VŠB – Technical University of Ostrava Faculty of Electrical Engineering and Computer Science Department of Information and communication Technology

(1)

Department of Information and communication Technology

Antěnní pole vaplikaci zabezpečení UWB Antenna Array in UWB Security Application

2019 Harihara Subramanian Ganesh

(2)

(3)

(4)

(5)

Acknowledgements

First and foremost, I would like to thank my supervisor, Ing. Marek Dvorsky, Ph.D, whose methodical guidance and valuable advice have been indispensable throughout the whole process of writing this thesis.

(6)

.

Abstract

The aim of this thesis is to design and implement a microstrip patch Vivaldi (UWB) array antenna with a single feed Wilkinson power divider network at an operating frequency of 5.5 GHz printed on a dielectric substrate ASTRA ® MT77 material which can reduce the security issues at each layer of UWB technology. The initial work of this thesis is to design and analyse the 2×1 and 4×1 Wilkinson power divider network which improved the isolation and matching of the ports. Later, on the designed array feeder network two and four rectangular radiating patches have been mounted in a linear configuration to achieve the required radiation properties. Finally with the help of simulated rectangular patch array antenna along its dimensions the structure has been improved to develop and construct the Vivaldi array antenna to attain the maximum bandwidth and gain in an optimum UWB bandwidth.

The antenna array is designed using standard equations and simulated by professional antenna development software called High frequency structural Simulator (HFSS). Among many antenna simulators, HFSS is selected as it allowed the inclusion of anisotropic ferrite material in the simulation process. The prototype array antenna is printed on a chosen dielectric substrate using the PCB plotter. Finally, the simulated reflection response and radiation characteristics of the designed Vivaldi array antenna are validated using experimental results obtained from the Network analyser and the Antenna Transmission & Measurement System respectively. It has been observed that the experimental array antenna exhibited very close radiation response to that of design objective.

Key Words

Microstrip Patch Antenna, UWB, Wilkinson Power Divider, HFSS, Rectangular Patch Array Antenna, Vivaldi Array Antenna, Network analyser and Bandwidth.

(7)

List of Tables

Table 3. 1. Comparison between microstrip transmission line and microstrip antenna ... 17

Table 4. 1: Design Dimensions of 2×1 Modified Wilkinson Power Divider ... 30

Table 4. 2. Simulated (2×1) Wilkinson Bandwidth ... 31

Table 4. 3: 2X1 Power Divider Total Bandwidths ... 35

Table 4. 4. Designing Dimensions of 4×1 Modified Wilkinson Power Divider ... 36

Table 4. 5. 4×1 Power Divider Total Bandwidth ... 38

Table 6. 1. Dimensions of Rectangular patch Antenna ... 45

Table 6. 2. Dimensions of 2-element Rectangular patch Antenna ... 46

Table 6. 3. Dimensions of 4-element Rectangular patch Antenna ... 46

Table 6. 4. Comparison of antenna characteristics between 2-element and 4-element ... 52

Table 7. 1. Dimensions of Antipodal Vivaldi Array Antenna ... 55

Table 7. 2: Comparison of antenna Characteristics between (2 and 4-element) AVA antenna . 58 Table 7. 3: Comparison of Side lobes and cross polarization between 2 and 4-element AVA in chosen frequency ... 58

(11)

List of Figures

Figure 2. 1: Antenna as a Transition device ... 5

Figure 2. 2: Transmission line thevenin equivalent of antenna in transmitting mode ... 6

Figure 2. 3: Various Shapes of Microstrip patch Antenna ... 7

Figure 2. 4: An N-Port Network ... 9

Figure 2. 5: Two Port Network Definition ... 10

Figure 2. 6: Coordinate System for antenna analysis [3] ... 11

Figure 3. 1: Microstrip Rectangular Patch Antenna [4] ... 14

Figure 3. 2: Radiation Mechanism associated with Micrsotrip Patch Antenna [4] ... 15

Figure 3. 3: Electric field distribution (side view) [4] ... 15

Figure 3. 4: Increase in length of the microstrip patch [4] ... 16

Figure 3. 5: Transmission-line model of rectangular Microstrip patch [4] ... 18

Figure 3. 6: Microstrip Line Feed ... 20

Figure 3. 7: Coaxial Probe Feed ... 21

Figure 3. 8: Aperture Feed ... 22

Figure 3. 9: Proximity Coupled Feed ... 22

Figure 3. 10: Pattern multiplication shown graphically [2] ... 24

Figure 4. 1: Block Diagram Power Divider ... 26

Figure 4. 2: Block Diagram Power Divider ... 27

Figure 4. 3: Typical Microstrip Wilksinon Power Divider ... 28

Figure 4. 4: Modified 2×1 structure of wilkinson power divider ... 29

Figure 4. 5: 2×1 wilkinson divider Return Losses ... 30

Figure 4. 6: Simulated 2×1 Wilkinson power division ... 32

Figure 4. 7: Simulated 2X1 Wilkinson Amplitude Balance ... 33

Figure 4. 8: Simulated 2×1 Wilkinson divider isolation loss ... 34

Figure 4. 9: VSWR of 2×1Wilkinson Power Divider ... 34

Figure 4. 10: Modified 4×1 structure of Wilkinson Power divider ... 36

Figure 4. 11: Return Loss of 4×1 Wilkinson divider ... 36

Figure 4. 12: Amplitude Balance of 4×1 Wilkinson divider ... 37

Figure 4. 13: Isolation Loss of 4×1 Wilkinson divider ... 37

Figure 4. 14: Power Division and VSWR of 4×1 Wilkinson divider ... 37

Figure 5. 1: Security issues at each layer of protocol stack ... 40

Figure 6. 1: Construction of 2-Element Rectangular Patch Microstrip Antenna Array... 45

Figure 6. 2: Construction of 4-Element Rectangular Patch Microstrip Antenna Array... 46

Figure 6. 3: Return Loss (S11) Comparison between 2-element and 4-element ... 48

Figure 6. 4: Comparison of Mismatch Losses between 2-element and 4-element ... 48

Figure 6. 5: Comparison of Reflection Coefficient between 2-element and 4-element... 49

Figure 6. 6: Comparison of VSWR between 2-element and 4-element ... 49

Figure 6. 7: Comparison of Gain between 2-element and 4-element ... 50

Figure 6. 8: Cross-Polarizations of 2-element Rectangular Patch Array Antenna ... 50

Figure 6. 9: Cross-Polarizations of 4-element Rectangular Patch Array Antenna ... 51

(12)

Figure 6. 10: 2D-radiation Pattern of 2-element ... 51

Figure 6. 11: 2D-radiation Pattern of 4-element ... 51

Figure 7. 1: Simulated Final Structure of 2-element AVA Array Antenna ... 54

Figure 7. 2: Simulated Final Structure of 4-element AVA Array Antenna ... 54

Figure 7. 3: Return Loss of 2-element and 4-element AVA ... 56

Figure 7. 4: Reflection Coefficient and Mismatch Loss of 2-element and 4-element AVA ... 56

Figure 7. 5: Gain measurement of 2-element and 4-element AVA and gain Comparative analysis of 2-element and 4-element between (4.5 to 10 GHz) ... 57

Figure 8. 1: Fabricated Front and back shape of the 4×1 AVA ... 59

Figure 8. 2: Testing of fabricated antenna using vector network analyzer [51] ... 59

Figure 8. 3: Comparison of simulated and experimental S11 response of 4×1 AVA antenna ... 60

Figure 8. 4: Comparison of VSWR between Simulated and practical 4×1 AVA antenna ... 61

Figure 8. 5: Comparison of Mismatch Match Loss between of 4×1 AVA Antenna ... 61

(13)

1

1 Introduction

According to the IEEE Standard Definitions, the antenna or aerial is defined as “a means of radiating or receiving radio waves” [1]. In other words, antennas act as an interface for electromagnetic energy, propagating between free space and guided medium. Amongst the various types of antennas that include wire antennas, aperture antennas, reflector antennas, lens antennas etc microstrip patches are one of the most versatile, conformal and easy to fabricate antennas. In this thesis, a microstrip rectangular patch array antenna with corporate feed network and Antipodal Vivaldi patch array antenna with corporate feed network beam steering capability is designed, fabricated and tested successfully.

1.1 Thesis Motivation

In the recent years, there has been rapid growth in wireless communication. With the increasing number of users and limited bandwidth that is available, operators are trying hard to optimize their network for larger capacity and improved quality coverage [1]. This surge has led the field of antenna engineering to constantly evolve and accommodate the need for wideband, low-cost, miniaturized and easily integrated elements [3]. A widely used antenna structure with above characteristics is microstrip antenna. The microstrip patch antennas are associated with several advantages of being low profile, versatile, conformal and low-cost devices. The advantages of microstrip antennas make them suitable for various applications like, vehicle based satellite link antennas [3], global positioning system (GPS) [4], radar for missiles and telemetry [3] and mobile handheld radios or communication devices [4]. But nonetheless, the microstrip antennas are also associated with some disadvantages, such as narrow bandwidth, low gain and the excitation of surface waves.

Over the years, a lot of research has been undertaken to overcome the disadvantages associated with these antennas. Some of the popular techniques proposed by researchers to widen the bandwidth are; increasing the height of antenna substrate [7], using aperture coupling method [7, 8] or using stacked patch structure [8]. Another method of using coplanar parasitic elements around or in line with primary driven element are also mentioned in the literature [ 9], where a geometry of four parasitic patches around four corners of a single probe fed patch yielded approximately 6 times the bandwidth of a single patch. The second important factor which falls as a disadvantage in the performance of this antenna is the gain. For microstrip patch antennas, gain can be increased by employing thin, low loss and low permittivity substrates [10].

Though gain has an important factor for security applications but to achieve the maximum high ultra - bandwidths and to match the impedance is much important factor. So here, in this thesis we have designed, fabricated and tested the high gain and wide bandwidth of Antipodal Vivaldi microstrip antenna (which is a form of UWB antennas) where the single microstrip antenna can be utilized for all the security applications.

(14)

2 In the design process of this array, we have first simulated and investigated the rectangular microstrip patch fed with corporate feed network. Power dividers play a very pivotal role in the splitting of microwave signals to feed the radiating elements. The array feeder is implemented using the Wilkinson type power divider because it provides a reduction in reflection of electromagnetic waves from each output arm of the divider network. In its basic form, Wilkinson power divider splits the input signal into two output signals with equal phase and amplitude. This class of dividers uses impedance matching transformers and isolation resistors to improve the transmission, reflection and isolation response of the device.

Later, we have mounted the rectangular patches above the power dividers and analyse the results. Once we have achieved the nominal results of rectangular patches, the rectangular structure was then extended to Antipodal Vivaldi antenna structure to get the maximum Ultra wide bandwidth and peak gain.

1.2 Thesis Objective

The objectives for this study are the following:

 Design a UWB array antenna that would provide a wide bandwidth of more than (<5 GHz) with a center frequency of 5.5 GHz

 Design a Wilkinson power divider feed network for UWB array antenna to match the impedance and to achieve the desired main-beam squint of the antenna

 Simulate and optimize the designed antenna array using HFSS Software (High frequency simulator software). Fabricate the antenna using PCB plotter. Use the

“network analyzer” and the “Antenna Test and measurement System” to observe the radiation characteristics and the beam scanning properties of the antenna

1.3 Thesis Overview

Before embarking on the antenna design process, it is essential to have a good theoretical background on this topic.

Chapter 2 of this thesis summarizes the reviewed literature, which helped the authors to gain a comprehensive yet thorough theoretical background on various types of antennas which includes array antennas.

Chapter 3 describes the comprehensive background about various components methods use in designing microstrip antennas. The process of designing array antenna started with array- feeder, used to excite the radiating elements.

Chapter 4 described the overall background of power divider. The first section of this chapter concentrated much on theoretical knowledge of Wilkinson power divider and the next section discuss the methods designing the integrated Wilkinson power divider of 2×1 as well as 4×1 and its simulation results, optimized frequency responses are also presented in this chapter.

Once the fed network is ready, the radiating elements of the array antenna can be selected and arranged to achieve the desired radiation characteristics.

(15)

3 Chapter 5 of this thesis section presents the designing methods of radiating elements for 2×1 and 4×1 rectangular microstrip patch array antenna. Once the results are achieved and its radiation characteristics are analyzed successfully. The rectangular patch structure has been enhanced to Vivaldi array to increase the bandwidth and the antenna beamwidth

Chapter 6 described about the brief knowledge on the networking and security issues at each protocol layers of Ultra-wide band Technology.

Chapter 7 will describe the thorough knowledge on the UWB antennas. The first section gives the theoretical knowledge on the growth of Vivaldi antenna. The second section describes the designing constraints of 2-elements and 4-elements radiating Vivaldi array antenna patches and the final section provides much knowledge on simulated results of Vivaldi array antenna with the help of HFSS

Chapter 8 describes the Measurement results and discussions of 4-element Vivaldi patch array antenna. Comparisons between experimental and simulated results are also tabulated in this chapter.

Chapter 9 described the conclusions drawn from this research and recommendations on

future work to be carried on this subject.

(16)

4

2 Antennas Theory

The history of antennas originates back to 1873 when James Clerk Maxwell presented „A Treatise on Electricity and Magnetism‟ [5]. This work drew from empirical and theoretical work that had already been carried out by scientists such as Gauss, Ampere, Faraday and others. Maxwell took the theories of electricity and magnetism and unified them. The equations he derived are presented below in differential form.

→ ^→ → (2.1)

→ ^→ → (2.2) → (2.3) → (2.4)

Where,

→ . /

→ ./

→ ( ) . /

→ . / . /

→ ( )

Maxwell‟s equations allow the calculation of the radiated fields from a known charge or current distribution. They also give a description of the behavior of the fields around a known current distribution or a known geometry. Maxwell‟s equations can then be used to understand the fundamental principles of antennas.

2.1 Definition of Antenna

The IEEE definition [2] of an antenna or aerial is:

“a means for radiating or receiving radio waves”

Radio waves are also referred to as electromagnetic waves, or light waves, as they travel at the speed of light and can be represented by sine waves. The distance a wave travels to complete one cycle is known as wavelength, λ, of the signal.

(17)

5 ( ) (2.5)

Where c is the speed of the light and f is the frequency (cycles per second).

In a vacuum or air the speed of light is approximately 3x10⁸ m/s. When a radio wave passes through a non-conducting medium other than air this shows the wave down and results in a shorter wavelength. This property is of great importance when designing antennas and is analyzed throughout this thesis.

An antenna can be viewed as a device which sends and receives electromagnetic waves.

Essentially, an antenna acts as an energy converter for a transmission line into free space radiation. Antennas are bidirectional so this relationship works exactly the same from free space to a transmission line. Figure 2.1 shows an antenna as a transition device [1]. The arrows displayed in figure 2.1 correspond to the electric field lines as the wave is transitioned into free space.

Figure 2. 1: Antenna as a Transition device

The transmitting antenna can be modeled as a Thevenin source, as shown in Figure 2.2.

It consists of a voltage generator and series impedance, the transmission line being represented by a characteristic impedance Zc and the antenna is represented by load ZA connected to the transmission line.

(18)

6 Figure 2. 2: Transmission line thevenin equivalent of antenna in transmitting mode

An antenna loss due to conductor dielectric or heat is represented by the loss resistance R_L. The radiation resistance, R_r represents the real part of the radiation impedance of the antenna. The reactance XA represents the imaginary part of the impedance associated with the radiation from the antenna.

2.2 Types of Antennas

There are numerous types of antennas developed for many different applications. They can be classified into four distinct groups

2.2.1

Wire Antennas

Wire antennas are probably the most recognizable, as they are ubiquitous and typified by TV aerials, car aerials etc. Wire antennas can include dipoles, loops, helical, sleeve diploes, Yagi-Uda arrays. Wire antennas generally have low gain and operate at lower frequencies (HF to UHF). They have the advantages of low cost, ease of fabrication and simple design.

2.2.2

Aperture Antennas

Aperture antennas have a physical opening through which propagating electromagnetic waves flow. For example, a horn antenna opening acts as a “funnel” directing the waves into the waveguide. The aperture is usually several wavelengths long in one or more dimensions. The pattern has a narrow main beam which leads to high gain. For a fixed aperture size, the main beam pattern narrows down as frequency increases. These types of antennas are very useful in aerospace and spacecraft applications because they can be easily flush-mounted on the skin of an aircraft or spacecraft application. Examples of these antennas include parabolic reflector, horn antennas, lenses antennas and circular apertures.

2.2.3

Array Antennas

Array antennas are made up of a remix of discrete sources which radiate individually.

The pattern of the array is determined by the relative amplitude and phase of the excitation

(19)

7 fields of each source and the geometric spacing of the sources, typical elements in an array is diploes, monopoles, and slots in waveguides, open-ended antennas and microstrip radiators.

2.2.4

Printed Antennas or Microstrip Patch Antennas

Printed antennas or Microstrip patch antennas can encompass all of the antennas mentioned in section 2.2.1 to 2.2.3, except they take a planar form. Printed antennas are made via photolithographic methods with both the feeding structure and the antenna fabricated on a dielectric substrate

2.3 Shapes of Microstrip Patch Antennas

In general, there are different shapes for microstrip patch antenna is available such as square, Rectangular, Diploe, Circular, Elliptical, Triangular, Disc sector, Circular ring and Ring sector

Figure 2. 3: Various Shapes of Microstrip patch Antenna

We have selected rectangular patch for designing our optimum array antenna with power divider since it is the most widely used configuration for patch antennas and due to low- profile structure it is specifically favoured for utilize in wireless applications.

The rectangular patch antenna also offers linear polarization with narrow beamwidth and directional radiations in nature. This makes them exceedingly reliable for security wireless handheld devices like GPS tracker, electronic pagers, mobile phones etc., also the communicating and remote measuring antennas in defence applications like missiles need to be thin and conformal thus rectangular patch antennas are often employed.

(20)

8

2.4 Antenna Analysis

The most fundamental antenna parameters are 1. Impedance Bandwidth

2. S-Parameters 3. Radiation Pattern 4. Directivity 5. Efficiency 6. Gain

7. Polarisation

2.4.1

Impedance Bandwidth

The term „impedance bandwidth‟ is used to describe the bandwidth over which the antenna has acceptable losses due to mismatch. The impedance bandwidth can be measured by the characterization of both the Voltage Standing Wave Ratio (VSWR) and Return Loss (RL) at the frequency band of interest. Both VSWR and RL are dependent on measuring the reflection coefficient (г). Г is defined as the ratio of the amplitude of the reflected voltage wave ( ) normalised to the amplitude of the incident voltage wave ( ) at a load [1]. Г can also be defined by using other field or circuit quantities and is defined by the following equation.

(2.6)

The VSWR is defined as the ratio between the voltage maximum and voltage minimum of the standing wave created by the mismatch at the load on a transmission line. The VSWR equation is shown in Equation 3.13

(2.7)

The return loss (RL) is the magnitude of the ratio of the reflected wave to that of the incident wave and is defined in dB as

(2.8)

The scattering parameter S₁₁is equivalent to г. It is common for S₁₁ to be defined in dB as:

(2.9)

The maximum acceptable mismatch for an antenna is normally 10% of the incident signal. For the reflection coefficient, this equates to г = 0.3162. For VSWR the impedance bandwidth lies between 1<VSWR<2 and for return loss is value must be greater than 10dB or S₁₁ < -10 dB.

(21)

9

2.4.2

S-Parameters

When designing antennas as part of a network, or on their own, it is advantageous to create a model which allows the designer insight into the performance of the system/antenna. It is common to extract useful data via a Vector Network Analyzer (VNA). The data is normally presented in the form of S-parameters. At low frequencies, simple circuit normally suffices but at high frequencies a distributed model is needed to account for a variety of possible physical effects, such as:

 Skin depth ( √ ⁄ )

 Energy propagation

 Radiation

 Reflections/ Transmission

 Fringing Fields (Coupling)

 Energy splitting/combining

Also S-Parameters enable broadband characterizations unlike impedance and admittance parameters, which need open or short-circuit terminations to characterize and are not conductive to broadband characterization. The S-parameters allow for a complete description of a N-Port Network as seen at its N-ports in Figure 2.4

Figure 2. 4: An N-Port Network

The S-Parameters are defined by measuring the voltage travelling waves between the N- ports. To explain this concept it is best to look at a two port network, as shown by Figure 2.5.

(22)

10 V1 = Voltage at port 1 a1 = Signal incident at port 1 V₂ = Voltage at port 2 b₁ = Signal reflected at port 2 I₁ = Current at port 1 a₂ = Signal incident at port 2 I₂ = Current at port 2 b₂ = Signal reflected at port 2

Z₀ = Characteristics Impedance Z1 = Port 1 Impedance Z₂ = Port 2 Impedance Figure 2. 5: Two Port Network Definition

It is also important to define the incident and reflected signals, which take the form:

√ (√ ) (2.10)

√ (√ ) (2.11) Combining equation (2.6) and (2.7) we get four possible results

 The input reflection coefficient, when port 2 is matched, |

 The reverse transmission gain, when port 1 is matched, |

 The output reflection coefficient, when port 1 is matched, |

 The forward transmission gain, when port 2 is matched, |

Typically, when using S-parameters to characterise antennas the reflection and forward transmission gain are most frequently used. Ideally reflection coefficients should tend towards zero (S₁₁ = S₂₂ = 0) as this means that there is no power being reflected back due to a good match to the characteristic impedance of the feeding structures, usually 50Ω. The forward transmission gain should ideally tend towards one, but this is generally not the case for low gain antennas where the path can be 20dB or greater.

(23)

11 It is normal to express the S-parameters in terms of decibels as follows:

( ) (2.12)

2.4.3

Radiation Pattern

An antenna radiation pattern is defined in the IEEE standard definitions of terms for Antennas [2].

“A mathematical function or a graphical representation of the radiation properties of the antenna as a function of space coordinates. In most cases, the radiation pattern is determined in the far-field region and is represented as a function of the directional coordinates. Radiation properties include power flux density, radiation intensity, field strength, directivity, phase or polarisation.”

Primarily, when measuring the radiation pattern, the property of most interest is the energy radiated relative to the antennas position. This is usually measured using spherical coordinates as shown in figure 2.6.

Figure 2. 6: Coordinate System for antenna analysis [3]

The antenna under test is placed at the origin and is rotated through ϕ = 0° - 360° and θ

= 0° - 180° while the power is measured in the far-field. As shown in figure 2.6 the x-z plane is considered the elevation plane. This is normally aligned with the electric field vector and is called the E – Plane. The x-y plane is normally aligned with the magnetic field vector and is termed the H-plane.

2.4.4

Directivity (D)

Antenna directivity in the IEEE standard definitions of terms for Antennas [4] as:

(24)

12

“The ratio of the radiation intensity in a given direction from the antenna to the radiation intensity averaged over all directions. The average radiation intensity is equal to the total power radiated by the antenna divided by 4π. If the direction is not specified, the direction of the maximum radiation intensity is implied.”

Essentially, this means that the directivity of an antenna is the ratio of the radiation intensity in a given direction over that of a isotropic source. This can be written as:

(2.13) Where, U = radiation density (W/unit solid angle)

U0 = radiation intensity of an isotropic source (W/unit solid angle) P_rad = Total radiated power (W)

If the antenna was to radiate in all directions (isotropic radiator) then its directivity would be unity. As an isotropic radiator cannot be realised practically, the most comparable antenna is a short dipole, which has a directivity of 1.5. Any other antenna will have a higher directivity than 1.5, which means their patterns are more focused in a particular direction.

2.4.5

Antenna Efficiency (ɳ)

Like other microwave components, antennas can suffer from losses. The total antenna efficiency takes into account the losses at the input terminals and within the structure of the antenna itself. The mismatch or reflection efficiency (ɳ_r) is directly related to the return loss (г) and can be defined as:

( ) (2.14)

The radiation efficiency (ɳ) is a measure of how much power is lost in the antenna due to conductor and dielectric loss. These losses reduce the radiation in my given direction and can be expressed as:

(2.15)

2.4.6

Antenna Gain

Antenna gain G, is the product of efficiency and directivity and is defined in the IEEE Standard Definitions of terms for antennas [21] as

“The ratio of the intensity, in a given direction, to the radiation intensity that would be obtained if the power accepted by the antenna were radiated isotropically. The radiation

(25)

13 intensity corresponding to the isotropically radiated power is equal to the power accepted by the antenna divided by 4π”.

This can be expressed as:

^{( )}

(2.16)

Unless specified, it is assumed that the antenna is receiving a signal in the direction of maximum gain. It is also common for the gain to be expressed in decibels and referenced to an isotropic source (G = 1), as shown in equation 3.23

( ) ( ⁄ ) (2.17)

2.4.7

Polarisation

The Polarisation of an antenna refers to the orientation of the electric field vector of the radiated wave and is defined in the IEEE Standard Definitions of terms for antenna [22] as:

“The property of an electromagnetic wave describing the time-varying direction and relative magnitude of the electric-field vector; specifically, the figure traced as a function of time by the extremity of the vector at a fixed location in space, and the sense in which it is traced, as observed along the direction of propagation”.

(26)

14

3 Microstrip Patch Antennas

In its basic form, microstrip antennas are similar to parallel plate capacitors. Both have parallel plates of metal layer and a sandwiched dielectric substrate between them. But in microstrip antenna, one of these metal plates is infinitely extended than the other, to form the ground plane whereas the smaller metal plate is described as radiating patch. Since the size of the patch is often proportional to frequency of the propagating signal, this class of antenna is classified as resonant antennas. This contributes to the basic shortcoming of the microstrip antennas related with its narrow bandwidth, usually only a few per cent [4] of the resonance frequency. So far, several shapes of microstrip patches, such as rectangular, circular, triangular, semicircular, sectoral and annular etc, are successfully used as radiating antenna elements employed in various communication and control devices [4].

Figure 3. 1: Microstrip Rectangular Patch Antenna [4]

Figure 3.1 shows the most basic type of microstrip patch antenna consisting of rectangular radiating elements. When the patch is excited by a feed line, charge is distributed on the underside of the patch and the ground plane. At a particular instant of time the attractive forces between the underside of the patch and the ground plane tend to hold a large amount of charge. And also the repulsive forces push the charges to the edge of the patch, creating a large density of charge at the edges. These are the sources of fringing field. Radiation from the microstrip antenna can occur from the fringing fields between the periphery of the patch and the ground plane [4].

Assuming no variations of the electric field along the width (W) and the thickness (t) of the microstrip structure, the electric field excited by the patch is shown in Figure3.3

(27)

15 Figure 3. 2: Radiation Mechanism associated with Micrsotrip Patch Antenna [4]

Figure 3. 3: Electric field distribution (side view) [4]

Radiation is described mostly to the fringing fields at the open circuited edges of the patch length. The fields at the end can be resolved into normal and tangential components with respect to the ground plane. The normal components are 180⁰ out of phase because the patch line is λ/2 long; therefore the far field radiation produced by them cancels in the broadside direction [5]. The tangential components (those parallel to the ground plane) are in phase, and the resulting fields combine to give maximum radiated field normal to the surface of the structure i.e., broadside direction. Therefore, the patch may be represented by two slots λ/2 apart as shown in figure below, excited in phase and radiating in the half space above the ground λplane.

(28)

16 Figure 3. 4: Increase in length of the microstrip patch [4]

Typically, to excite the fundamental TEM mode, the length L of the rectangular patch remains slightly smaller than λ/2, where λ is the wavelength in the effective dielectric medium [3]. In terms of free space wavelength (λ

0), λ is expressed by [6],

√ (3.1)

Where is the effective dielectric constant of a microstrip line and is given as

0 1 (3.2)

The value of stays between 1 (dielectric constant of air) and the dielectric constant of the substrate, because the electromagnetic fields excited by the micro strip resides partially in the air and partially in the substrate. However, to enhance the electromagnetic (EM) fields in the air, which account for radiation, the width (W) of the patch needs to be increased which makes the radiating EM fields can also be enhanced by decreasing the or by increasing the substrate thickness (h). It is of note that, since „W‟ and „h‟ are constrained by the input- impedance and unwanted-surface-waves respectively, a compromise is required while selecting antenna dimensions. Since microstrip patches are often feed or integrated with micro strip transmission-lines or circuits, the design requirement of these devices are also important. At microwave frequency band, typical comparison of micro strip antenna with that of microwave transmission-line is given in Table 3.1.

(29)

17 Table 3. 1. Comparison between microstrip transmission line and microstrip antenna

Parameter Microstrip transmission line

Microstrip Antenna

H Thin Thick

High Low

W Small Large

Result Maximum Propagation Maximized radiation

For broadside radiation, microstrip antenna array are designed to produce the main beam of the antenna normal to the patch plane. The far field component can be written as

( ), ( ) (3.3) Where, ( ) ⁰

1

. / (3.4)

Although microstrip antennas have come a long way in past two decades, the disadvantages associated with it is still an area of exploration.

3.1 Advantages of Microstrip Patch Antennas

 They are light in weight and low profile

 They can be made conformal to the host surface

 Their ease of mass production using printed circuit technology leads to a low fabrication cost.

 They are easier to integrate with other microstrip Circuits [7].

 They support both linear polarization and circular polarization [7].

 They can be realized in a very compact form, desirable for personal and mobile communication hand held devices.

 They allow for dual and triple band operations [8].

3.2 Disadvantages of Microstrip Patch Antennas

 Narrow bandwidth [4,5,8]

 Lower power gain [8]

 Lower power handling capability [9]

 Polarization impurity

 Surface wave [10]

3.3 Methods of Analysis

The general structure of a microstrip antenna is that of two dimensional radiating patches on a thin dielectric substrate and therefore may be categorized as two-dimensional

(30)

18 planar component for analysis purposes. There are many methods of analysis for microstrip antennas. The most popular methods are based on the transmission line model, cavity model and full-wave analysis. In this work transmission line model is used to calculate the dimensions and input impedance associated with the radiating patches. But the simulating software, HFSS, used in this study uses finite element method to carry out the full wave analysis of the designed structure. Brief descriptions of these methods are as follows

3.3.1

The Transmission Line Model

Although the transmission line model yields less accurate results, it is a very simple model and provides a good physical insight of the basic antenna performance. In this model, the microstrip patch element is viewed as a transmission line resonator with the field only varying along the length (no transverse field variations), and the radiation occurs mainly from the fringing fields at the open circuited ends. The patch is represented by two slots that are spaced by length of resonator. This model was originally developed for rectangular patches but has been extended for generalized patch shapes. Many variations of this model have been used to analyze the microstrip antenna [11]. Since the normal analysis was derived for rectangular patch, some authors [12] have modified it to suite to other patch shapes like rectangular, and others have modified it to suite to triangular patch shapes [13].

(A) Rectangular Patch Antenna (B). Transmission model equivalent

Figure 3. 5: Transmission-line model of rectangular Microstrip patch [4]

Figure 3.5 shows a typical rectangular patch antenna and its equivalent circuit using transmission-line model. Each radiating slots corresponding to the edges of the patch is represented by parallel equivalent admittance, Y, which is given by

(3.5)

Where G and B represents the conductance and the susceptance produced by the slot or the radiating edge of the microstrip patch. These conductance and susceptance are typically expressed as [4],

0 ( ) 1 (3.6)

, ( )-

(3.7)

(31)

19 Where W is the width of the microstrip patch and h is the height of the substrate, is the free-space wavelength of the propagating electromagnetic wave. Thus the resonant input impedance can be given as,

( ) ₍ ₎ (3.8)

Where G₁₂ is the mutual conductance between the slots which are representative of patch edges and is given by,

∫ [^.^/] ( ) (3.9) Where, J₀ is the Bessel function of the first kind and order zero.

Also the resonant frequency of the microstrip patch for the dominant mode TM010 is given as,

( )

√ √

( )√ √ (3.10)

Where , is the increase in patch length because of fringing. Although the transmission line model is easy to use, many patch configurations cannot be analyzed using this model due to its inability to consider the field variation orthogonal to the direction of propagation. In this study, this model is used to calculate the patch dimensions and input impedance.

3.3.2

The Cavity Model

Cavity model is more complex than transmission line model and provides more accurate results. In this model the region between the patch and the ground plane is treated as a cavity, which is surrounded by magnetic walls around the periphery and by electric walls from the top and bottom sides. Since thin substrates are used, the field inside the cavity is assumed to be uniform along the thickness of the substrate [4]. In cavity model, the analysis is made simple, by expressing the electromagnetic fields within the patch substrate, as a summation of the various resonant modes of the two dimensional radiator (i.e., the patch in this case). Since the normal substrates that are used to produce the microstrip patch antennas are thin, the usual assumption is that the field inside the cavity is uniform along the thickness of the substrate [14].

3.3.3

Full Wave Model

Full wave models are very versatile and can provide very accurate results. Method of Moments [15], the Finite Difference Time Domain method [19], the Finite element method (FEM) [16] all belong to this category, they are suitable for volumetric configurations. The Finite Element Method (FEM) is more popular amongst these methods and in this method the region of interest is divided into any number of finite surfaces or volume elements depending upon the planar or volumetric structures to be analysed. These discretized units, generally referred to as finite elements, are well defined geometrical shapes, such as triangular elements for planar configurations and tetrahedral and prismatic elements for three dimensional configurations [14].

The software used in this study, called (Ansoft HFSS v13) is a FEM based

(32)

20

3.4 Excitation Techniques of Microstrip Antennas

The microstrip antennas can be excited or fed directly either by coaxial probe or by a microstrip line. It can also be excited indirectly using electromagnetic coupling (proximity) or by aperture coupling method, in which there are no direct metallic contact between the feed line and the patch. Since feeding technique influences the input impedance, it is often exploited for matching purposes. Also as the antenna efficiency depends on the transfer of power to the radiating element, feeding technique plays a vital role in the design process. The four most popular feeding techniques are discussed below.

3.4.1

Full Wave Model

Microstrip line feed patch antennas are easy to fabricate as both the feed line and the radiating elements are printed on the same substrate. The impedance matching associated with this class of antennas are also simpler compared to other methods. Although these antennas have low spurious radiation, often the radiation from the feed line increases the cross polarization level. Also, the thick substrate associated with these antennas (for bandwidth improvement) introduces surface waves that deteriorate the antenna performance. In millimeter- wave range, the size of the feed line is comparable to the patch size, leading to increased undesired radiation. Typically, the patches are feed from its edge and the edge impedance should be matched with the impedance of the feed line for maximum power transfer. Since the input impedance of the patch gradually decreases from maximum at the edge (150 to 300 Ω) to minimum at the centre, insets are often used to connect the feed line to a relatively low impedance spot of the patch. This popular technique is shown in Figure3.6. The input impedance related to the length of the inset is given by,

( ) ₍

) . / (3.11)

Where y₀ is the inset length from slot at the feeding edge of patch, is the length of the patch, G1 and G12 are self and mutual conductance‟s.

Figure 3. 6: Microstrip Line Feed

(33)

21 Other methods of impedance matching between the feed line and radiating patch is achieved by introducing a single or multi-section quarter-wave-transformers. In this study, optimized insets are introduced to match the rectangular radiating elements with microstrip feed lines.

3.4.2

Coaxial Feed

The coaxial or probe feed arrangement is one in which the centre conductor of the coaxial connector is soldered to the patch. The main advantage of this feed is that it can be placed at any desired location inside the patch to match its input impedance; other advantages are that it can be easily fabricated and has low spurious radiation [1]. The main disadvantage of a coaxial feed antenna is the requirement of drilling a hole in the substrate to reach the bottom part of the patch. Other disadvantages include narrow bandwidth, difficult to theoretically model, and its asymmetrical non planar configuration.

Figure 3. 7: Coaxial Probe Feed

3.4.3

Aperture Coupling

This is a method of indirectly exciting the patch, where the electromagnetic fields are coupled from the microstrip feed line to the radiating patch through an electrically small aperture or slot, cut in the ground plane. Figure 3.8 illustrate such an aperture coupled microstrip rectangular antenna. The coupling aperture is usually centered under the patch, leading to lower cross polarization due to symmetry of the configuration. The shape, size, and location of the aperture decide the amount of coupling from the feed line to the patch [3], this can lead to improved bandwidth as shown by some authors [17 and 18]. The slot aperture can be either resonant or non-resonant [19]. The resonant slot provides another resonance in addition to the patch resonance thereby increasing the bandwidth, but at the expense of back radiation.

Therefore, the slot should not resonate within the operating frequency band of the antenna. The advantage of this feeding technique is that the radiator is shielded from the feed structure by the ground plane. Another important advantage is the freedom of selecting two different substrates;

one for the feed line and another for the radiating patch. Since both substrates can be optimized simultaneously, the need for a compromise between radiation and propagation requirements can be avoided. This flexibility in choosing the appropriate substrates also minimizes unwanted

(34)

22 surface waves, spurious coupling between antenna elements and thereby increasing the efficiency as well as the bandwidth of the antenna. However, the fabricating process of this class of antenna is difficult and can easily deteriorate the performance due to small errors in the alignment of the different layers. In this study an aperture coupled stacked patch antenna is also designed for beam scanning purposes.

Figure 3. 8: Aperture Feed

3.4.4

Proximity Coupling

This is another coupling method which does not involve the direct contact of the feed line. But unlike aperture coupling, this method uses electromagnetic coupling between the feed line and the radiating patches, printed on separate substrates [1]. Figure3.9 shows proximity coupled rectangular patch antenna. Typically, the bottom feeder substrate is thin and made of high dielectric constant, whereas the top patch substrate is thick and made of low dielectric constant. The advantage of this coupling is that it yields the largest bandwidth compared to other coupling methods; it is somewhat easy to model and has low spurious radiation. Some authors have shown the improvement in bandwidth [20]. The disadvantage is that it is more difficult to fabricate

Figure 3. 9: Proximity Coupled Feed

(35)

23

3.5 Antenna Array

Typically the array arrangements of microstrip antennas are used to improve the efficiency, directivity and gain for the radiating system. That is because the radiation from the single antenna element is often very wide in pattern with large, beam angles. This is not good for point to point communications, which requires antennas that are more directives in nature.

Also a single radiating element often generates radiation patterns with unacceptable bandwidth, efficiency and gain parameters. All these and more make the utilization of a single element antenna not recommendable. The implementation of antennas in array configuration overcomes these drawbacks. In this section concept of arrays is briefly described

Antenna arrays are basically a collection of radiating elements, geometrically arranged in a specific manner, to generate the required radiation pattern. Each antenna in the array is called as an element and it can be anything from simple dipole antenna, monopole antenna, horn antennas, or as in this case microstrip patches. These antenna arrays are classified into linear arrays, planar arrays or 3- dimensional arrays depending on the positioning of the antenna elements.

The two basic types of antenna array are: uniform and non-uniform. Uniform arrays are the simplest one-dimensional array antenna, where the signal inputted in each identical element consists of identical amplitude and equal differential phase distribution. This class of array has the narrowest main lobe and considerable amount of side lobes. On the other hand, non-uniform array antenna with unequal amplitude distribution yields a more controlled side-lobe level. The phased array is special type of antenna array, where the spatial distributions of the radiated fields are electronically scanned to enhance the desired signal, by introducing differential phase (and/or magnitude) in the input signal of the radiating elements. Phased-array antennas have been developed mainly for radar applications but are being used more now for space-based communications applications because of their advantages in scanning, re-configurability, weight and power [2]. Also because of the development in the integration technology of small microwave circuits, has led the deployment of these antennas in ground, ship, air and space communication [2]. Also they find application in other areas apart from communication; like the non-destructive testing (NDT) etc.

In phased array antenna, several types of phase shifters are used to control the phases of the antenna elements to achieve beam steering. A detail description of a linear microstrip phased array antenna designed in this study [chapter 5] with the help of Wilkinson power divider which is discussed in the following section.

3.5.1

Array Factor

The total field of an antenna array can be calculated by pattern multiplication [2]

( ) , ( )- , ) ( )

(36)

24 This is equal to the field of a single element multiplied by the factor, called array factor [2], as shown in figure 3.10. The array factor is a function of array geometry and the excitation phase

Figure 3. 10: Pattern multiplication shown graphically [2]

By varying the separation between the array elements and introducing the progressive phase shift (β), the radiation pattern of the antenna array can be controlled. This type of control is normally referred to as phase scan or beam steering of the antenna array. The arrays are of uniform type, if the elements are fed with microwave signal with same magnitude (current amplitudes), and same progressive phase difference (β). But they are called non-uniform, if the elements are fed with signals of different magnitude and same or different β [2]. The array factor for any N-element array is given by the following equation [2].

∑ ^{( )} (3.13) Where, (3.14)

⁄

(37)

25

3.5.2

Mutual Coupling

The inspection method, though intuitive cannot be used for more complex arrays with more number of elements and complex phase differences and distance of separation. Moreover, all these derivations have been based on the assumption that all the elements are isolated from each other and from external sources. However, in practice, this is rarely the case. Each element interacts with all the other elements in the array creating what is known as mutual coupling which causes changes in the current magnitude, phase and distribution on each element. The end result is that the total array pattern is different from the no-coupling case.

Mutual coupling [3] not only depends on the proximity of the elements, but also on the frequency and scan direction. Also, it was found that mutual coupling decreases as the spacing between elements increases and that the coupling strength is predicted by the far-field pattern of each of the elements. The input impedance of the „m‟th element in the presence of all other elements and with mutual coupling included is expressed as

(3.15)

This is referred to as the driving-point impedance. It is evident that input impedance of each element depends on the mutual impedances with the other elements in the array and the terminal currents.

The design of a linear phased microstrip rectangular patch as well as Vivaldi array patch is carried out here using simulating software HFSS, mutual coupling are optimized using the software.

In phased array antenna, several types of array feeders and phase shifters can be used to electronically steer the beam of the array antenna. In order to select the suitable power divider and phase shifter for our optimum designed array antenna, basic knowledge about these devices are essential. In the following section, the Wilkinson power divider is designed and simulated successfully before mounting the radiating the patch elements.

(38)

26

4 Power Divider

The design process of the antenna array requires the dimensions of associated array feeder. Array feeder is a power divider network that provides required transmission, reflection and isolation properties. In this study, the four-way power divider network is designed using microstrip transmission line that supports quasi-TEM mode of propagation. In the literature, various types of microstrip power dividers have been employed in the feeding of the microstrip antennas. The most popular ones being the quadrature hybrid, annular of ring, Wilkinson and Y- Junction power divider. By definition, a -3dB power divider is ideally a passive lossless reciprocal three port device that divides power equally in magnitude and phase. Figure 4.1 sows a three port lossless and reciprocal power divider with all matched ports.

Figure 4. 1: Block Diagram Power Divider

The general expression of the S-parameter of the S-parameter matrix related this device is:

, - (

) (4.1)

Since all the three ports of this power divider are matched, S_ii = 0, and the network is nonreciprocal, S_ij = S_ji, the modified S-matrix can be written as,

, - (

) (4.2)

Also as the network is lossless then this matrix should be unitary and yields,

(39)

27 Considering the above equations we see that the second column must have at least two out of three S parameters to be equal to zero. But if two of them were zero then one of the equations in the first column will be violated. Thus, we can conclude that it is impossible to satisfy all the three criteria of any power divider, such as, lossless, matched and reciprocal.

Since relaxing anyone of these constraints makes the other two achievable.

In this study, we will start with the design process of a two way array feeder network using Wilkinson type of power divider. Thus, a brief introduction of this type of power divider is presented here. The Wilkinson power divider invented around 1960 splits the signal in two equal amplitude equal phase output signals [23]. Wilkinson relied on quarter wave transformers to match the split ports to common port. Since a lossless reciprocal three port network cannot have all ports simultaneously matched, Wilkinson introduced a resistor for the purpose of matching. Apart from matching, the resistor also improved the isolation between the output ports. Figure 4.2 shows the equivalent circuit and the microstrip implementation of a wilkinson two-way

Figure 4. 2: Block Diagram Power Divider

Many authors like Lee et al [24] proposed formulae for implementing a multi section power divider by modifying the wilkinson‟s design to increase the device bandwidth. Others [25] proposed a 9-way power divider consisting of 3-way power dividers in 2 stages to, further improve the bandwidth. Antsos et al [25] implemented a 14- way power divider in the K-band, based on the novel technique of improving the Wilkinson‟s design.

Apart from the number of ports, keeping the ports constant and dividing the power unequally also has been done by some authors. Like Antsos et al [25] have designed an unequal power divider based on different impedances employed for different arms of the power divider.

(40)

28

4.1 Wilkinson Power Divider

Wilkinson power divider can be used for both equal and unequal power division. The main advantages of Wilkinson power divider is its ability to isolate the output ports and minimize the mutual affect due to mismatched terminations. But this power divider also has some limitations. The design process of the microstrip array feeder, using Wilkinson power divider, is presented in this chapter. The limitations associated with this class of divider network and methods proposed to overcome them are also discussed.

The theory of a Wilkinson power divider may be studied in [26]. A transmission line of Z_i is split into two lines of √ and vice versa after a section of λg/4 from √ back to Z_i. The resistance of the value 2Zi should be connected in the course of the second transition.

The typical construction image of Wilkinson power divider is shown in figure 4.3.

Design constraints and difficulties accompany the conventional realization of the power divider with microstrip. First the angle ψ , see figure 4.3, of the outgoing branches is limited since the impedance conversion has to be achieved without severe losses. Second, this latter mentioned transition from one width to another generally exhibits higher losses than a simple mitred curve with only one width. Finally, in theory, the resistance needs to be between both ports of different widths, which yields an ambiguous situation for the design procedure and often leads to more optimization runs from case to case.

Figure 4. 3: Typical Microstrip Wilksinon Power Divider

(41)

29

4.2 Design and Construction of 2×1 Wilkinson power divider feed network for antenna array

The modified structure of Wilkinson power divider (see Fig 4.4) was designed on Astra^®MT77 substrate material [27] from ISOLA with a relative dielectric constant ɛ_r of 3.0 and a loss tangent of 0.0017 respectively. The characteristics impedances related the micro-strip lines required for this design process are then calculated using the following formulas [28].

0

1 (4.3) ^{√ (}⁾

⁽⁾

( ) [

] (4.4) And

( ), ( ) - (4.5)

Note that instead of the substrate dielectric constant (ɛ_r) the effective value of the dielectric constant (ɛreff) is used. The calculated values of the widths, tabulated below are used for the design of modified Wilkinson power divider. Also, In order to match the impedance of 50Ω, the power dividers are simulated and analyzed with SMA connectors by using the accurate dimensions of real SMA model available on the market.

Figure 4. 4: Modified 2×1 structure of wilkinson power divider

(42)

30 Table 4. 1: Design Dimensions of 2×1 Modified Wilkinson Power Divider

Parameters Description Dimension

Lg Length of the ground 24.164 mm

Wg Width of the ground 28.414 mm

P1w Width of Port 1 3.5 mm

P1l Length of Port 1 4.8 mm

P2w Width of Port 2 1 mm

P2l Length of Port 2 7 mm

P3w Width of Port 3 1 mm

Dis Distance between Ports 21.565 mm

Rin Inner Radius 3 mm

Rout Outer Radius 6 mm

Wres Resistor Width 5 mm

Lres Resistor Length 2.5 mm

Dis_1 Separation Distance 12.443 mm

Fw Feed Width 2 mm

Fl Feed length 3.5 mm

The desired simulated power divider was measured between 1 to 10 GHz to be sure to cover the entire band of the operation for the device. The results for the return loss at each port are shown in figure 4.5

Figure 4. 5: 2×1 wilkinson divider Return Losses

(43)

31 Table 4.2 shows the maximum return loss achieved at each port, the frequency at which the maximum return loss occurred, and the band over which it performed within the 10 dB minimum

Table 4. 2. Simulated (2×1) Wilkinson Bandwidth

S-Parameter

Centre Frequency (GHz)

Max Return Loss (dB)

Band>10

dB Bandwidth

Fractional Bandwidth (%)

S11 5.5 25.21 4.97 GHz -

6.16 GHz 1.19 GHz 21.636

S11 7.86 20.33 7.24 GHz -

8.39 GHz 1.15 GHz 14.631

S22 6.54 26.69 5.35 GHz -

7 GHz 1.65 GHz 25.229

S33 6.54 27 5.43 GHz -

7 GHz 1.57 GHz 24

( )

Since the Wilkinson power divider is a reciprocal device, S12 should be equal to S21, S31 should equal to S31 and S23 should equal S32. Figure 4.6 shows the values for S12, S21, S13 and S31 between 1 to 10 GHz.

(44)

32 Figure 4. 6: Simulated 2×1 Wilkinson power division

The results in figure 4.6 show similar responses between S12 and S21 and likewise S13 and S31, indicating that the device is reciprocal. Figure 4.6 also indicates that the power divider does not quite evenly divide the power. The bandwidth over which these fluctuations occur is approximately 1 to 10 GHz. By taking the difference between S31 and S21, the amplitude balance between the two outputs can be determined. Figure 4.7 shows the amplitude balance of the optimum Wilkinson divider.

(45)

33 Figure 4. 7: Simulated 2X1 Wilkinson Amplitude Balance

These results are a direct result of the inaccuracies in the hand cut traces of the Wilkinson. The quarter-wavelength section leading to port two was cut narrower than intended, leading to slightly higher impedance than the section leading to port three. Consequently, the difference in impedance between the two quarter-wavelength sections caused not only an unequal power split, but also contributed to a small amount of loss in resistor.

The originally deemed acceptable value of amplitude balance for Wilkinson power divider must be below 0.4. Based on figure 4.7, an amplitude balance below 0.4 is reliably achieved from approximately 1 to 10 GHz.