MODELING THE F REGION IONOSPHERIC PEAK HEIGHT VARIATIONS OVER MALAYSIA BY ANTENNA PATTERN SYNTHESIS TECHNIQUE ZETI AKMA BINTI RHAZALI Thesis submitted in fulfilment of the requirements for the award of the degree of Doctor of Philosophy in Electrical Engineering Faculty of Electrical and Electronic Engineering UNIVERSITI MALAYSIA PAHANG SEPTEMBER 2014
24
Embed
MODELING THE F REGION IONOSPHERIC PEAK HEIGHT …umpir.ump.edu.my/13178/1/FKEE - ZETI AKMA RHAZALI - CD 9649.pdf · MODELING THE F REGION IONOSPHERIC PEAK HEIGHT VARIATIONS OVER MALAYSIA
This document is posted to help you gain knowledge. Please leave a comment to let me know what you think about it! Share it to your friends and learn new things together.
Transcript
MODELING THE F REGION IONOSPHERIC PEAK HEIGHT VARIATIONS OVER
MALAYSIA BY ANTENNA PATTERN SYNTHESIS TECHNIQUE
ZETI AKMA BINTI RHAZALI
Thesis submitted in fulfilment of the requirements
for the award of the degree of
Doctor of Philosophy in Electrical Engineering
Faculty of Electrical and Electronic Engineering
UNIVERSITI MALAYSIA PAHANG
SEPTEMBER 2014
vi
ABSTRACT
The ionospheric F region over Malaysia is still an issue to many radio communication
enthusiasts. The actual height of this layer is not well defined. Models available are
mostly reliable for temperate zones. This thesis describes the determination of the actual
F layer height and proposed a unique model to represent the height variations. The
ionospheric F region is observed via ionograms, produced by the ionosonde operated at
Parit Raja (2oN, 103oW, dip 14.3o), Batu Pahat, Malaysia. The ionogram gives the
virtual height representation of the ionosphere. POLAN ionogram inversion program is
used to determine the real height of the ionospheric layer. The observations are held
during period of moderate to low solar activity of solar cycle 23 (2005 to 2010).
However, in this work, only hourly data of March, June, September, and December,
2006 and 2007, are examined. The data are statistically analysed to summarize their
main characteristics. The actual height of the F layer is determined from the median
values and the coefficient of variability quantifies the height deviations. To derive the
mathematical representation of the variations, the least-squares regression technique is
used to fit functions to the median data. The best fit function is the descriptive model
that describes the variations. A new model of ionospheric height variations is also
proposed on the same basis. The ionospheric height time variation which is a cyclic
event is re-represented in polar coordinate form. The cyclic representation which
approximates an antenna radiation pattern allows the development of antenna equivalent
model of peak height variations. The association of ionospheric height variations to the
radiation pattern of antenna array is the novelty idea of this study. The observation
results indicate that the median height of peak electron density, hmax, varies from 420
km in June to 550 km in other months during noon time. The night-time average heights
rest around 300 km for all months. The daytime peak is found highest in December
solstice season and lowest in June season while post sunset peaks are not seen during
this period. The descriptive mathematical model of diurnal variations shows that the
variations fit well into a four-term Fourier series model. The two-element arrays with
array spacing in the x-direction of /8, and phase of /3, and with element spacing in
the y-direction of /4, and phase of zero, is the optimal array configuration which
signifies the variations. The results show that an array of two-element arrays antenna is
suitable to represent the ionosphere peak height variations over Malaysia during
moderate to low solar activity period.
vii
ABSTRAK
Rantau F ionosfera merentasi Malaysia masih menjadi isu kepada kebanyakan
pengguna komunikasi radio. Ketinggian sebenar lapisan ini belum dapat dipastikan.
Model-model yang ada pula cenderung untuk kegunaan di zon bersuhu. Tesis in
menjelaskan penentuan ketinggian sebenar lapisan F dan mensyorkan sebuah model
unik sebagai paparan kepelbagaian ketinggian. Rantau F ionosfera diperhatikan melalui
ionogram, hasil cerapan yang diperolehi daripada ‘ionosonde’ yang beroperasi di Parit
Raja (2oN, 103oW, dip 14.3o), Batu Pahat, Malaysia. Ionogram mempaparkan
ketinggian maya lapisan ionosfera. Perisian POLAN digunakan untuk menukar kepada
ketinggian sebenar lapisan ini. Pencerapan telah dilakukan dalam tempoh aktiviti solar
sederhana hingga rendah, pada kitaran solar 23 (2005-2010). Bagaimanapun, untuk
kajian ini, hanya ionogram dalam sela masa jam bagi bulan Mac, Jun, September, dan
Disember, 2006 dan 2007, diteliti. Analisa statistik digunakan untuk menentukan ciri
penting data tersebut. Ketinggian lapisan ionosfera diperolehi daripada nilai median
manakala pekali variasi menentukan corak kepelbagaian. Pembangunan model
matematik menggunakan teknik ‘least-squares regression’. Sesuatu fungsi matematik
akan disesuaikan kepada nilai median tersebut. Model matematik diperolehi apabila
fungsi matematik tersebut memberikan penghampiran terbaik. Atas dasar yang sama,
pembangunan model bagi menggambarkan kepelbagaian ketinggian ionosfera juga
dilaksanakan. Kepelbagaian ketinggian lapisan ionosfera terhadap masa adalah suatu
keadaan yang silih berulang. Ia boleh dipaparkan dalam koordinat polar. Paparan
sebegini menghampiri paparan corak radiasi antenna tatasusun justeru pembangunan
model kepelbagaian ketinggian lapisan ionosfera adalah perlu. Kesinambungan di antara
kepelbagaian ketinggian lapisan ionosfera dan corak radiasi antena tatasusun merupakan
idea terbaru yang terhasil dari kajian ini. Keputusan pemerhatian menunjukkan
ketinggian ionosfera, hmax berubah dari 420 km pada bulan Jun kepada 550 km pada
bulan-bulan yang lain semasa waktu tengahari. Purata ketinggian pada waktu malam
pula berada pada paras 300 km. Puncak tertinggi dicapai pada bulan Disember dan
terendah dalam bulan Jun. Model matematik untuk kepelbagaian harian menunjukkan
bahawa model jujukan Fourier 4-terma sangat bersesuaian dengan kepelbagaian ini.
Antena tatasusun dua-elemen dengan jarak susunan /8 dalam arah-x dan fasa /3, dan
jarak elemen /4 dalam arah-y dengan fasa sifar, adalah konfigurasi tatasusun yang
paling optimum untuk menggambarkan kepelbagaian ketinggian ionosfera. Keputusan
kajian menyatakan tatasusun dua-elemen tatasusun adalah paparan yang paling sesuai
untuk kepelbagaian ketinggian ionosfera untuk tempoh aktiviti solar sederhana hingga
rendah.
viii
TABLE OF CONTENTS
Page
SUPERVISOR’S DECLARATION ii
STUDENT’S DECLARATION iii
DEDICATION iv
ACKNOWLEDGEMENTS v
ABSTRACT vi
ABSTRAK vii
TABLE OF CONTENTS viii
LIST OF TABLES xii
LIST OF FIGURES xiii
LIST OF SYMBOLS xvii
LIST OF ABBREVIATIONS xix
CHAPTER 1 INTRODUCTION
1.1 Background 1
1.2 Problem Statement 3
1.3 Objectives 5
1.4 Scope of Work 7
1.4.1 Ionospheric Observation
1.4.2 Data Analysis and Trending
1.4.3 Modeling
7
8
8
1.5 Thesis Outline 8
1.6 Summary 10
CHAPTER 2 LITERATURE REVIEW
2.1 Early Discovery 11
2.2 Ionospheric Research 12
2.2.1 Research in Equatorial Region
2.2.2 Research in Malaysia
16
21
ix
2.3 Ionospheric Electron-Density Profile 22
2.3.1 Peak Electron Density, Nmax
2.3.2 Height of Peak Electron Density, hmax
23
28
2.4 Associating Ionospheric Height Variation to Antenna 30
2.4.1 Array Pattern Synthesis Techniques
2.4.2 Parameterising Ionospheric Height Variations by Array
Design Techniques
32
33
2.5 Summary 34
CHAPTER 3 CHARACTERISING THE IONOSPHERIC F REGION
3.1 Introduction 36
3.2 From Observation to Modeling 37
3.3 Data Collection 38
3.3.1 Radio Sounding Technique
3.3.2 Experimental Setup
3.3.3 The Ionogram
38
40
41
3.4 Data Management and Pre-processing 43
3.4.2 Ionogram Inversion using POLAN Program
3.4.2 Technique used to Manage Empty Data Bins
3.4.3 Data Assimilation
3.4.4 Data Mining
43
49
52
55
3.5 Exploratory Analysis of Data 57
3.6 Modeling the Ionospheric Peak Height Variations 59
3.6.1 The Descriptive Model
3.6.2 Ionospheric Peak Height Variations Model
59
60
3.7 Summary 61
CHAPTER 4 OBSERVATIONAL RESULTS OF THE IONOSPHERE
OVER MALAYSIA
4.1 Introduction 64
4.2 Ionospheric Electron Density Studies 64
4.2.1 Electron-Density Profile
4.2.2 Peak Electron Density, Nmax
4.2.3 Height of Peak Electron Density, hmax 4.2.4 Temporal Variations of Nmax
and hmax
64
68
71
76
x
4.3 Ionospheric Features and Characteristics over Parit Raja 78
4.3.1 F3 layer Features of the Noon time Ionosphere
4.3.2 Spread F and Sporadic E Events
78
83
4.4 Influence of Solar Activity 85
4.4.1 Midday vs. Midnight Profiles 86
4.5 Influence of Geomagnetic Activity 88
4.5.1 Geomagnetic Storms Impact on Plasma Distribution 89
4.6 Ionospheric Variability 92
4.7 International Reference Ionosphere (IRI) Model Verification 94
4.8 Summary 97
CHAPTER 5 ANTENNA PATTERN EQUIVALENCE OF PEAK
HEIGHT VARIATIONS
5.1 Introduction 104
5.2 Characterisation of the Height of Peak Electron Density Variations 105
5.2.1 Introduction
5.2.2 Fourier Components as Approximation to Median Height of
Peak Electron Density Variations
5.2.3 Mathematical Representation
5.2.4 Cyclic Diagram
105
105
112
114
5.3 Antenna Pattern Equivalence of Peak Height Variations 115
5.3.1 Development of the Model
5.3.2 Array of an Array
5.3.3 Closeness of Fit
115
126
133
5.4 Summary 139
CHAPTER 6 CONCLUSIONS
6.1 Introduction 143
6.1.1 Ionospheric Height Variations
6.1.2 Cyclic Diagram
6.1.3 Parameterisation of Ionospheric Height Variations
143
144
144
6.2 Summary 145
6.3 Future Work 147
xi
REFERENCES 149
APPENDICES
A Solar Cycle 23/24 158
B Sample Ionograms 159
C Curve Fitting using Fourier Series Approximation 164
D Catalogue of Array Patterns 188
E List of Publications 193
xii
LIST OF TABLES
Table No. Title Page
2.1 Ionosonde stations in equatorial region 17
3.1 Standard modes of analysis incorporated in POLAN 45
3.2 Solar activity indicator 53
3.3 Geomagnetic activity condition 55
4.1 The daytime and night-time Nmax and hmax values for five
consecutive days of January 2005
88
4.2 Optimal value of median Nmax, and hmax and their corresponding
standard deviations
97
5.1 RMSE values for all representative months 108
5.2 Performance comparison for the four representative months
under different geometrical configuration
125
5.3 Performance comparison for the four representative months
under different geometrical configuration
134
5.4 Standard deviations 135
5.5 Numerical results of the antenna equivalence of hmax variations
(2006) giving the magnitude, G dB, and difference, dB
141
5.6 Numerical results of the antenna equivalence of hmax variations
(2007) giving the magnitude, G dB, and difference, dB
142
5.7 Results showing the coefficients of Fourier series
approximation to March median hmax variations
164
5.8 Results showing the coefficients of Fourier series
approximation to June median hmax variations
170
5.9 Results showing the coefficients of Fourier series
approximation to September median hmax variations
176
5.10 Results showing the coefficients of Fourier series
approximation to December median hmax variations
182
5.11 Array factor of N = 2, 3, and 4 for different element spacing, d 188
5.12 Array factor of N = 2 for different element spacing, d, and
progressive phase shift,
190
xiii
LIST OF FIGURES
Figure No. Title Page
2.1 (a) An illustration of the equatorial plasma fountain effect
(Bilitza, undated) (b) Latitudinal variation of electron density at