my future: a reminder that temporary setbacks

are part of the challenge.

It has been a great pleasure to work among Communication Commission of Kenya

(CCK) staff at the Frequency Spectrum Management and Radio Communication

Systems group. I would especially like to thank Mr. Derrick Khamali, J. Kamunge,

S.Mwichuri, Dr. G. Mugeni and Ema for their tremendous technical and logistical

support during the research period. I am grateful to the Director General of CCK Mr.

-Francis Wangusi, Director Frequency Spectrum Management Mr. S. Kibe for allowing

me to carry out my research in the organization.

Special thanks also to Madam Rachel Abiah of Kenya Bureau of Standard (KEBS), Mr.

Moses Kinyanjui of Radiation Protection Board of Kenya (RPB) and Mr. Benson

Mungai of Jomo Kenyatta University of Agriculture and Technology (JKUAT) for

providing 'me both with valuable insights into related technical fields, and with

intriguing discussions of all subjects great and small.

 iv

Getting this far has only been possible thanks to my wife Hannah, my sons Edwin, Jeff

and Charles, who have provided so much warmth and support through the years.

Finally, I would like to acknowledge the academic support by the teaching staff of

Physics Department in Kenyatta University for their constructive critique and

suggestions during departmental seminars.

 v

TABLE OF CONTENTS

DECLARATION

DEDICATION ii

ACKNOWLEDGMENT iii

TABLE OF CONTENTS v

LIST OF TABLES ix

LIST OF FIGURES xi

LIST OF APPEDICES xiii

ABBREVIATIONS AND ACRONYMS xiv

ABSTRACT xvii

1 INTRODUCTION 1

1.1 Background of the study 1

1.2 Statement of the problem ·2

1.3 Objectives 3

1.3.1 General objective 3

,,1.3.2 Specific objectives 3

1.4 Research justification 3

1.5 Research scope 4

2 LITERATURE REVIEW 6

2.1 Introduction 6

2.2 GSM (Global System for Mobile Communications) 7

2.3 GSM evolution 9

2.4 Technical details 11

2.4.1 Network structure 12

2.4.2 Base station subsystem 12

 vi

2.4.3 GSM carrier frequencies 14

2.4.4 Voice codec 15

2.4.5 Subscriber Identity Module (SIM) 15

2.5 Bio-effects of RF radiation exposure 16

2.6 Radiation associated to base transceiver station (BTS) 17

2.7 Research into bio-effects at low levels of exposure 18

2.8 Radio frequency radiation case studies 20

2.9 Summary of some of low-level studies reviewed by working groups 22

3 RESEARCH METHODOLOGY 24

3.1 Introduction 24

3.2 Primary data collection 24

3.2.1 Preliminary investigation 24

3.2.2 Broadband and narrowband measurements 24

3.3 Secondary Data 25

3.4 Location of the data collection ·26

).5 Materials and methods 26

3.5.1 Research instruments 26

3.5.2 Method 30

3.6 Research Scope 31

3.6.1 Measurements sample sites 31

3.6.2 Greenfield and rooftop 33

3.6.3 Indoor micro 35

3.6.4 dBm definition 36

3.6.5 dBi definition 37

3.7 Generating RF radiation profiles within cellular base stations 37

 vii

3.7.1 Functionalities ofTEMs drive test kit 38

3.7.2 Procedure of generating RF radiation profiles 38

3.8 Technical Challenges for RF radiation measurements 39

4 DATA COLLECTION, ANALYSIS AND INTERPRETATION 41

4.1 Introduction 41

4.2 Electromagnetic radiation measurements results 42

4.2.1 Measurements for Site 001 (Sarit Centre) 43

4.2.2 Radiation measurements for Site 002 (Kilimani) 48

4.2.3 Radiation measurements for Site 003 (Kayole) 52

4.2.4 Radiation measurements at Site 004 (Aga Khan) 54

·4.2.5 Radiation measurements for Site 005 (Jamuhuri) 58

4.2.6 Radiation measurements for Site 006 (Olmeut) 59

4.2.7 Radiation measurements for Site 007 (Ndumberi Road.) 60

4.2.8 Radiation measurements for Site 008 (Kitengela) 62

4.2.9 Radiation measurements for Site 009 (Ruai) 63

4.2.10 Radiation measurements for Site 010 (Kahuho) 64

I'

4.2.11 Radiation measurements for Site 011 (Umoja) ·65

4.2.12 Radiation measurements for Site 012 (Kiandutu) 66

4.2.13 Radiation measurements for Site 013 (Kibiko) . 67

4.2.14 Radiation measurements for Site 014 (Zimmerman) 68

4.2.15 Radiation measurements for site15 (Kijabe) 69

4.2.16 Radiation Measurements for site 016 (Tuskys) 70

4.3 Evaluation of measured results against the recommended ICNIRP levels 71

4.4 Generating RF radiation profiles within cellular base stations 76

5 SUMMARY, CONCLUSIONS AND RECOMMENDATIONS 76

 viii

5.1 Summary and conclusions 80

5.2 Recommendations 82

REFERENCES 83
 ix

LIST OF TABLES

Table 2.1 Cellular frequency spectrum used in Kenya 6

Table 2.2 Uplink cellular frequency spectrum as allocated by CCK 7

Table 2.3 Downlink cellular frequency spectrum as allocated by CCK 8

Table 2.4 Low-level studies review 23

Table 3.1 Location coordinates and category of the selected base stations 33

Table 4.1 Measurements results from site 001 (Sarit Center) 45

Table 4.2 Ground Floor maximum RF Radiation for site 001 (Sarit Center) 46

Table 4.3 First Floor maximum RF Radiation for site 001 (Sarit Center) 47

Table 4.4 Second and Third Floor maximum RF Radiation for (Sarit Center) 48

Table 4.5 Measurements results from site 002(Kilimani) 51

Table 4.6 Maximum RF radiation for site 002 (Kilimani) 52

Table 4.7 Radiation for site 003 (Kayole) 53

Table 4.8 Measurements results from site 004 (Aga Khan) 56

Table 4.9 Maximum RF radiation for site 004 (Aga Khan) 57

,- Table 4.10 Maximum RF radiation for site 005 (Jamuhuri) 58

Table 4.11 Maximum RF radiation for site 006 (Olmeut) 59

Table 4.12 Maximum RF radiation for site 007 (Ndumberi Rd) 61

Table 4.13 Maximum RF radiation for site 008 (Kitengela) 62

Table 4.14 Maximum RF radiation for site 009 (Ruai) 63

Table 4.15' Maximum RF radiation for site 010 (Kahuho) 64

Table 4.16 Maximum RF radiation for site 011 (Umoja) 65

Table 4.17 Maximum RF radiations from site 012 (Kiandutu) 66

Table 4.18 Maximum RF radiations from site 013 (Kibiko) 67

Table 4.19 Maximum RF radiations from site 014 (Zimmerman) 68

 x

Table 4.20 Maximum RF radiation for Site 015 (Kijabe) 69

Table 4.21 Maximum RF radiation from Site 016 (Tuskys) 70

Table 4.22 Effective RF radiation exposure 76

 xi

LIST OF FIGURES

Fig 2.1 Structure of a GSM network 12

Fig 2.2 GSM cell site antennas 13

Fig 3.1 FSH 18- Spectrum analyzer with probe ·27

Fig 3.2 Global positioning system (GPS, Garmin ) 28

Fig 3.3 Field Strength Meter (RAHAM) 28

Fig 3.4 Computer used to process the measured results 29

Fig 3.5 TEMS Pocket poisoning tool 29

Fig 3.6 Selected Base station locations within Nairobi and its environs 32

Fig 3.7 Greenfield mast at site 002 (Kilimani) 34

Fig 3.8 Rooftop installation at site 007 (Ndumberi Rd) 34

I'

Fig 3.9 Indoor installation at site 016 (Tuskys) 35

Fig 3.10 Generating RF radiation profiles within cellular base stations 39

Fig 4.1 GSM 900 Spectral scans taken from 001 (Sarit Centre) 43

Fig 4.2 . GSM 1800 spectral scans taken from 001 (Sarit Centre) 44

Fig 4.3 UMTS 2100 spectral scans' taken from 001 (Sarit Centre) 44

Fig 4.4 GSM 900 Spectral scans taken from 002 (Kilimani) 49
 xii

Fig 4.5 GSM 1800 Spectral scans taken from 002 (Kilimani) 49

Fig 4.6 UMTS 2100 spectral scans taken from 002 (Kilimani) 50

Fig 4.7 GSM 900 spectral scans taken from 004 (Aga Khan) 54

Fig 4.8 GSM 1800 spectral scans taken from 004 (Aga Khan) 55

Fig 4.9 UMTS 2100 spectral scans taken from 004 (Aga Khan) 55

Fig 4.10 A photograph of a rooftop base station in Ndumberi Rd 60

Fig 4.11 Comparison of Radiations from site (001) Aga Khan Hospital 72

Fig 4.12 Comparison of Radiations from site 002 (Kilimani) 73

Fig 4.13 GSM 900 RF exposure 74

Fig 4.14 GSM 1800 RF exposure 74

Fig4.15 UMTS 2100 RF exposure 75

,-

Fig4.16 2G RF radiation profile from Upper Hill cluster of base station 77

Fig 4.17 A graph ofrecorded levels against the count (samples) 78

Fig 4.18 2G RF radiation profile from first floor of BP Shell building 78

Fig 4.19 . 3G RF radiation profile from first floor of BP Shell building 79

 xiii

LIST OF APPENDIXES

Appendix 1 86

Appendix 2 89

Appendix 3 91
 xiv

ABBREVIATIONS AND ACRONYMS

3GPP 3rd Generation Partnership Project

4GLTE 4th Generation Long term Evolution

BTS Base Transceiver Station

BSC Base station controller

BSS Base Station Subsystem

CAK Communication Authority of Kenya

CCK Communication Commission Of Kenya

CEPT European Conference of Postal and Telecommunications

EDGE Enhanced Data Rates for GSM Evolution

EFR Enhanced Full Rate

ErR Equipment Identification Register

EME , Electro- Magnetic Energy

EMF Electro-Magnetic Fields

ETSI .European Telecommunications Standards Institute (ETSI)

GPRS General Packet Radio Services

GPS Global Positioning System

 xv

GSM Global System for Mobile Communication

HLR Home Location Register

HSDPA High-Speed Downlink Packet Access

HSUPA High-Speed Uplink Packet Access

ICNIRP International Commission on Non-Ionizing Radiation Protection

ITU International Telecommunication Union

LPC Linear Predictive Coding

MHz Megahertz

MMS Multimedia Messaging Service

MPE Maximum Public Exposure

MSC Main Switching Center

NFP: Network Facility Provider

NIEMR Non-ionizing Electromagnetic Radiation

NIR Non-ionizing Radiation

NSS Network and Switching Subsystem

ODe Ornithin De Carboxylase

OSS Operations Support Subsystem

 xvi

RF Radio Frequency

RFR Radio Frequency Radiation

SAR Specific Energy Absorption Rate

SIM Subscriber Identity Module

UMTS Universal Mobile Telephone services

VLR Visitors Location Register

W-CDMA) Wide Band Code Division Multiple Access

WHO World Health Organization

WLANs Wide Local Area Networks

W 1m2 Watts per square Meter

USIM Universal Subscriber Identity Module

 xvii

ABSTRACT

The World Health Organization (WHO), International Commission on Non-ionizing

Radiation Protection (ICNIRP) has stated that continuous exposure to Radio Frequency
Radiation (RF) beyond certain levels could cause health consequences such as cancer,
congenital anomalies, epilepsy, and persistent headache and may destroy reproductive
organs etc. In this regard and following the proliferation of cellular base stations across
Kenya this study was carried out to determine and evaluate the levels of RF radiation
that a typical person would be exposed to in a given enviromnent in Kenya. This study
measured radio frequency radiation levels at 16 selected cellular base station
transmitters within Nairobi, Kenya. The results were evaluated against the
recommended levels set by ICNIRP. Detailed measurements of power density of the
radio frequency radiation were performed over the cellular frequency bands GSM 900,
GSM 1800 and UMTS 2100, which are the frequency bands in which the cellular
transmitters in Kenya operate. Measurements were taken at the point perceived to have
high population density near the base stations paying more attention to the direction of
maximum radiation of antennas used. The highest levels at each point of measurements
were recorded and evaluated against the ICNIRP exposure guidelines. ICNIRP 1998
specifies a maximum non-occupational exposure limit of 4.5W/m2 at GSM 900 Band,
9W/m2 at GSM1800 frequency band and 10W/m2 at UMTS 2100 frequency band.
Based on our measurements from 16 selected cellular base stations within Nairobi and
its environs the worst case of RF exposure at GSM 900 was 0.080W/m2 from an indoor
micro base station situated in Sarit Centre. This value is 1.7% of the ICNIRP maximum
public exposure (MPE) limits at the frequency band. The highest level at Sarit Centre
was attributed to the close proximity of the radiating antenna to the point of
measurement and the electromagnetic waves may have been scattered by walls and
other obstacles thus creating high radiation levels away from the base station antennas.
At GSM 1800 band, the worst RF exposure recorded was 0.325W/m2 from the base
station at Sarit Centre .This value is 3.6% of the ICNIRP MPE limits in this frequency
band. The highest RF exposure at UMTS 2100 recorded was 7.06mW/m2 recorded at
t"anindoor site 016 (Tuskys Supermarket) in Nairobi CBD. This value is only 0.07% of
the' ICNIRP MPE limit at the frequency band in consideration. The study reveals that
the radiation exposure levels in Nairobi and environs emitted by the four main mobile
phones operators i.e. Safaricom, Airtel, Orange and Essar are several orders of
magnitude below the maximum permitted limits.
 1

CHAPTER ONE

INTRODUCTION

1.1 Background of the study

Since the introduction of mobile phone in Kenya the health implication of radio fre.quency

(RF) radiation from cellular base station has been a subject of great debate and concern

among the Kenyan citizens. Some interested groups opine that radiations from cellular

base station are harmful to health. Some believe that exposure to these radiations for long

period may lead to diseases like cancer, congenital anomalies, epilepsy, and persistent

headache and may destroy reproductive organs. Some international bodies believe that

certain areas of the body like eyes and testicles are particularly vulnerable to RF heating

because of the relative lack of available blood flow to dissipate the excessive heat load

(Hyland, 2000).

At relatively low levels of exposure to RF radiation, that is, levels lower than those that

~ould produce significant heating; the evidence for harmful biological effects is

ambiguous and unproven. Such effects have sometimes been referred to as "non thermal"

effects. It is generally agreed that further research is needed to determine the effects and

their possible relevance, if any, to human health (Loque et al., 2004). Others believe that

there is risk of RF radiation to pregnant women. A pregnant woman and the foetus are both

vulnerable because of the fact that these RF radiation continuously react with the

developing embryo. When a pregnant lady either uses mobile phone or when illuminated

with RF radiation, the developing child can be affected. The developmental malformation

may affect the child's brain. Human brain is said to be the most vulnerable portion to the
 2

Non-ionizing Electromagnetic Radiation (NIEMR). Some of the known effects are

neurological effects, increase in Ornithin De Carboxylase (ODC) activity, effects on

enzymes and free radicals decreasing the brain metabolism (Persson, 1997; Thomas et ai.,

2007).

Statement of the problem

The proliferation of wireless antennas across Kenya may have created a human health

hazard for the public. Members of public stay or work in close proximity to RF radiation

transmitting antennas without knowledge or understanding of the health risks. There is no

"rule-of-thumb" regarding exposure distances; each antenna system is unique. RF radiation

is tasteless, odorless and invisible. Many times, members of public do not even know the

RF radiation transmitting antenna is present because the installation is "hidden" behind

panels or walls, out of public sight.

Although some mobile phone system manufacturers and GSM operators across the world

insists that there are no discernible effects from the base station radiation, the International

Commission on Non-ionizing Radiation Protection (ICNIRP) insists that beyond certain

emission levels there are health consequences. It is due to these health consequences that,

this work aimed at investigating the intensity of the RF emission from some selected

cellular base stations in Nairobi (Kenya) area and environ. It is important to note that, since

the introduction and use of cellular networks in Kenya we have witnessed mushrooming of

base stations than never before. Base stations have been erected in green fields, on top of

buildings and inside buildings; this has raised concern about RF and health.
 3

1.3 Objectives

1.3.1 General objective

The objective of this study was to measure radio frequency radiation levels at selected

cellular base station transmitters, and to evaluate the results against the internationally

recommended levels.

1.3.2 Specific objectives

The specific objectives of the study were as follows.

1. To carry out the radio frequency radiation measurements in GSM 900, GSM 1800

and UMTS 2100 frequency bands at selected base stations within Nairobi and

environ.

II. Determine the highest RF levels at each base station as a percentage of International

Commission of Non-ionizing Radiation protection (ICNJRP 1998) recommendations

for public limiting exposure.

iii. Generate and map Electromagnetic Radiation profiles within

(a) A cluster of Greenfield and rooftop base stations and

(b) Within floors of a building with indoor cellular network coverage.

1.4 Research justification

It is important to note that, since the introduction and use of cellular networks in Kenya we

have witnessed mushrooming of base stations than never before. Base stations have been

erected in green field, on top of buildings, inside buildings etc. In this regard people live

in fear of developing complications associated with RF radiation emissions. It is extremely

 4

complicated to approach electromagnetic radiation from a theoretical angle due to the

many variables in the environment. That is why measurement on a specific location is

often essential to chart the prevailing electromagnetic fields and to localize possible

sources of hazardous emissions. There has been lack of empirical data on RF radiation

from cellular base station transmitters that can inform policy designers in Kenya. There has

been no locally published research that could inform policy designers, mobile operators

and members of public whether RF radiation from cellular base station transmitters in

Kenya is within internationally recommended levels set by the International Commission

on nonionizing radiation protection (ICNIRP) recommendations.

In some areas mobile phone operators meet a lot of opposition from the residents when

they plan to erect a base station. Residents usually demand to know if the base station in

their proximity will have effect on them. Mostly the answer given is not convincing

because of lack of data in public domain for reference. This research will make a major

contribution to the body of knowledge on the non-ionizing RF radiation from cellular base

stations in Kenya. The analysis of the data obtained from the measurements carried out has

revealed the state of compliance to internationally recommended RF radiation levels.

Future researchers and stake holders could use the results of this research to develop an RF

radiation protection framework for Kenya.

1.5 Research scope

The scope of the research was to generate data that can be used to conclusively draw

position of radio frequency emission from cellular base stations in Kenya. This data was
 5

captured from existing cellular base station transmitters within Nairobi and outside

Nairobi. Measurement from sixteen cellular base stations were taken and then evaluated

against ICNIRP 1998 Maximum Public Exposure (MPE) limits to give a clear picture of

the RF exposure contributed by cellular base stations. Twenty base stations were targeted

and choice of the sites was random but selective with main focus on areas with high

population and hence high concentration of base stations. The choice was also made to

capture emissions from all the licensed cellular operators in Kenya. The categories of the

base stations were also considered when selecting the sites. We were however able to

conduct measurements in sixteen base stations due to time limit allocated by the CAK

engineers.
 6

CHAPTER TWO

LITERATURE REVIEW

2.1 Introduction

Radio spectrum refers to the part of the electromagnetic spectrum corresponding to radio

frequencies that is; frequencies lower than around 300 GHz (wavelengths longer than

about I mm). Electromagnetic waves in this frequency range are called radio waves and

are commonly used for radio communication and various other technologies. Because of

the potential for electromagnetic interference between users, the generation of radio waves

is strictly regulated by the government in most countries, coordinated by an international

standards body called the International Telecommunications Union (ITU). Different parts

of the radio spectrum are allocated for different radio transmission technologies and

applications. In some cases, parts of the radio spectrum is sold or licensed to operators of

private radio transmission services (for example, cellular telephone operators or broadcast

television stations). Ranges of allocated frequencies are often referred to by their

provisioned use for example, cellular spectrum (Colin, 2003). Table 2.1 shows the three
I'

main cellular frequencies sub bands used in Kenya and the frequencies within the bands

allocated for downlink and uplink.

Table 2.1 Cellular frequency spectrum used in Kenya

Cellular sub bands Uplink Down link

GSM 900 880 -915 MHz 925 -960 MHz
GSM 1800 1710 -1800 MHz 1805 -1880 MHz
UMTS 2100 1885 - 2025 MHz 2110 - 2200 MHz

Table 2.2 and Table 2.3 show the cellular frequency spectrum allocation in Kenya.

Downlink frequencies refer to the frequencies that base stations use to transmit information
 7

to the mobile phone while the uplink refers to the frequencies used by the mobile phone to

transmit to the base station. Communication Commission of Kenya (CCK) has allocated

the frequencies to different mobile phone operators as shown in the tables 2.2 and 2.3.

Table 2.2 Uplink cellular frequency spectrum as allocated by CAK

GSM 900 MHz UPLINK

Frequency MHz Bandwidth (MHz) Operator
880-882.5 2.5 Orange
882.5 - 890 7.5 Essar
890 - 900 10 Safaricom
900 - 905 5.0 Orange
905 - 915 5.0 Airtel
GSM 1800 MHz UPLINK
1710 - 1720 10.0 Econet
1720 - 1730 10.0 Airtel
1730 - 1740 10.0 Orange
1740-1750 10.0 Safaricom
1750 -1770 10.0 SNO
1770 - 1780 10.0 Bellwestem
1780 - 1785 5.0 CDMA2000
1785 - 1800 5.0 IBURST
UMTS 2100 MHz UPLINK
1910 - 1920 10 3G (TDD)
1920 - 1930 10 Safaricom
1930 - 1940 10 Airtel
1940 -1950 10 Essar
l'950 - 1955 5 SNO
1955 - 1960 5 CDMA
1960 - 1965 5 SNO
1965 - 1975 10 Orange
1965 - 1980 5 SNO
2010 - 2025 5 3G (TDD)

2.2 GSM (Global System for Mobile Communications)

GSM (Global System for Mobile Communications, originally Groupe Special Mobile), is a

standard set developed by the European Telecommunications Standards Institute (ETSI) to

describe protocols for second generation (2G) digital cellular networks used by mobile
 8

phones. It became the de facto global standard for mobile communications with over 80%

market share.

Table 2.3 Downlink cellular frequency spectrum as allocated by CAK

GSM 900 MHz Downlink

Frequency MHz Bandwidth (MHz) Operator
GSM 900 MHz Downlink
925 - 927.5 2.5 Orange
927.5 - 935 7.5 Essar
935 - 945 10.0 Safaricom
945 - 950 5.0 Orange
950 - 960 10.0 Airtel
GSM 1800 MHz Downlink
1805 - 1815 10.0 Essar
1815 - 1825 10.0 Airtel
1825 - 1835 10.0 Orange
1835 - 1845 10.0 Safaricom
1845 - 1865 10.0 SNO
1865 - 1875 10.0 Bellwestern
1875 - 1900 25.0 CDMA 2000
1900 - 1910 10.0 Popote wiress
UMTS 2100 MHz Down
link
2110-2120 10.0 Safaricom
2120 - 2130 10.0 Airtel
2130 -2140 10.0 Essar
2140 -2145 5.0 SNO
2'145 - 2150 5.0 Broadband
2150 - 2155 5.0 SNO
2155 - 2165 10.0 Orange
2165-2170 5.0 SNO

The GSM standard was developed as a replacement for first generation (l G) analog

cellular networks, and originally described a digital, circuit-switched network optimized

for full duplex voice telephony. This was expanded over time to include data

communications, first by circuit-switched transport, then packet data transport via GPRS

(General Packet Radio Services) and EDGE (Enhanced Data Rates for GSM Evolution).
 10

Silberhorn (Germany), Stephen Temple (UK), Philippe Dupuis (France), and Renzo Failli

(Italy). In 1989 the Groupe Special Mobile committee was transferred from CEPT to the

European Telecommunications Standards Institute (ETSI) (Hillebrand, Friedheim, (2001).

In parallel, France and Germany signed a joint development agreement in 1984 a~d were

joined by Italy and the UK in 1986. In 1986 the European Commission proposed reserving

the 900 MHz spectrum band for GSM.

Phase I of the GSM specifications were published in 1990. The world's first GSM call was

made by the former Finnish prime minister Harri Holkeri to Kaarina Suonio (mayor in city

of Tampere) on July 1, 1991, on a network built by Te1enokia and Siemens and operated

by Radiolinja. The following year in 1992, the first short messaging service (SMS) was

sent and Vodafone UK and Telecom Finland signed the first international roaming

agreement. Work began in 1991 to expand the GSM standard to the 1800 MHz frequency

band and the first 1800 MHz network became operational in the UK by 1993. That year,

Telecom Australia became the first network operator to deploy a GSM network outside

Eurcpe and the first practical hand-held GSM mobile phone became available.

In 1995, fax, data and SMS messaging services were launched commercially, the fir t

1900 MHz GSM network became operational in the United States and GSM subscribers

worldwide exceeded 10 million. In the same year, the GSM Association was formed. Pre-

paid GSM SIM cards were launched in 1996 and worldwide GSM subscribers passed 100

million in 1998.
 11

In 2000, the first commercial GPRS services were launched and the first GPRS compatible

handsets became available for sale. In 2001 the first UMTS (W-CDMA) network was

launched and worldwide GSM subscribers exceeded 500 million. In 2002 the first

Multimedia Messaging Service (MMS) were introduced and the first GSM network in the

800 MHz frequency band became operational. EDGE services first became operational in a

network in 2003 and the number of worldwide GSM subscribers exceeded billion in

2004.

By 2005, GSM networks accounted for more than 75% of the worldwide cellular network

market, serving 1.5 billion subscribers. In 2005 the first HSDPA capable network also

became operational. The first HSUPA network was launched in 2007 and worldwide GSM

subscribers exceeded two billion in 2008. The GSM Association estimates that

technologies defined in the GSM standard serve 80% of the global mobile market,

encompassing more than 5 billion people across more than 212 countries and territories,

making GSM the most ubiquitous of the many standards for cellular networks (Moulyand

Marie, 2002).

2.4 Technical details

Figure 2.1 shows the four discrete sections of a GSM network. The four sections are Base

Station Subsystem, Network and Switching Subsystem, GPRS Core Network and

Operations support system. Figure 2.1 shows the structure of a GSM network.
 12

2.4.1 Network structure

The network is structured into a number of discrete sections:

i. Base Station Subsystem (BSS) - the base stations and their controllers as explained in

section 2.4.2

ii. Network and Switching Subsystem (NSS) - the part of the network most similar to a

fixed network, sometimes just called the "core network"

iii. GPRS Core Network - the optional part which allows packet-based Internet

connections

iv. Operations support system (OSS) - network maintenance

GPRS PS

BSS

EIR

Mobile station Base station subsystem Network and switching subsystem'

Fig: 2.1 Structure of a GSM network

2.4.2 Base station subsystem

Fig 2.2 shows a typical Base Station subsystem setup' with panel antennas, a tower mount

amplifier and a parabolic transmission antenna.

 13

GSM is a cellular network, which means that cell phones connect to it by searching for

cells in the immediate vicinity. There are five different cell sizes in a GSM network macro,

micro, pico, femto, and umbrella cells. The coverage area of each cell varies according to

the implementation environment. A Macro cell can be regarded as cell where the base

station antenna is installed on a mast or a building above average rooftop level. A Micro

cell is a cell whose antenna height is under average rooftop level; it is typically used in

urban areas. Picocells is a small cell whose coverage diameter is a few dozen meters; it is

mainly used in indoors. Femtocells are cells designed for use in residential or small

business environments and connect to the service provider's network via a broadband

internet connection. Umbrella cells are used to cover shadowed regions of smaller cells and

fill in gaps in coverage between those cells.

GSM panel antenna

Tower mount amplifier

Parabolic transmission antenna

Fig 2.2 GSM cell site antennas

A Cell horizontal radius varies depending on antenna height, antenna gain, and

propagation conditions from a couple of hundred meters to several tens of kilometers. The

longest distance the GSM specification supports in practical use is 35 kilometers. There are
 14

several implementations of the concept of an extended cell, where the cell radius could be

double or even more, depending on the antenna system, the type of terrain, and the timing

advance.

Indoor coverage is also supported by GSM and may be achieved by using an indoor

picocell base station, or an indoor repeater with distributed indoor antennas fed through

power splitters, to deliver the radio signals from an antenna outdoors to the separate indoor

distributed antenna system. These are typically deployed when significant call capacity is

needed indoors, examples can be found in shopping centers or airports. However, this is

not a prerequisite, since indoor coverage is also provided by in-building penetration of the

radio signals from any nearby cell.

2.4.3 GSM carrier frequencies

In some parts of the world where GSM900, GSM 1800 and UMTS2100 bands were

already allocated, the 850 MHz and 1900 MHz bands were used instead (for example in

Canada and the United States). In rare cases the 400 and 450 MHz frequency bands are

assigned in some countries because they were previously used for first-generation systems.

Most 3G networks in Europe operate in the 2100 MHz frequency band.

Regardless of the frequency selected by an operator, it is divided into timeslots for

individual phones. This allows eight full-rate or sixteen half-rate speech channels per radio

frequency. These eight radio times lots (or burst periods) are grouped into a TDMA frame.

Half-rate channels use alternate frames in the same timeslot. The channel data rate for all 8
 lS

channels is 270.833 Kbitls, and the frame duration is 4.615 milliseconds. The transmission

power in the handset is limited to a maximum of 2 Watts in GSM 850/900 and 1 Watt in

GSM 180011900.

2.4.4 Voice codecs

GSM has used a variety of voice codecs to squeeze 3.1 kHz audio into between 6.5 and

13 kbitls. Originally, two codecs, named after the types of data channel they were

allocated, were used, called Half Rate (6.5 kbitls) and Full Rate (13 kbitls). These used a

system based on linear predictive coding (LPC). In addition to being efficient with bitrates,

these codecs also made it easier to identify more important parts of the audio, allowing the

air interface layer to prioritize and better protect these parts of the signal.

GSM was further enhanced in 1997 with the Enhanced Full Rate (EFR) codec, a

12.2 kbitls codec that uses a full-rate channel. Finally, with the development of UMTS,

EFR was refactored into a variable-rate codec called AMR-Narrowband, which is high

quality and robust against interference when used on full-rate channels, or less robust but

still relatively high quality when used in good radio conditions on half-rate channels

(Salgues, 1997)

2.4.5 Subscriber Identity Module (81M)

One of the key features ofGSM is the Subscriber Identity Module, commonly known as a

SIM card. The SIM is a detachable smart card containing the user's subscription

information and phone book. This allows the user to retain his or her information after
 16

switching handsets. Alternatively, the user can change operators while retaining the

handset simply by changing the SIM. Some operators will block this by allowing the phone

to use only a single SIM, or only a SIM issued by them; this practice is known as SIM

locking.

2.5 Bio-effects of RF radiation exposure

Mobile phone use in Kenya has increased substantially, the number of subscribers having

reached 29.7 million by October 2012 (CCK, 2012). As a result, there has been a

corresponding rise in the public interest expressed with regard to health issues associated

with the exposure of humans to Electro-magnetic Fields (EMF). Both mobile phones and

their Base Transceiver Stations (BTS commonly known as 'base stations') emit

radiofrequency electromagnetic energy (EME), which is a form of radiation technically

referred to as non-ionizing radiation (NIR). Ionizing radiation (such as X-rays and Gamma

Rays), on the other hand are capable of causing ionization and producing molecular

changes, some of which can damage biological tissue.

Exposure to RF radiation which is non-ionizing is categorized into two: Occupational and

general-public exposure;

i. In occupational exposure, persons exposed as a consequence of their employment are

fully aware of the danger of such exposure and are supposed to take necessary

precautionary measures.

ii. Exposure that is not employment related such as radiation from mobile phones and BTS

is classified under general-public exposure (ICNIRP, 1998).

 17

Various organizations such as the United state of America based, Federal Communications

Commission (FCC) and International Commission of Non-Ionizing Radiation Protection

(lCNIRP) have set RF exposure limits. Exposure to Radio frequency radiation above the

reference limits is termed as hazardous. Accruing health effects due to such exposures are

thermally and non-thermally induced (Loque et al., 2004). Absorption of RF energy by

biological tissues and the heating thereof is facilitated by electrical properties of bio-matter

and the body's thermoregulatory mechanism (Hyland, 2000). The amount of heat produced

in the exposed tissues depends primarily on exposure time and the intensity of radiation

penetrating the system. The World Health Organization (WHO) has associated brain

cancer with RF exposures and has determined that the most thermally vulnerable organs

include genitals and eyes (Maneesh et al., 2009). Other effects include a stinging sensation

and a feeling of heat in the facial skin (Sandstrom et al., 1998). Non-thermal effects such

as headache, dizziness, fatigue, stress, difficulties in concentrating and nausea have been

reported (Krewski , 2007). In this work, the intensities of RF radiation from selected BTS
,.
in Kenya were investigated and assessed based on the established safety standards on non-

ionizing radiation.

2.6 Radiation associated to base transceiver station (BTS)

Due to the ever-increasing mobile-phone usage in Kenya, increasing numbers of BTS are

being setup to support the growing network coverage. Other wireless networks, such as

those that facilitate high-speed Internet access (such as WLANs) are also on the increase.

Such exposure, which is dependent upon the proximity of the human being to the BTS and
 18

other such relevant environmental factors, is comparable to the RF exposure caused by

radio or Television-broadcast transmitters and if not properly checked can cause problems

to human and animal life. Under ideal conditions the strength of the RF field is at its

greatest in the immediate vicinity of the source and diminishes quickly in relation to

distance from the source. Network access in the immediate vicinity of a BTS should often

be restricted due to the fact that the RF signal may exceed the permitted exposure limits

(CCK,2012).

BTS antennae (typically mounted on buildings or towers upto a height of 15 to 50 meters

above ground) emit a fan-shaped transmission beam, which is typically vertically narrow

and horizontally broad, and whose intensity decreases rapidly in relation to the distance

from the BTS. The point at which the beam comes into contact with the ground is also

assumed to be sufficiently low as to render its strength significantly lower than the

recommended maximum radiation limits.

2.7 Research into bio-effects at low levels of exposure

Electromagnetic fields of all frequencies represent one of the most common and fastest

growing environmental influences, about which anxiety and speculation are spreading. All

populations are now exposed to varying degrees of EMF, and the levels will continue to

increase as technology advances. Extensive research has been conducted into possible

health effects of exposure to many parts of the frequency spectrum (Moni, 2010).

International scientific reviews were conducted mainly with the purpose of identifying;

i. Health effects that can be substantiated from the literature, and

 19

ii. Biological effects that are suggestive of possible health effects but require further

research to determine if exposure to EMF at low levels of exposure normally encountered

in the living and working environment has any impact on health.

World health organization (WHO), through its International EMF Project, has conducted a

series of in-depth international reviews of the scientific literature on the biological and

health effects of exposure to electromagnetic fields. The following are some of these

reports:-

i. Biological Effects of RF Electromagnetic Fields and Related Health Risks (WHO

Munich, 1996).

ii. Electromagnetic Fields and Non-Specific Health Symptoms (WHO Graz, 1998)

iii. Sensitivity of Children to EMF exposure (WHO Istanbul, 2004)

iv. EMF Hypersensitivity (WHO Prague, 2004).

v. Guiding public health policy in areas of scientific uncertainty (WHO Ottawa, 2005)

""A number of international organizations have formulated guidelines establishing limits for

occupational and residential EMF exposure. The exposure limits for EMF fields developed

by the International Commission on Non-Ionizing Radiation Protection (ICNIRP) - a non-

governmental organization formally recognized by WHO, were developed following

reviews of all the peer-reviewed scientific literature, including thermal and non-thermal

effects. The standards are based on evaluations of biological effects that have been

established to have health consequences.

 20

2.8 Radio frequency radiation case studies

Gregory Nelson reported a measurement study conducted by the University of the West

Indies to assist the Telecommunications Regulatory Commission of the Virgin Islands

(UK}(TRC). The study involved the measurement of the Radiofrequency Radiation (RFR)

levels in the British Virgin Islands (BVI). Measurements were taken at various sites on the

main Islands of Tortola, Virgin Gorda, Anegada and Jost Van Dyke at places- where

complaints originated. The exercise comprised conducting measurements with a broadband

field strength meter for comparison to the general public exposure limits set by the

International Commission on Non-Ionizing Radiation Protection (ICNIRP). Spectrum

scans of cellular bands were also collected to clarify the contributions of cellular

frequencies to overall levels measured at locations. At all locations measurements were

taken with the broadband field strength meter, a spectrum analyzer was used to capture

data on cellular contributions. Overall measurements did not exceed 8%, where not more

than 6% was due to the cellular bands. In 52% of the cases, cellular radiation contributed

between 40% to 80% of the overall levels measured. The study revealed that the maximum

RF emission from a broadcasting transmitter was 8% ofMPE recorded at Tortola (Gregory

etal.,2010).

In New Delhi India Seven telecom operators, including RCom, Tata Teleservices

Maharashtra, Airtel and Vodafone, were found violating radiation norms laid down by the

government after a random measurement checks conducted at a location in Mumbai.

Eleven BTSs (Base Transceiver Stations) of Reliance Communications, TTML, Airtel,

Vodafone, Idea Cellular, Airtel and Loop Telecom were found radiating beyond
 21

permissible limits of the new radiation norms when measurements were carried out in

some houses facing nearby BTS antenna. Orders were issued to the 7 operators to stop

radiating from these BTSs with immediate effect. Functioning of the BTSs was resumed

only after the sites are made compliant and information to that effect was verified by

Telecom Enforcement and Resource Monitoring (TERM) Cell. The new radiation norms,

applicable from September 2012, lowered limit of emission from telecom towers by one-

tenth compared to earlier ICNIRP MPE limits in India. Under the new guidelines,

emissions from telecom antenna mounted on mobile towers should be equivalent to

frequency range in which the antenna operates. Like, an antenna operating in the frequency

range of 400 MHz will be allowed to emit 400 watt of energy per 2,000 square metre of

area. It sets minimum distance of a tower (with two antennas) to be 35 metres from a

residential building (Rajan, 2012).

A study of the electromagnetic environment on existing residential and rural land at Minto

(Australia) was performed to determine the impact on the Minto urban development

proposed by Landcom. The electromagnetic radiation (EMR) levels over the entire

redevelopment site emanating from the Mobile Phone Base Transceiver Site (BTS) on the

adjacent Sydney Water land, Eagleview Road, Minto and all other fixed radio transmitter

installations in the vicinity of the development area were measured. The power flux

densities due to all base station transmitters at each measurement location were calculated.

An assessment of the likely impact of these electromagnetic radiations (EMR) levels

within the terms of the Campbelltown City Council Development Control Plan No. 107

(Siting of Communication Facilities,. Telecommunication towers) - Electromagnetic

 22

Radiation was determined. Campbell town City Council Development Control Plan No.

107 "Siting of Communication Facilities (telecommunication Towers)" Part 1 paragraph

12 Electromagnetic Radiation states: Communication facilities should meet the following

precautionary siting requirements:

(i) Should be setup at least 300 meters from any residential dwelling unless the annual

2
average exposure at the nearest residential dwelling is less than 0.2 uW/cm

(ii) Should be setup at least 300 meters from any school, childcare centre, hospital or aged

care centre and the annual average exposure at these land uses must be less than 0.2
2
uW/cm.

The combined EMR of all existing radio communications transmitters at selected measured

locations within the Minto urban development project were less than 30% of the Campbelt

town City Council Development Control Plan No. 107 requirements for siting of

communication facilities. The highest recorded level was 8.5% of the MPE limits of

Campbelt town council limit exposure (DCP 107). These levels also complied with the

Australian Communications Authority Standard (ACAS) regulations on human exposure to

radiofrequency electromagnetic energy (Bradshaw, 2005).

2.9 Summary of some of low-level studies reviewed by working groups

Fig 2.4 summarizes some of the low level studies reviewed by working groups. The

columns shows the Authors, exposure conditions, test system used, the theme of the study

and the outcome of each study,

 23

Table 2.4 Low-level studies review

Author Exposure Test system Theme outcome

Clearyet 2450 MHz RIF Cytolytic Cell At SAR>25W/kg reduction in

al1997 (SAR greater & Lymphocyte prolifer CTLL-2 proliferation. At
less than 25 W s (CTLL-2) ation lower SARs increase in CTLL-
Vkg isothermal 2 Proliferation immediately
RF 24 h exposure after exposure but reduced
24h post-exposure

Kwee et al 960 MHz GSM or Cell culture Cell For GSM linear correlation
1998 ELF in a TEM of pro lifer between power level and
cell for 3 different transformed ation growth change. For ELF
power levels human linear correlation between the
epithelial length of exposure time to
amnion cells obtain maximum effect and
field strength

Velizarov 960 MHz (GSM) Cell line Cell No change in Cell proliferation
et al1999 in TEM Cell size not prolifer under different temperatures
specified ation

Reeves RF Exceeding 34 Patients Psychol Warmth sensation associated

2000 permitted ogical with power density. Abnormal
exposure and tissue destruction negatively
laborat correlated with power density.
ory No Neurological or
paramet Ophthalmologic findings
ers
 24

CHAPTER THREE

RESEARCH METHODOLOGY

3.1 Introduction

This chapter details the methods used to address the research problem. as outlined in the

previous chapter. It covers the methodology, the various research instruments used and

their rationale thereof, the procedure followed and the location where the research was

conducted as well as the targeted subjects.

3.2 Primary data collection

3.2.1 Preliminary investigation

Preliminary investigations included indentifying the site location and inspecting the site to

note any relevant characteristics that may impact upon readings (e.g. the presence of

obstructions - trees, buildings; reflecting objects - cars, water, metallic structures). Next, as

necessary, the field strength m ter and isotropic broadband probe was used to determine

spatial variation of RF fields around the site in order to identify the position' of the

maximum field strength. Given the transmitting frequencies of current cellular

"
technologies, and the location of transmitting antennas, measurements were taken in the far

field of the transmitter. GPS coordinates w re recorded along with any other relevant

details about the sites.

3.2.2 Broadband and narrowband measurements

The field strength meter and isotropic broadband probe was used to measure radiation

within the range 300 kHz to 50 GHz, thus determining radiation levels of many possible
 25

cellular radio frequency sources. At the identified measurement locations, the field strength

meter and probe was used to capture field strength data for 6 minutes, with I-second

sampling intervals. Within each sampling interval the maximum, minimum, and average

values for that I-second interval was logged in by the meter. The field meter was mounted

at approximately 1.5 meters above the local ground level using the non-metallic tripod

(ICNIRP, 1998). Multiple readings were taken at a sample point since at each site multiple

antennas exist to cover different sectors. Taking this into account, readings were taken for

each sector, provided that physical access to a suitable measurement point was possible. A

Spectrum Analyzer was used to measure radiation by frequency within the range 800 MHz

to 2200 MHz, thus determining radiation levels by frequency for many possible cellular

radio frequency sources in the area. While the Field strength meter logged in 300 kHz to

50 GHz RF radiation data for a site, spectrum analyzer was used to log RF radiation data

across the cellular frequency ranges being measured. The analyzer was placed on a

continuous sweep, and the data recorded for subsequent analysis. For this study scans were

collected to cover the GSM services and 3G services (where applicable) to confirm the

existing levels.

3.3 Secondary Data

The secondary data constituted information from scholarly literature review of publications

on RF radiation, ICT sector reports and publications, international journals and

publications and internet data. The secondary data played a vital part in the development

of the constructs used in primary data collection.

 26

3.4 Location of the data collection

The research was conducted within the City of Nairobi and its environs using instruments

and other resources from Communication Commission of Kenya (CCK) and cellular

network operators. For the purpose of this study, the state of RF radiation in Nairobi City

and its environs was assumed to be representative of RF radiation expectations in

developing countries and particularly Kenya presently and in the near future. The choice of

the cellular base station was random but selective.

3.5 Materials and methods

3.5.1 Research instruments

A research instrument is a testing device for measuring a given phenomenon or a set of

guidelines for observation (Crossman, 2004). A wide range of research instruments was

utilized for primary data collection. The choice of instrument depended on the situation

where measurements are being made. All instruments used for measuring RF fields had the

following basic components covering the frequency range of interest»

i. Field strength meter

ii. Spectrum analyzer

iii. An isotropic probe to sample the field.

iv. Laptop/PC with the relevant software to process the measured results.

v. GPS (Global Positioning System) recording location coordinates.

vi. Digital camera

Fig 3.1 shows an RF spectrum analyzer and associated probe used for the RF radiation

measurements. A Spectrum Analyzer was used to measure radiation by frequency within

 27

the range 800 MHz to 2200 MHz, thus determining radiation levels by frequency for many

possible cellular radio frequency sources in the area.

Fig 3.2 shows a Global Positioning System (GPS ) used to indentify the location of the

cellular base station. The GPS records the coordinates of the base station.

Fig 3.1: FSH 18- Spectrum analyzer with probe

 28

Fig 3.2: Global Positioning System (GPS, Garmin)

Fig 3.3 shows the field strength meter used to indetify the location of the highest

exposure .. The field strength meter and isotropic broadband probe was used to

measure radiation within the range 300 kHz to 50 GHz, thus determining radiation

levels of many possible cellular radio frequency sources.

Fig 3.3: Field Strength Meter (RAHAM)

 29

Fig 3.4 shows the computer used to process the data collected by the spectrum

analyser.

Fig 3.4 Computer used to process the measured results.

Fig 3.5 TEMS pocket positioning tool

 30

3.5.2 Method

The primary aim of this study was to determine the RF radiation exposure levels resulting

from all signals frequencies produced by the particular cellular base stations under survey.

In this case measurements were taken for frequencies within the GSM 900, GSM 1800 and

UMTS 2100 bands. Mobile telephone communication signals are both transient and partly

random in their occurrence and distribution. In this context, we were interested in

determining the RF radiation levels at many locations and more particularly, we wanted to

estimate highest levels for each location in the area surrounding the base station. Relevant

spectrum analyzer data were recorded and subsequently analyzed. Measurements were

performed within all the sectors available at various locations close to the base station, by

continuously scanning the frequency bands and logging the signal level for the cellular

base station.

Where more than one carrier (Safaricorn, Airtel, Orange, and Essar) were detected

combined activity from all carriers was determined. Measurements were performed at
I'

sixteen different cellular base stations in Nairobi and environ. Measurements were taken at

distances ranging from 5 m to 350 m in areas around the base station accessible to the

public and the highest levels recorded. The RF radiation levels of cellular base stations

attributable to the major carriers were investigated, although not all were present at each

base station. There were various categories of base station considered i.e. Greenfield,

Rooftop and Indoor Micro. For the Greenfield and Rooftop the base station antennae were

mostly positioned on towers between 20 to 60 meters in height. Most commonly,

directional panel antennae were used. Such antennae divide the area around the base
 31

station into three sectors, with each sector usually covering a ] 20 degree arc in azimuth.

Hence, three sets of antennae are normally in use at a typical base station. Indoor Micro

Base stations were found to employ Omni-directional antennae at Sarit centre and Tuskys

supermarket base stations surveyed. An Omni-directional antenna provides 3600 coverage

of the area serviced by the base station.

A fixed antenna (Probe) was mounted on the spectrum analyzer and automated mobile

measurements were made whilst walking around the cellular base station under survey.

Both signal data and position information of the base station [using Global Positioning

System (OPS)] were recorded. For each sector we were able to make measurements of all

active frequencies at each particular measurement sample location using the spectrum

analyzer. All measurements were made at a height of 1.5 meters which is the average

height of a man above ground level. Measurement antennae were oriented to obtain

maximum signal strength for the particular frequency band being measured. Relevant

spectrum analyzer data were recorded automatically and saved in the analyzer memory and

later downloaded into a computer.

3.6 Research Scope

3.6.1Measurements sample sites

Measurements were performed at 48 different locations from sixteen different base stations

within Nairobi and environ. At least three points of measurements were chosen around

each of the 12 Greenfield and 2 Rooftop sites. Two to four points of measurements were

chosen from each of the two indoor micro sites depending on the number of floors. The
 32

measurements were performed between 19th Feb 2013 and 28th of April 2013 at the

locations shown in table 3.1 below. All measurements were made at a height of 1.5 meters

above ground level and at positions frequented by the public. Figure 3.6 is a map showing

the physical location of each sampl,e base station where radio frequency radiations were

measured. The coordinates in Table 3.1 were used to generate the Google map in Figure

3.6. Table 3.1 shows the base station identity, name, category and the base station location

coordinates as recorded by the GPS.

j
The sixteen cellular base stations were categorized into

I. Greenfield: Where the mast is erected on an open field

II. Rooftop: The mast and the antennas used are mounted on a rooftop of a building

111. Indoor micro. Omni directional antennas are mounted on the ceiling In the

building.

Figure 3.7 shows a base station panel antenna mounted on a 45 m tall Greenfield structure

at Kilimani. Figure 3.8 shows a base station panel antennas mounted on a structure built on

a rooftop of a building at Ndumberi road.

Fig 3.6: Selected Base station locations within Nairobi and its environs
 33

Figure 3.9 shows a sample of an indoor Omni directional antenna mounted on the ceiling

of the Tuskys supermarket.

Table 3.1: Location coordinates and category of the selected base stations

Site ID Site Name BTS Category Site Coordinates

001 Sarit Indoor Micro 011}15' 41.16" S, 361}36' 25.93" E

002 Kilirnani Greenfield 01u IT 22.10" S, 36u 46' 53.70" E

003 Kayole Greenfield 011}16' 56.42" S, 361}54' 32.95" E

004 AgaKhan Rooftop 01u 15' 24.98" S, 36u 48' 41.36" E

005 Jarnuhuri Greenfield 01° 18' 02.41" S, 36° 46' 24.94" E

006 Olrnuet Greenfield 011}23' 26.80" S, 361}47'40.70" E

007 Nduberird Rooftop 01u16' 52.90" S, 36°49'34.00" E

008 Kitengela Greenfield 011}38'41.40" S, 361}57' 30.1 0" E

-
009 Ruai Greenfield 01u 16 '33.42" S, 36u 40'30.94" E

010 Kahuho Greenfield 011}12' 48.44" S, 361}40' 30.94" E

011 Urnoja Greenfield 01u 16' 56.42" S, 36u 53' 11.95" E'

012 Kiandutu Greenfield 011}08' 49.80" S, 361}57' 40.80" E

013 Kibiku Greenfield or 21' 10.40" S, 36u 38' 34.93" E

0-14 Zirnerman Greenfield 011}12' 44.44" S,36° 53' 37.95" E

015 Kijabe Greenfield 001}57' 38.20" S, 36u 36' 16.50" E

016 Tuskys pioneer Indoor Micro 01u 17' 18.11" S, 361}49' 36.84" E

3.6.2 Greenfield and Rooftop

When evaluating conformance of exposure limits, for these two categories usually only the

maximum signal is considered. This value was found using spectrum analyzer and an

isotropic antenna using the so-called panning approach (ICNIRP, 1998). The antenna had a

gain of 3dBi. The desired frequency ranges were chosen. In our case we have (GSM900
 34

down link, GSM 1800 down link and UMTS 2100 down link. At each frequency band the

antenna was slowly tilted in all directions until maximum signal values were obtained.

Panel antenna

Parabolic antenna

Fig 3.7 Greenfield mast at site 002 (Kilimani)

Panel antenna

Fig 3.8: Rooftop installation at site 007 (Ndumberi Rd)

 35

Indoor Omni antenna

Fig 3.9 Indoor installation at site 016 (Tuskys)

3.6.3 Indoor micro

In the indoor however, walk test (measurements taken as one walks about the target area)

was done through the entire room. The highest reading was found by rotating the probe.

The logged data from each narrowband sweep was stored in dBm units. In order to

compare the values obtained to those for the ICNIRP 1998 recommendations it was

necessary to perform several calculations to convert the measurements into percentages of

the exposure limits for the general public. Due to the amount of data collected, the

calculations were done via a computer program created for the measurements. An outline

of the procedure is provided below.

For each set of measurements taken at a location, the logged data was retrieved by the

computer program. Data included antenna gains and cable insertion losses for all relevant
 36

frequencies, as well as the actual scanned frequencies, and measured dBm values for

measurement intervals. The antenna parameters and insertion cable losses for the feeder

cable were provided by the manufacturer. This al 0 facilitated the conversion of the units

to percentage conformance by frequency. Measurement intervals contained numerous

sweeps of each spectrum band. For each ingle sweep. at each frequency for which data

was stored. the dBm readings w 1'0 env rted to W/m,2. according to the formula,

S (W/m2) =1 O(P-G)/IO /1000 x 41t/).. 1 - uee_. .os~ ~_••• - -. ~ 3.1

where

S = Power density in W1m2

P =Measured radiation levels in dBm

G =Receiving antenna gain in dBi at the particular frequency measured

A. = Wavelength of transmitter frequency in meters

)"?141t =Effective area of the isotropic antenna

3~6.4dllm definition

dBm OJ' declbel-milliwatt is an Iectrical power unit in decibels (dll), referenced to 1

milliwatt (m W),

The power in decibel-milliwatts (p(c!Om») is equal to the base 10 logarithm of the power in

rnilliwatts (P(mW»):

P (dBm) = 10 . log 10 ( P (mW) / l m W)~•.~-••.- -.-~r~~~.---_

..,-.s_.- .._-~.••....•.
_..__~_~ -_-,,~------ 3.2
The power in milliwatts (PemW» is equal to lmW times 10 raised by the power in decibel-

milliwatts (P(dBm») divided by 10:

 37

P (mW) = 1m W . 1 if(dblll)/10 -----------------------.-------------------------------------------------3.3

3.6.5 dBi definition

The expression dBi is used to define the gain of an antenna system relative to an ideal

isotropic radiator at radio frequencies.

The symbol is an abbreviation for "decibels relative to isotropic."

The gain G of antenna A, in dBi, is:

G = 10 log 10 (I A / I oJ -----------------.------------------------------------------------------------3.4
where I A is the current radiated by the antenna and

I Q is the current radiated by the ideal isotropic radiator

An isotropic radiator has a gain of 0 dBi.

Additionally, at each frequency the exposure limit was determined and results were used to

express the exposure limits as percentages for the general public. This provided an idea of

the contribution of cellular transmissions to the overall exposure.

3.7 Generating RF radiation profile within cellular base stations

The Electromagnetic radiation profiles within Base stations were generated using

Telecommunication Expense Management System (TEMS) Pocket positioning tool. The

tool consists of

i. TEMs handset (e.g. Ericsson TEMs phones complete with charger, data cable, & USB

hub)

ii. Laptop installed with TEMs 14 investigation software

 38

iii. GPS with external antenna and data cable

Fig 3.9 shows the Telecommunication Expense Management System (TEMS) drive test kit

used to record the signal profile within a cellular base station.

3.7.1 Functionalities of TEMs drive test kit

TEMS drive test kit has capability of collecting and analyzing the following data during

measurements

i. Signal field intensity (Rx levels and RSCP for 2 G and 3G respectively)

ii. Energy to Noise (EcINO) ratio

iii. Receive Signal Strength Indication (RSSI)

iv. Quality of Service (QoS) information etc.

In this study the TEMs drive test kit wa used to measure the Signal field intensity, analyze

and generate the radiation profile within the base station. The system used had

i. Measurement resolution of 0.01 dB

ii. Measurements accuracy of ± 1.0 dB (typical)

iri, Acquisition time of 20 milliseconds (Typical: i.e. about 50 samples are measured each

second).

3.7.2 Procedure for generating RF radiation profiles

Fig 3.10 illustrates the procedure used to generate the electromagnetic levels pattern within

the cellular base stations. MRU (mobile receive unit) and MCG (mobile call generator) are

the Ericsson TEMS phones.

 39

A call was made to mobile receive unit (MRU) by a mobile call generator (MCG). When

MRU receives a call, it answers automatically and starts to playa Perceptual Evaluation of

Speech Quality (PESQ) speech sentences that last for 1 hour. During the 1 hour period

TEMS pocket positioning tool records the downlink data in TEMS investigation log file as

one perform a drive /walk test. The profiles recorded are presented in section 4.4.

TEMS G
p
Investigation
S

Fig 3.10 Generating RF radiation profile within a cellular base station

3.8 Technical Challenges for RF radiation measurements

Measurement of the emissions from cell towers presented particular technical challenges.

Protocols for the measurement of RF energy for the purpose of human exposure

assessment often recommend the use of an "isotropic broadband probe" because this type

of sensor responds equally to energy arriving from any direction, and over a broad

frequency range, as does the human body. These instruments were used because they

permit a quick and simple measurement. Unfortunately, the "broadband" characteristic of

the isotropic broadband probe means that it measured any signals across a wide range of

frequencies. The readings produced by the instrument were a combination of all signals

present. In a large number of cases, the other signals present near the selected cell tower

were as strong as the cellular signals from the selected base station being measured.
 40

One of the most significant RF measurement problems, and one responsible for some of

the greatest inaccuracy, involves an instrument erroneous response that could have

occurred when two or more strong signals were present at the same time. A very large

proportion of antenna sites (cell and broadcast) have multiple strong signals. Instrument

design can minimize this problem, but many of the commonly used isotropic broadband

meters perform very poorly in this multi-signal environment. The result is a reading that

is much higher than actual.

A spectrum analyzer was used for identification of RF sources, and for assessment of the

relative magnitude of signals in different frequency ranges. The use of this instrument

with a calibrated antenna allowed a sensitive and precise "channel power measurement"

across selected frequency ranges, or measurement of the strength of an individual signal;

however the spectrum analyzer used could not identify the direction of the source of the

radiation especially in areas with a high population of base stations.

 41

CHAPTER FOUR

DATA COLLECTION, ANALYSIS AND INTERPRETATION

4.1 Introduction

Mobile telephone communication signals are both transient and partly random in their

occurrence and distribution. In this context, the study aimed at determining the radio

frequency radiation levels at some selected cellular base stations within Nairobi and

evaluates the RF levels against the maximum exposure limits set by ICNIRP 1999.

Measurements were taken for frequencies within the GSM900, GSM1800 and UMTS2100

bands. Relevant spectrum analyzer data were recorded automatically and subsequently

analyzed. Measurements were performed within ail the sectors available, at various

locations near the base station, by continuously scanning the frequency bands and logging

the signal level for the cellular base station. Where more than one carrier (Safaricom,

Airtel, Orange, and Essar) were detected combined activity from all carriers was

determined. Measurements were taken at various di tances ranging from 5 m to 350 m.

Two factors were considered when choosing the distances i.e accessibility and how far the
I'

base station in consideration was from the neighboring base station. In most instances the

sites were randomly but selectively chosen focusing on the population density,

concentration of the base stations and the choice was also made to capture RF emissions

from all the licensed cellular operators in Kenya and possible categories of base stations.

The RF radiation levels of cellular base stations attributable to the major carriers were

investigated, although not all were pre ent at each base station. There were three categories

'of base station considered i.e. Greenfield, Rooftop and Indoor Micro. For the Greenfield
 42

and Rooftop the base station antennae were mostly positioned on towers between 20 to 60

. meters in height. Most commonly, directional panel antennae were used. Such antennae

divide the area around the base station into three sectors, with each sector usually covering

a 120 degree arc in azimuth. Hence, three sets of antennae are normally in use at a typical

base station.

Indoor Micro Base stations were found to employ Omni-directional antennae at site

OOl(Sarit centre) and site 016 (Tuskys supermarket) base stations surveyed. An Omni-

directional antenna provides 3600 coverage of the area serviced by the base station. A fixed

antenna (Probe) was mounted on the spectrum analyzer and automated mobile

measurements were made whilst walking around the GSM base station under survey. Both

signal data and position information of the base station [using Global Positioning System

(GPS)] were recorded. For each sector measurements were made of all active frequencies

at each particular mobile measurement sample location.

4.2 Electromagnetic radiation measurements results

The RF radiation was measured for the four mobile phone operators i.e. Safaricom Kenya

Ltd, Airtel Kenya, Orange and Essar Kenya in sixteen different cellular base stations. All

the base stations measured were within Nairobi and environ. In addition to the radiation

levels measured, the location of each base station was taken using a GPS. The

measurement results recorded by the spectrum analyzer were presented as spectral scans.

The radiation levels in the spectral scans are recorded in dBm. Since the ICNIRP

Maximum Public Exposure (MPE) limits are expressed in W/m2, the power density in
 43

mW/m2 were calculated and tabulated as shown for each base station below. For the intents

and purposes of this study spectral scans for the three categories of base stations (i.e.

Greenfield, rooftop and indoor micro) are presented only for three base stations (Sarit

center, Kilimani and Aga Khan Hospital).

4.2.1 Measurements for Site 001 (Sarit Centre)

These measurements were taken at The Sarit Centre in Westland, Nairobi within a

building. The base station coordinates are 01°15' 41.16" S, 36° 48' 06.85 "E. All

measurements were taken 2 meters above the ground and 3m to 5m from the antennas on

the ceiling using spectrum analyzer model FSH 18-100598 in peak mode. The following

figures are examples of Spectral scans recorded from an indoor cellular base station at Sarit

Center. Figures 4.1, 4.2 and 4.3 show spectrum scans with the highest RF radiation levels

for GSM 900, GSM 1800 and UMTS 2100 downlink bands recorded at Sarit Center.

Fig 4.~: GSM 900 Spectral scans taken from 001 (Sarit Centre)
 44

The x-axis of the spectral scan represents the frequency range in the band with the start and

stop frequency indicated at the bottom of the scan. The y-axis of the scan indicates the

radiation levels detected by the spectrum analyzer. The highest levels in each band are

indicated by markers Ml, M2, M3 and M4 respectively in the scan. For example the

highest level in GSM 900 is -6.52 dBm at 958 MHz (marker Ml Fig 4.1). In GSM 18.00

the highest level is -21.18 dBm at 1833.5 MHz (marker Ml Fig 4.2). In UMTS 2100 band

the highest recorded level is -25.18 dBm at 2157.5MHz (marker MI Fig 4.3).

Fig 4.2: GSM 1800 spectral scans taken from 001 (Sarit Centre)
I'

Fig 4.3: UMTS 2100 spectral scans taken from 001 (Sarit Centre)
 45

Table 4.1 represents the levels in dBm and in m W1m2 recorded at each floor of the building

in Sarit Center.

Table 4.1 Measurements results from site 001(Sarit Center)

Floor Peak Cellular Measured Measured Maximum ExactPOM ICNlRP
freq Operator levels power density exposure Limits
MHz (dBm) (W/ml) (W/ml) w/m1
Gnd 938 Safari com -13.07 3.05 mW/m2 2 m above the 4.5 w/m2
Floor 942 Safaricom -13.81 2.58 mW/m2 3.05 mW/ m2 floor and 3 to 4.5 w/m2
953 Airtel -29.30 74.66 ~W/m2 . 5 m under the 4.5 w/m"
960 Airtel -31.88 41.83 JlW/m2 antenna 4.5 w/m2
1825 Orange 1.45 325.55 mW/m2 325.55 mounted on 9w/m2
1819 Airtel -18.92 2.96mW/m2 m'W/m" the ceiling 9w/m2
1835 Safaricom -23.06 1.17 mW/m2 9w/m2
2127 Airtel -18.17 7.65 m'W/m" 7.65 mW/m2 10 w/m2
2163 Orange -21.72 2.21 mW/m2 10 w/m2
First 938 Safaricom 1.15 80.3i m W/m" 80.31 mW/m2 4.5 w/m2
Floor 2m above the
942 Safaricom -2.64 33.8 JlW/m2 4.5 w/m2
floor and 3 to
1834 Orange -40.06 23.21 IlW/m2 23.21 JlW/m2 9w/m2
1832 Orange -41.67 15.98 JlW/m2 5m under the 9w/m2
1872 Safari com .42.10 15.10 JlW/m2 antenna 9w/m2
1842 Safari com -45.17 7.22 ).1W/m2 mounted on 9 w/m2
2129 Airtel -32.37 183.78JlW/m2 183.78 the ceiling 10 w/m2
2116 Safaricom -38.85 40.83 JlW/m2 JlW/m2 10 w/m2
2164 Orange -39.45 37.19 ~iW/m2 10 w/m2
-
2nc 938 Safaricom -14.43 2.22 mW/m2 4.5 w/m2
Floor 942 Safari com -17.03 1.23 .mW/m2 2.22 mW/m2 4.5 w/rn!
f--. 2m above the
945 Orange -17.54 1.10 mW/m2 4.5 w/m2
floor and 3 to
958 Airtel -38.00 1.02 mW/m2 4.5 w/m2
5m under the
929 Essar -39.78 6.35 ).1W/m2
antenna
I'
1825 Orange -12.87 12.04 mW/m2 12.04 mW/m2 . mounted on 9w/m2
1815 Airtel -15.11 7.12 mW/m2 the ceiling 9 w/m2
1835 Orange -39.11 28.92 JlW/m2 9 w/m?
2126 Airtel -21.14 2.43 mw/rn" 2.43 mW/m2 10 w/m2
2118 Safaricom -3i .64 215.18 !lW/m 2 10 w/nf
2114 Safaricom -32.37 181.18 ).1W/m2 10 w/rrf
2159 Orange -35.10 100.80 ).1W/m2 10 w/m2
938 Safaricom -30.80 51.216 JlW/m2 5l.216 JlW/m 4.5 w/m2
942 Safaricom -31.68 37 ..4 IlW/m2 4.5 w/m2
944 Safaricom -32.20 38.2 ).1W/m2 2m above the 4.5 w/rn"
3rd .
953 Airtel -40.14 6.1?yW/m2 floor and 3 to 4.5 w/m2
Floor 1825 Airtel -27.32 421.2 JlW/m2 42l.2 JlW/m2 5m under the 9w/m2
3rd 1835 Orange -34.56 82.46 JlW/rn" 9 w/m'
antenna
Floor 1838 Safari com -53.37 1.09 ).1W/m2 mounted on 9w/m2
2127 Airtel -28.43 454.45IlW/m2 454.45).1W/m2 the ceiling 10 w/m2
2118 Safaricom -32.02 197.15IlW/m2 10 w/m2
2112 Safaricorn -32.23 186.79 ).1W/m2 10 w/m2
2159 Orange -46.42 7.44 J,.lW/m2 10 w/m2
 46

Table 4.2 summarizes the highest exposure levels at the ground floor of Sarit center

radiated by each cellular operator in each frequency band.

Table 4.2: Ground Floor maximum RF Radiation for site 00l(Sarit Center)
FLOOR RADIATION
Freq range Safaricom mxm Airtel mxm level Orange mxm level
(MHz) level
Ground 825-975 3.05 rnw/m? 74.66 ~W/m2 Undetected
1725-1875 1.17 mW/m2 2.96 mW/m2 325.55 mW/m2
2025- 2175 Undetected 7.65 mW/m2 2.21 mW/m2

The radiation measurements taken from site OOl(Sarit Center) at different frequency bands

indicate a number of variations among the three cellular phone operators. For instance at

the GSM900 frequency band, Safaricorn had the highest radiation measurement of 3.05

mW/m2 at 938 MHz.

In the GSM1800 band Orange had the highest radiation measurement of325.55 mw/m- at

1825 MHz, followed by Airtel with 2.96 mW/m2 at 1819 MHz and Safaricom had the

lowest peak radiation measurement of 1.]7 mw/rn" at 1835 MHz. The highest level at

ground floor was attributed to the close proximity of the radiating antenna to the point of

measurement and the electromagnetic waves may have been scattered by walls and other

obstacles thus creating high radiation levels away from the base station antennas.

In the UMTS2100 band Safaricorn RF radiation was not detected while Airtel had the

highest RF radiation measurement of 7.65 mw/m? at 2127 MHz followed by Orange with

2.21 rnw/m" at 2163 MHz. It is clear from the above findings that each of the cellular

phone operators had at least one of the points where it had an RF radiation higher than the

rest.
 47

Table 4.3: First floor maximum RF radiation for site 00l(Sarit Center)

FLOOR RADIATION

Freq (MHz) Safaricom Airtel mxm Orange mxm Essar mxm

mxm level level level level
First 825-975 80.31 mW/m2 Undetected Undetected Undetected

1725-1875 I5.J 0 11W1m2 r Undetected 23.211lW/m2 Undetected

2025- 2175 40.831lW/m2 183.781lW/m2 37.191lW/m2 Undetected

From the Table 4.3 it is evident that in the GSM900 band it is only Safaricom radiation of

80.31 mW1m2 that was detected at 938 MHz in the first floor. There were no radiations

from Airtel and Orange respectively detected.

Still on the first floor of site 00 I in the GSM 1800 frequency band, radiations from two

cellular phone operators were detected: Safaricom with strength of 15.10 11W1m2 at 1872

MHz and Orange with strength of 23.21 IlW/m2 at 1834 MHz. In this frequency band

Orange had the highest observed trength of radiation standing at 23.21 IlW/m2 whereas

Airtel radiation could not be detected.

In the UMTS 2100 frequency band the rad iations from all the three mobile phone

operators were detected. The highest radiation was that from Airtel with the strength of

183.78 IlW/m2 at 2129 MHz followed by Safaricom with radiation strength of 40.83

11W/m2 at 2116 MHz and finally Orange with a recorded radiation of 37.19 11W/m2 at 2164

MHz.
 48

Table 4.4: Second and third floor maximum RF radiations for site 001 (Sarit Center)
FLOOR RADIATION
Freq range Safaricom mxm AirteJ mxm Orange mxm Essar mxm
(MHz) level level level level
Second 825-975 2.22 mW/m2 1.02 mW/m2 1.10 mW/m2 6.35 ~W/m2
1725-1875 Undetected 7.12 mW/m2 12.04 mW/m" Undetected

2025- 2175 215.18 ~IW/m2 2.43 mW/m2 100.80 ~W/m2 Undetected

Third 825-975 51.216 ~tW/m2 6.15 ~W/m2 Undetected Undetected

1725-1875 1.09 ~W/m3 421.2 ~W/m2 82.46 ~W/m2 Undetected

2025- 2175 197.15/lW/m2 454.45 ~W/m2 7.44 ~W/m2 Undetected

It is evident from table 4.4 above that in second floor in the GSM 900 band Safaricom had

the highest radiation with the power density of 2.22 mw/m" at 938 MHz while radiation

emission levels was lowest for Essar with peak radiation of 6.35 /..l W/m2 at 929 MHz. In

the GSM 1800 band, only radiations from Airtel and Orange with power densities of 7.12

mW/m2 and 12.04 mW/m2 at 1815 MHz and 1825 MHz respectively were detected,

whereas in the GSM2100 band Safaricorn, Orange and Airtel radiations were detected with

Airtel having the highest peak radiation of 2.43 mW1m2 and Safaricorn the lowest peak

power density of215.18 /..lW/rn2 at 2126 MHz and 2118 MHz respectively.

4.2.2 Radiation measurements for Site 002 (Kilimani)

The RF radiation measurement was then taken at site 002 (Kilimani) with the following

location coordinates 010 IT 22.1" S, 36° 46' 53.7 "E. Site 002 which is a Greenfield kind

of base station located along Kilichwa Road in Kilimani Nairobi. The measurements were

taken 5 m away from the base station in the land lord's premise 50 m and 150 m points of

measurements inside the nearby residential flats. The findings are illustrated and explained

below. Figures 4.4, 4.5 and 4.6 show spectrum scans with the highest RF radiation levels
 •

49

for GSM 900, GSM 1800 and UMTS 2100 downlink bands recorded at site 002

(Kilimani). The highest level in GSM 900 was -13.78 dBm at 942 MHz (marker Ml Fig

4.4). In GSM 1800 the highest level was -30.72 dBm at 1836.5 MHz (marker Ml Fig 4.5).

In UMTS 2100 band the highest recorded level was -19.87 dBm at 2111.3 MHz (marker

Ml Fig 4.6).

Fig 4.4: GSM 900 Spectral scans taken from site 002 (kilimani)

Fig 4.5: GSM 1800 Spectral scans taken from 002 (Kilimani)
.
 50

Fig 4.6: UMTS 2100 spectral scans taken from 002 (Kilimani)

Table 4.5 represents the levels in dBm and in mW/rn" recorded at each distance from the

Kilimani base station while table 4.6 summarizes the highest exposure levels at each

distance from the base station radiated by each mobile phone operator in each frequency

band.

According to the findings in the above table 4.6 above , at 5 meters in the land lords

premise it is clear that Safaricom had overall highest radiation level in all the three

categories of cellular frequency bands. In GSM900 band Safari com had the highest RF

level of 6.96 mW1m2 followed by Airtel with 2.39 m W1m2 at 942 MHz and 960 MHz

respectively both at 50m away from the base station. There was no trace of radiation from

Orange and Essar mobile operators in the areas considered. Safaricom also had the highest

power densities in both GSM 1800 and UMTS 2100 frequency bands with levels of 1.31

mWzm? and 5.09 mWzrn'' at 1839 MHz and 2111 MHz respectively. Airtel trailed

Safaricom in both cases. It is also important to note that at 50 m in the GSM 1800
 51

frequency band only Safaricom radiation that was detected. All the others could not be

traced.

Table 4.5 Measurements results from site 002(Kilimani)

Dist. Peak Mobile Measured Measured Maximum Exact ICNIRP
From freq Operator levels power density exposure POM expo ure
Mast MHz (dBm) (W/ml) (W/m2) Limits
942 Safaricom -11. 78 4.12 mWm2 4.12 mWm2 Land 4.5 w/m2
939 Safari com -12.81 3.23 mW/m2 lords 4.5 w/m2
960 Airtel -22.26 365.84/-!W/m2 premise 4'.5 w/m2
958 Airtel -22.95 325.60 ~lW/m2 4.5 w/m?
1836 Safaricom -28.72 316.74I1W/m2 316.74 9w/m2
5m 1839 Safaricom -32.96 119.71I1W/m2 I1W/m2 9w/m2
1843 Safaricom -37.47 42.56I1W/m2 9w/m2
1819 Airtel -38.96 29.42I1W/m2 9w/m2
2111 Safaricom -17.87 5.09 rnW/rn" 5.09 mW/m2 10 w/m2
2128 Airtel -21.26 2.19 mW/m2 10 w/m"
2118 Safaricom -22.23 1.89 mW/m2 10 w/m2
942 Safaricom -9.50 6.96 mW/m2 6.96 mW/m2 Residentia 4.5 w/m2
939 Safaricom -11.53 4.33 mW/m2 I flats 4.5 w/m2
937 safari com -13.64 2.66 mW/m2 4.5 w/m2
945 safaricom -13.95 2.52 mW/m2 4.5 w/m2
960 Airtel -14.32 2.39 mW/m2 4.5 w/m2
952 Airtel -14.85 2.08 mW/m2 4.5 w/m"
50m 1837 safari com -22.57 1.31 mW/m2 1.31 mW/m2 9w/m2
1839 safaricom -25.37 685.43/lW/m2 9 w/m2
1841 safaricom -27.26 445.50I1W/m2 9 w/m2
2127 Airtel -21.68 2.15 mW/m2 2.15 mW/m2 10 w/m2
2112 safari com -27.12 605.82 JlW/m2 10 w/m2
2118 safari com -27.24 592.66 llW/~ 10 w/m2
I'
958 Airtel -16.97 1.32 mW/m" 1.32 mW/m2 Residentia 4.5 w/rn"
941 safari com -24.68 210.92I1W/m 2
I flats 4.5 w/m"
938 safaricom -26.68 132.18/!W/m2 4.5 w/m2
944 safaricom -30.29 58.33I1W/m2 4.5 w/m2
1837 safari com -23.S3 1.05 mW/m2 1.05 mW/m2 9w/m2
1841 safaricom .32.01 149.30 f.1W/m 2 9w/m2
200m 1819 Airtel -30.29 215.8711W/rn2 9w/m2
1816 Airtel -40.53 20.49I1W/m2 9 w/rn"
2128 Airtel -23.93 1.28 mW/m2 1.28 mW/m2 10 w/m2
2116 safari com -27.15 603.93I1W/m2 10 w/m2
2113 safaricom -27.27 585.81IlW/m2 10 w/m2
 52

Table 4.6: Maximum RF radiation for site 002 (Kilimani)

DIST.(M) RADIATION
Freq range Safaricom AirteJ rnxrn level Orange Essar rnxm
(MHz) mxm level mxm level level
825-975 4.12 mWIl12 0.366 mW/m2 undetected undetected
5m 1725-1875 0.316 mW/m2 29.42 I-lW/rj1z undetected undetected
2025- 2175 5.09 rl1W/m~ 2.19 mW/m~ undetected undetected
825-975 6.96 mW/m= 2.39 mW/m2 undetected undetected
50m 1725·1875 1.31 mW/ll1z undetected undetected undetected
--~
2025- 2175 0.606 JlI W 1m3' 2.15 mW/mz undetected undetected
825-975
- 0.211 mW/m; 1.32 mW/m'z undetected undetected
200111 1725-1875 1.051l1Wltn2 0.216 mW/m2 undetected undetected
2025·2175 0.604 m W/m" 1.28 mW/m2 undetected undetected

In the distance of 200 m where measurements w re taken, it is evident that Airtel had the

highest radiations of 1.32 rnWltnl and 1.28 mW/m2 at 958 MHz and 2128 MHz

respectively. In both cases Safarieorn's highest radiation within the two frequency bands

was far much lower than those of Airtel, However in the GSM1800 frequency band,

afaricom had the highest power densiti s of 1.05 mW/rn" at iS3? MHZ compared to

Airtel' 215.87 ~ W/m2 at 1816 MHl. tt can also be observed that only afaricom and

Airtel radiations could be detected within the three distance of 5 m, 50 m and 200 m

re pectively in site 002. This means thai Lt ere was no nearby Orange and Essar cellular

base station.

4.2.3 Radiation measurements for 'ite 003 (Kayole)

This site is situated in Nairobi Kayole estate near a church, the ite location coordinates are

01° 16' 56.42" S, 36° 54' 32.9~·(E. In the areas around the base station where

measurements were taken there was no RF radiation detected from Orange, however all the

other three op rater wet' represented. Measurements were taken at 50 m and 200 m

points from the base station. At 50 m pint, mea urernents were taken in the church

compound. 200 m point taken at residential flats. It was not possible to take
 53

measurements at 5 m from the base station because there was no access to the area where

the base station is installed. The findings are illustrated and explained next.

Table 4.7 summarizes the highest exposure levels at each distance considered from the

Kayole base station radiated by each mobile phone operator in each frequency band.

Table 4.7: Radiation for site 003 (Kayole)

DIST.(M) RADIATION
Freq range Safaricom Alrtel Maxm Orange Maxm Essar Maxm
(MHz) Maxm Level level level level
50 m 825-975 4.24 mW/m2 81.60 ~Wlm2 Undetected Undetected

1725-1875 0.363 mW/m2 20.79 llW/m2 Undetected Undetected

2025- 2175 6.19 J,.tW/m2 Undetected Undetected Undetected

200m 825-975 0.155 mW/\11~ 24.92 J.l.W/m2 Undetected Undetected

1725-1875 0.398 mW/m2 Undetected Undetected 24.43 IlW/m~

2025-2175 0.700 mW/m2 15.08 J,lW/m2 Undetected Undetected

In the GSM900 frequency band at a distance of 50 m Safaricom had the highest peak

power density of 4.24 mW1m2 at 937 MHZ followed by Airtel with power density of 81..60

J.lW1m2 at 958 MHz, At the same point of measurements in the GSM 1800 frequency band
I'

Safaricom had the highest power density of 0.363 m'W/rn! at 1838 MHz while Airtel had a

peak power density of 20.79 J.l W1m2 at 1866 MHz. In the UMTS21 00 frequency band it

was only Safaricom RF radiation of 6.19 m W1m2 at 2 I 12 MHz which was detected at the

same 50 m point, At 200 m point of measurements, again Safaricom had the highest peak

radiation of 0.155 rnw/rn- at 937 MHz in the GSM 900 frequency band followed by Airtel

with power density of 24.92 J.lW/m2 at 958 MHz. Orange and Essar radiations were not

detected. In the GSM 1800 frequency band Safaricom radiation of 0.398 rnw/m" at 1837

MHz was the highest compared to the Essar radiation with power density of 24.43 J.lW1m2
 54

at 1805 MHz? In the UMTS 2100 frequency band only Safaricom and Airtel RF radiation

were detected with the power densi-ties of 0.700 m W/m" and 15.08 JlW/m2 at 2118 MHz

and 2127 MHz respectively. From the results it is evident that the radiations from the

nearest Orange base station were not detectable.

4.2.4 Radiation measurements at Site 004 (Aga Khan)

This site is Neighboring Aga khan Hospital. It is a rooftop type of facility. Measurements

were taken at 5 m from the base station on the rooftop of the building, 80 m point taken at

CGHU Secondary school, 200 m inside Aga khan Hospital. The site coordinates are 10 15'

_24.98" Sand 36048' 41.36"E.Fig 4.7, Fig 4.8 and Fig 4.9 shows spectrum scans with

the highest RF radiation levels for GSM 900, GSM 1800 and UMTS 2100 downlink bands

recorded at a rooftop base station at site 004 (Aga Khan). The highest level in GSM 900 is

-13.75 dBm at 949 MHz (marker Ml Fig 4.7). In GSM 1800 the highest level is -37.38

dBm at 1838 MHz (marker Ml Fig 4.8). In UMTS 2100 band the highest recorded level is

-25.94 dBm at 2158 MHz (marker Ml Fig 4.9).

Fig 4.7: GSM 900 spectral scans taken from 004 (Aga Khan)
 ss

Fig 4.8: GSM 1800 spectral scans taken from 004 (Aga Khan)

Fig 4.9:UMTS 2100 spectral scans taken from 004 (Aga Khan)

Table 4.8 represents the levels in dBm and in mW1m2 recorded at each distance from the

Aga Khan base station while table 4.9 summarizes the highest exposure levels at each
 56

distance from the base station radiated by each mobile phone operator in each frequency

band.

Table 4.8 Measurements results from site 004 (Aga Khan)

Dist. Peak Mobile Measured Measured Maximum ICNIRP Exact

From freq Operator levels power density exposure Exposure POM
Mast MHz (dBm) (W/m2) (W/m2) Limits
949 Orange -11.75 4.21 mw/m- 4.21 mw/m- 4.5 w/m? Rooftop 5m
931 Essar -15.91 1.56 mw/m" 4.5 w/m2 from the
935 Safari com -21.08 477.06 ~W/m2 4.5 w/m" mast
939 Safari com -24.37 225.571.1W/m2 4.5 w/m"
1838 Safaricom -35.38 68.49 uw/m? 68.49 uw/m! 9 w/m"
1871 Safari com -39.13 29.93 uw/m" 9 w/m"
Sm 1835 Safari com -40.65 20.29 uw/m" 9 w/m2
. 1844
2158
Safaricom
Orange
-45.52
-23.94
6.68 uw/m"
1.3;2 mW/m' 1.32 mW/m2
9w/m2
10 w/m"
2162 Orange -25.68 884.43 JlW/m2 10 w/m2
2112 Safaricom .45,60 I 8.60p.W/m2 10 w/m2
2128 Airtel -.::t632 7.39/.lW/m2 10 w/m'
946 Orange -17.14 1.21 mw/rn! 1.21 mw/m- 4.5 w/m2 In CGHU
949 Orange .24.07 246.881lW/m2 4.5 w/m- Secondary
936 Safaricom -32.34 35.76 ~W/m2 4.5 w/m! school,
1839 Safaricom -34.29 88.10 uw/rn" 88.1 0 uw/m" 9 w/rn?
1869 Belwestern -40.27 22.97 flw/m2 9w/m2
100m 1840 Safaricom -40.32 22.00 uw/m" 9w/m2
1870 Belwestern -40.29 I 22.89 J.1w/~2", 9 w/m"
2114 Safari com -32.70 167.94 uw/rn! 167.94 uw/rn" 10 w/m'
2118 Safari com -33.54 138.93 uw/rn? 10 w/m2
2159 Orange -43.89 J6.761lw/m2 10 w/m"
2164 Orange -44.06 12.86 uw/rn" 10 w/m"
950 Orange -9.04 7.88 mw/m! 7.88 mw/m- 4.5 w/m2 Inside Aga
926 Orange -20.44 542.22 JlW/m2 4.5 w/m- Khan
I'
947 Orange -21.40 - 454.62 JlW/m2 4.5 w/rn" Hospital
945 Orange -29.77 65.89 uw/m? 4.5 w/rn"
1839 Safari com -31.86 154.2 uw/m" 154.2 uw/rn" 9 w/m-
1840 Safari com -34.66 .~
81.02 uw/m? 9 w/m2
200m 1868 Belwestern -36.73 51.85 uw/rn" 9w/m2
1843 Safaricom -40.68 20.32 uw/rn? 9w/m2
2162 Orange -25.92 836.88 uw/m" 836.88 11W1m2 10 w/m"
2tl2 Safaricom -25.91 800.46 uw/m" 10 w/m"
2117 Safaricom -41.23 23.63 uw/m? 10 w/m?
 57

Table 4.9: Maximum RF radiation for site 004 (Aga Khan)

RADIATION
DlST. (1\'1) Freq Range Safaricom mxm Airtel mxm Orange mxm Essar mxm level
(MHz) level level level
5m 825-975 0.478 mW/m2 Undetected 4.21 mW/m2 1.56 mW/m2

1725-1875 68.49 JlW/m2 Undetected Undetected Undetected

2025- 2175 8.60 ).lW/rn2 7.39 ).lW/m2 1.32 mW/m2 Undetected

100 m 825-975 35.76 llW/m2 Undetected 1.21 mW/m2 Undetected

1725-1875 88.10 ).lW/n;2 Undetected Undetected Undetected

2025- 2175 0.168 mW/m2 Undetected 16.76 llW/m2 Undetected

200m 825-975 Undetected Undetected 0.542 mW/m2 Undetected

1725-1875 0.154 mW/J112 Und tected Undetected Undetected

2025- 2175 0.800 mWltn2

- Undetected 0.837 mW/m2 Undetected
I -~ I

In GSM 900 band the highest radiation level was from Orange with a power density of

4.21 mW/m2 at 949 MHz within a distance of 5m from the base station. Within the same

distance Safaricorn registered 0.477 mW/m2 9.t 935 MHz, 1.56mW/m2 radiation of Essar

was detected at 93\ MHz,. irtel was not detected.

I'

In the GSM 1800 band only Safarlcorn was detected in all the distances from the base

station with the highest power density of 0,154 mW/m2 at 1839 MHz within a distance of

200 m from the base station. In this band Rf radiation rrom other operators were not

represented. In the UMTS 2100 band -a aricorn, orange and Airtel were represented with

Orange having tho highest power density of 1.32 mW/m1 at 2159 MHz within a distance of

5 m from the base station. Safaricom highest power density in this frequency band was

0.80 mW/m" at 2112 MHz within 200 m from th base station. In the same frequency band
 58

Airtel had the lowest radiation level of7.39 IlW/m2 at 2128 MHz within 5m away from the

base station.

4.2.5 Radiation measurements for Site 005 (Jamuhuri)

This is a Green Field type of facil ity with coordinates of 010 18' 2.41" Sand 360 46'

24.94"E. This base station is located in Jamuhuri high school compound. 50 m and 100 m

measurement points were taken at the neighboring residential flats while the highest levels

at the 200 m point were at the school compound. Table 4.10 summarizes the highest

exposure levels at each distance considered from the Jamuhuri base station radiated by

each mobile phone operator in eacl frequency band.

Table 4.10: Maximum RF radiation for site 005 (Jarnuhuri)

RADIATION
DIST. Freq Range Safaricom mxm Airtel mxm Orange mxm Essarmxm
(M) (Mhz) levels levels levels levels
50m 825-975 2.34 niW/ml 22.68 m W/m" Undetected 4.1 mW/m2

1725-1875 28.89 mW/m2 Undetected 0.653 mW/m2 Undetected

2025- 2175 2.55 mW/m2 Undetected 0.551 mW/m2 Undetected

100 m 825-975 0.468 m Wzrn" Undetected Undetected 31.26 m'W/m"

I'

1725-1875 0.233 mW/m' Undetected 6.73 mW/m2 Undetected

2025- 2175 6.63 mW/m' Undetected 1.42 mW/m' Undetected

200 m 825-975 5.96 ~W!mZ 7.33 ~W/m2 Undetected 0.834 mW/m2

1725-1875 2.87 uWhn2 Undetected 0.524 mW/m2 Undetected

2025- 2175 21.05 f.LW/m' Undetected 12.94 f.LW/m' Undetected

The results in Table 4.10 shows that Airtel had the highest power density of 22.68 mW/m2

in the GSM 900 band at a distance of 5 m from the base station followed by Essar with

power density of 4.1 mW/m2 the same distance at 958 MHz and 930 MHz respectively.
 59

Orange had the highest radiation level of6.73 rnw/m" in the GSM 1800 band at 1834 MHz

100 m away from the base station, while Safaricom had the highest power density of 6.63

mWzrn"in the UMTS 2100 band at 2119 MHz at the same distance.

4.2.6 Radiation measurements for Site 006 (Olmeut)

'These measurements were taken from a base Transceiver station situated in an area near

Ongata Rongai. This site lies next to Nairobi National park in an upcoming Olmuet

residential estate. All the measurement points apart from 300 m point were taken in the

residential estate. The 300 m point was taken at a nearby shop. The base station

coordinates are: 01° 23' 26.8 S, 36° 47' 40.7" E.

Table 4.11: Maximum RF radiation for site 006 (Olmeut)

J>~IATION
DIST.(M) Freq range Safaricom Airtel mxm Orange mxm Essarrnxm
(MHz) mxm level level level level
501 825-975 30.8 ~lW/m2 0.51 ~W/m2 Undetected Undetected

1725-1875 1.17 mW/m2 Undetected Undetected Undetected

2025- 2175 Undetected Undetected Undetected Undetected

I'

100m 825-975 24.50/iWfm2 Undetected Undetected Undetected

1725-1875 2.45 ~lWim2 Undetected Undetected Undetected

2025-2175 Undetected Undetected Undetected Undetected

209m 825·975 19.41 ~W/mz 0.132mW/m2 Undetected Undetected

1725·1875 1.17 ~WhU2 Undetected Undetected Undetected

~
2025- 2175 Undetected Undetected Undetected Undetected

300m 825-975 15.8 !1W/m2 I, Undetected Undetected Undetected

1725-1875 4.74/iW/m; Undetected Undetected Undetected

2025· 2175 Undetected Undetected Undetected Undetected

"
 60

Table 4.11 summarizes the highest exposure levels at each distance considered from the

Olmuet base station radiated by each mobile phone operator in each frequency band. It is

evident from the findings that Safaricom has the highest radiations in the site at various

distances from the base station. It is also clear that in the GSM 900 band the strength of the

radiation seems to be decreasing with an increase in the distance from the base station.

This is supported by the measurements for Safaricom as recorded at 5 m, 100 m, 200 m

and 300 m where the maximum radiation measurements are 30.8 IlW/m2, 24.5 IlW/m2, and

19.4 11W/m2 and respectively. Safaricom radiation is the most dominant followed by

Airtel. Orange and Essar RF levels were not detected. It is also important to note that

UMTS 2100 RF radiations were not detected; this is probably because 3G networks had

not been deployed by the time the measurements were done.

4.2.7 Radiation measurements for Site 007 (Ndumberi rd.)

The measurements were taken from a roof top mobile base station situated at the River

road- Ndumberi road junction Nairobi. The area adjacent to the base of the mast was being

used by lodging workers as a laundry area even though the antennae were less than 12

meters above their heads. The coordinates of the site are 010 16' 52.9" S, 360 49' 34.0" E.

Fig 4.10: Photograph of a rooftop base station in Ndumberi Rd

 61

Table 4.12 summarizes the highest exposure levels at each distance considered from the

Ndumberi Rd base station radiated by each mobile phone operator in each frequency band.

Table 4.12: Maximum RF radiation for site 007 (Ndumberi Rd)

RADIATION
DIST.( Freq Range Safaricom mxm Alrtel mxm Orange mxm Essar mxm
M) (MHz) level level level level
5m 825-975 2.82 mW/mz 1.00 mW/m2 3.16 mW/m- Undetected

1725-1875 0.95 mW/m2 Undetected 0.316 rnW/m" Undetected

2025- 2175 0.13 mW/m2 Undetected Undetected Undetected

144 m 825-975 0.15 mW/mz 6.42/lW/m2 Undetected Undetected

1725-1875 7.46/lW/m2 5.28 mW/m2 Undetected Undetected

2025- 2175 Undetected Undetected Undetected Undetected

222m 825-975 9.84 !!W/ml b.81 IlW/m2 Undetected Undetected

1725~1875 47.2IlW/m2 Undetected Undetected Undetected

2025- 2175 Undetected Undetected Undetected Undetected

At this base station, the highest radiations mea ured at 5 m distance was 3.16 mW/m2 at

948 MHz from Orange followed by Safaricorri's 2.82 mW/m2 at 937 MHz. Power density

of 1.00 mW/m2 from Airtel was detected at 952 MHz.

The highest radiation within a distance of 144 m was 5.28 mW/m2 at 1822 MHz emitted by

Airtel. No RF radiation from Orange and Essar were detected. It is worth noting that 30

radiation was only detected from Safaricom at a distance of 5m from the base station. The

amount of the UMTS 2100 (30) detected was 0,13 mW1m2 at 2117 MHz.
 62

4.2.8 Radiation measurements for Site 008 (Kitengela)

This base station lies in the moderately populated Kitengela town. 100m measuring points

was in the Nomad Hotel, 208 m and 300 m points were in the Residential flats. The

recorded base station coordinates are 01038' 41.4" S, 0360 57' 30.1" E.

Table 4.13 summarizes the highest exposure levels at each distance considered from the

Kitengela base station radiated by each mobile phone operator in each frequency band.

, Table 4.13: Maximum RF' radiation for site 008 (Kitengela)

RADIATION
DIST. Freq Range Safaricom mxm Airtel mxm Orange mxm Essar mxm level
M (MHz) level level level
100 m 825-975 4,1) mw/m" Undetected Undetected 0.598 mW/m2

1725-1875 3.38 mW/m2 U. detected Undetected Undetected

208 m 825-975 0.12 m'W/rn" 0.80 llW/m2 1.58 llW/m2 Undetected

1725-1875 1.17 mw/m' 1.67 mW/m< Undetected Undetected

300m 825-975 30.80 flW/m2 0.913 mW/m2 0.816 m W/m" Undetected

1725-1875 11.80 llW/m' Undetected Undetected Undetected

I'

Within 100m from the base station Safaricom had the highest radiation levels in both GSM

900 and GSM 1800. The levels recorded were 4.11 mW/m:il and 3.38 m W/m2 at 938 MHz

and 1844 MHz respectively. Essar radiation was detected at 931 MHz with 0.598 mW1m2

power density.

At 208 rn Airtel recorded the highest radiation of 1.67 mW/m2 at 1824 MHz. Radiation

. from Essar was not detected at all frequency bands within the same distance. Within 300m

from the base station the radiation detected from Safaricom was 30.80 /lW/m2, Airtel 0.913
 63

mw/m-and Orange 0.816 m'W/rn" at 937 MHz, 954 MHz and 946 MHz respectively all in

GSM900. In GSM 1800 only radiation from Safaricom was detected, other operators were

not detected. It is evident that there was no UMTS (3G) radiation detected from all the four

operators in the distance considered.

4.2.9 Radiation measurements for Site 009 (Ruai)

This base station is situated in Ruai near the Nairobi eastern bypass Kangundo road

junction. Measurements were made at several points where public are known to frequent.

All the measurement points were within the surrounding residential estate. It is a

Greenfield Cellular Base Transceiver station with the following coordinates 010 16' 33.42"

S, 36040' 30.94" E.

Table 4.14 summarizes the highest exposure levels at each distance considered from the

Ruai base station radiated by each mobile phone operator in each frequency band.

Table 4.14: Maximum RF radiation for site 009 (Ruai)

RADIATION
DIST.(M Freq Range Safaricom Airtcl mxm Orange mxm Essar mxm level
) (MHz) mxm level level level
150 m 825-975 0.J5 mW/rn2 Undetected 0.454 m'W/m" 1.18 mW/mz

1725-1875 1.34 mWhn2 Undetected 6.22 mW/m2 Undetected

300m 825-975 1.68 mW/m' 0.558 mW/m2 0.368 mW/in2 Undetected

The radiations from the above site how that Essar registered the highest radiation levels in

GSM900 of 1.18 m W/m" at 934 MHz within a distance of 150 m. At the same distance

there was no radiation detected from Essar in the GSM1800 band, Orange registered the
 64

overall highest radiation level of 6.22 m W1m2 in GSM 1800 band at the same distance from

the base station.

Radiation from Safaricom was the highest at a distance of 300m from the base station. The

maximum Safaricom radiation recorded level at this distance was 0.558 mW1m2 at 938

MHz followed by 0.558 m W1m2 from Airtel at 953 MHz.

4.2.10 Radiation measurements for Site Q10 (Kahuho)

This is a Greenfield Cellular Base Transceiver station lays in Kahuho Village in Kabete.

The coordinates recorded for this base station were 010 12' 48,44" S, 36 40' 30.94" E. Q

Table 4.15 summarizes the highest exposure levels at each distance considered from the

Kahuho base station radiated by each mobile phone operator in each frequency band.

Table 4.15: Maximum RF radiation for site 010 (Kahuho)

~~,., ,.,.,.-----
.. ."",,,,,,, "'~
RADiATION·
DIST. (M) Freq Range Safarirom mxm Airfel mxm" Orange mxm Essar mxm
(MHz) level level level level
150 m 825-975 0.123 mW/m2 2.01 f.4W/m2 Undetected Undetected
I'

1725-1875 L71111W/m; Undetected 2,46f.4W/m2 Undetected

194 J11 825-975 o.96I11W!il~ 0.18 mW7J11~ 0.43 mW/rn~ Undetected

I 1725·1875 Undetected .- Undetected 2.53 mW/m2 Undetected

I
335 J11 825-975 77.50 'Lt\A/lm2 0.80 ~W/m2 Undetected Undetected

1725·1875 Undetected I Undetected 1.24 ~lW/J112 Undetected

1..,

The highest radiation levels were recorded in two main areas at this base station. At this

base station, the deteriorating trend of the radiation strength as the distance from the origin

. increases is also evident. For instance Safaricorn has a radiation measurement of 0.123

mW/m2at 150 m, 0.96 mW/m:l at 194 In and 77.5 ~tW/m2at 335 m when taken in GSM 900
 65

band. This confirms that ideally the radiation continues to degrade as the distance increases

from the source.

No RF emissions were detected from Essar and UMTS 2100 in all the distance considered

at the time of measurements. This implies that there is nearby base station transmitting at

30 and Essar had neither deployed 2G nor 3G in the neighborhood.

4.2.11 Radiation measurements for Site 011 (Umoja)

This Base station lies in the densely populated Umoja estate in Nairobi town.

Measurements were taken at 175m and 280m within the densely populated Umoja estate It

is a Greenfield cellular Base Transceiver station with the coordinates of 010 16' 56.42" S,:

36053' 11.95" E.

Table 4.16 summarizes the highest exposure levels at each distance considered from the

Umoja base station radiated by each mobile phone operator in each frequency band.

T-able 4.16: Maximum RF radiation for site 011(Umoja)

~ RADIATION
DIST. Freq Range Safaricom Airtel mxm Orange mxm Essarmxm
(M) (MHz) mxm level level level level
175 m 825-975 0.79 mW/m2 0.63 IlW/m2 Undetected 0.265 m'W/m"
1725-1875 1.63 mW/m2 Undetected 1.23 IlW/m2 Undetected

280 m 825-975 0.36 mW/m2 0.35 mW/m2 0.17 mW/m2 Undetected

1725-1875 4.71 J,lW/m2 0.39 mW/m2 1.6001 W/m2 Undetected

The radiation measurements from Umoja also indicate that highest recorded power density

at GSM 900 was 0.79 mW/m2 from Safaricom. It was recorded at a distance of 175 m
 66

away from the Greenfield cellular Base Transceiver station. There was no recorded

radiation at GSM900 from Orange. At GSM 1800 Safaricom also registered the overall

highest radiation level of 1.63 m W1m2 at 1841 MHz within a distance of 175 m from the

base station closely followed by Orange with a level of 1.60 mW1m2 at 1827 MHz. This

highest level from Orange was measured at a distance of 280 m from the base station. No

radiation was detected for 3G from all the four cellular operators. The maximum level

taken was far much bellow the maximum exposure limit set by ICNIRP.

4.2.12 Radiation measurements for Site 012 (Kiandutu)

The radiation measurements from a Greenfield cellular Base Transceiver station based in

Kiandutu slum in Thika town w re also taken. Thika town is 42 kilometers from airobi

city and base station's coordinates are 0 I 08' 49.8" and 36057' 40.8" E. The findings are

illustrated and explained below.

Table 4.17 summarizes the highest »cposure levels at each distance considered from the

Kiandutu base station radiated by each mobile phone operator in each frequency band.
I'

Table 4.17: Maximum RF radlatlons from site 012 (Kiandutu)

RADIATION
DST. (M) Freq Range Safaricom mxrn Airtel Orange mxm Essar mxm level
(MHz) level mxmlevel level
102m 1725-1875 1.31 mW/n'i~ I Undetected 1.241J.W/m2 Undetected
i
211 m 1725-1875 1.18 flW/rni I Undetected 98.10 nW/m2 Undetected

300m 1725-1875 O.154ft1Wi\11.~ -1L

Undete ted 3.61 mW/ml Undetected
-

From the above measurements results it is evident that both Safaricom and Orange uses

GSM 1800 frequency band to provide mobile communi ation at the area around the base
 67

station. The highest overall Safaricom radiation had the power density of 1.31 mW1m2 at

1844 MHz compared to the Orange overall power density of 3.61 mWzrn" at 1829 MHz

within a distance of 102 m and 300 m respectively.

Kiandutu so far is the only base station where radiation from GSM900 is not detected.

UMTS 2100 was also not available during the time of measurement.

4.2.13 Radiation measurements for Site 013 (Kibiko)

The measurements were taken from a Greenfield cellular Base Transceiver station near

Ngong Township. Measurements were taken from two point's i.e. 100 m point lies in

residential estate while the 350 m point lies on the nearby earth road. The base station

coordinates are: 01021' 10.40" S, 360 38' 34,93" E.

Table 4.18 summarizes the highest exposure levels at each distance considered from the

Kibiko base station radiated by each mobile phone operator in each frequency band.
I'

Table 4.18: Maximum RF radiations from site 013 (Kibiko)

RADIATION
DIST. Freq Range Safaricom mxm Airtel mxm Orange mxm Essarmxm
(M) (MHz) level level level level
100 m 825-975 0.455 mW/m2 81.60 I!W/m~ 31.00 I!W/m~ Undetected

1725-1875 0.65 mW/m2 Undetected Undetected Undetected

350 m 825-975 97.lO I!W/m< O.4Q mW/m2 Undetected 17.00 I!W/ml

1725-1875 0.12 mw/m? Undetected Undetected Undetected

It is evident that, in the above Greenfield cellular base transceiver station at Kibiko,

Safaricom has the highest radiations. For instance Safaricom has the highest recorded

.radiation of 0.65 mW1m2 at 1838 MHz closely followed by Airtel with power density of
 68

0.40 m W1m2 at 952 MHz. Essar and Orange has the least radiation levels of 17.00 Jl W1m2

and 31.00 JlW/m2 at 933 MHz and 947 MHz re pectively. It is also evident that the

network coverage around this base station uses GSM 900 frequency band and GSM 1800

only.

4.2.14 Radiation measurements for Site 014 (Zimmerman)

This is a Greenfield mobile base transceiver station in the corner of the schools playing

ground. The school is located in the densely populated Zimmerman estate. Measurements

at the 150 m point was done in front of class 3E while the 100 m point was in the middle of

the playground. The coordinates of the base station are 0 I 0 12' 44.44" S, 360 53' 37.95" E

Table 4.19 summarizes the highest exposure levels at each distance considered from the

Zimman base station radiated by ach mobile phone operator in each frequency band.

Table 4.19: Maximum RF radiations from site 014 (Zimmerman)

,~~-
RADJATlON
DST. Freq Range Safaricom mxm Airtel mxm Orange mxm Essar mxm level
on (Mllz) level level level
150 m 825-975 0.0.66 mW/m2 1.38 mW/m~ Undetected 1.21 mW/m~ .
1725-1875 0.297 m'W/m" 2.81 mW/rn· 0.31 mW/m2 Undetected

200m 825-975 0.49Il1W/mz U ndetected 98 J,lW/m2 lfndetected

I 1725.i875 0.81 mWinl2 Undetected 781lW/m2 Undetected

At the above site, Safari com uses G 'M 900 and GSM 1800 bands with the highest power

density measured of 0.81 m W/m2 at 18 0 ~1Hz. Orange provide coverage in GSM 900 and

GSM 1800 with the highest p wer d nsity of 0.31 m WI! ,2 at 1878 MHz. Airtel radiation

wa the overall highest both in GSM 900 and GSM 1800 with the power density of 1.38

I11W/m2 and 2.81 mW/m2 respectively,

 69

4.2.15 Radiation measurements for site 015 (Kijabe)

Some measurements were also taken from site 015 situated at Kijabe hilltop. The base

station coordinates are 0° 57' 38.2" Sand 36°49'36.84" E. Kijabe is a few kilometers

distance from Nairobi city. The measurements were taken at point of population 206 m

from a Greenfield cellular base Transceiver station at the hilltop. Findings are illustrated

and explained below.

Table 4.20 summarizes the highest exposure levels at each distance considered from the

Kijabe base station radiated by each mobile phone operator in each frequency band.

Table 4.20: Maximum RF radiation for Site 015 (Kijabe)

DIST. Freq Range

ISafaricorn mxm Airtel mxm
RADIATION
Essar mxm
Orange mxm
(M) (MHz) level level level level

206m 825-975 1.30 mw/rn" 0.83mW/m" . Undetected Undetected

1725-1875 0.32 mW/m< Undetected Undetected Undetected

~ ..•..,...

This base station lies off Kijabe Township in an extremely steep and bushy terrain within
I'

the escarpment. Only one metering point was available. It is evident from the above

findings that Safaricom had the highest radiation at Kijabe with power density of 1.30

mW/m2 at 959 MHz while Airtel had the lowest detected peak level of 0.83 mW/m2 at 957

MHz.

At the same distance of 206 m in GSM ] 800 only Safaricom was detected at 1837 MHz

with the power density of 0.37 /lW/n? In all the frequency bands no RF radiations were

detected from Orange and Essar at this point.

 70

4.2.16 Radiation Measurements for site 016 (Tuskys Supermarket)

This site is based at Tuskys Supermarket (Pioneer house) with coordinates of 1°17' 18.11"

S, and 36° 49' 36.84"E. It is an indoor micro type of facility situated in the busy Nairobi

CBD area. All measurements were taken at 2 meters above the ground and 3m to 5m from

the antennas on the ceiling.

Table 4.21 summarizes the highest exposure levels at floor of the Tuskys base station

radiated by each mobile phone operator in each frequency band.

Table 4.21: Maximum RF radiation from Site 016 (Tuskys Supermarket)

RADiATION
Floor Freq Range I Safaricom Airtel mxm Orange mxm Essar mxm
(MHz) mxm level level level level

I 10.18I1W/m" 1.56 mW/m4

-F.!
God floor 825-975 2.22 111W/m" 1.10 mW/m'
1725-1875 Un'! '''''ted; (mW/n>' ]2.Q4 mW/m4 Undetected
2Q25-2175 0.26 mW/tn 2.43 mW/m.! Undetected Undetected

1st floor 825-975 3.04mW/nl 74.67I1W/ml Undetected 2.12 mW/m"

1725-1875 1.16 mW/mz 2.96 tnW/m4 Undetected Undetected

2025- 2175 Undetected 7.65 mW/m- 2.20 mW/m2 Undetected

, -~ .•~

This base station is an indoor BTS situated in the Tuskys pioneer sup rmarket in Nairobi

CBD. The measurements taken at Tuskys also indicate the same trend of deteriorating

radiation levels as the distance from the origin increases. For instance at the first floor

where the facility is, the radiation is 3.04 m'W/m" and at the ground floor it is 2.22 m W/m"

This is a confirmation that the radiations actually reduce in strength a the distance from

the origin increases. Airtel is represented in all the frequency bands in both floors. Essar

has deployed GSM 900 only.

 71

4.3 Evaluation of measured results against the recommended ICNIRP levels

After getting the various measurements, the study sought to compare the radiation levels

with the ones recommended by the International Commission on Non-ionizing Radiation

Protection for exposure limitation. In general it can be observed that the radiation

measurements taken in Nairobi and environs are far much lower than the ones

recommended internationally. For instance in The Sarit Centre building where the highest

exposures were 325.55 rnW/m~ at 1825 MHz radiated by Orange, the maximum

internationally recommended standard e, posure is 9.0 W/mz. Converting the 325.55

gives a value of 0.3255 W/m'l,. This is 3.62% of the ICNIRP

recommendation for maximum public exposure. This is an indication that the radiations

considered to be the highest in the ground floor of Sarit centre building are far much lower

than the internationally recommended threshold.

For instance when we look at the graph in figure 4,11 illu trating the radiation

measurements taken from Aga Khan Hospital at a distance of 5m, all the mobile phone
I'

companies had radiations that are below 4.21 rnWzrn?compared to the 4.5 W/m2 limits that

is set by the International Commis ion on Non-ionizing Radiation Protection (ICNIRP).

This is an indication that the exposures are too low to cause harm to people. A close look

at another point in site 002 (Kilimani) with the highest radiation of 6.96 mW/m2 in GSM

900 compared to the recommended international standards shows the same trend as the one

mentioned above. The recommended international exposure under the same conditions is

4.5 W/m2. Converting 6.96 mW/m2 to W/m2 gives a value of 0.00696 w/m2 which is

approximately 0.] 5% of the MPE limit.

 72

In GSM 1800 the highest recorded RF radiation level was 1.31 mW1m2 from Safaricom at

1837 MHz this value translate to about 0.015% ofICNIRP limit of 4.5 W/m2.

In UMTS2100 band the highest recorded value was 2.19 mW1m2 which is approximately

0.022% oflCNIRP MPE limit of 10 W/m2. These values are still far much lower than the

internationally recommended exposure limits. This is an indication that the exposures

experienced within the base station are still safe and far much less harmful to human. For

instance the measurements taken at a distance of 200 m from the site for all the mobile

phone companies indicate that the exposures are very small and negligible as shown in

figure 4.12. It is clear from figure 4.12 that the local exposure is still at a very low level at

a distance of200 m. The international recommended maximum exposure is 4.5 W/m2.

Radiation levels

• • • •
• Radiation v'(/m2

___ Maximum w/m24.5

W/III:J

• • • •
Mobile operators

Fig 4.11: Comparison of Radiations from site (001) Aga Khan Hospital

Comparison of the radiations from Kijabe Hill also exhibits the same features of being

lower than what is internationally considered to be the highest exposure. The highest

exposure at GSM 900 is 1.30 m W/m2 The recommended highest exposure for such a
 73

distance of 206m is 4.5 W1m2 converting 1.30 mW/m2 gives a value of 0.0013 W/m2 This

is a very small and almost negligible figure that is unlikely to cause any significant harm to

people. In general, all the radiations that were measured at various points and locations

within Kijabe confirm that they are not high enough to measure up with the internationally

given thresholds.

Radiation levels

• • • •

w/m2
~Radiation w/m2

• • • •
Mobile operators

Fig 4.12: Comparison of Radiations from site 002 (Kilimani)

Figures 4.13, 4.14 and 4.15 show the graphical representation of maximum RF exposure in

GSM 900, GSM 1800 and UMTS 2100 respectively for the four main mobile operators in

Kenya. The horizontal axis represent sample base stations and vertical axis represent

maximum exposure levels at each base station for each mobile phone operator. For

example in GSM 900 band Safaricom radiated above 8 mW/m2 in Sarit center, Airtel and

Essar radiated above 8 mW1m2 at Jamuhuri . At Kiandutu no radiation was detected from

all the operators.

 74

GSM 900 exposure

•• • ••• . ••
• • • _SFC
. I
• L
• • L.
_ART

_ORG
_ ESR

Sample BTSSites

Fig 4.13: GSM 900 RF exposure

In figure 4.14 Orange radiated above 8 mW1m2 in GSM 1800 band at Sarit and Turskys

pioneer. RF radiation from all the other operators were bellow 8 mW1m2 in all the base

stations

GSM 1800 Exposure

.
•
• 1. •••• •
1 _SFC
I 1 I. I I III LI 1 • • 1 I
_ART

_ORG
_ ESR

Sample BTSsites

Fig 4.14: GSM 1800 RF exposure

 75

In figure 4.15 the radiation levels detected for the four main operators at all the base

stations were less than 8 mW/m2.

UMTS 2100 Exposure

_ 8
N

E
•....... 7
~ 6
.§. 5
III

~ 4
~ 3
III

~EI
:; 2
III
8. 1 • • I
~ 0
II I I

Fig 4.15: UMTS 2100 RF exposure

In general it is evident from the table 4.22 above that, all the RF radiation levels recorded

at yarious points and locations within Nairobi and it's environ had RF radiation strength

lower than 1% of the ICNIRP recommended exposure limits to the public. Only site 001

(Sarit center) that recorded above 1% of the ICNIRP recommendation. At Sarit Center

Orange registered 3.62% of the 9 W/m2 recommended by ICNIRP at GSM1800, however

this is still very low. The tabulated values confirm that the RF radiations are not high

enough to measure up with the internationally given thresholds. This is a confirmation that

the mobile phone companies are operating within the recommended limits of ICNIRP
 76

Table 4.22: Effective RF radiation exposure

FREQ Measured Highest Highe t ICNIRP %of

Base Station ID (MHz) levels levels power MPE Limits MPE
(dBm) (dBm) Density (W/m2)
(mW/m2)
001 (Sarit) 1825 -1.50 ±3 1.45 325.550 9.0 3.620

002 (Kilimani) 942 -12.50 ±3 ·9.50 6.960 4.5 0.150

003 (Kayole) 2112 .20.03 ±3 -17.03 6.180 10.0 0.062

004 (Aga Khan) 950 -12.04 =*:3 -9,04 7.880 4.5 0.175

005 (Jamuhuri) 958 ~7.52 ±3 -4.52 22.680 4.5 0.500

006 (Olmuet) 938 -16.07 ±3 -13.07 3.050 4.5 0.068

~
007 (Nduberi) rd) 1822 -19.43 ±3 .16.43 5.280 9.0 0.059
-
008 (Kitengela) 938 -14.75±3 -11.75 4.113 9.0 0.048
I
009 (Ruai) 1830 -2J.76±3 .18.76 6.218 9.0 0.06<)

010 (Kahuho) 1833 ·22.68 ±3 -19,68 2.530 9.0 0.028

OIl (Umoja) 1841 -24.63 :t:3 -21.63 1.630 9.0 0.018

012 (Kiandutu) 1829 -21.11 :t:3 -18.11 3.611 9.0 0.040

013 (Kibiko) 1838 -28,61 ±3 -25.61 0.649 9.0 0.007

014 (Zimmerman) 1821 -31.10 ±3 .19.07 2.873 9.0 0.032

01) (Kijabe) 939 -19.76 ±3 -16.76 1.300 4.5 0.029

016 (Tuskys 1825 -15.87±3 -12.87 12.030 9.0 0.134

pioneer)

4.4 Generating RF radiation profile within cellular base stations

A TEMs drive test kit described previously in section 3.7 was used to generate profiles for

two categories of base stations namely;

i. A cluster of Greenfield bas stations at Upper hill area in Nairobi

ii.An indoor base station in BP Shell building in Nairobi town center

 77

Figures 4.16, 4.18 and 4.19 show samples of 2G and 3G profiles generated at Upper hill

and BP Shell. The profiles give an idea of electromagnetic field distribution in the area

around the base stations.

c=Jt c
t
,~
~
t. -
T_ Key

f. o BTS location
'~Rd UPI"',Hal
\f. Railway Yard c ~
i l'"
" ~
1 N9on9Rd /~ I
I
% .,/
I '/oG ~
; \ I.~~'(}

\t -:
••••• 06
I
"..,2
1OiI-= \,
NgurnoEstat! SUn Vi
Estate
,
d

!
~
R",.NoIoobGollC>.b i\ 0
/r/~
\ t
•.••.•~
••••s;/'••

~,"
.,.eRd ~
.<'
Makll\a
Estate I r$'

Madaraka
Estate
Golfe"", ••
Estate
.i ~'"
~ 1

J I i
1
e
1
~
1

i
·• Below -95 (1) D.D%
>= -95 to < -85 (14-4) 2.6%
·• >= -85 to < -75 (2264) 4D.5%
·• >= -75 to < -65 (2487) 4-4.5%
·• Above -65 (697) 12.5%

Fig 4.16: 2G RF radiation profile from Upper hill clusters of base station.

From figure 4.16 only one sample had level below -95 dbm which is approximately 0% of

the total samples. 44.5% (2487) of the samples had registered levels between -75 dbm and

-65 dbm. Only 12.5% (697) of the samples had levels above -65 dbm
 78

RxLevSub Distribution

1200 ~----------------------------~~~~-----r
100%

90%
1000
80%

70%
800
60%

----------+ 50%

40%
400
30%

20%
200
10%

o 0%
-110 -106 -102 -98 -94 -90 -86 -82 -78 -74 -70 -66 -62 -58 -54 -50 -46 -42
dBm

Fig 4.17: Recorded levels against the count (samples)

I - I
o BTS location

2ndftoor_Eriooson UMTS _ ••I (...

• Ba_·105(O) 0.0"
• _.10510<·95(11)0.3'110
_·95 10<·85(1III!5jSO.1"
• --8510<·75(1013) ~.8"
__ .7510<.115(370) 111.7,.

• -'-'·115(151) 7.1"

15 ••

Fig 4.18: 2G RF radiation profile from first floor of BP Shell building (indoor BTS).

From figure 4.18: 45.8% (1013) of the samples had levels between -85 dbm to -75 dbm

and only 7.1% (158) of the samples had levels above -65 dbm.
 79

Key

o BTS location

•
•
-
lni ~-·I
'--'- •
--~ ~
"';;1-
po 'f!o

L..-"I--I~.~
r
••I
•
T _.f
I-
. ~
(
~

~_I...
c

•••,- 4
• :......-
"
-
e IL
._"

,k4'4
I
-
r-I
I

,
I I I I I I I I .;Iir
.•...
UU_Ac::tNoeS.~RSCP_O
• a.&o.w -105.00 (3730) 2••.. 7••. r--

- - .-~='~~r·~
.-.
-.. .~
s.& ...
• >- -105.00 to < -95.00 (!SB!!S)
.•.
-- .. ....
>- -95.00 to < -.815.00 (1915) 1.2.... ~
• >- -8.5.00 to < -75.00 (2.045) 1e ....
a_l.
• >- -75.00 to < -05.00 (11'3) 7.5"
·0 0: ....••. ~
~..iI ' ••• !~
• Abovoo -es.oe {OJ 0.0"' '
• Ii. t!.1
I
! ~ ~ J ! ~ J !
, ,. no
I

Fig 4.19: 3G RF radiation profile from first floor of BP shell building (indoor BTS)

From the above profiles it is evident that most of the higher radiation levels are close to the

base station antennas, however the highest levels may not necessarily be near the base

station, this is because the electromagnetic waves maybe be scattered by walls and other

obstacles thus creating high radiation levels away from the base station antennas.
 80

C APTER FIVE

SUMMARY CONCLUSION AND RECOMMENDATIONS

5.1 Summary and Conclusion

The main purpo e of this study was to rovid data on the RF radiation emission levels at

selected cellular bas stations and e -aluate th r suits against the required limit for non-

occupational xpo ure with respect t the ICNIP . tandard for R xpo ure 1998. The

results clearly demcnstrat d that the RF emlsslena from the cellular base stations wer

several orders of magnitude below th lit it spec lfled by ICNJRP ..

In the GSM 900 Safaricorn worst case Rf' radiatlon emis ion was 80.30 mW/mt recorded

at Sari center (Site 001). This value is L78%1 of lCNIRP 4,5 W/m2 Maximum Public

Exposure limits (M PE). Airtel w< rst case RF radiation emission was 22.68 m W1m2

re orded at Jarm huri (site 00-). This value is 0.50% of 1 NIRP 4. - W/m2 maximum

public e 'po lire limits. range re orded the worst case RF radiation emi sion of 7.88

mW/ni' at Aga Khan (site 004). This I 'vel is 0.18% of lCNIRP 4.5 W/rn2 maximum public

xposure limits, Essars worst case RF radiation was 31,26 t11 W1m2 at Jamuhuri (site 005).

This tran late to 0.69% of ICNIRP . W/m~ maximum public exposure limits

In GSM 1800 Safaricom, Airtel, Orange and ~:sar recorded the wor t case RF radiation

emissions of 1.17 mW/m2, 7.J2 mW/m2~ 32~.55 mW/m2 and 0.024 mW/m'J. respectively.

Safaricom, Airtel and Orange r corded these worst case RF radiation levels at arit (site

00 I), while Essar recorded the lev I at Kayole (site 003). These RF radiation emissions
 81

translate to 0.013%, 0.08%, 3.62%, and 0.0003% of ICNIRP 9.0 mW/m2 maximum

exposure limits at GSM1800 band respectively.

Safaricom, Airtel and Orange recorded 6.63 mW/m2, 5.91 mW/m2 and 2.21 mW/m2 in the

UMTS 2100 band respectiv ly. Safaricom had the wor t case RF radiation emission at

Jamuhuri (site 005), while Airtel and Orange recorded their worst levels at Sarit (site 001).

RF radiation emissions from Essar were not detected meaning that at the time of

measurement Essar had not deployed 3G technology in Nairobi. Only 7 out of 16 cellular

base stations recorded some values in UMTS 2100 band. Most of these values were from

afaricom. The worst case RF radiation levels in this cas ar over 1500 times less than the

lCNJRP maximum exposure limit of 10 mW/m2 in the frequency band.

It is evident from OUI' measurements results that the overall worst case RF radiation

emission of 325.55 mW/m2 (3.62% of ICNIRP MPE limits) was from an Orange cellular

base station at Sarit (site 001). Most of the other base station selected had RF radiation

levels of less than 1% of the ICNIRP recommended exposure limits to the public. The

values confirm that the RF radiations are not high enough to measure up with the

internationally given thresholds. This is a confirmation that the mobile phone companies

are operating within the recommended limits of ICNIRP. Ba ed on our measurements

against ICNIRP (1998) maximum public exposure .limits, the public i afe from harmful

RF radiation. It is considered that the reported RF radiation levels are a good

repres .ntation of what might be ~.'P cted at other base stations in Kenya urban and rural

areas.
 82

5.2 Recommendations

Since there are other RF sources such as Broadcasting stations e.g. VHF TV, UHF TV,

AM radio, FM radio and Paging etc. It is recommended that other researchers can focus on

these grey areas,

Based on the technical challenges encountered it is recommended that future researchers

use equipment that have capabi lities of detecting the direction of the 'Duree of radiation

since there are several sources especially in urban areas that affect the accuracy of the

results.

Penal code Cap 243 only provide policy on ionizing radiation such as Xvrays, Gamma rays

ete leavlng out non-iontzing l'adiatitm; - '!jl therefere reeemrnend t~ policy to be put in place

to protect the public from eny llttuH: hW'Hlfulradia.LiQn.

 83

REFERENCES
Anita G. G. (2005). Genetic damage in mobile phone users, some preliminary findings.
Indian Journal of Medical Genetics, 11: 99-103.

Barnes H. 1. (1999). Cell Phones: Are they Safe? American Society of Safety Engineers,
December Edition, 21.

Bradshaw CO. (2005). Electromagnetic radiation survey of the Minto Urban (Australia)
rural project

CCK. (2012). Quarterly Sector Statistics Report, 211d

quarter June 2012, Communications
Commission of Kenya, Nairobi. Httpr/rwww.cck.go.ke/consumers

Cleary .S.F, Lui .L.M and Marchant .R.F (1997). In Vitro Lymphocyte proliferation
induced by Radio Frequency electromagnetic radiation under Isothermal conditions,
Bioelectromagnetic, 11: 47-56

Colin R. (2003). Competition and regulation in utility markets. Edward Elgar Publishin,
p. 175. ISBN 978-1-84376-230-0

Crossan, (2004).Towards an understanding, Research Philosophy, Nurse Researcher 11 No

.1, Indiana University, retrieved from www.slis.indiana.edu/faculty

Ezeowu O. (1990). Statistics for professional students, theory problems and solution, 2nd
Edition hydra publisher Ltd Onitsha Nigeria, pp. 103-104

Gregory N., Sean Rocke, Dr. Richelle Adams, Orette Backer (Feb 2010). British virgin
Island radiation measurement project

Hillebrand, Friedheim, (2001). GSM and UMTS, the Creation of Global Mobile
Communications. John Wiley and Sons. ISBN 978.,0-470-84322,.2.

http://www.arpansa.gov.au/rad iationprotecti on/basestationf urvey /index.cfm

http://www.articlesbase.com/communicationarticles/. October 03 2010

Hyland, G. J. (2000).How exposure to GSM base station radiation can adversely affects
human, Physics and biology of mobile telephony. Lancet, 356: 1833~1836.

ICNIRP, (1998). Guidelines for limiting exposure to time-varying electric, magnetic and

electromagnetic fields (up to 300GHz). Health Physics Society 74: 494-522.

Kelly C. (2005). Radiofrequency (RF) radiation. Health Physics Society (HPS), pp. 1-10.
 84

Krewski, (2007). Current Trends in Health and Safety Risk Assessment of Work-Related
Exposure to EMFs, Milan, Italy, February 14-16,2007

Kwees. And Raskmark p. (1998). Changes in c 11proliferation due to environmental non-

ionizing radiation: Microwave radiation, Bioelectrochemistry and Bioenergetics, 44: 251-
255

Loque J.N., Hamburger i., ilv rrnan P.M. and Chiachierini R.P. (2004). Congenital
anomalies and paternal occupational exposure to short wave, microwave, infrared and
acoustic radiation, Health prospective, 112.: 1733-1801.

Maneesh M., Anil, P. K., Jayalekshmi, H., Bhargav, K., and Rohith, V. (2009).Radio
frequency electromagnetic radiation (RF-EMR) from GSM (O.9/1.8GHz) mobile phones
induces oxidative stress and reduce sperm motility in rats, Clinical Science, 64: 561~565.

Moni, (20] 0). Are mobile phone-tower-bts-safe

Mouly and Marie, (2002). The ~M System for Mobile Communications. Telecom
Publishing. ISBJ T 978-0-94 592~ 15~O

Persson B.R. (1997). Blood brain barrier permeability in rats exposed to electromagnetic
fields used in wireless communication and wireless network, Kluwer Academic Publishers
Hingham MA USA; pp. 455- 461.

Rajan Mathews, (2012). A irtel, eliance communications Vodafone among 7 found

violating radiation norms

Rhodes and Schwarz (2004/200). 'I est and measuren ent product" Catalogue
200412005, pg 5-11

S~lgues S.B. (April 1997). Les telecoms mobiles GSM DeS. Hennes (2nd Ed.). Hermes
Sciences Publications. ISBN 2866016068

Sandstorm M., Mild K. H., Wi! {I J" ilnd Oftedal G. (1998). Symptoms a sociated with
mobile phone use, Results of the Swedish •.N 1'\ -e ian Surve '. Occupational Medicine, 48:
357-360.

Siegmund R. and Redl '. (Februar 1995). An Introduction to GSM. Artech House.
ISBN 978¥O-89006~785-7.

Siegmund R. and Redl S. (April I 98). M and Personal Communications Handbook.

Artech HOl(.~e. ISBN 978"O-89006~957.

Th mas H, Uwe ,-1, Peter S (2007). Non-stop pulsed radiation inside home. Znd
International workshop on Riolo tim/ effects of electromagnetic fields, pp. 775-780
 85

Velizarov, Raskmark, Kwee (1999) .. The effect of radiofrequency fields on cell

proliferation are non-thermal, Bioelectrochemistry and Bioenergetics, 48, 177-180

WHO, (1996). International Seminar on Biological Effects ofRF Electromagnetic Fields

and Related Health Risks, WHO Munich, Germany, 1996.

WHO, (1997).International Seminar on Biological Effects of Static and ELF Electric and
Magnetic Fields and related Health Risks, WHO Bologna, Italy. 4-5 June 1997

WHO, (1998).International Workshop on Electromagnetic Fields and Non-Specific Health

Symptoms. WHO Graz, Austria. 19-20 September 1998

WHO, (2000). Electromagnetic fields and public health: mobile telephones and their base
station, Fact Sheet No. 193. Geneva, Switzerland: World Health Organization

WHO, (2002).Workshop on' Adverse Temperature Levels in the Human Body, WHO
Geneva, Switzerland. 21-22 March 2002

WHO, (2004).International Seminar and Working Group meeting on EMF

Hypersensitivity Prague, Czech Republic. 25-26 October 2004

WHO, (2004).Workshop on Sensitivity of Children to EMF Exposure, WHO Istanbul,

Turkey. 9-10 June 2004
 86

APPENDIX 1

GSM 900 Spectral Scans

87
88
 89

APPENDIX 2

GSM 1800 Spectral Scans

90
 91

APPENDIX 3:

UMTS 2100 Spectral Scans

92
93