”f" M LIBRARY
ichigan State
200% University
This is to certify that the
thesis entitled

Bandwidth Extension of a Body Worn Antenna Vest

presented by

Raenita Ann Fenner

has been accepted towards fulfillment
of the requirements for the

Master of degree in Electrical and Computer
Science Engineering

 

 

@mem

MdL) or Professor” 3 Signature
8 - 23”— o ;L

Date

 

MSU is an affirmative-action, equal-opportunity employer

-O-I-0-0-0-I-c-u-I-o-.-n-o-o-l-a-ICOOI-;~_--.-.-.-.-u-I-o=_-:--o-o-c-o—c-n-o-o-

 

PLACE IN RETURN BOX to remove this checkout from your record.
TO AVOID FINES return on or before date due.
MAY BE RECALLED with earlier due date if requested.

DATE DUE DATE DUE DATE DUE

 

 

BANDWIDTH EXTENSION OF A BODY WORN ANTENNA VEST
By

Raenita Ann Fenner

A THESIS
Submitted to
Michigan State University
in partial fulfillment of the requirements
for the degree of
MASTERS OF SCIENCE
Electrical and Computer Engineering

2007

 

ABSTRACT
BANDWIDTH EXTENSION OF A BODY WORN ANTENNA VEST
By

Raenita Ann Fenner

This thesis presents the work accomplished for the research and development of
an antenna to extend the bandwidth of a body worn antenna vest (BWAV) developed at
Southwest Research Institute (SwRI) from 50-3000 MHz to include the 20-150 MHz
band. The BWAV is a wearable vest that is partly composed Of antennas, which while
being worn by a human operator, can be used for direction finding and various military
and civilian communications. The vest was developed by Sle to Operate in the band 50
f 3000 MHz. In this thesis, an antenna is designed to cover the band 20 — 150 MHz.

This thesis includes an introduction which reviews the past work done on the
BWAV, a literature review of BWAV designs and similar applications, and an overview
of the BWAV antenna design process. Next, the simulation tools (FEKO, GA-FEKO,
and their necessary components) and simulation methods are described. Subsequently,
summaries Of the antennas designed and simulated are presented. The summaries include
assessments of the designed and simulated antennas and their possible incorporation in
the BWAV with the chosen dimensions. Thereafter, various result parameters (V SWR,
radiation patterns, specific absorption ration (SAR), current density) are discussed.
Lastly, conclusions are given based on the simulated results of this thesis and the goals of

the BWAV developed by Sle.

‘

 

   
   
   
    
   
  
  

M‘br‘ m» 58w?
-. ‘r wifiv‘éws. U:
r 0. “5")“ t *W‘
i l.“ ; . ‘ ' ,
1' u M v Par-mu
”or rln’l'hir
W by
' use and Support
RaenitaAnn Fem
2007

      
    

W.';..qu.c.m..« we-..

.I.- .

WAN”. .‘rav .-. ‘

I ‘ air Jim. {'4'

.‘f
1 _‘
." -

 

    
  
   
  
  
  
     

Acknmvlrdgtm
Hunt: 'Jvc at n ma rm 3» “his law tub in Curr-pitting this “hhufi
a \iUlCif. rum. w uh u . e lulurr Wrap; mews. who». Ibflmflwfl ,
.Hr WW,
Lthr- and i lulVC mu.) ,uujIlr' tn Ih enlz. " ‘

inrxr. t .‘xthIiJit'1'"lfi;ll)i.l'lly‘M'-l\13r Dr. Edward w .
sr-tmr is)» '.-‘J_.;t- canttd‘nt‘r. and. supp-3n Our tinny W“
m... rim; ‘. .t my L. l. .a ttl'c mcmbcn.‘ .. Terrance M fl“
Balmuhrnltntl‘mze . In. lN' , :‘::“;‘ simi- “WJ’W . -.

I l.:‘-L tn ‘lmtll {if b’arhn-mFW 0!. Pan, M h
support an: "f"i'."-~ drum n. lit-lwm WC I.
graduate s. llv-nl and heme yucca-ht an m school. . 1 '

. ‘Jll
,. ..\. er "6.

l want to cairn”. a mum. you In Swfll Sign! EM

Diwsion fur suxponmg n on m: GEM fellowship in mm“

  
  
  
  
  
  
  
  
 
  
  

{1’0

pcrionn lY-‘JC‘ILii « 2: one ‘J their (“’qu [also want I) H
allowing me (0 me (”3 Hle) to pcdoml I signified Md

Thank vet: to Anthony l’ltfmmfi for all d m w
has. and being there 1&1!“qu this am . ,. mi

lastly. I want to thank all of my {Mun

may: alwaymuppmingdlofmym
Mmdbdnguccflemuamplatafl
.nmuyt‘uuftxdwayshung-

I

Acknowledgements
From the starting point of this research to completing this document, has been filled with
a variety of moments — success, failure, happiness, sadness. I have definitely not made it
alone and I have many people to thank.

First, I would like to thank my advisor Dr. Edward Rothwell for being a constant
source knowledge, guidance, and support. Our many meetings helped tremendously.
Also, thank you to my committee members, Dr. Terrance Brown and Dr. Shanker
Balasubramaniam, for their help and advice.

I have to thank Dr. Barbara O’Kelly and Dr. Percy Pierre for their constant
support and tireless efforts in helping me and others achieve our dreams of attending
graduate school and being successful in graduate school.

I want to extend a thank you to Sle — Signal Exploitation and Geolocation
Division for supporting me on the GEM fellowship for two years, and allowing me to
perform research on one of their projects. I also want to thank Jack Ross for generously
allowing me to use GA-FEKO to perform a significant portion of my research.

Thank you to Anthony Plummer for all of your help and support in formatting the
thesis, and being there throughout this experience.

Lastly, I want to thank all of my family and friends. First, I want to thank my
parents for always supporting all of my hopes and aspirations from kindergarten to the
present, and being excellent examples in what the human spirit can achieve. I would like
to thank my sister for always being supportive of my goals. Thanks to all of the faculty
at Morgan State University that helped and inspired me to attend graduate school.

Finally, thanks to all of my close friends for keeping me in their thoughts and prayers.

V

 

TABLE OF CONTENTS

 

 

 

 

 

 

 

 

 

 

 

 

 

 

 

 

 

 

 

 

 

 

 

 

 

 

 

 

 

 

 

 

 

List of Tables ...................................................................................... viii
List of Figures ....................................................................................... ix
Chapter 1; Intrndnotinn l
1.1 Problem Overview 1
1.1.1 Current Technology/Military Initiatives 1
1.1.2 BWAV Prototype 1
1.1.3 Future Goals and Objectives 2
1.2 Literature Review 3
1.2.1 Purpose of the Literary Review 3
1.2.2 BWAV Literature Review 4
1.2.3 Wearable Antennas 5
1.2.4 Antenna Elements 6
1.3 Overview of the Design Process 8
Chapter 2: Simulation Methods 9
2.1 Introduction 9
2.2 FEKO 9
2.2.1 Introduction to FEKO 9
2.2.2 MoM and FEKO 11
2.3 FEKO Simulation Methods 14
2.4 OPTFEKO 15
2.4.1 Introduction — What is OPTFEKO? 15
2.4.2 OPTFEKO Usage 16
2.5 GA-FEKO 17
Chapter 3: Antenna Designs 18
3.1 Introduction 18
3.2 Elliptical Loops 18
3.2.1 Background on Elliptical Loops 18
3.2.2 Elliptical Loop Design 19
3.3 Fractals 19
3.3.1 Background on Fractals 19
3.3.2 Fractal Design 22
3.4 Meander Line Bowtie Antenna 22
3.4.1 Background on Meander Line Bowtie Antenna 22
3.4.2 Meander Line Bowtie Antenna Design 23
3.5 Self—Structuring Antennas (SSA) 25
25

3.5.1 Background on the SSA

 

vi

 

3.5.2 SSA Designs 25

 

 

 

 

 

 

 

Chapter 4: Results 30
4.1 Introduction 30
4.2 Elliptical Loop, Minkowski Fractal, and Meander Line Bowtie Antenna Results

30
4.3 Full-Size Meander Line Bowtie Antenna Results 32
4.4 SSA Require 33
4.4.1 Flat SSA in Free Space Results 33
4.4.2 Conformal SSA in Free Space Results 55
4.4.3 Conformal SSA in Front of Dielectric Body Results 56
Chapter 5; (‘nnr‘lncinnc 85

 

Bibliography: .................... 87

 

vii

 

  

 

“guru 7 1 .‘\l:t1i.w'~‘.‘\kli-H'/.vu. , .. ,. .. . ‘A _

 

{191.er '7 "11:13.: ‘ ' "-- i'u' H r \V-uzin.‘

i'wgurc Jul ‘55. .1)!" v,‘ ‘

 

      
     
  
   
    
  
  
  
 
 
  
 
  

i‘lgu'k .~ (Lint! HL. . “AA . <1‘.’C(‘.\FIOLTC.,. . .. .,.,,,...,,,...,..,.....

)Téguq} if (I‘lti‘flii .; \Ni ' {fax V;- Cl')“ In PRC ch-ou'h-Cunmuoow
Figure 1'- ~ ( nu-tli'm .11 ‘x‘A M It .c-. :21 Iii-31:1 tnc Bmty.......-..............

t..1\\r.n'"nlqzll mlrdyfimw

'l

Figmc .1-

Figure .1 l vuwa r l‘mth I: Lupmn 14:0an

Lilli. Banal"; -~. Jenna.

8.. .. a.-.-umu~u

Hg are 1-1' .‘m’ s... wanna Bowtie Antenna VS‘WR vs. m A
Figure Jr '1 l 111 Si. .. . Mgm ter Linc Bowtie Anm “1'
MHz; . ..... . .. ............ ....-...... ........
Figure 4—4 - VSWR m Frequency 10! HI! 88A In ”8”
Hm 4-5~ Azimuthanotal (imp of Sweat Fun-ad. m
Punt: 4-0 . Polar Plot for Tom “mun-akin -
Hm 4-7-Color Code for Current Density mm...
”#8 - 01m Deon‘tyon SSA mPluet-
'mucummwonssa‘mrum:
Immw- CwDeasttyon SSAh'fi-M
_. “9...“: ~GMMIMSM'W ."

LIST OF FIGURES

 

 

 

 

 

 

 

 

 

Figure 3— l-Elliptical Loop I 20
Figure 3-2 — Minkowski Fractal 21
Figure 3—3-Meander Line Bowtie Antenna 24
Figure 3-4-SSA Design 27
Figure 3-5 - Conformal SSA in Free Space 28
Figure 3-6 - Conformal SSA Cross Section in Free Space 28
Figure 3-7 - Conformal SSA in front of Dielectric Body 29
Figure 3-8 - Conformal SSA in front of Dielectric Body Cross Section .......................... 29
Figure 4-1 VSWR vs. Frequency of Elliptical Loop, Minkowski Fractal, and Meander
Line Bowtie 1‘ "t eeeee 31
Figure 4-2 - Full-Size Meander Bowtie Antenna VSWR vs. Frequency Plot .................. 34
Figure 4-3— Full-Size Meander Line Bowtie Antenna VSWR vs. Frequency (60-150
MHz) , 35
Figure 4-4 - VSWR vs. Frequency for Flat SSA in Free Space ....................................... 36

Figure 4-5- Azimuthal Total Gain of Several Frequencies of Flat SSA in Free Space 38

 

 

 

Figure 4-6 - Polar Plot for Total Azimuthal Gain for SSA in Free Space ........................ 39
Figure 4-7-Color Code for Current Density Plots 40
Figure 4—8 - Current Density on SSA in Free Space, 20 MHz 41
Figure 4-9 - Current Density on SSA in Free Space, 30 MHz 42
Figure 4—10 — Current Density on SSA in Free Space, 40 MHz ....................................... 43
Figure 4—11 - Current Density on SSA in Free Space, 50 MHz ....................................... 44
Figure 4-12 - Current Density on SSA in Free Space, 60 MHz ....................................... 45
ix

 

‘

Figure 4—13 - Current Density on SSA in Free Space, 70 MHz ....................................... 46

Figure 4-14 - Current Density on SSA in Free Space, 80 MHz ....................................... 47

Figure 4-15 - Current Density on SSA in Free Space, 90 MHz ....................................... 48
Figure 4-16 - Current Density on SSA in Free Space, 100 MHz ..................................... 49
Figure 4-17 - Current Density on SSA in Free Space, 110 MHz ..................................... 50
Figure 4-18 - Current Density on SSA in Free Space, 120 MHz ..................................... 51
Figure 4-19 - Current Density on SSA in Free Space, 130 MHz ..................................... 52
Figure 4-20 - Current Density on SSA in Free Space, 140 MHz ..................................... 53
Figure 4—21 - Current Density on SSA in Free Space, 150 MHz ..................................... 54
Figure 4-22 - VSWR vs. Frequency for Conformal SSA in Free Space .......................... 57

Figure 4-23 - Azimuthal Total Gain of Several Frequencies of Conformal SSA in Free
Space 58

 

Figure 4-24 — Azimuthal Total Gain of Several Frequencies of Conformal SSA in Free

 

Space, Polar Plot 59
Figure 4-25 - Current Density on Conformal SSA in Free Space, 20 MHz ..................... 60
Figure 4—26 - Current Density on Conformal SSA in Free Space, 40 MHz ..................... 61
Figure 4-27 - Density on Conformal SSA in Free Space, 50 MHz .................................. 62
Figure 4—28 - Current Density on Conformal SSA in Free Space, 60 MHz ..................... 63
Figure 4-29 - Current Density on Conformal SSA in Free Space, 70 MHz ..................... 64
Figure 4—30 - Current Density on Conformal SSA in Free Space, 80 MHz ..................... 65
Figure 4—31 — Current Density on Conformal SSA in Free Space, 90 MHz ..................... 66
Figure 4—32 - Current Density on Conformal SSA in Free Space, 100 MHz ................... 67
Figure 4-33 Current Density on Conformal SSA in Free Space, 110 MHz ..................... 68
x

‘

Figure 4—34 - Current Density on Conformal SSA in Free Space, 120 MHz ................... 69

Figure 4-35 — Current Density on Conformal SSA in Free Space, 130 MHz ................... 70

Figure 4-36 - Current Density on Conformal SSA in Free Space, 140 MHz ................... 71
Figure 4—37 - Current Density on Conformal SSA in Free Space, 150MHz .................... 72

Figure 4-38 - VSWR vs. Frequency for Conformal SSA in Front of Dielectric Body ..... 75
Figure 4-39 - Azimuthal Total Gain of Several Frequencies of Conformal SSA in Front of
Dielectric Body 76

Figure 4—40 - Azimuthal Total Gain of Several Frequencies of Conformal SSA in Front of
Dielectric Body, Polar Plot 77
Figure 4-41 - Current Density on Conformal SSA in Front of Dielectric Body, 20 MHz 78
Figure 4-42 - Current Density on Conformal SSA in Front of Dielectric Body, 40 MHz 79
Figure 4-43 — Current Density on Conformal SSA in Front of Dielectric Body, 60 MHz 80
Figure 444 - Current Density on Conformal SSA in Front of Dielectric Body, 80 MHz 81

Figure 4-45 - Current Density “on Conformal SSA in Front of Dielectric Body, 100 MHz
82

 

Figure 4-46 - Current Density on Conformal SSA in Front of Dielectric Body, 150MHz
83

 

Figure 4-47 - SAR vs. Frequency for Conformal SSA in Front of Dielectric Body ........ 84

Images in this thesis/dissertation are presented in color

xi

 

Chapter 1: Introduction

1.1 Problem Overview

1.1.1 Current Technology/Military Initiatives

Many technologies are born from the need to meet new military initiative needs. Two
recent military initiatives, the Joint Tactical Radio System (JTRS) and the Joint Threat
Warning System (JTW S), are pushing communication technologies to new bounds. For
example, JTRS has been formulated as a family of software defined radios that will give
warfighters voice, data, and video information over an extremely wide bandwidth [1].
With the combination of JTRS and JTW S, new communication technologies are needed
to meet wideband and hands-free operation goals. One technology that will greatly
enhance communication of warfighters on the ground in terms of JTRS and JTWS
systems is a wideband body-worn antenna vest (BWAV).

The project reported in this thesis was developed as a joint venture between the
author and engineers in the Signal Exploitation and Geolocation Division at Southwest
Research Institute (SwRI) in San Antonio, TX. The objective of this project was to
research and develop an antenna element that will be able to serve the various

specifications of a BWAV.

1.1.2 BWAV Prototype

SwRI developed a prototype of a BWAV in 2004 [2]. The prototype was designed to be
a widebanded conformal antenna system that can aid in the use of man-portable radio

communications and threat-warning equipment. The BWAV prototype was initially

‘

designed to operate in the 50-2000 MHz frequency range; the final intention is for the
BWAV prototype to cover the entire VHF and UHF frequency ranges of 30-3000 MHz.
This will enable the BWAV prototype to be used with JTRS and JTWS systems in the
future.

The BWAV prototype developed at Sle is composed of two different antenna
arrays. One array is an array of four bowtie elements that cover the 50-650 MHz
frequency range. The second array is comprised of six tapered slot antenna elements that
cover the 650-2000 MHz frequency range. The bowtie elements are 22 cm long and
composed of nickel/silver metallized ripstop nylon fabric. The bowtie elements are sewn
into the shell of the vest underneath the arms. The tapered slot elements are etched on
the ground plane of a printed circuit board (PCB) microstrip feed line, and the boards are
located in pockets on the inside of the vest.

Several parameters were tested on the BWAV prototype while it was on a
phantom (a mannequin with salt water inside) and a live subject. The BWAV prototype
was tested for its return loss characteristics, farfield azimuthal gain-by-comparison, and
farfield azimuthal radiation patterns. The return loss was measured in an anechoic
chamber via a network analyzer, while the farfield azimuthal gain-by-comparison farfield
azimuthal radiation patterns were measured on an outdoor antenna range with a ground

level rotary.

1.1.3 Future Goals and Objectives

Using the BWAV prototype from Sle as a benchmark, it was decided that more work
was needed to make the BWAV fully functional all the way down to 20 MHz.
Unfortunately, the design of a BWAV able to operate over the entire VHF and UHF

2

 

frequency range has proven to be very difficult. The main difficulty in the design of the
BWAV arose in the lower part of the VHF frequency range. The lower VHF frequency
range was a problem area because it was difficult to design an antenna element big
enough for such large wavelengths, while small enough to fit on a vest wearable for
humans. Thus, the main goal of this project was to research and design an antenna
element that will be able to sense signals from the 20-150 MHz range, and be able to fit

on an antenna vest.
1.2 Literature Review

1.2.1 Purpose of the Literary Review

The first step in the design process for the design of the BWAV was to do a substantial
literature review. The literature review was dual-purposed. One purpose was to find
BWAV designs done by others. By doing a comprehensive search for other BWAV
designs, the author could see'where current work in BWAV design was going, see what
has been achieved in BWAV design, and find the goals and objectives of others. The
second purpose was to find antenna elements that would be suitable antenna elements for
the BWAV. Suitable antenna elements are those considered to be small in physical size
(in order to fit human wearable vest), but elements that could also adequately sense at
low frequencies (lower VHF range). The research in turn gave the author an idea of
where to start in the antenna design process. The following sections are summaries of the
literature found in the following categories: BWAV designs, wearable antennas, and

small antenna elements.

k

1.2.2 BWAV Literature Review

A literature review was done on BWAV in order find the status of research and design of
BWAV, and to find the goals and objectives of others. Two main designs were found —
the Combat Wear Integration Antenna System (COMWIN) [3] and the Direction Finding
System using a Body-Wom Antenna [4]. Summaries of their designs are in subsections

1.2.2.1 and 1.2.2.2.

1.2.2.1 COMWIN Antenna System

The COMW IN antenna system described in [3] was designed to be used in the new era of
military equipment for warfighters. Thus, COMWIN was designed to be able to receive
and transmit voice, video, and data over a broad range of frequencies. COMWIN was
also designed to have a low visual signature to protect warfighters using communication
equipment and to be used in conjunction with the hand-held version of JTRS. The
COMW IN antenna system can cover the 2-2400 MHz frequency range via three antenna
subsystems — a vest antenna, helmet antenna, and a whole body antenna. The vest
antenna covers the 30-500 MHz range and is essentially a slot antenna design. The
helmet antenna covers the 500-2400 MHz range, and is comprised of two equal radiating
surfaces separated by a gap appended to a standard military Kevlar helmet. The whole
body antenna consists of metallic soles in the shoes, a flack jacket, and pants that cover
the 2-30 MHz frequency range. The three antenna subsystems are integrated together
such that the soldier can switch to the appropriate antenna subsystem via a rotary dial,
which can be placed on the wrist. During measurement testing, it was found that the
COMWIN had a voltage standing wave ratio (VSWR) of less than 3:1 between 5-1200

MHz. Future work is being done such that no human input is needed to switch to the

4

appropriate antenna subsystem and comparing the antenna subsystems’ receiving and
transmitting capability against standard antennas. The COMW IN system was interesting
because the design implemented several antenna elements to cover the entire frequency

range and it gave a benchmark or standard for BWAV performance.

1.2.2.2 Direction Finding System Using a Body-Wom Antenna

The other BWAV design found during the literature review was in [4]. The Apostolos
BWAV is an antenna vest that can sense the direction of incoming electromagnetic
signals. This BWAV was designed, for example, to assist soldiers, policeman, or border
patrollers to find enemies or criminals by sensing their mobile devices. The antenna
elements on the vest are an array of meander line loaded antenna elements which are
made out of flexible conducting fabric. The Apostolos design was intended to be used
With the BWAV wearer standing up or lying on the ground; the Apsotolos design was
also designed to cover the 2 MHz — 40 GHz frequency range. Unfortunately, this design
was not very helpful because there was no data on performance capabilities shown and

the design had an emphasis on the composition of the direction finding system.

1.2.3 Wearable Antennas
This section seeks to review the literature found on wearable antennas. Wearable
antennas were researched in great depth in order to find an antenna element that has been
designed for the VHF and UHF frequency range, but has also already been tailored to be
worn on a human.

During the literature review, several papers were on found on the wearable

antenna subject. The papers cited many similar applications for the antenna designs like

 

military or police use, integration into smart clothes, and cellular phones. Various types
of patch and microstrip antennas dominated the type of antenna elements implemented;
examples of the antenna elements implemented are the E shaped microstrip and the U
shaped slot [5, 6]. Similar materials were also used in the construction of the antennas
like various conducting fabrics (e.g. — knitted copper, Flectron) and copper tape [7, 8].
Many of the antennas were mounted or sewn onto fleece sweaters or felt which doubled
as a substrate in most cases. Unfortunately, many of the wearable antenna designs found
during the literature search were designed between the 380 MHz — 3.0 GHz frequency
range. Thus, it would have been very difficult to implement one of these designs into this

project where the operating frequency range was 20-150 MHz.

1.2.4 Antenna Elements

The purpose of this section is to expound upon the possible antenna elements that were
found during the literature search that would lend themselves for this project. Thus, a
search was done for antennas that have a large bandwidth, but that are relatively small in
size, too. The three main types of antennas that were found during the search are folded
loops, fractals, and meander lines. The following subsections summarize the literature

that supported using folded loops, fractals, and meander lines for this application.

1.2.4.1 Folded Loops

Most of the literature found on folded loops was in [9-11]. The research done on folded
loops was conducted especially to find an antenna element that would be able to be
scaled down as handsets and cellular phones are scaled to smaller sizes. The folded loop

was reported to be a two-wire transmission line folded at a quarter-wavelength to form a

x

half-wave dipole that appears to be a one-wavelength loop antenna. In the literature, the
dimensions of the folded loop were all less than 10 mm and the operating frequency was
1.860 GHz. The reported measured and simulated radiation patterns were similar to those
of a traditional dipole, and the measured and simulated return loss was less than -20 dB at
the operating frequency.

The folded loop was found to be an attractive element for this project because of
its good performance (radiation patterns, return loss, etc.), its built-in space saving
design, and the fact that its design takes into account the human body effect via the self-
balanced design. In addition, the folded loop has a relatively large bandwidth of up to 98
MHz, and the adjustment of the strip width and space between folded elements can

further alter the bandwidth of the element.

1.2.4.2 Fractals and Meander Lines
During the literature review, both fractal and meander line antennas became antenna
elements that would be explored for this project. Fractal and meander line antennas were
found to be attractive elements for this project because of their natural space saving
capabilities. Almost any traditional antenna element (e.g. — dipole, loop, patch, etc.) can
be turned into a fractal, a meander line, or both (not simultaneously). Furthermore, along
with any traditional antenna element able to become a fractal or meander line, fractal and
meander line antennas often have similar radiation patterns to those of their traditional
counterparts [12, 13]. For example, a meander line dipole and a dipole will have very
similar radiation patterns.

Another key reason to research fractals and meander lines for this project is
because they offer more electrical length in the same physical area as their traditional

7

1‘

antenna counterparts yield; this important aspect usually enables fractals and meander
lines a greater input impedance than a traditional antenna that takes up the same area.
Since achieving adequate input impedance is very difficult for antennas that are small
compared to a wavelength, this feature of fractals and meander lines was a main reason
for researching them for this project [8, 14, 15]. In addition, fractals and meander lines

are low profile in nature and can be designed to multi-banded or widebanded [16].

1.3 Overview of the Design Process

The goal or objective of this project was to design an antenna for a BWAV that would be
able to receive or transmit from 20-150 MHz and be small enough to fit on a vest
wearable for a soldier. The first step taken to solve this problem was to identify antennas
that would be suitable elements for the BWAV. These antenna elements were identified
through the literature review. Next, the identified antenna elements were designed in
FEKO electromagnetic software, and were compared against each other via their VSWR
(voltage standing wave ratio) in free space and against a dielectric body. The antenna
element that behaved the best was then optimized in OPT FEKO or GA-FEKO. The final

antenna element was then simulated for its radiation patterns and gain.

 

 

Chapter 2: Simulation Methods

2.1 Introduction

The purpose of Chapter 2 is to explain the simulation methods and programs used or
implemented while working on the BWAV project. This chapter is especially important
because simulation within the electromagnetic solver FEKO was the heart of this project;
the simulation process can be considered the analog of a fabrication procedure in terms of
importance. In order to find a suitable antenna element for the BWAV, several antennas
were designed and simulated in FEKO after a careful population of antenna elements
were chosen during the literature review.

In this chapter one will find sections on FEKO usage, OPT FEKO, and GA—
FEKO. Each section will give a brief overview of the programs used, the simulation

algorithms, and simulation parameters used during the antenna design process.
2.2 FEKO

2.2.1 Introduction to FEKO

As from [17], the term FEKO is an abbreviation of the German phrase F Elderchnung bei
Korpern mit beliebiger Oberflache which means in English Fields computations
involving bodies of arbitrary shape. In essence, FEKO is electromagnetic solver which
can perform electromagnetic analyses of problems of an arbitrary shape. FEKO uses the
Method of Moments (MOM), Physical Optics, and the Uniform Theory of Diffraction to

perform its analyses. All of FEKO calculations are performed in the frequency domain

 

with the 3 jar time convention. In addition, FEKO supports time domain harmonic
sources, plane wave incidence, various voltage gap formulations, magnetic ring currents,
and impressed currents and patterns.

FEKO was used for this project for several reasons. One reason FEKO was
implemented in this project was because of the ease to input dielectric bodies and their
properties; this was very important because all of the antenna elements would have to be
simulated in front of a dielectric body similar to a human torso. Another reason FEKO
was used in this project was because of the ease of controlling and changing the
segmentation and wire radii which highly control the run time and results of simulations.
The optimization features of OPT FEKO and the program GA-FEKO (a optimization
program developed at John Ross and Associates which uses genetic algorithms) were
reasons to use FEKO, too. Also, FEKO was chosen because of its popularity at MSU and
the ease to find others using the program.

The FEKO user interface is composed of three main components - CADFEKO,
EDITFEKO, and POSTFEKO. CADFEKO was not used in this project and will not be
discussed any further.

There are four steps necessary to create and view a geometry in FEKO. The first
step is to create *.pre file which contains the geometry and solution parameters;
EDITFEKO is the text—editor used to create and edit the *.pre file. Next, PREFEKO
processes the *.pre file and creates a *.fek file for FEKO. Then FEKO, the actual field

calculation code, process the *.fek file and creates and *.out file which contains all of the

;

 

solution data. Finally, POSTFEKO can be used for all post processing of data in the

*.out file.

2.2.2 MoM and FEKO

FEKO, the actual code that computes electric and magnetic field values, uses a
combination of the integral equation method and the method 'of moments to calculate
fields. The first step that FEKO takes is to calculate the electric surface currents [8] on
perfectly conducting surfaces or the equivalent electric and magnetic surface currents on
a dielectric surface. From the surface currents or equivalent surface currents, FEKO can
then calculate parameters such as the input impedance, gain, and radiation patterns of
antennas.

To calculate the surface currents or the equivalent surface currents, FEKO
implements the electric field integral equation (EFIE) by default; the EFIE enforces the
tangential boundary conditions of the electric field. FEKO can also implement the
combined field integral equation (CFIE), but this must be enforced by the user. The
EFIE is often used to solve radiation and scattering problems, and can solve problems
where the geometry is an open or closed body. The purpose of the EFIE is to cast the
solution for the unknown surface current in an integral equation where the induced

surface current is a part of the integrand [2]. One form of the EFIE from [1] is below:

VV - Z +k2K
jargo 5(1).

fiin’” = —fi><

where the variables in (1) are defined in Figure l.

11

 

 

Variable Definition

 

Incident Electric Field

 

Surface of Scatterer

C4111:

 

Normal Vector of S that points

3>

into the Domain

 

Perrnittivity of Free Space

on
o

 

Angular Frequency

 

Pr8

Free Space Wavenumber

 

Magnetic Vector Potential

{bl

 

 

 

Table 1 - Variable Definitions in (1)

MoM is simply the numerical method by which FEKO solves the EFIE. The

EFIE can be thought to be in the following form:

F(8)=h (2)

In (2), F is a linear operator, h is an excitation, and g is the response function which is the
surface current or equivalent surface current. The method of moments technique solves

equation 2 by expanding g into a linear combination of N terms like in (3).

l2

 

N
g(z') = 20.3,. (z’) (3)
n=1

In equation 3, an is an unknown coefficient and gn( Z ') is a basis function which has the

same domain as g(z '). The basis functions for perfectly conducting surfaces in FEKO are
triangles for surfaces, and segments for wires. For dielectrics, FEKO uses triangles for
surfaces or cuboids for volume regions for the basis functions [8]. (3) is then substituted

into equation 2 to give (4):

FEKO then obtains the a,l coefficients by solving a system of linear equations. Many
MoM solvers find the an coefficients from (4) by solving N linearly independent
equations formulated via point matching or collocation. Point matching or collocation

evaluates (4) at N different points. Therefore, the system of equations is

N
ZInF(gn)z=zm = hm,m =1,2.....,N

n=1 (5)

or, written as a matrix equation,

[Z... 111. l = [V.. 1.6)

where

Zmn — Flgn ) _ ’In _ an’Vm — h(zm)
Z—Zm (7)
1,, in equation (6) can then be solved for by matrix inversion or matrix inversion
techniques such as Gaussian elimination, conjugate gradient, bi-conjugate gradient, etc.
Therefore, the solution of (5) or (6) can then be substituted into (3) to find g(z') , the

surface current or equivalent surface current.

2.3 FEKO Simulation Methods

FEKO was the main simulation tool used to design the BWAV. The simulation method
or algorithm used by the author is straightforward. The first step taken was to design one
of the chosen antenna elements in EDITFEKO. As per the FEKO user manual, the wire
segmentation was kept to less than 71/ 10 and the wire radius was kept to four times the
wire segmentation. Once the element was designed in EDITFEKO, the design was
simulated by FEKO from 20‘ - 150 MHz using ten frequency points; only ten frequency
points were used initially to keep the run time down. After all of the elements had been
simulated for their initial test simulation, the elements were compared via their VSWR
VSWR was chosen as the initial comparison parameter because if the VSWR compared
to a 50 Q line was not close to being less than 2:1, the antenna would not have a good
input impedance and would not be able to be fed in practice. The antenna with the best
VSWR performance was then simulated for a characterization simulation which consisted
of 100 frequency points from 20 — 150 MHz. The purpose of the characterization

simulation was to get an idea of the complete performance of the antenna in free space.

k

In order to optimize the antenna’s performance, OPTFEKO was used to optimize various

features of the antenna. Finally, the antenna was simulated in front of a dielectric cuboid.
2.4 OPI‘FEKO

2.4.1 Introduction — What is OPTFEKO?

As in [17], OPTFEKO is a code within the FEKO package that has the ability to optimize
or vary parameters so as to obtain a specific performance from a geometry. For example,
OPTFEKO can vary the physical length of an antenna to obtain the optimal antenna
radiation pattern for a specific application. OPTFEKO can optimize or vary any
parameter that can be written symbolically in the *.pre file; those parameters are called
optimization parameters. An aim function is used within OPT FEKO to determine how
close a solution is to a certain goal or target. Example aim functions are gain, radiation
patterns, impedance, and resonance. In addition, OPTFEKO uses various optimization
methods that can be specified by the user. Examples of the optimization methods that
FEKO uses are the conjugate-gradient method, the quasi—Newton method, and the
discrete points method.

OPTFEKO needs two files to perform an optimization — a *.pre file and a *.opt
file. The *.pre file contains the optimization variables in symbolic form already written
in terms of the other components of the geometry. The *.opt file specifies which
variables within the *.pre file will be optimized and their minimum and maximum values.
The *.opt file also contains the optimization method, optimization parameters, and the
aim function. Once the *.pre file and the *.opt file have been formulated, OPTFEKO is

implemented to perform the actual optimization. Finally, OPTFEKO will create various

‘

 

other *.pre files with different values of the optimization variables that have to be run in

FEKO to obtain the results of the optimization process.

2.4.2 OPTFEKO Usage

OPTFEKO was used as tool for optimizing various antenna parameters in this project.
Antenna parameters that were optimized with OPTFEKO were antenna length, width,
and element spacing. The first step in the optimization was to choose appropriate bounds
for the optimization variables such that the antenna would not exceed the size for a
BWAV. The impedance/reflection and resonance aim functions were used primarily for
optimization. The impedance/reflection aim function was used because the
impedance/reflection aim function minimizes the reflection coefficient; minimizing the
reflection coefficient of the antenna was found to be a good approach because it also
maximizes transmitted or received power, since the input impedance of the antenna
becomes that of standard equipment. The resonance aim function was used because the
resonance aim function rrrinimizes the imaginary part of the input impedance; minimizing
the imaginary part of the input impedance was found useful because small antennas are
often highly reactive.

Various optimization methods were implemented to see which method performed
best. Finally, after OPTFEKO was run and the various *.pre files were run with the
optimized variables, the results were compared to see which optimization variable had

the most impact on the antenna’s performance.

 

2.5 GA-FEKO
GA-FEKO is a binary genetic algorithm optimizer that interfaces with FEKO. GA-
FEKO works by first allowing the user to create a template that contains encoded
variables. The encoded variables can be any physical antenna parameter the user desires.
GA-FEKO then creates a population of *.pre FEKO files. A population is all of the
possible *.pre files created from generating all the possible combinations of encoded
variables. Next, GA-FEKO creates an initial generation of *.pre files, which is a random
selection from the population. GA—FEKO then runs FEKO on the initial population and
ranks the *.pre files by fitness; fitness is the antenna parameter or parameters that the user
is trying to optimize. For example, VSWR, input impedance, and radiation patterns are
all possible fitness functions. Thereafter, GA-FEKO moves to create another generation
by mutating some *.pre files (randomly changing their encoded variables), allowing
certain *.pre files to crossover to the next generation (creating a new *.pre file by
combing two *.pre files from the previous generation in a specified manner), and
allowing certain *.pre files to exactly go to the next generation. The rate of mutation and
crossover are both chosen by the user.

GA-FEKO was used in this project to optimize an antenna called the self-
structuring antenna (SSA). The SSA can be reconfigured in many different ways, and
GA-FEKO was the method of optimizing for VSWR to find the optimal configuration of

the SSA.

 

Chapter 3: Antenna Designs

3.1 Introduction

The intent of Chapter 3 is to present information about the antennas simulated for the
BWAV. Background information on the antennas themselves, physical parameters and
dimensions are presented within the chapter. The antennas simulated were loops,

fractals, meander line bowties, and self-structuring antennas (SSA).
3.2 Elliptical Loops

3.2.1 Background on Elliptical Loops

Loop antennas are one of the most basic, prevalent, and versatile antennas; loops are
generally low-cost and easy to build. Loop antennas can be found in HF, VHF, and UHF
antenna applications [18], and can be designed in a variety of geometric shapes — circles,
squares, triangles, ellipses, etc.

Loop antennas were a starting antenna design in this project due to their
popularity at Sle. Loop antennas can be found in many of the direction finding
systems at Sle. Since the antenna being developed was for the BWAV, which will be
used in various direction finding applications, loop antennas were a clear starting point
for the present research. In addition to the easy integration into direction finding
technology, the elliptical loop could easily be physically implemented in the existing

SwRI BWAV prototype.

 

3.2.2 Elliptical Loop Design

The BWAV prototype at SwRI has pockets sewn into the inside of the vest that lend
themselves for an antenna fabricated on a printed circuit board. Thus, the loop design
took into account the size and shape of the pockets of the BWAV prototype. The pockets
of the BWAV prototype are rectangular and 76 mm by 152 mm. Thus, the loop antenna
should take the shape of an ellipse with major axis 152 mm and minor axis 76 rmn. The
feed point of the elliptical loop was chosen along one of the sides coinciding with the

major axis. A diagram of the elliptical loop designed in FEKO is in Figure 1.
3.3 Fractals

3.3.1 Background on Fractals

Fractal antennas, which are antennas with repeated self-similar geometry, are becoming
common antenna elements to use when antenna miniaturization is a goal. Fractals are
widely used in personal communication systems, satellites, and various other wireless
communications. Fractals are attractive antenna elements because they can be wide-
banded or multi-banded, and low-profile.

Fractals were used in this project as a way to improve upon the elliptical loop. In
essence, it was thought by the author that implementing a fractal loop would give more
electrical length in the same 76 mm by 152 mm area of the elliptical loop; it was hoped
that this extra length would improve upon the performance of the elliptical loop.
Essentially, the square fractal was an effort to see if implementing an antenna for the

BWAV prototype’s pockets for the 20-150 MHz frequency range would be feasible.

 

 

 

 

 

Figure 3-1-Elliptica1 Loop

20

 

 

 

 

 

Figure 3-2 — Minkowski Fractal

21

 

3.3.2 Fractal Design

The fractal antenna design implemented in this project was based on a square 2nd order
Minkowski fractal loop in [15] and [12]. The Minkowski fractal loop is a deterministic
fractal. A deterministic fractal is fractal generated by repeating the same scaled down
geometry over and over [18]. In this case, the self-repeating geometry was chosen to be a
square like in [15] and [14]. The author decided to maintain the same self-repeating
geometry as the fractals found in [15] and [14] because of the popularity and ease of
using squares as a self-repeating geometry. If the second order Minkowski fractal yielded
a significant improvement in VSWR over the elliptical loop, then more fractal iterations
would have been added.

The square fractal design fits in a square 60 mm area. The square Minkowski
fractal was fed by breaking one of its wires in the center of the design. A diagram of the

Minkowski fractal is in Figure 3.2.
3.4 Meander Line Bowtie Antenna

3.4.1 Background on Meander Line Bowtie Antenna

The meander line bowtie antenna is a combination of a meander line antenna, which can
be made simultaneously physically small and electrically large, and bowtie antennas,
which are naturally widebanded. Meander line bowtie antennas are found in various
military and communication applications. The wideband nature of meander line bowties
fits well with the BWAV because the BWAV design is supposed to be wideband (20-150

MHz). In addition, the use of the meandering wires helps increase the electrical length of

22

 

the antenna, which in turn helps improve parameters like input impedance and VSWR at

the lower frequencies. The design implemented in this project was highly based on [14].

3.4.2 Meander Line Bowtie Antenna Design

The meander line bowtie antenna design used in this project was based on [14], with the
only exception was that the antenna was made into a full bowtie instead of a half bowtie
as in [14].

The design of the meander line bowtie antenna is largely based on the geometry
of an isosceles triangles. First, the bowtie angle, a — the one unequal angle of the
isosceles triangle, and the antenna height, H, are chosen. The bowtie angle is very
important because it controls the total width of the antenna. Next, the number of
meanders, N, is chosen; once the number of meander lines is chosen, the segment length,
e, is found by dividing the length of one of the equal isosceles triangle legs by N. Once
all H, N, a, and e are known, a meander line bowtie antenna can easily be designed using
EDITFEKO. The parameters H, N, a, and e are indicated in Figure 3.3.

The meander line bowtie design went through several iterations. The first design
was intended to fit on a human torso and thus was 240 mm x 500 mm. After many
FEKO simulations and optimizations on the bowtie angle and segment length, it was
decided by the author that in order to acquire optimal antenna VSWR and input
impedance, the meander line bowtie antenna should be extended to a full body length.
Therefore, the meander line bowtie’s dimensions were changed to 965.2 mm x 1828 mm

(3.8 ft. x 6 ft.).

23

 

 

i

f
‘s

34
I

 

 

 

 

Figure 3-3—Meander Line Bowtie Antenna

24

 

3.5 Self-Structuring Antennas (SSA)

3.5.1 Background on the SSA

As stated in [19], the SSA is a “number of wires or patches interconnected by

”

controllable switches. The opening and closing of the switches creates many different
antenna configurations; binary genetic algorithms like those in GA-FEKO are used to
find optimal configurations for specific applications.

The SSA lends itself to be used in applications where the antenna environment is
changing because the SSA can adapt itself to give optimal performance. This trait of the
SSA is the main reason for choosing it for this project because the BWAV bandwidth is
so wide that it is very difficult to design one specific antenna for the entire bandwidth;
the SSA has the ability to choose from many different antenna configurations as the

frequency is changed to achieve optimal VSWR, input impedance, gain, and radiation

patterns.

3.5.2 SSA Designs

The SSA design implemented in this project is based on [19]. The SSA design is
essentially a quilt of wires and controllable switches. The dimensions were chosen such
that it would occupy the entire body length. The dimensions were thus 1016 mm x 1825
mm. The SSA design 61 switches total. GA-FEKO was also used to generate the
population and test for fitness for the SSA. A diagram of the SSA in free space is in

Figure 3-4, where the red square is the source.

25

 

Since the SSA has a width that will not fit entirely on one side of the average
human male, a conformal SSA was designed in FEKO as shown in Figures 3.6 and 3.7.
The conformal SSA also has a total of 61 switches. In addition, the conformal SSA was
simulated in front of a dielectric body is shown in Figure 3-5 and 3.6. In Figures 3.5-3.8,
the red squares are sources.

The dielectric body is simply a dielectric cuboid in FEKO. The dielectric
properties were chosen to meet the Federal Communications Commissions’ Office of
Engineering and Technology standards for human phantoms used for the testing of

human exposure to radio frequency energy [20]; the guidelines state that the relative

permittivity be less than 5, the dielectric loss tangent be less than .05, and 1000 kg/m3 be

used for mass/density. Therefore, the relative permittivity was chosen to 4, the dielectric

loss tangent was chosen to be .04, and the mass/density was chosen to be 1000 kg/m3 for

the dielectric body for this project.

26

 

 

 

Controllabl Switches

 

 

 

Figure 3-4-SSA Design

27

 

 

 

—

1‘—

ii

Figure 3-5 - Conformal SSA in Free Space

 

 

 

 

 

 

 

 

{—_\

 

 

 

Figure 3-6 - Conformal SSA Cross Section in Free Space

 

 

 

 

 

 

 

 

Figure 3—7 - Conformal SSA in front of Dielectric Body

 

 

Figure 3-8 - Conformal SSA in front of Dielectric Body Cross Section

 

 

 

29

 

Chapter 4: Results

4.1 Introduction

The intent of Chapter 4 is to present the simulated results of the antennas described in
Chapter 3. The objective of all the antennas simulated was to achieve similar or better
results than SwRI’s BWAV prototype. SwRI was able to achieve a maximum VSWR of
10 across the 50 — 3000 MHz frequency range; therefore, the primary goal was to achieve
a VSWR with respect to a 50 Q line of less 10 for the 20 — 150 MHz frequency range.

To get a better sense of the overall performance of the simulated antennas, other
antenna parameters are presented within the chapter. For the SSA antennas, the radiation
patterns are presented so the reader can get a sense of how the elements radiate. In
addition, the current distribution and SAR (specific absorption ratio) data are presented

for the SSA elements.

4.2 Elliptical Loop, Minkowski Fractal, and Meander Line Bowtie Antenna Results

The elliptical loop, Minkowski fractal, and meander line bowtie antenna were simulated
only to detemrine their VSWR. The VSWR vs. frequency plot is shown in Figure 4.1 for
all three antennas. With careful observation of the scale of the y-axis, one can see that
the VSWR is in the thousands for almost the entire 20 -150 MHz frequency range for all
three antennas, and is clearly not an acceptable result. The poor performance of these
antennas is attributed to their deficiency in size; all three antennas are significantly

smaller than the wavelengths (2 m — 15 m) of the frequency range.

30

‘2‘

 

VSW R

VSWR [1803]

 

20 40 60 so 1 00 1 20 140 1 60
Frequency {MHz}
- Elliptical Loop — Minkowski Fried
— Momdor Lino em
Figure 4—1 VSWR vs. Frequency of Elliptical Loop, Minkowski Fractal, and

 

 

 

Meander Line Bowtie Antennas

31

 

4.3 Full-Size Meander Line Bowtie Antenna Results

The dimensions of the meander line bowtie antenna described section in 4.2 were
rescaled to be 1828 mm x 965 mm (6 ft. x 3.16 ft) to create the full-size meander line
bowtie antenna. The meander line bowtie antenna from section 4.2 from was chosen to
be extended to a full body size because it had the best VSWR performance of the
antennas studied in section 4.2.

The full size meander line antenna was tested for VSWR in both free space and in
front of a dielectric body. The dielectric body electrical properties were chosen based on
the Federal Communications Commission (FCC) flat phantom specifications for radio

frequency (RF) exposure [20]. The FCC specifies that flat phantoms have a relative

permittivity, er, less than 5 and a dielectric loss tangent, 8, of less than .05. Therefore, the

electric properties of the dielectric body were er = 4 and 8 = .04. In addition, the

dielectric body was placed 50 mm behind the meander line bowtie antenna, and was 846
mm W x 1825 mm L x 355 mm D in size. The VSWR vs. frequency plot for the full-size
meander line antenna in free-space and in front of a dielectric body is in Figure 4.2; a
zoomed in portion of Figure 4.2 from 60 - 150 MHz is in Figure 4.3. From observation of
Figures 4.2 and 4.3, one again can see that the VSWR performance for the full-size
meander line bowtie antenna is not acceptable. These poor results can be attributed to the
fact that full-size meander bowtie antenna width is still not big enough compared to the

wavelengths of the frequency range to yield a suitable result. For instance, the meander

32

‘

 

line bowtie width ranges from .06}. - .48)» from 20 - 150 MHz and the meander line

bowtie length ranges from .12)» - .91}. from 20 — 150 MHz.
4.4 SSA Results
4.4.] Flat SSA in Free Space Results

4.4.1.1 Flat SSA in Free Space VSWR Results

The SSA is able to adapt itself as frequency changes for a particular parameter; for this
project, the parameter was chosen to be VSWR compared to a 50 9 line because a good
VSWR (less than 2:1 typically) ensures a decent input impedance (50 Q + jO Q) for the
antenna. Therefore, for every frequency under test there is a different SSA configuration,
and thus an optimal VSWR.

The VSWR vs. frequency plot for the flat SSA in free space is shown in Figure
4.4. The VSWR data produced was achieved by running GA-FEKO between 12-20

generations with a population of 40-70 antenna configurations. The flat SSA has a total

1
of 61 switches, which creates 2n = 26 = 2.30e18 possible SSA configurations.

All of the frequency points in Figure 4.4 meet the SwRI VSWR criteria of less
than 10:1 for the entire frequency band of 20— 150 MHz. Optimally, the VSWR should be
less than 2:1 for the entire frequency band, and 78% of the frequency points have a
VSWR of less than 2:1 for the frequency band. Better VSWR results are possible for the
remaining frequency points if GA—FEKO is run for additional generations to try to reach

a VSWR of less than 2:1.

33

 

VSWR

 

 

 

 

 

 

 

 

 

 

 

 

 

 

so
”I
§ 30
32
g 20
>
10” l
o A A

20 40 60 80 11” 120 140 160
FreqrunylMHz]
I-Fmspoeo -lnFrortofDio|octricBodyl

 

 

 

 

Figure 4-2 - Full-Size Meander Bowtie Antenna VSWR vs. Frequency Plot

34

 

 

VSWR [1 e+03]

 

0.0
60 70 80 90100110120130140150160

FrewlMl-IZI
- rm - lnFroflofDidoctIicBody

 

 

 

Figure 4-3- Full-Size Meander Line Bowtie Antenna VSWR vs. Frequency (60-150

MHz)

35

‘

 

 

VSWR vs. Frequency (SSA in Free Space)

A
$2.5 / \
t” / \ A /
\v/ \ /\ /
\/ W

O , 50 100 150
Frequency (MHz)

 

 

 

U.)

 

 

N

 

 

 

 

 

 

 

Figure 4-4 - VSWR vs. Frequency for Flat SSA in Free Space

36

 

 

4.4.1.2 Flat SSA Radiation Patterns in Free Space

The radiation patterns are presented to give the reader a sense of how the flat SSA in free
space radiates. Figure 4.5 shows the total farfield azimuthal radiation patterns for 20, 40,
80, 100, 120, and 150 MHz; Figure 4.6 shows the total farfield azimuthal gain on a polar
plot for the same frequencies. For all of the frequencies, nulls in the pattern appear near
75 degrees and 250 degrees. A general pattern for the behavior of the maximum null
depth cannot be found over the frequency band; this is clearly observed in Figure 4.6
where one can more easily detect sharp changes in the patterns as frequency changes.
This behavior can most likely be explained by the fact that at each frequency there is a

different antenna configuration with different VSWR and input impedance.

4.4.1.3 Flat SSA Current Distribution in Free Space

Figures 4.8—4.21 show the current density on the wire segments of the SSA for various
frequencies within the 20-150 MHz frequency band. The current density is presented to
give the reader a sense of the current hot spots along the antenna. Unfortunately, no
major conclusions can be drawn from frequency to frequency since there is a different
antenna configuration at each frequency. Figure 4-7 is a general color code for all of the

current density plots in this thesis.

37

 

 

 

Gain

 

 

 

 

 

 

 

 

 

 

 

Gain [:18]

 

 

 

 

 

 

 

 

 

 

 

 

 

 

O 50 100 150 200 250 300 350 400

AngloPhi[deg]
—20MHz —40NHZ —80MHz —100M-lz —120MHz
—150MHz

 

 

 

 

 

 

Figure 4—5— Azimuthal Total Gain of Several Frequencies of Flat SSA in Free Space

38

 

 

 

Gain

 

 

 

v
o
v
-
.
p

.I

‘
"----

 

 

 

— 100MHz - 120MHz

—40MHZ —80NHZ

- 20MHZ
—150MHZ

 

 

 

Figure 4-6 - Polar Plot for Total Azimuthal Gain for SSA in Free Space

39

 

-30
~50
-70
-90
~110
—l30
-150
-l70
-l90
-210

 

 

 

Figure 4-7-Color Code for Current Density Plots

40

 

 

——F_i—_——

 

 

 

 

 

 

 

 

 

Figure 4-8 - Current Density on SSA in Free Space, 20 MHz

41

 

 

 

 

 

 

 

 

 

 

 

 

 

Figure 4-9 - Current Density on SSA in Free Space, 30 MHz

42

 

 

 

 

 

 

 

 

 

 

 

Figure 4-10 - Current Density on SSA in Free Space, 40 MHz

43

 

Figure 4—11 — Current Density on SSA in Free Space, 50 MHz

 

 

 

 

Figure 4-12 - Current Density on SSA in Free Space, 60 MHz

 

 

 

45

 

 

Figure 4-13 - Current Density on SSA in Free Space, 70 MHz

 

 

 

46

 

Figure 4—14 - Current Density on SSA in Free Space, 80 MHz

 

 

 

47

 

 

 

 

 

Figure 4-15 — Current Density on SSA in Free Space, 90 MHz

 

 

 

48

 

Figure 4-16 - Current Density on SSA in Free Space, 100 MHz

 

 

 

49

 

 

Figure 4-17 - Current Density on SSA in Free Space, 110 MHz

 

 

 

50

 

Figure 4-18 - Current Density on SSA in Free Space, 120 MHz

 

 

 

51

 

Figure 4-19 — Current Density on SSA in Free Space, 130 MHz

 

 

 

52

 

Figure 4-20 - Current Density on SSA in Free Space, 140 MHz

 

 

 

53

 

Figure 4-21 - Current Density on SSA in Free Space, 150 MHz

 

 

 

54

4.4.2 Conformal SSA in Free Space Results

4.4.2.1 Conformal SSA in Free Space VSWR Results

The best VSWR vs. frequency plot for the conformal SSA in free space is shown in
Figure 4.22. The VSWR data produced was achieved by running GA-FEKO between 20-
30 generations with a population of 40-70 antenna configurations. The conformal SSA
also has a total of 61 switches like the flat SSA.

All of the frequency points in Figure 4.22 meet the SwRI VSWR criteria of less
than 10:1 for the entire frequency band of 20-150 MHz. Optimally, the VSWR should be
less than 2:1 for the entire frequency band, and approximately 85% of the frequency
points have a VSWR of less than 2:1 for the frequency band. This is a significant
increase in percentage over the flat SSA. This can be attributed to the fact that more
generations with larger populations were run in GA-FEKO to find the optimal solution.
Likewise in the case of the flat SSA, better VSWR results can be achieved for the
remaining frequency points if GA-FEKO is run for additional generations to try to reach

a VSWR of less than 2: 1.

4.4.2.2 Conformal SSA Radiation Patterns in Free Space

The radiation patterns are presented to give the reader a sense of how the conformal SSA
in free space radiates. Figure 4.23 shows the total farfield azimuthal radiation patterns
for 20, 40, 80, 100, 120, and 150 MHz; Figure 4.24 shows the total farfield azimuthal
gain on a polar plot for the same frequencies. For all of the frequencies, nulls in the

pattern appear between 100-125 degrees and 250-350 degrees. The nulls in the radiation

55

patterns are also significantly less deep than the nulls of the flat SSA in free space; the
maximum null for the flat SSA is approximately 25 dB, while the maximum null depth of

the conformal SSA is 12 dB for most frequencies.

4.4.2.3 Conformal SSA Current Distribution in Free Space

Figures 4.25-4.37 show the current density on the wire segments of the SSA for various
frequencies within the 20-150 MHz frequency band. From observation of figures, the
current distribution varies a lot as frequency varies. This observation shows further that

at each frequency, there is a different configuration of the SSA.

4.4.3 Conformal SSA in Front of Dielectric Body Results

4.4.3.1 Conformal SSA in Front of Dielectric Body VSWR Results

The conformal SSA was modeled as described in section 3.6. The VSWR vs. frequency
plot for the conformal SSA in free space is shown in Figure 4.38. The VSWR data
produced was achieved by running GA-FEKO between 20-30 generations with a
population of 40-70 antenna configurations. The conformal SSA also has a total of 61

switches like the flat SSA and the conformal SSA in free space.

56

 

 

VSWR

4.5

VSWR vs. Frequency (Conformal SSA in Free Space)

 

 

 

 

 

 

 

 

 

 

 

 

 

 

 

 

 

 

 

l I

 

 

 

20 40 6O 80 100 120 140
Frequency (MHz)

 

Figure 4—22 - VSWR vs. Frequency for Conformal SSA in Free Space

57

 

 

Gain

 

 

-_--- ---4

 

 

Gin [dB]

 

 

 

 

 

 

 

 

 

 

 

 

 

 

 

0 50 100 150 200 250 300 350 400
Anglo Phi [deg]

 

 

—' 20 MHz — 40 MHZ — 80 MHZ — 100 MHz
— 120 MHz 150 MHz

 

 

Figure 4-23 - Azimuthal Total Gain of Several Frequencies of Conformal SSA in Free

Space

58

 

 

......
.....

{horas}

~10
~15
~20
23

~

  
 

 

 

_L
‘29
I
I
4--_
I
O

-o
a""

 

— 20 MHZ — 40 MHZ — 80 MHZ — 100 MHZ
— 120 MHZ 150 MHZ

 

 

Figure 4-24 - Azimuthal Total Gain of Several Frequencies of Conformal SSA in Free

Space, Polar Plot

59

 

Figure 4-25 - Current Density on Conformal SSA in Free Space, 20 MHz

 

 

 

60

 

Figure 4-26 - Current Density on Conformal SSA in Free Space, 40 MHz

 

 

 

61

 

Figure 4-27 - Density on Conformal SSA in Free Space, 50 MHz

 

 

 

62

 

Figure 4—28 - Current Density on Conformal SSA in Free Space, 60 MHz

 

 

 

63

 

Figure 4-29 - Current Density on Conformal SSA in Free Space, 70 MHz

 

 

 

 

Figure 4-30 - Current Density on Conformal SSA in Free Space, 80 MHz

 

 

 

65

 

Figure 4-31 - Current Density on Conformal SSA in Free Space, 90 MHz

 

 

 

66

 

Figure 4—32 - Current Density on Conformal SSA in Free Space, 100 MHz

 

 

 

67

 

Figure 4—33 Current Density on Conformal SSA in Free Space, 110 MHz

 

 

 

68

 

Figure 4-34 - Current Density on Conformal SSA in Free Space, 120 MHz

 

 

 

69

 

Figure 4-35 - Current Density on Conformal SSA in Free Space, 130 MHz

 

 

 

70

 

Figure 4-36 - Current Density on Conformal SSA in Free Space, 140 MHz

 

 

 

71

 

Figure 4-37 - Current Density on Conformal SSA in Free Space, 150MHz

 

 

 

72

All of the frequency points in Figure 4.21 meet the SwRI VSWR criteria of less
than 10:1 for the entire frequency band of 20-150 MHz, and 83% of the frequency points
meet the optimal standard of less than 2:1 for VSWR. From 40-150 MHz, the VSWR
performance of the conformal SSA in front of the dielectric body is similar to that of the
conformal antenna in free space. The one outlier is the VSWR at 20 MHz which has a
significantly higher VSWR than the other frequency points. An optimal VSWR was
difficult to achieve in front of the dielectric body at 20 MHz because it was difficult to
find an appropriate fitness function of VSWR and input impedance to find the best SSA

configuration.

4.4.3.2 Conformal SSA Radiation Patterns in Front of Dielectric Body

The radiation patterns are presented to give the reader a sense of how the conformal SSA
in front of a dielectric body radiates. Figure 4.39 shows the total farfield azimuthal
radiation patterns for 20, 40, 50, 60, 100, and 150 MHz; Figure 4.0 shows the farfield
azimuthal gain on a polar plot for the same frequencies. The nulls of the conformal SSA
in front of the dielectric body are very similar to the nulls of the conformal SSA in free
space. The null of the conformal SSA in free space are between 100-125 degrees and
250-350 degrees. The nulls of the conformal SSA in front of the dielectric body appear
between 80-150 degrees and 250-325 degrees. The radiation patterns of the conformal
SSA in front of the dielectric body are also smoother than radiation patterns of the

conformal SSA in free space; this is clearly evidenced in Figure 4.39.

73

4.4.3.3 Conformal SSA Current Distributions in Front of Dielectric Body

Figures 4.41—4.46 show the current density on the wire segments of the conformal SSA in
front of the dielectric body for various frequencies within the 20-150 MHz frequency
band. From comparison of the current density plots, one can see that again the current
distribution along the wires is varying greatly as frequency varies. This can be explained

once again by the fact that there is a different antenna configuration at each frequency.

4.4.3.4 Conformal SSA SAR Results

Specific absorption ratio or SAR is way to detect how much RF energy is absorbed by a
dielectric body. The calculation of SAR is important for this project for two reasons. One
reason the SAR data is important is because one of the major goals of the BWAV is for
the BWAV to be worn by human beings. If the BWAV is to be worn by humans, then it
must pass specific guidelines set by the FCC for SAR to ensure the safety of potential
BWAV users. The second reason the SAR is important is because it gives an idea of
how much power is being absorbed by the dielectric body and not radiated into space.
The FCC stipulates that the SAR must be less than 1.6 W/kg per one gram of tissue when
being tested against a body phantom whose parameters were discussed in section 3.6.
The SAR data presented was a initial step in taking meaningful SAR data that can be
compared to FCC guidelines. For SAR to be meaningful, one must now the maximum
radiated power of the antenna, and drive the antenna at that power to acquire the proper
SAR data. At the time this data was collected, the desired radiated power was not
specified to the author by SwRI. Nonetheless, a plot of SAR vs. frequency is shown in

Figure 4.47.

74

 

 

VSWR

VSWR vs. Frequency (Conformal Antenna in Front of Dielectric

 

 

 

 

 

 

 

 

 

 

 

 

 

 

 

 

 

 

 

 

 

 

 

 

 

 

Body)
10 .
9
8 f\
7 ‘\
5 ‘ \
4 \
3 \:
2 \i -
1 ‘WJ We“:
0 i l
o 20 4o 60 so 100 120 140 160

Frequency (MHz)

 

Figure 4-38 - VSWR vs. Frequency for Conformal SSA in Front of Dielectric Body

75

 

 

Gain

Gain [as]

 

O 50 100 150 200 250 300 350 400

 

 

Anglo Phi [dog]
— 20 MHZ — 40 MHZ — 50 MHz — 60 MHZ
—' 100 MHZ 150 MHZ

 

 

 

 

Figure 4-39 - Azimuthal Total Gain of Several Frequencies of Conformal SSA in Front of

Dielectric Body

76

 

 

 

 

 

 

270

 

— 20 MHZ — 40 MHZ -- 50 MHZ — 60 MHZ
— 100 MHZ 150 MHZ

 

 

 

 

Figure 4-40 - Azimuthal Total Gain of Several Frequencies of Conformal SSA in Front of

Dielectric Body, Polar Plot

77

 

 

 

 

 

Figure 4-41 - Current Density on Conformal SSA in Front of Dielectric Body, 20 MHz

78

 

 

 

 

 

Figure 4-42 ~ Current Density on Conformal SSA in Front of Dielectric Body, 40 MHz

79

 

 

 

 

 

Figure 4-43 ~ Current Density on Conformal SSA in Front of Dielectric Body, 60 MHz

80

 

 

 

 

 

Figure 4-44 - Current Density on Conformal SSA in Front of Dielectric Body, 80 MHz

81

 

 

 

 

 

Figure 4-45 - Current Density on Conformal SSA in Front of Dielectric Body, 100 MHz

82

 

 

 

 

 

Figure 4-46 - Current Density on Conformal SSA in Front of Dielectric Body, 150MHz

83

 

 

50013-05
45013-05
4.00E-05
35013-05
3.00E-05
325013-05
2.00E-05
1 .50E-05
1.00E-05
5.00E-06
0.00E+00

SAR vs. Frequency (Conformal Antenna in Front of Dielectric

 

 

 

 

 

 

 

 

 

 

 

 

 

 

 

 

 

 

 

 

 

 

 

 

 

 

 

Bod)

\

\

\

I \

/\ I \

/\ l \ L

/ A
\l V\.

 

Figure 4-47 - SAR vs. Frequency for Conformal SSA in Front of Dielectric Body

84

 

Chapter 5: Conclusions

This thesis has presented the research and development of an antenna to extend the
bandwidth of a BWAV originally designed at SwRI. The main objective was to
development an antenna that could be implemented on the BWAV to extend the
preexisting bandwidth of 50-3000 MHz to include 20-150 MHz.

Initially, an introductory chapter was presented which included reviews of past
work done on the BWAV, a literature review of BWAV designs and similar applications,
and an overview of the BWAV. Secondary, the simulation programs and simulation
techniques were explained. Next, the candidate antennas chosen for the BWAV were
summarized by presenting background information on the antennas, the rationale for
choosing the antennas, and their dimensions and workings. The VSWR results of the
candidate antennas were presented and discussed; one particular antenna element was
examined further for its potential use on the BWAV — the self-structuring antenna or
SSA.

The SSA performance was discussed thoroughly in terms of VSWR, radiation
patterns, and current density in free space. The SSA was also reshaped as a conformal
antenna that could perceivably be bent around a human. The conformal SSA was also
simulated for its VSWR, radiation, patterns, current density in free space and in front of a
dielectric body whose parameters were chosen to match FCC RF phantom guidelines.
The conformal SSA was also simulated for SAR to see if it could pass FCC SAR safety

regulations, also.

85

In conclusion, the SSA had excellent VSWR in both the flat and conformal
configurations, which makes the SSA a legitimate choice for implementation in the
BWAV. For the SSA to be implemented in the BWAV the SSA would have to be
optimized for a wearable garment and optimized to be used in direction finding

applications in the future.

86

[1]

[2]

[3]

[4]

[5]

[6]

[7]

[8]

[9]

[10]

[11]

Bibliography

J. Pike, "Joint Tactical Radio System Programmable, Modular Communications
System," Alexandria, VA, 2007.

E. C. Kohls, A. Abler, P. Siemsen, J. Hughes, R. Perez, and D. Widdoes, "A
Multi-Band Body-Wom Antenna Vest," in IEEE Antennas and Propagation
International Symposium, 2004, pp. 447-450.

R. C. Adams, "Testing and Integration of the COMWIN Antenna System," in
Proceedings of MILCOM, 2002, pp. 637-641.

J. T. Apostolos, "Direction Finding System Using Body-Wom Antenna." vol.
US. Patent 6 771 224 B2, Aug. 3, 2004.

P. Salonen, J. Kim, Y. Rahmat-Samii, and H. Hurme, "Dual-Band E-Shaped
Patch Wearable Textile Antenna," in IEEE Antennas and Propagation
International Symposium, 2004, pp. 463-469.

P. Salonen, Y. Rahmat-Samii, H. Hurme, and M. Kivikoski, "Dual-Band
Wearable Textile Antenna," in IEEE Antennas and Propagation International
Symposium, 2004, pp. 463-469.

P. Salonen and H. Hurme, "A Novel Fabric WLAN Antenna for Wearable
Applications," in IEEE Antennas and Propagation Symposium, 2003, pp. 700-
703.

M. Tanaka and J. Jang, "Wearable Microstrip Antenna," in IEEE Antennas and
Propagation International Symposium, 2003, pp. 704-707.

S. Hayashida, H. Morishita, Y. Koyanagi, and K. Fujimoto, "Wideband Folded
Loop Antenna for Handsets," in IEEE Antennas and Propagation International
Symposium, 2002, pp. 2-5.

S. Hayashida, T. Tanaka, H. Morishita, Y. Koyanagi, and K. Fujimoto, "Built-in
Half Size Folded Dipole Antenna for Handsets," in IEEE Antennas and
Propagation International Symposium, 2004, pp. 2759-2762.

H. Morishita, Y. Kim, Y. Koyanagi, and K. Fujimoto, "A Folded Loop Antenna
System for Handsets," in IEEE Antennas and Propagation International
Symposium, 2001, pp. 440-443.

87

[12]

[13]

[14]

[15]

[16]

[17]

[18]

[19]

[20]

S. R. Best, "A Comparison of the Resonant Properties of Small Space-Filling
Fractal Antennas," in Antennas and Wireless Propagation Letters, 2003, pp. 197-
200.

J. P. Gianvittorio and Y. Rahmat-Samii, "Fractal Antennas: A Novel Antenna
Miniaturization Technique," IEEE Antennas and Propagation Magazine, vol. 44,
pp. 20-36, 2002.

M. Ali and S. Stuchly, "A Meander-Line Bow-tie Antenna," in IEEE Antennas
and Propagation International Symposium, 1996, pp. 1566-1569.

F. Arazm, R. K. Baee, C. Ghobadi, and J. Norinia, "Square Loop Antenna
Miniaturiazation Using New Fractal Geometry," in Twentieth NRSC, 2003, pp.
B4- 1-8.

D. H. Werner, R. L. Haupt, and P. L. Werner, "Fractal Antenna Engineering: The
Theory and Design of Fractal Antenna Arrays," IEEE Antennas and Propagation
Magazine, vol. 41, pp. 37-58, 1999.

FEKO User Manual, Suite 4.1. Stellenbosch, South Africa, 1998-2003.

C. A. Balanis, Antenna Theory: Analysis and Design. Hoboken, NJ: John Wiley
& Sons, Inc, 2005.

C. M. Coleman, E. J. Rothwell, J. E. Ross, and L. L. Nagy, "Self-Structuring
Antennas," IEEE Antennas and Propagation Magazine, vol. 44, pp. 11-23, 2002.

Federal Communications Commission - Office of Engineering Technology,
"Radio Frequency Safety," July 2007.

88

 

 

 

 

 

 

 

 

 

 

 

2
.. . m 7
m w
. . t 6
,, m 5
.v. 9
. u 2 .
. t. 0
. will 3 H
. sllo.
NAH2 .
My 1
.. Willi! 3 . .