Trang 1 NGUYEN CONG THUANMINISTRY OF EDUCATION AND TRAINING HANOI UNIVERSITY OF SCIENCE AND TECHNOLOGY --- Nguyen Cong Thuan COMPUTER SCIENCESTUDY AND DESIGN OF 8-PORT RECONFIGURABLE PHA
INTRODUCTION
Application and Technical Area
Positioning technology is essential in modern life, serving as the backbone for various applications, including navigation, tracking, and location-based services While outdoor positioning using satellite systems like GPS, GLONASS, and Galileo is widely adopted, indoor positioning systems have garnered significant research interest over the past decade Indoor localization offers numerous innovative services, such as guiding visitors in museums, preventing theft of valuable devices, locating products in supermarkets, navigating through malls, and optimizing power consumption for devices.
Indoor positioning systems are essential due to the limitations of outdoor positioning methods, which are hindered by the scattering and attenuation of microwaves from roofs and walls Various technologies have been developed for indoor positioning, including infrared (IR), Bluetooth, radio-frequency identification (RFID), wireless local area networks (WLAN), ultra-wideband, ultrasound, magnetic positioning, and audible sounds Notably, WLAN-based solutions have gained significant attention due to their extensive range and the widespread availability of compatible devices.
The proliferation of WLAN devices, now numbering in the billions, is driving significant advancements in WLAN-based indoor positioning technology This growing trend suggests that the application and adoption of such positioning systems are poised for widespread implementation in the near future.
Problem Statement and Technical Issue
Indoor localization techniques using WLAN are categorized into RSS scene analysis, ToA, TDoA, RToF, and AoA The primary challenge of the RSS scene analysis technique is its high instability in indoor environments In contrast, ToA, TDoA, and RToF rely on precise clock synchronization among devices Additionally, the AoA technique necessitates a directional antenna design to accurately estimate the relative angle of an object in relation to reference points.
Recent advancements in antenna designs for Angle of Arrival (AoA)-based indoor localization have been notable G Giorgetti and A Cidronali introduced a switched-beam directional antenna featuring six circular antennas to cover distinct areas within a room Similarly, M Rzymowski et al developed a design utilizing a twelve-element electrically steerable parasitic array radiator antenna, which includes a central active monopole This configuration allows for the adjustment of parasitic elements via switches, enabling twelve different directional lobes Kamarudin et al proposed reconfigurable antennas using PIN diodes to modify lumped components, facilitating the transition between four beams Additionally, Bui et al created a switch-beam array antenna based on a 4x4 Butler matrix to generate beams at four specific angles However, these designs primarily rely on switching among a limited number of predefined beams, which restricts resolution in object localization and can lead to significant positioning errors.
Research Aim and Objective
For effective indoor localization, achieving high accuracy in object positioning is crucial, necessitating antenna designs that offer precise beam steering capabilities Current systems predominantly utilize switched-beam antenna structures, where the resolution of the angle of arrival is largely influenced by the number of antenna elements Consequently, enhancing beam scanning resolution requires an increase in both the number of antennas and switching components, such as PIN diodes and FETs, leading to greater complexity and cumbersome system designs.
This thesis focuses on enhancing the accuracy of Angle of Arrival (AoA)-based indoor localization by developing a phased array antenna operating in the Wi-Fi band (2.4GHz to 2.484GHz) The goal is to achieve precise beam steering without the need to increase the number of antennas, with key objectives centered around optimizing performance and efficiency in indoor positioning systems.
+ To determine the structure of phased array antenna for AoA technique
+ To develop the 360 o continuous reflection type phase shifter that is controllable
I have developed a phased array antenna that utilizes an advanced structure and phase shifter, enabling high-resolution steering of the main beam without the need for additional antenna or switching elements.
Thesis Outline
This thesis work has been organized as follows:
Chapter 2 provides an overview of fundamental concepts related to phased array antennas, including Array Geometry, Array Factor, Grating Lobes, and Mutual Coupling It also explores the principle of beam steering and various methods for feeding antennas and controlling phase differences Furthermore, key antenna parameters pertinent to the thesis are outlined to support subsequent sections.
Chapter 3 discusses the structure of phased array antennas used for indoor localization After that, power dividers for feed networks are designed, follow by ed principle and design of the proposed phase shifter Next, antenna elements are chosen in order to create the total beam suitable for indoor localization issue Finally, the structure of my positioning system is mentioned
Here is a rewritten paragraph that complies with SEO rules:"In the fourth chapter, we delve into the simulation, fabrication, and measurement processes of our design Utilizing ADS software, we simulated power dividers and phase shifters, while CST Microwave Studio enabled the design of antennas and phased array antenna systems For fabrication, we employed FR4 and Roger4003c substrates Subsequently, the characteristics of power dividers, phase shifters, and antennas were measured using a Network Analyzer, and the radiation patterns of the antennas were examined in an anechoic chamber."
Chapter 5 complete the thesis by providing conclusions and perspectives for s future works
LITERATURE REVIEW
Basics of Microwave Engineering
In microwave engineering, key concepts include the Smith chart, waveguides, wave modes, and impedance matching; however, this thesis focuses on three main issues: transmission line impedance, microstrip discontinuity, and the scattering matrix Understanding transmission line impedance is crucial for calculating the equivalent impedance of circuits and utilizing specific lengths to create short or open points in microwave circuits Microstrip discontinuities present significant challenges, particularly at higher frequencies, with Section 2.1.2 detailing typical types of discontinuities and techniques to address them Lastly, the scattering matrix is essential for characterizing microwave circuits effectively.
In a lossless transmission line of length \( l \) terminated with a load \( Z_L \), an incident wave described by \( V_o + e^{-j\beta z} \) is produced from a source located at \( z < 0 \) The relationship between voltage and current for this traveling wave is characterized by the impedance \( Z_0 \), known as the characteristic impedance of the line.
5 impedance of the line However, when the line is terminated in an arbitrary load
Z L , the ratio of voltage to current at the load must be Z L
Figure 2-1 A transmission line terminated in a load impedance [6] :
The input impedance is simple the line impedance seen at the beginning (z = - l) of transmission line
To determine Z in , we first must determine the voltage and current at the beginning of the transmission line (z = -l).
While L is the reflection coefficient of a load, determined by its impedance Z L , and the impedance toward the source o L o
Combining two expressions (2-4), (2-5), we get:
We have the Euler’s equations: l j l e l j l e l j l j
Using Euler’s relationship, we can likewise write the input impedance without complex exponentials: l jZ Z l jZ
tan tan sin cos sin cos
Next, we consider some special case of lossless terminated transmission line
If the length of transmission line is exactly one-half wavelength, l/ 2, we find that:
In other words, if the transmission line is precisely one-half wavelength long, the input impedance is equal to the load impedance, regardless of Z 0 or
If the length of the transmission line is exactly one-quarter wavelength,
When a transmission line is exactly one-quarter wavelength long, the input impedance is inversely related to the load impedance In the case of a short circuit where the load impedance (Z_L) is zero, the input impedance at the start of the λ/4 transmission line can be determined accordingly.
That means the quater-wave transmission line enables us to transform a short- circuit into an open-circuit and vice versa
If the transmission line is electrically small, its length lis small with respect to signal wavelength , we find that:
In other words, if the transmission line length is much smaller than a wavelength, the input impedance Z in will always be equal to the load impedance Z L
Surface waves are electromagnetic waves that propagate along the dielectric interface of microstrip circuits, generated at any discontinuity These waves can couple with other microstrips, leading to decreased isolation between networks and resulting in crosstalk, coupling, and attenuation Discontinuities, caused by abrupt changes in the geometry of the strip conductor, alter the electromagnetic field distributions, affecting capacitance and inductance Common discontinuities in practical microstrip layouts include bends, open-ends, gaps, steps, and T-junctions.
Bends are the most frequently encountered discontinuities The 90° bend, can be modeled by T-network circuit for a short line length, as shown in Figure 2-2
Figure 2-2: Bend: (a geometry; (b) equivalent circuit[7] )
The values of the components are as follows [7]
The accuracy on the capacitance is quoted as within 5% over the ranges of 15
2 r and0.1W/h5, and on the inductance is about 3% for0.5W /h2
To compensate the excess capacitance, we can use increased inductance or decreased capacitance techniques by cutting the corner (Figure 2-3 ).
Open-ends in microstrip lines occur when the termination is not closed, leading to fringing fields that extend beyond the physical boundaries of the line This phenomenon causes the fields to behave like a shunt capacitor or a short transmission line, influencing the overall performance of the microstrip.
A closed-form expression is below with the accuracy better than 2% for the range of 0.01W /h100 and r 128
The simplest way to compensate for the increase in line length is to reduce the length of the designed line by the correct amount
Gaps can be equivalently represented as a π capacitor circuit (Figure 2-5), while the shunt (C p ) and series (C g ) capacitors can be defined by: e o g e p
The accuracy of these expression is within 7% for 0.5W /h2and 15
In narrow gaps, the capacitance value \(C_p\) approaches zero while the gap capacitance \(C_g\) increases, with practical series capacitors ranging from 0.01pF to 0.5pF Conversely, in larger gaps, the capacitance \(C_g\) is negligible, leading to the gaps behaving like an open-end circuit.
Step Widthin is also affected by fringing effect in microstrip lines and has equivalent circuit as shown in Figure 2-6, while capacitor and inductors can be calculated as:
. 0 ) ( re re c rei c ci wi w w w w w w re re c re
The inductances per unit length of the microstrips, denoted as L1 and L2, correspond to their respective widths W1 and W2 Additionally, Zci represents the characteristic impedance, and εrei indicates the effective dielectric constant for each width Wi The speed of light in free space is denoted by c, and h refers to the substrate thickness measured in micrometers.
The discontinuity capacitance C affects the effective length of distributed elements, resulting in an increase for the wide line W1 and a corresponding decrease for the narrow line W2 To counteract this excess capacitance, we can adjust the length of the W1 line to achieve the desired electrical length.
T-Junction is also a component, usually encountered in microwave circuits as impedance matching, power divider and hybrid coupler T-junction can be compensated by cutting a part at the intersection area as shown in Figure 2-7
Figure 2-7: T-junction discontinuity compensation configuration[8]
In conventional circuit theory, voltage and current are the primary quantities, with relationships expressed through impedances (Z), admittances (Y), and hybrid or ABCD parameters A two-terminal circuit element can be defined by its complex impedance, indicating the voltage-current ratio at its terminals More complex multi-terminal networks can be represented by equivalent circuits, such as tee, pi, or ladder networks However, at microwave frequencies, traditional Z, H, Y, and ABCD parameters become unmeasurable due to the comparable lengths of cables and signal wavelengths, leading to variations in current magnitude and phase across the cable This complicates the unique specification of potential differences between terminals, and achieving short or open circuits across a broad frequency range is challenging, especially with active devices like transistors Consequently, simpler and refined analysis techniques, such as scattering parameters introduced by Campbell and Foster in 1922, are sought for effective microwave circuit handling.
Scattering parameters are defined based on the incident and reflected waves at network ports It is logical to consider the incident waves as independent variables, represented as V m , and to express their influence on the reflected wave V n at port n through a series of scattering coefficients.
In two port networks, other parameters such as Z, Y, H, ABCD can be transformed from scattering parameters These equations is expressed by D.M.Pozar in [6]
Fundamental Parameters of Antennas
In antenna engineering, several key parameters influence the design of phased array antennas for indoor localization systems This thesis focuses on three critical parameters: Return Loss and Standing Wave Ratio, which are used to assess the operational frequency of an antenna, and the Radiation Pattern, which graphically represents the antenna's radiation properties in relation to spatial coordinates Additionally, the impact of the relative angle between two antennas on transmission loss is discussed, particularly in the context of polarization.
2.2.1 Return Loss and Voltage Standing Wave Ratio
In the transmission process, a source generates a wave that travels along a transmission line to an antenna Ideally, the antenna should radiate all the energy into the surrounding space; however, impedance mismatch causes some of the energy to be reflected, which decreases transmission efficiency This reflected energy is wasted, making it essential to minimize reflection at the antenna port during antenna design.
V + and V - , respectively The reflection coefficient is given by
In microwave theory, the scattering matrix is essential for describing a network, particularly in the context of antennas, which are considered single-port devices The reflection coefficient, denoted as S11, is equivalent to the S11 parameter, representing how much power is reflected back Engineers typically express S11 in decibels (dB) for practical applications.
The absolute value of S 11 (dB) is called the Return Loss: RL = |S 11 (dB)|
The Voltage Standing Wave Ratio (VSWR) is a crucial parameter used to indirectly assess the amplitude of reflected waves in a transmission line It represents the ratio of the maximum voltage to the minimum voltage along the line at a consistent frequency.
13 the maximum voltage is the sum of amplitudes of incident and reflected voltage, and the minimum voltage is the subtraction of amplitudes of incident and reflected amplitude
In quantitative evaluations, antennas with a maximum reflected power of 10% are deemed acceptable, which translates to a return loss of 10 dB or a Voltage Standing Wave Ratio (VSWR) of approximately 2 Consequently, the designed antenna must achieve a return loss greater than 10 dB or a VSWR lower than 2 at the designated operating frequency.
Before discussing radiation patterns, it's essential to understand the three regions surrounding an antenna: the reactive near-field, the radiating near-field, and the far-field regions, as illustrated in Figure 2-8.
The reactive near-field region surrounds an antenna and is characterized by predominantly reactive fields, making the relationship between the electric (E) and magnetic (H) fields complex and unpredictable This region typically extends to an outer boundary defined by the formula 0.62 D³/λ, where D represents the largest dimension of the antenna and λ denotes the wavelength.
The radiating near-field region lies between the reactive near-field and far-field regions of an antenna This area may be negligible if the antenna's maximum dimension is small relative to the wavelength Typically, the radiating near-field is defined by the range 0.62 D³/λ < R < 2D²/λ, where λ represents the wavelength and D is the largest dimension of the antenna, which must be significantly larger than the wavelength for this region to exist.
The far-field region is crucial as it defines the radiation pattern of an antenna, making it essential for effective wireless communication over long distances This operational area is where most antennas function, highlighting its significance in wireless technology.
The radiation pattern of an antenna remains consistent regardless of distance and is primarily influenced by the radiated fields, with the electric (E) and magnetic (H) fields being orthogonal to each other and the direction of propagation, similar to plane waves The far-field region is typically defined as the area beyond a distance of 2D²/λ from the antenna, where λ represents the wavelength and D denotes the largest dimension of the antenna, which should be significantly larger than the wavelength.
Figure 2-8: Fields regions of an antenna
The antenna radiation pattern is a mathematical function or graphical representation that illustrates the radiation properties of an antenna in relation to space coordinates, specifically observed in the antenna’s far field This pattern reflects the 2D or 3D spatial distribution of radiated energy based on the observer's position along a constant radius path Typically, radiation patterns are normalized to their maximum value, resulting in normalized radiation patterns that can be displayed on either linear or logarithmic scales However, logarithmic scale representation in decibels (dB) is often preferred, as it provides a clearer view of low-value sections of the pattern, including side lobes, minor lobes, and back lobes For instance, the normalized radiation patterns of a 10-element linear array antenna can be visualized on both linear and logarithmic scales.
Figure 2-9: Radiation pattern of array antenna: (a) in linear scale; (b) in dB
In practical applications, 3D patterns are often captured through a series of 2D representations Typically, the essential information can be conveyed effectively using just a few 2D plots, making it unnecessary to scan entire 2D surfaces to generate a complete 3D pattern.
Polarization for an antenna in a given direction is defined as the polarization of the E-field transmitted by the antenna Polarization may be classified as linear, circular and elliptical
Linearly polarized antennas have an E-field directed along a line, which can be classified as horizontally polarized when parallel to the ground and vertically polarized when perpendicular to it In contrast, circular polarization features an E-field that rotates around the origin over time, with equal strength in all directions If the rotation is counter-clockwise, it is termed Right Hand Circularly Polarized (RHCP), while clockwise rotation is referred to as Left Hand Circularly Polarized (LHCP) Elliptical polarization resembles circular polarization but exhibits varying strengths in different directions, resulting in an elliptical locus of strength.
Polarization significantly affects communication between antennas For instance, a horizontally polarized antenna cannot communicate with a vertically polarized antenna However, if the horizontally polarized antenna is rotated to become vertically polarized, communication is possible Additionally, when two linearly polarized antennas are rotated at an angle relative to each other, the resulting power loss due to polarization mismatch is quantified by the Polarization Loss Factor.
The power loss factor (PLF) is defined as cos 2 θ, indicating that two antennas with circular polarization experience no power loss from polarization mismatch In contrast, a combination of a linearly polarized antenna and a circularly polarized antenna results in a consistent polarization mismatch of 0.5, equating to -3dB, regardless of their relative orientation.
Figure 2-10: PLF according to different transmitter/receiver polarizations
Phased Array Antenna
A phased array antenna is a sophisticated system comprising multiple antenna elements that are fed coherently, allowing for variable phase and amplitude control to shape the radiation pattern This technology, first introduced in military applications during the 1940s, significantly enhanced the reception and transmission capabilities of antennas Moreover, it enables electronic steering of the antenna system, facilitating targeted reception and transmission.
Phased array antennas transmit information directionally without mechanical movement, offering capabilities unattainable by single antenna types For long-distance communication, antennas with high directivity and gain are essential, as standard single elements typically exhibit low directivity and wide radiation patterns Consequently, high-directivity antenna arrays are employed to fulfill these requirements Since their inception, phased arrays have found diverse applications, initially in military settings for decades, and more recently in civilian radar and communication systems, leading to growing interest in their commercial use.
Typically, antenna elements in an array are identical, which simplifies design and implementation, although this is not a strict requirement These elements can be configured in various geometries to produce different beam patterns As the dimensions of the array geometry increase, the system's complexity also rises The selection of geometry is primarily influenced by the specific application of the array, with further details discussed in section 2.3.1.
The total beam of an antenna array is formed by the combined fields emitted from each antenna element, where constructive interference enhances signals in desired directions and destructive interference cancels signals in undesired directions, creating directive patterns The impact of these element fields on the overall beam is defined as the array factor, and beam steering relies on modifications to this array factor.
Two undesirable phenomena, grating lobes and mutual coupling, in array antenna are introduced in section 2.3.3
In addition to array geometry, array factor, grating lobes, and mutual coupling, it is crucial to consider feed techniques in antenna design The excitation wave for each antenna element is delivered from the source via feed networks, which also serve as the platform for integrating control signals.
Microwave circuits utilize specific components to effectively control the amplitude and phase of waves, facilitating the generation of desired beam patterns Section 2.3.4 details the feed networks designed for each antenna within the array.
The geometrical configuration of an array can take various forms, including linear, planar, circular, or spherical Each configuration influences the effective field distribution and the mutual coupling characteristics, resulting in distinct performance outcomes for each arrangement.
Figure 2-11: Phased array antenna geometry: (a) Linear, (b) Planar, (c) Circular, (d)
A linear phased array consists of antenna elements arranged in a straight line along an axis, typically using identical antennas spaced evenly to simplify calculations and beam control This configuration, known as a uniform array, features elements of the same magnitude with progressively phased signals The spacing between these antenna elements, referred to as "element spacing" and denoted as d, is crucial for performance In a linear phased array, the main beam can only scan along the x-axis, with the angle θ representing the arrival angle of the radio waves.
A planar phased array antenna consists of multiple antennas arranged in a plane with uniform spacing, featuring N elements in each column and M elements in each row By manipulating the magnitude and phase of the incoming waves for each antenna, the main beam can be directed along both the x and y axes, enabling comprehensive 2D angular scanning in horizontal and vertical directions This capability makes planar phased array antennas ideal for applications in smart antennas and beamforming antenna systems.
In a circular phased array, antennas are positioned on a circular ring of radius 'a' with 'N' equally spaced elements, similar to a planar phased array This configuration allows the circular phased array to steer beams in two dimensions Essentially, while it resembles a one-dimensional linear array, its unique circular arrangement distinguishes it from traditional planar arrays.
A spherical phased array features antenna elements positioned on the surface of a sphere, either uniformly or non-uniformly spaced This unique design allows for radiation in any direction, providing omnidirectional or isotropic coverage Despite its advantages, the complexity in modeling, designing, and fabricating spherical phased arrays has resulted in them receiving less attention compared to other geometries.
Here is a rewritten paragraph that captures the essence of the original text, optimized for SEO:"The Array Factor (AF) is a critical parameter in array antenna design, varying uniquely for each array configuration It is influenced by several key factors, including the number of antennas, geometrical arrangements, relative magnitude, phase difference, and element spacing As a mathematical representation, the AF describes the relationship between the total field of the array and the field of a single element By multiplying the field of a single antenna (Es) with the AF, the total field (Etotal) at the far-field of the array can be accurately calculated, as demonstrated in [10]."
In an analysis of two dipole antennas with an element spacing of d = λ/4, the total fields vary significantly based on different phase excitations, as illustrated in Figure 2-12.
Determining a universal formula for Array Factor is challenging due to its dependence on various parameters Typically, the Array Factor formula is derived under specific conditions The most straightforward scenario involves a uniform amplitude and spacing linear array, where the distances between adjacent elements and the power supplied to the antennas are consistent In this case, the Array Factor can be expressed as:
In antenna arrays, the number of antenna elements (N) and the spacing between these elements (d) are crucial factors The phase difference (β) of the incident waves received by the antennas also plays a significant role, while the angle θ varies from 0 to 2π The peak value of the array factor is observed at a specific angle, optimizing the array's performance.
DESIGN OF PHASED ARRAY ANTENNA
Structure
In order to have a suitable structure of phased array antenna for indoor localization, I need to pay attention to three issues: array geometry, element spacing and feed network
For indoor positioning, the focus is solely on user location, making a 2D linear array structure ideal for this purpose While alternative architectures like planar, circular, and spherical arrays exist, they introduce unnecessary complexity and cumbersome designs Therefore, a linear array with uniform amplitude and spacing is preferred due to its straightforward beam steering principle.
In a linear array antenna system, optimal element spacing is crucial to minimize mutual coupling, avoid grating lobes, and enhance directivity As outlined in section 2.3.3, element spacing should ideally range from 0.33 to 0.5 wavelengths to mitigate mutual coupling effects To prevent grating lobes while scanning the main beam angle from -45° to 45°, the spacing must be less than 0.58 wavelengths Additionally, research by V Rabinovich and N Alexandrov illustrates that directivity is directly related to element spacing; when spacing is less than one wavelength, directivity is proportional to the ratio of spacing to wavelength Therefore, for optimal performance, the element spacing in this antenna system is set to half a wavelength.
Figure 3-1: Directivity as a function of the element spacing of linear array antenna [11]
The parallel feed network is chosen for its advantages over serial feed and butler matrix systems The subsequent sections of the thesis will detail the components of the parallel feed, specifically focusing on the power divider and phase shifter.
Power Divider
A parallel feed network, as detailed in section 2.3.4, consists of power dividers and phase shifters that split an incident wave into multiple ports For a uniform amplitude and spacing linear array design, power must be evenly distributed among the output ports When the outgoing waves maintain identical phases, phase compensation becomes unnecessary, ensuring that the output waves have both the same amplitude and phase In microwave applications, minimizing loss allows for greater power transmission to antennas, thereby enhancing the transmission range Additionally, mismatches in microwave circuits can lead to reflection waves that may interfere with the output ports; however, ensuring high isolation between the power divider's output ports can prevent these reflected waves from impacting one another.
Power dividers and combiners are essential passive components in the microwave field, designed to efficiently distribute or combine signals As power dividers, they split an input signal into two or more lower power outputs, while as combiners, they merge multiple input signals to generate a higher power output These devices typically offer in-phase output signals and can achieve both equal (3dB) and unequal power division ratios Notably, T-junction and Wilkinson configurations are among the most widely used power dividers and combiners in the industry.
The T-junction divider is the simplest form of a power divider, functioning as a three-port network with one input and two outputs However, a significant drawback is the lack of isolation between the output ports, which can lead to unexpected signals if one port is not properly matched Additionally, T-junction dividers struggle to simultaneously achieve optimal power loss and impedance matching at the output ports Consequently, when utilizing T-junction dividers, users have two design options to consider.
The Lossless Divider design features three transmission lines, Z0, Z1, and Z2, as illustrated in Figure 3-2a This configuration allows incoming power from the Z0 line to be divided into two smaller outputs through the Z1 and Z2 lines Detailed analysis of the lossless divider is available for further exploration.
The output line impedances, Z1 and Z2, can be tailored to achieve different power division ratios in power dividers For instance, a 3dB equal split power divider can be constructed using a 50-ohm input line and two 100-ohm output lines This lossless divider features a scattering matrix at the design frequency, where the non-zero values of S22, S23, S32, and S33 are influenced by the division ratio In a 3dB lossless divider, these values are 0.5, -0.5, -0.5, and 0.5, respectively.
Figure 3-2: T-junction divider: (a) Lossless; (b) Resistive
The second design features a Resistive Divider composed of three resistors, as illustrated in Figure 3-2b This resistive divider can be effectively analyzed using circuit theory With the appropriate resistor configuration, it is possible to achieve matching at all ports, although the two output ports may lack isolation The corresponding scattering matrix is also provided.
The Wilkinson Power Divider (WPD) effectively addresses the disadvantages of T-junction dividers, such as impedance matching, power loss, and isolation, while maintaining a lossless appearance when output ports are matched The WPD structure, illustrated in Figure 3-3, consists of two quarter-wave transmission lines with equal impedance and a lumped-element resistor at the end The analysis of this circuit is typically conducted using the "even-odd" mode analysis technique, with the scattering matrix calculations detailed in reference [6] Ultimately, the scattering matrix for the WPD is provided.
The input power is efficiently divided between two output ports without any transmission line losses Additionally, there are no reflected waves from the output ports, ensuring complete isolation between them.
Figure 3-3: The Wilkinson power divider: (a) Microstrip line form, (b) Equivalent
I chose the 8-way power divider for my antenna array's feed network due to the advantages of Wilkinson Power Dividers (WPD) over T-junction dividers As introduced by David M Pozar in [6], the N-way WPD features a transmission line whose impedance varies with the number of output ports, ensuring matched conditions and isolation across all ports However, a significant challenge arises for designs with three or more outputs, as the need for crossovers in the resistor configuration complicates planar fabrication.
Figure 3-4: An N-way, equal-split Wilkinson power divider[6]
Figure 3-5 An 8-way equal-split Wilkinson power divider :
Alternatively, I use the cascade structure, shown as in Figure 3-5 Two way WPD will be connected together by 50 transmission lines in a 3-stage structure The Ω
50 transmission lines allow to arbitrarily adjust distances between output ports in Ω order to satisfy the dimension of phase shifters and element spacing between antennas in phased array antenna system.
Phase Shifter
In a parallel feed network, the phase shifter plays a crucial role by generating the necessary phase differences for waves directed towards antennas It must fulfill three essential requirements: first, it should be capable of achieving a full 360-degree phase shift to accommodate all phase difference needs in a phased array antenna system Second, the ability to continuously shift the phase allows for precise steering of the main beam Lastly, in a uniform amplitude and spacing linear array, a phase shifter with low insertion loss variation is vital, as any significant loss would complicate the beam steering process, treating the linear array as a nonuniform amplitude system.
O ut pu t 1 O ut pu t 2 O ut pu t 3 O ut pu t 4 O ut pu t 5 O ut pu t 6 O ut pu t 7 O ut pu t 8
A phase shifter is a crucial component in phased array antenna systems, enabling the adjustment of wave phases for each antenna to facilitate beam scanning or shape reconfiguration Based on the desired output, phase shifters are categorized into two types: analog phase shifters and digital phase shifters.
Digital phase shifters utilizing semiconductor components like PIN diodes and FET switches offer predefined states for accurate phase shifting While these phase shifters deliver high precision, their structural complexity increases to accommodate high-resolution requirements As the demand for higher resolution rises, the number of predefined states and corresponding switching elements also expands, resulting in a larger controller size.
An analog phase shifter utilizing varactors or Schottky diodes allows for continuous phase shift adjustments, leveraging the diodes' capacitance that varies with bias voltage to function as electrically variable capacitors in tuned circuits This capability enables high-resolution phase shifting without the need for hardware modifications However, this approach comes with drawbacks, including lower accuracy and a relatively narrow bandwidth.
To implement a microwave phase shifter, there are 4 basic ways:
Switched line phase shifters are a common form of digital phase shifters that utilize delay lines and switching elements to generate varying time delays The phase shift produced is solely influenced by the length of the transmission line, ensuring high stability across different temperatures and over time A basic schematic representation of a switched line phase shifter can be found in Figure 3-6 a.
Figure 3-6 Type of phase shifter: (a) Switched Line; (b) Switched Network; : s
Switched network phase shifters, as illustrated in Figure 3-6b, function similarly to switched line phase shifters; however, they utilize networks composed of inductors and capacitors instead of delay lines This design results in less variation in size compared to switched line phase shifters, making switched network phase shifters particularly suitable for low-frequency applications.
Loaded line phase shifters utilize a shunt reactance that can be electrically adjusted by a PIN diode or FET to achieve the desired phase shift They offer benefits such as simplicity and low insertion loss for phase shifts under 45 degrees However, achieving larger phase shifts necessitates high susceptance values, which can lead to increased insertion loss, making this type of phase shifter suitable only for shifts less than 45 degrees.
Reflection type phase shifters utilize a 3dB hybrid coupler and two tunable loads, enabling continuous phase shifts exceeding 360 degrees This design ensures low insertion loss, as demonstrated in previous studies [17][19].
For an effective phased array antenna system, a reflection-type phase shifter offering a full 360-degree continuous phase shift is the optimal choice The following section will introduce my proposed reflection-type phase shifter design.
3.3.3 Reflection Type Phase Shifter a Principle of Reflection Type Phase Shifter
The Reflection Type Phase Shifter (RTPS) is an electrically adjustable device designed to shift the phase of a wave at the output port while eliminating reflected waves at the input port It integrates a 3dB Hybrid Coupler with various components, including capacitors, resistors, transmission lines, and varactors, to achieve this functionality.
The 3dB Hybrid Coupler is a four-port network designed with all ports matched to a reference impedance of 50 ohms When a signal is introduced at port 1, it is transmitted to ports 3 and 4 while being suppressed at port 2 This unique functionality allows for effective signal distribution and isolation within the network.
34 equal to one of incoming wave at port 1, and phase of waves at port 3 and 4 shift
90 o and 180 o compared with one at port 1, respectively
In mathematical model, 3dB Hybrid Coupler is characterized by a scattering matrix that represents voltage relationship between incoming (V + ) and reflected (V - ) waves at different ports
Here, S ij is ratio of the reflected wave at port to the incoming wave at port i j when there is only input wave from port j
To be more specific, we have:
Considering the reflection coefficient ( ) generated by the load, the reflection coefficient is determined from the load impedance and the source impedance j
The RTPS model, illustrated in Figure 3-8, consists of two key components: the 3dB Hybrid Coupler and the reflective load In this setup, port 1 serves as the incoming port, while port 2 functions as the outgoing port, with loads Z L3 and Z L4 connected to ports 3 and 4, respectively The phase shift of the RTPS is defined as the phase difference between the outgoing wave at port 2 and the incoming wave at port 1.
Suppose there is only incoming wave at port 1 When passing through the 3dB Hybrid Coupler, from (3-8), waves at ports 3 and 4 are:
Because of appearance of Z L3 load at port 3, there is wave reflection with reflection coefficient Γ3 Therefore, from (3-7) and (3-8), the reflection wave at port
Similarly, we have the reflected wave at port 4: 4 4 4 4 1
The reflected waves at ports 3 and 4 can be also consider as incoming waves from ports 3, 4 transmitting to ports 1 and 2
We set V 3 ' andV 4 ' as incoming wave at ports 3 and 4, respectively We have:
These two waves transmit to ports 1 and 2 through 3dB Hybrid Coupler, and outgoing waves from ports 1 and 2 are calculated fro (3-5), (3-9) m and (3-10) as follow:
For the phase shifter, we wish that there is no reflected waves at the port 1 Therefore, 3 4 enables us to eliminate the reflected wave at port 1
From (3-7) and (3-11), choosing Z L3 = Z L4 makes 3 4 and outgoing wave at port 2 asV 2 j V 1 Finally, the forward voltage gain S21 of RTPS will be as follow:
(3-12) b Design of Reflection Type Phase Shifter
The reflection type phase shifter (RTPS) is an innovative device that enables continuous phase adjustment of waves without requiring hardware modifications, utilizing the capacitance variations of varactors As illustrated in Figure 3-9, the RTPS comprises a 3dB hybrid coupler and reflection loads positioned at ports 3 and 4, allowing waves to enter through port 1 and exit via port 2.
Figure 3-9 Schematic Diagram of RTPS :
When putting reactive loads to ports 3 and 4, the reflection coefficient are:
From (3- ) 12 and (3-13), the phase shift of RTPS with reactive loads is:
The phase shift values (∠S21) are significantly influenced by the capacitance range of varactors, which ultimately constrains the achievable phase shift range To theoretically attain a full 360° phase shift, the capacitance of varactors would need to be infinite, a scenario that is not feasible Consequently, it is impossible for RTPS to achieve a complete 360° phase shift using a single varactor with reflection loads.
To address the issue, I utilize a reflection load structure, as illustrated in Figure 3-10 This approach enables the selection of an appropriate capacitance range, followed by the identification of suitable varactors and their corresponding voltage.
Figure 3-10 Reflection Load of RTPS :
The equivalent impedance of the load is calculated as below:
+ From section 2.1.1, because Z T2 is a quarter-wavelength transmission line, the equivalent impedance of Z T2 and D 2 is:
+ Similarly, ZT1 is also a quarter-wavelength, so the equivalent impedance is:
+ On the other hand, the impedance of the capacitance diodes are: d
In order to get 360 o phase shift, we have:
Thus, by calculation of ZT2 satisfying (8), we will have 360 o phase shift From (3- ), we have: 21
At 2.45 GHz, I choose the varactor SMV1247, package type SC-79 with Ls =
0.7nH To simplify the voltage controller of varactors, can choose the capacitance I range of SMV1247 from 8.5pF to 0.7pF corresponding to the voltage range from 0V to 5V From (3-21), Z T2 = 15Ω
Antenna Element
When selecting antenna elements for a Wi-Fi based indoor positioning system, two critical parameters must be considered: operating frequency and radiation pattern The antennas must effectively operate within the Wi-Fi frequency band of 2.4GHz to 2.484GHz to ensure optimal power transmission to smartphone antennas Additionally, the return loss of these antenna elements should exceed 10dB within the same frequency range Furthermore, the radiation pattern is essential, as the main beam is a function of this pattern, influencing the overall performance of the positioning system.
In the design of phased array antennas, the angle of the main beam can vary from -45° to 45°, necessitating that the antenna element has a half-power beamwidth greater than 90° This specification ensures that the radio signal strength is primarily influenced by the array factor While omnidirectional antennas can fulfill this requirement, they typically radiate a significant portion of power to the rear, leading to unwanted secondary main beams due to reflections To address this issue, I propose using a Microstrip patch antenna, which not only meets the required beamwidth but also features minimal back lobes.
When designing a microstrip antenna, three popular analysis methods are utilized: transmission line, cavity, and full wave The transmission line model is the simplest but lacks accuracy and poses challenges in modeling coupling Conversely, the cavity model offers greater accuracy while being more complex, yet it still struggles with coupling modeling despite its successful applications Full wave models provide high accuracy and versatility, capable of handling single elements, finite and infinite arrays, stacked elements, and arbitrary shapes, but they are the most complex to implement.
With advancements in computer hardware, simulation tasks are becoming quicker and more efficient Consequently, for single antennas, the use of cavity and full-wave models is often unnecessary Instead, the transmission line model is typically employed to provide approximate parameters, followed by tuning in the simulation to achieve the final design.
This thesis presents the design of a microstrip patch antenna tailored for the Wi-Fi frequency band of 2.4GHz to 2.484GHz The antenna features an input impedance of 50Ω and is constructed using an FR4 substrate, characterized by a dielectric constant of ε = 4.3, a loss tangent of tan = 0.02, and a thickness of 62 mil (1.58mm).
The width is computed with the following equation [10], [25], [26], where is f center frequency (2.45GHz):
The actual length of the microstrip patch antenna is given by: mm L
W h h mm W h h W L f mm L eff eff eff eff eff
EXPERIMENTAL RESULT
Wilkinson Power Divider
The WPD simulation is performed on Keysight Technologies’s Advanced Design System (ADS) software The substrate used to simulate and fabricate is FR4 with following parameters:
For a 2-way WPD, theoretically, we know that the total input power is equally divided into two output ports Suppose that input is port 1, and outputs are port 2 and
3 (Figure 4-1a) S 21 = S 31 -3dB= as shown in Figure 4-1b
Figure 4-1: The 2-way WPD in theory: (a) Schematic Circuit; (b) Forward gains S 21 , S 31
Microstrip circuits are influenced by the substrate's loss tangent coefficient and any discontinuities present in the microstrip The loss tangent indicates the dielectric loss, with a value of tanδ = 0 signifying a lossless material.
In lossy materials, the loss tangent exceeds 0, indicating greater energy loss, with a higher loss tangent correlating to increased losses For instance, the loss tangent of FR4 substrate is approximately 0.025, which theoretically results in a forward gain of less than one for WPD on FR4 Additionally, the microstrip characteristics contribute to this phenomenon.
The phenomenon of discontinuity, discussed in section 2.1.2 and referenced in sources [6] and [8], impacts practical applications by introducing parasitic components that alter circuit parameters To achieve simulation results that closely align with measured outcomes, it is essential to incorporate discontinuity components like Bend and Tee in the simulation Subsequently, the dimensions of the transmission lines are fine-tuned using the optimization tool in ADS software to attain the desired results The two-way WPD on FR4 is illustrated in Figure 4-2, with S21 and S31 both measuring -3.23 dB.
Figure 4-2: The 2-way WPD: (a) Schematic Circuit; (b) Forward gains S 21 , S 31
The 8-way WPD is constructed from 2-way WPDs and 50 transmission lines, following the cascade structure outlined in section 3.2.2 In microwave circuits, closely placed Ω transmission lines can lead to coupling, which impacts circuit characteristics This coupling is challenging to calculate and estimate, and it is not represented in schematic simulations Therefore, EM simulation and EM co-simulation are employed to ensure that the final simulated results meet expectations.
Figure 4-3: 8-way Wilkinson Power Divider
The 8-way WPD is constructed on an FR4 substrate and subsequently measured using the PNA N5222A machine, which requires calibration prior to measurement The forward gain for each port is illustrated in Figures 4-4 to 4-6, with the red thin line representing schematic simulation, the blue bold lines indicating EM co-simulation, and the blue dashed lines reflecting the fabricated results of the 8-way WPD.
In schematic simulations, the forward gains across all output ports are uniform at -10.882 dB, deviating from the theoretical value of -9 dB due to losses incurred while propagating on an FR4 substrate with a loss tangent coefficient of tanδ = 0.025 In contrast, the electromagnetic co-simulation shows varied forward gains at the output ports, indicating discrepancies in performance.
-11.031dB -10.493dBto As mentioned above, the difference comes from the coupling between transmission lines
The forward gains of the 8-way WPD on FR4 material closely align with simulation results at the designed frequency of 2.45GHz, with measurement discrepancies remaining below 1dB These differences arise from fabrication tolerances and material heterogeneity, which affect mismatch and loss compared to simulations In contrast, significant discrepancies occur at frequencies below 2.4GHz and above 2.5GHz due to fabrication errors, but these do not impact the final result since the system is specifically designed to operate at 2.45GHz.
Figure 4-4: Forward gain at port 2, 3, 4: S 21 , S 31, S 41
Figure 4-5: Forward gain at port 5, 6, 7: S 51 , S 61, S 71
Figure 4-6: Forward gain at port 8, 9: S 81, S 91
Figure 4-7: Isolation between output ports
Figure 4-7 illustrates the schematic simulated results (red thin line), electromagnetic co-simulation (blue thick line), and measured results (blue dashed line) The power divider exhibits an isolation of approximately 20 dB between its output ports, which is comparable to similar products available in the market.
Reflection Type Phase Shifter
The RTPS simulation is also performed on Keysight Technologies’s ADS software The substrate used for simulation and fabrication is Roger4003C with the following parameters:
The previous design of the RTPS on FR4 substrate exhibited significant variations in measured insertion loss, necessitating complex compensation that added bulk to the system This issue arose from the high loss tangent coefficient of FR4, which resulted in substantial signal loss To address this challenge, a substrate with a lower loss tangent was selected, specifically Roger4003c, which is readily available in our resources Consequently, the RTPS is now implemented on Roger4003c to improve performance and reduce complications.
ADS software includes essential components like microstrip lines, terminals, DC voltage, and stubs However, since the SMV1247 varactors are not natively supported, it's crucial to develop an equivalent model before proceeding with the design and simulation of the RTPS Skyworks provides a SPICE model for the SMV1247 in its technical documentation, which can be imported into ADS software This allows for the creation of an equivalent model of the SMV1247, incorporating elements such as Ls and Cp.
R s values corresponding to SC-79 packages, L s = 0.7nH C, p = 0.54pF R, s = 4.9
Figure 4-8: Equivalent model of SMV1247: (a) on SPICE; (b) on ADS
The C-V characteristic curve of the SMV1247 using the ADS model at 100MHz, depicted in Figure 4-9b, closely resembles the characteristic curve found in the SMV1247 technical paper (Figure 4-9a) Therefore, this ADS model of the SMV1247 will be utilized in subsequent simulations.
Figure 4-9: C- curve of SMV1247 on: (a) technical document; (b) ADS V
To initiate the simulation of the RTPS, ensure all necessary components are in place Similar to the 8-way WPD discussed in section 4.1, the RTPS is implemented in both schematic simulation and EM Cosimulation, as illustrated in Figure 4-10 Additionally, microstrip discontinuity components, including bends, tees, and width steps, are incorporated The optimization and tunnel tools in ADS software facilitate dimension adjustments of the circuit to achieve the desired outcomes Ultimately, the RTPS is fabricated based on these simulations.
Roger4003c and measured by the PNA N5222A machine №te that the PNA machine is always calibrated before measuring
Figure 4-10 Schematic circuit of Reflection type phase shifter :
Figure 4-11 Reflection type phase shifter fabricated on Roger4003c :
The results of the phase shifter are shown as Figure 4- , Figure 4-13, where 12 the green dash line, blue dash dot line and red circle line are schematic simulation,
EM Cosimulation and measured results, respectively
Figure 4-13: Phase shift of RTPS
Figure 4 illustrates the forward gains (S12 21) across three scenarios: schematic simulations, electromagnetic field simulations, and actual measurement results It is observed that the energy transmitted through the phase shifter is slightly lower than that predicted by both the schematic and electromagnetic simulations This discrepancy may stem from fabrication errors and suboptimal contact between the varactors and the microstrip line, which affect the resistance values in the varactors' equivalent circuit, leading to significantly higher power losses than anticipated in the simulations.
Reverse Voltage (V) ar gS 21 (d eg re e)
57 about 2.5dB with nearly 0.5dB of variation The low insertion loss variation enables us to make a phased array antenna with uniform amplitude and spacing
Figure 4-13 shows phase shifts in three cases, the schematic, electromagnetic simulation and measurement results The phase shifters are capable of shifting about
The phase shifter exhibits non-linear behavior compared to the other two cases; however, with a 12-bit DAC, I can still achieve precise control over the phase shifter, allowing for small resolution adjustments.
Microstrip Patch Antenna
The microstrip patch antenna was designed using CST Microwave Studio software and subsequently fabricated with FR4 material, as illustrated in Figure 4-14 To assess the antenna's bandwidth, the parameters VSWR and S11 were measured utilizing the PNA N5222A network analyzer At the operational frequency, the targeted VSWR and S11 values were evaluated.
S 11 are less than 2 and -10dB, respectively №te that, the network analyzer is calibrated before doing measurement.
Figure 4-14: Fabricated microstrip patch antenna
The VSWR and S 11 of patch antenna versus frequency are shown in Figure 4-15, where the red thin lines and blue dash lines are simulated and measured characteristics, respectively
Figure 4-15 Microstrip patch antenna paramters: (a) VSWR; (b) Return Loss :
It can be seen that in simulation, my microstrip patch antenna can operate at Wi-
The microstrip patch antenna operates within a frequency range of 2.424GHz to 2.485GHz, which is slightly limited compared to the ideal Wi-Fi band of 2.4GHz to 2.484GHz This discrepancy arises from material heterogeneity and fabrication errors Despite not covering the entire Wi-Fi spectrum, the antenna remains functional across several Wi-Fi channels.
In general, the radiation pattern of an antenna is three-dimensional Because it is impractical to measure a three-dimensional pattern, a number of two-dimensional
59 patterns are measured, as mentioned in section 2.2.2 Patterns can be obtained by
To analyze antenna patterns, one angle is fixed while the other is varied For a single antenna, the typical azimuthal pattern is set at θ = π/2, with the elevation pattern at ϕ = 0 In the case of phased array antennas, only the θ = π/2 azimuthal pattern is measured to determine the main beam's direction, as the ϕ = 0 elevation pattern does not yield significant information.
As discussed in section 2.2.2, antennas will be observed in the far-field region
For optimal performance, measurement systems should ideally be situated outdoors to eliminate reflection waves; however, this is often impractical As a solution, indoor anechoic chambers have been created to simulate outdoor conditions These chambers are lined with RF absorbers that effectively absorb RF waves, preventing them from reflecting off the walls The reflection coefficient of these anechoic chambers is approximately [insert specific value here].
Figure 4-16: Structure of measuring chamber
Structure of the measurement room is shown in Figure 4- A horn antenna, 16
HF 906 serves as both a source antenna and a test antenna, with its receiving capabilities relying on the reciprocity property of antennas, which ensures that the transmitting and receiving radiation patterns are identical This characteristic eliminates the need to swap between the two antennas The network analyzer, PNA, is utilized to measure the ratio of received power to transmitted power.
The N5222A test antenna is rotated using a controlled rotating table with a 5-degree angle step I developed a controller for this system and created a Windows program to facilitate communication with the controller, retrieve measurement results from the PNA N5222A, and visualize two-dimensional patterns using OpenGL.
Figure 4-17: Measurement in Anechoic chamber in reality
Figure 4-18: Radiation Pattern of microstrip patch antenna: (a) on simulation; (b) comparison between simulated and measured result
The Figure 4- shows the radiation pattern of my microstrip patch antenna 18From the Figure 4-18a, we can see that the half power beam width is larger than 90 o ,
Here is a rewritten paragraph that meets SEO rules:"The radiation pattern of the designed antenna closely resembles the simulated results, as evident in Figure 4-18b This similarity indicates that the antenna's radiation pattern meets the necessary requirements for an indoor positioning system, making it a suitable solution for this application."
Phased Array Antenna
The phased array antenna is constructed by assembling key components, including the power divider, phase shifter, antenna, and controller Its radiation patterns are evaluated in an anechoic chamber at the Department of Technical Logistics of the Ministry of Public Security, similar to the microstrip patch antenna at 2.45 GHz The antenna is designed to steer its main beam across angles ranging from -45° to 45° in 5° increments The measured results are then compared with simulated results in the same coordinate system, as illustrated in Figure 4-19.
The direction of the main beam in measurements can effectively track simulated results; however, discrepancies arise due to phase shifts from variations in electrical length during pairing, as well as amplitude changes in the Wilkinson power divider and reflection-type phase shifter Additionally, fabrication tolerances contribute to beam angle errors between measured and calculated values, with beam scan angles maintaining a tolerance of ±5º (Table 4-1) Overall, the measurement results align well with the simulated outcomes.
In comparing simulated results to real-world measurements, it is observed that the gain of side lobes is significantly higher in practice This discrepancy arises from the differences in spatial conditions between simulation and measurement While simulations are conducted in open space, eliminating reflections that can affect the radiation pattern, actual measurements, even in an anechoic chamber, are still influenced by reflections from the chamber walls and other objects present.
62 and affect to measured results Compared with main lobe, side lobe level is less than about 10dB in most case (Table 4-1)
Table 4-1: Comparison of main beam angle and side lobe level in simulation and measurement
Main beam angle Side lobe level Main beam angle Side lobe level
Table 4-2 presents a comparison of the performance of my antenna design against previous indoor localization antenna designs Notably, earlier designs utilized a system of switching predefined beams, resulting in a limited number of beams available for localization.
The design of indoor antennas is constrained by the limited number of antenna elements, which typically amounts to 63 This limitation affects the number of lobes, as larger antennas are impractical for indoor use Consequently, the scanning steps in previous research have been relatively large, including angles of 60°, 30°, 90°, and 18° However, this work introduces a method to control the Array Factor of the antenna array through the phase of incoming waves, allowing the number of beams to be independent of the number of antenna elements Additionally, the implementation of a full 360° continuous phase shifter permits arbitrary phase adjustments, enabling continuous steering of the main beam with a refined angle step of 5°.
Table 4-2: Comparison with previous antenna design for indoor localization
Antenna Scanning range № of beam Average
Figure 4-19: Radiation pattern of phased array antenna at different angles
CONCLUSION AND FUTURE WORK
Conclusions
This thesis focuses on enhancing the resolution of Angle of Arrival (AoA)-based indoor positioning systems through the design of an 8-port phased array antenna utilizing a reflection-type phase shifter It begins with an analytical foundation to select the optimal phased array antenna structure for indoor applications Subsequently, the design process includes the development of the feed network, comprising a power divider and phase shifter, followed by the design of the antenna elements.
The 8-way power divider is composed of 2-way Wilkinson Power Dividers to limit the loss and increase the isolation between output ports The 8-way power divider, deployed on FR4 substrate, has insertion loss about 11dB and isolation about 20dB
A reflection-type phase shifter has been designed and fabricated using a proposed load topology that incorporates two varactors (SMV1247) and two transmission lines on a Roger4003c substrate This innovative phase shifter enables continuous phase control across a full 360-degree range, achieving an insertion loss of approximately 2.5 dB with a variation of 0.5 dB.
Antenna elements are microstrip patch antennas on FR4 material that can operate in frequency range from 2.424GHz to 2.484GHz, and their half power beam width larger than 90 o
The radiation pattern of the phased array antenna was evaluated in an anechoic chamber, with the main beam direction adjustable from -45° to 45° in 5° increments The measured results indicate that the main beam direction closely aligns with the desired simulation outcomes However, some squint is observed, attributed to amplitude variations in the Wilkinson power divider and reflection-type phase shifter, as well as minor electrical length differences in the assembly Typically, the side lobe level is approximately 10 dB lower than the main lobe.
Future Works
I proposed some improvements for my system and summary it as following:
To assess the effectiveness of my phased array antenna, I will develop an indoor localization application utilizing Angle of Arrival (AoA) algorithms These algorithms, including Beamscan, ESPRIT, and MUSIC, are designed to minimize the impact of reflections within indoor environments.
Effective control of side-lobes is crucial for minimizing reflections in a room over the long term Consequently, designing variable gain amplifiers is vital for managing the amplitude of waves directed to each antenna Utilizing beamforming algorithms for smart antennas ensures comprehensive control of side-lobes.
In their 2016 paper presented at the 9th Regional Conference on Electrical and Electronics Engineering (RCEEE 2016), Cong Thuan Nguyen, Trung Kien Dao, and Thanh Huong Nguyen discuss the design of an innovative 8-port antenna array feed network This design incorporates a programmable phase shifter specifically tailored for the ISM 2.4GHz band, showcasing advancements in antenna technology and its applications in modern communication systems.
Thanh Huong Nguyen, Cong Thuan Nguyen,“A Continuous 360 o Reflection Type Phase Shifter with Low Loss Variation at 2.4GHz for Indoor Localization”,
2018 International Conference on Information and Communication Technology and Digital Convergence Business (ICIDB 2018), 19-21 January 2018, Hanoi, Vietnam
[1] L Brás, N B Carvalho, P Pinho, L Kulas, and K Nyka, “A Review of Antennas for Indoor Positioning Systems,” Int J Antennas Propag., vol 2012, pp 1 14, –
[2] G Giorgetti, A Cidronali, S Gupta, and G Manes, “Single-anchor indoor localization using a switched-beam antenna,” IEEE Commun Lett., vol 13, no
[3] M Rzymowski, P Woznica, and L Kulas, “Single-Anchor Indoor Localization Using ESPAR Antenna,” IEEE Antennas Wirel Propag Lett., vol 15, pp 1183–
[4] M R Kamarudin, Y I Nechayev, and P S Hall, “Onbody Diversity and Angle- of-Arrival Measurement Using a Pattern Switching Antenna,” IEEE Trans Antennas Propag., vol 57, no 4, pp 964 971, Apr 2009 –
[5] T D Bui, B H Nguyen, Q C Nguyen, and M T Le, “Design of beam steering antenna for localization applications,” in Antennas and Propagation (ISAP), 2016
[6] D M Pozar, Microwave engineering, 4th ed Hoboken, NJ: Wiley, 2012
[7] -S G Hong and M J Lancaster, J Microstrip filters for RF/microwave applications New York: John Wiley, 2001
[8] I J Bahl, Lumped elements for RF and microwave circuits Boston: Artech House, 2003
[9] R J Mailloux, Phased array antenna handbook, vol 2 Artech House Boston,
[10] C A Balanis, Antenna theory: analysis and design, 3rd ed Hoboken, NJ: John Wiley, 2005
[11] V Rabinovich and N Alexandrov, Antenna Arrays and Automotive Applications New York, NY: Springer New York, 2013
[12] D L Miroslav, J e L Milan, and O e L Borislav, Analysis of SDMA & smart antenna techniques for existing and new mobile communication systems 2003
[13] C H Tseng, C J Chen, and T H Chu, “A Low-Cost 60-GHz Switched-Beam Patch Antenna Array With Butler Matrix Network,” IEEE Antennas Wirel Propag Lett., vol 7, pp 432 435, 2008 –
[14] C C Chang, R H Lee, and T Y Shih, “Design of a Beam Switching/Steering Butler Matrix for Phased Array System,” IEEE Trans Antennas Propag., vol 58, no 2, pp 367 374, Feb 2010 –
[15] T A Denidni and T E Libar, “Wide band four-port butler matrix for switched multibeam antenna arrays,” in 14th IEEE Proceedings on Personal, Indoor and Mobile Radio Communications, 2003 PIMRC 2003., 2003, vol 3, pp 2461 2464 – vol.3
[16] J Butler, “Beam-forming matrix simplifies design of electronically scanned antennas,” Electron Des., vol 12, pp 170 173, 1961 –
[17] T Lambard, O Lafond, M Himdi, H Jeuland, and S Bolioli, “A novel analog 360° phase shifter design in Ku and Ka bands,” Microw Opt Technol Lett., vol
[18] S Bulja and D Mirshekar-Syahkal, “Analysis and design of a new reflection- type 360° phase shifter with combined switch and varactor,” Microw Opt Technol Lett., vol 52, no 3, pp 530 535, Mar 2010 –
[19] F Burdin, Z Iskandar, F Podevin, and P Ferrari, “Design of Compact Reflection-Type Phase Shifters With High Figure-of-Merit,” IEEE Trans Microw Theory Tech., vol 63, no 6, pp 1883 1893, Jun 2015 –
[20] D W Gardner and M A Wickert, “Microwave filter design using radial line stubs,” in IEEE Region 5 Conference, 1988:’Spanning the Peaks of Electrotechnology’, 1988, pp 68 72 –
[21] B C Wadell, Transmission line design handbook Artech House, 1991
[22] H A Atwater, “The design of the radial line stub: A useful microstrip circuit element,” Microw J., pp 149 156, 1985 –
[23] L Weijun, C Xiaojuan, L Xiaoxin, M Xiaolin, L Xinyu, and W Xiaoliang,
“A radial stub test circuit for microwave power devices,” Chin J Semicond., vol
[24] R Sorrentino and L Roselli, “A new simple and accurate formula for microstrip radial stub,” IEEE Microw Guid Wave Lett., vol 2, no 12, pp 480–
[25] M A Afridi, “Microstrip Patch Antenna Designing at 2.4 GHz Frequency,” -
[26] G Casu, C Moraru, and A Kovacs, “Design and implementation of microstrip patch antenna array,” in Communications (COMM), 2014 10th International Conference on, 2014, pp 1 4 –
[27] Z Zhang, Antenna design for mobile devices Singapore: John Wiley & Sons (Asia), 2011
Here is a rewritten paragraph that captures the essence of calibration:Calibration is a crucial process that involves comparing the measurement values of a device under test with those of a calibration standard of known accuracy, which has a higher level of precision than the device being calibrated By doing so, the calibrated device can be adjusted to match the standard device's readings, ensuring more accurate results in future measurements.
When using scattering instruments like Precision Network Analyzers (PNA) and Vector Network Analyzers (VNA), it is crucial to recognize that cable loss can vary significantly due to environmental factors, cable design, and the quality of connections with meters and devices To ensure accurate measurements, a calibration step is essential prior to taking readings Additionally, maintaining a stable cable position and secure connections during both calibration and measurement processes is vital for obtaining reliable results.
Figure A-1: Elements of 85052D calibration kit
The PNA N5222A Network Analyzer operates on the RF Platform and utilizes the 85052D calibration kit, which supports a frequency range from DC to 26.5GHz This kit features both male and female 3.5mm connectors for calibrating two ports and includes SHORT, OPEN, BROADBAND, and ADAPTER 50 loads, as illustrated in Figure A-1 The calibration process varies depending on the port number selected for measurement.
For measuring antenna performance, a one-port I mode calibration is utilized, as the antenna comprises a single port element This calibration process is facilitated by the Calibration Wizard found in the network analyzer, which guides users through specific steps for accurate results.
+ Select the number of ports is 01
+ Choose the 85052D calibration kit and 3.5 female connector
+ Sequentially connect OPEN, SHORT and BROADBAND to measure
Figure A-2 OPEN, SHORT, BROADBAND elements of 85052D calibration kit :
To accurately measure the characteristics of a power divider and phase shifter, it is essential to utilize a minimum of two ports and perform calibration in a two-port mode This calibration is facilitated by the Calibration Wizard in the PNA network analyzer, which begins by selecting the number of ports as two.
+ Choose the 85052D calibration kit and 3.5 female connector
+ Sequentially connect OPEN, SHORT and BROADBAND to port 1 then port
+ Finally, connect ADAPTER 50to 2 ports then measure
Appendix B: Antenna Radiation Pattern Measurement System
The Antenna Radiation Pattern Measurement (ARPM) system includes 4 main parts as shown in Figure B-1
Figure B-1: Block diagram of ARPM system
The test antenna is mounted on a rotating structure that allows for azimuth surface rotation This rotary motion is powered by a stepped motor, ensuring precise movement Figure B-2 illustrates the design created in SolidWorks alongside a real-world image of the structure.
+ The stepped motor is controlled by controller through the driver L298 module
At the same time, controller also communicate with computer
The central component of this system is a computer program that interfaces with a controller to rotate the structure and obtain S21 measurements using the PNA N5222A machine All results are stored in a text file and visualized as a radiation pattern using OpenGL.
+ Final part is the PNA N5222A machine that is used to measure the loss between reference antenna and test antenna
Appendix C: Main beam angle versus DC bias look up table
Beam Angle PS1 PS2 PS3 PS4 PS5 PS6 PS7 PS8