Design and Simulation of a Planar Crossed-Dipole Global Navigation Satellite

Transcription

1 Design and Simulation of a Planar Crossed-Dipole Global Navigation Satellite System (GNSS) Antenna in the L1 Frequency Band A thesis presented to the faculty of the Russ College of Engineering and Technology of Ohio University In partial fulfillment of the requirements for the degree Master of Science Mahesh Katragadda December Mahesh Katragadda. All Rights Reserved.

2 2 This thesis titled Design and Simulation of a Planar Crossed-Dipole Global Navigation Satellite System (GNSS) Antenna in the L1 Frequency Band by MAHESH KATRAGADDA has been approved for the Department of Electrical Engineering and the Russ College of Engineering and Technology by Chris G. Bartone Professor of Electrical Engineering and Computer Science Dennis Irwin Dean, Russ College of Engineering and Technology

3 3 ABSTRACT KATRAGADDA, MAHESH, M.S., August 2012, Electrical Engineering Design and Simulation of a Planar Crossed-Dipole Global Navigation Satellite System (GNSS) Antenna in the L1 Frequency Band (113 pp.) Director of Thesis: Dr. Chris G. Bartone The purpose of this research was to design and investigate the performance of a planar cross-dipole Global Navigation Satellite System (GNSS) receiver antenna to operate in GNSS L1 frequency band with multipath mitigation performance. First, a crossed-half-wavelength dipole antenna was designed to resonate centered at 1.6 GHz, and then various configurations were investigated and simulated. In an attempt to improve the antenna s multipath mitigation performance, these various configurations included the addition of concentric rings in various numbers and sizes with the inclusion of cavity banking. Roger s dual copper clad RO3010 material was used as a substrate for the design. The simulated cavity backing consisted of a thin block of Styrofoam and Cumming s RGDS-124 (microwave absorbing material) placed between the substrate and the ground plane (finite and infinite), simulated to increase the antenna performance. The prototype crossed-dipole antenna configurations were tuned and optimized based on the simulated results obtained from the Agilent s Advanced Design System Momentum 2009U1 software. The S- parameters and radiation pattern results obtained from these simulations were compared and analyzed in MATLAB. The simulated antenna prototype performance is expected to meet the performance requirements for L1 GNSS receiver antenna.

4 Dedicated to my Parents 4

5 5 ACKNOWLEDGEMENTS It is my pleasure to thank each and every individual who made this thesis possible. First and foremost, I am very grateful to have Dr. Chris Bartone as my thesis supervisor. I am glad to say that his support, his enthusiasm, his inspiration, his patience and his ability to explain things simply and clearly gave me the confidence to explore my research interests. I would like to take this opportunity to thank him for providing me with the research assistantship during the first two quarters. I have gained lots of knowledge from Chris at each and every stage of my thesis; his sound advising, teaching, and lots of good ideas are invaluable. I would have been lost without him. I would like to thank Dr. Dill, Dr. David Matolak, and Dr. Roger Radcliff for accepting my services as a Graduate Assistant in the Electrical Engineering and Computer Science Department. I am also thankful to Bryan Jordan, Donner Davis and Mike Finney for offering me a student part-time in the computer labs of Stocker Center and the Voinovich School of Leadership and Public Affairs. Thanks for your financial support, guys. I am also grateful to all staff and colleagues at the Russ College of Engineering who helped me throughout my graduate study life. I specially thank Dr. Simbo Odunaiya, Dr. Sanjeev Gunawardena and Dr. David Ingram for being a part of my thesis defense committee. Finally, I would like to thank my parents and brother; without their support I wouldn t be standing in front of you today. I would like to dedicate my thesis to them.

6 6 TABLE OF CONTENTS Abstract... 3 Acknowledgements... 5 List of Figures List of Tables List of Acronyms Introduction Background: GPS Fundamentals: Space Segment (SS) Control Segment (CS) User Segment (US) GLONASS Fundamentals: Galileo Fundamentals: Signal Characteristics: S-parameter Analysis: Scope: GNSS Antenna GNSS Antenna Performance Requirements: Resonant Frequency ( ): Voltage Standing Wave Ratio (VSWR) and Return Loss (RL): VSWR Bandwidth (BW): Antenna Polarization: Antenna Axial Ratio (AR):... 31

7 Antenna Pattern: Desired to Undesired Ratio: Antenna Phase Reference: Challenges: Objective: Performance Requirements: GNSS Antenna Design Antenna Selection Criteria: Advantages and Limitations of the Microstrip Antennas: Various Configurations: Overview: Multipath Mitigation and CAN: Feeding Technique and Feed Locations: Design Considerations for a Printed Dipole Antenna Resonant Frequency ( ) and Dipole Dimensions (L and W) Dipole Bandwidth: Radiation Pattern: Polarization of Planar Dipole Antenna: Substrate Selection Criteria: Design of Concentric Rings: Antenna Design Summary: Simulation Setup and Performance Analysis of Planar Crossed-Dipole Antenna Configurations In Free-Space ADS Momentum 50

8 Overview: ADS Layout: Simulation Setup in ADS Momentum: Substrate Material Properties: Properties of the Absorbing Material and the Finite Ground Plane: Design Configurations: Performance Analysis of Planar Dipole in Free-Space: Design and Simulation of a Single Half-Wave Dipole in Free-Space: Analysis of Single Half-Wave Dipole Results: Validation Half-Wave Dipole Antenna Simulation in Free-Space: Performance Analysis of the Planar Cross-Dipole in Free-Space: Design and Simulation of a Crossed-Dipole in Free-Space Crossed-Dipole Free-Space Configurations with Rings: Analysis of Results Obtained From the Free-Space Simulations: Simulation and Measurement of L1 Planar Crossed-Dipole Antenna on Infinite and Finite Ground Planes and With a CAN Simulation Setup for the Planar Crossed-Dipole Design with an Infinite, Finite Ground Plane and a CAN and Configuration Descriptions: Performance of the Planar Crossed-Dipole Design Configurations with an Infinite Ground Plane: Analysis of the Results Obtained from Planar Crossed-Dipole Design Configuration with the Infinite Ground Plane Simulations: Finite Ground Plane Simulations...78

9 Simulation Setup for the Planar Cross-Dipole Design with a Finite Ground Plane: Performance of the Planar Crossed-Dipole Antenna with a Finite Ground Plane: Analysis of the Results Obtained from Planar Crossed-Dipole Design Configuration with the finite Ground Plane Simulations: Analysis of D/U Ratio Obtained from Antenna Configurations with a Finite Ground Plane: Conclusion Recommendations References Appendix A: Free-Space Simulation Results Appendix B: Infinite Ground Plane Simulation Results

10 10 LIST OF FIGURES Figure 1.1 GNSS L1 Signal Characteristics Figure 2.1 Impedance Smith Chart for an Antenna Resonating at 1.59 GHz Figure 2.2 VSWR Plot for an Antenna Resonating at GHz Figure 2.3 Comparisons of Theoretical Radiation Pattern and the Typical Radiation Pattern of a GNSS Antenna Figure 2.4 Principle of Choke Ring GNSS Antenna [28] Figure D View of a (L x W) Planar Dipole Antenna Printed on a Dielectric Substrate Figure 3.2. Cavity Backing of a Crossed-Dipole Printed on RO Figure 3.3 Antenna Fed with the Semi Rigid Cable from the Bottom Figure 3.4 Concentric Rings used for Multipath Mitigation Figure 3.5 Crossed-Dipole Antenna Prototype with Concentric Rings Designed Using Rogers RO3010 and Cavity Backing Figure 4.1 Screenshot of the Crossed-Dipole Antenna Layout in ADS 2009U1 Showing Various Commands Available in Momentum Figure 4.2 Defining the Substrate Layers and Layout Layers for the Infinite Ground Plane Simulations in ADS Momentum 2009U Figure 4.3 Inner Edge Configuration of a Planar Dipole Strip Figure 4.4 Geometry of a Single-Dipole Striped on RO3010 in Free-Space Figure 4.5 Simulation Results of a Single Half-Wave Dipole Antenna in Free-Space68 Figure 4.6 Validation of Half-Wave Dipole in Free-Space Figure 4.7 Top View of Cross-Dipole Antenna on the Top Substrate Layer in Free- Space... 71

11 11 Figure D and 3-D Layout views of an Antenna Configuration 012B with the Finite Ground Plane Figure Simulation Results of a Planar Crossed-Dipole GNNS Antenna Prototype with a Finite Ground Plane Figure Normalized Gain Patterns for the Various Planar Crossed-Dipole GNSS L1 Antenna Configurations with the Finite Ground Plane Figure D/U Ratio Plots for various Antenna Configurations with a Finite Ground Plane Figure A.1 Design and Radiation Pattern of a Crossed-Dipole Antenna Prototype with and without Rings in Free-Space Figure A.2 Normalized Gain Elevation Pattern for the Antenna Configurations ( ) in Free-Space Figure B.1 2-D and 3-D Layout views of a Antenna Configuration 012B with the Infinite Ground Plane Figure B.2 S-parameter and 3-D Radiation Plots for the dipole 012B with an Infinite Ground Plane Figure B.3 Normalized Gain Patterns for the Various Planar Crossed-Dipole GNSS L1 Antenna Configurations with an Infinite Ground Plane

12 12 LIST OF TABLES Table 1.1 Description and Operational Status of GNSS around the World Table 1.2 Signal Characteristics of a GNSS (L1, E1, B1) bands Table 2.1 Performance Requirements of a Planar Crossed-Dipole L1 GNSS Antenna Prototype Table 4.1 Properties of Rogers RO3010 Substrate Material Table 4.2 Properties of the Styrofoam, Absorbing Material and Ground Plane Material Table 4.3 Description of the Antenna Configurations Table 4.4 Dimensions of the CAN used in the Design Table 5.1 S-parameter Results of Planar GNSS L1 Prototype Antennas Simulated with Finite Ground Plane Table A.1. S-parameters of Planar GNSS L1 Prototype Antennas Simulated in Free- Space Table B.1 S-parameter Results of a Planar Cross-Dipole GNSS L1 Antenna Prototypes Simulated with Infinite Ground Plane

13 13 LIST OF ACRONYMS ADS AFSC AR ARP BW C/A CBOC CDMA CP D D/U db dbi dbic DGPS DOD FDMA GLONASS GNSS GPS HZA I IEEE ILMA Advanced Design System Air Force Space Command Axial Ratio Antenna Reference Point Bandwidth Coarse Acquisition Composite Binary Offset Carrier Modulation Code Division Multiple Access Circular Polarization Desired Signal Desired to Undesired ratio Decibels db measured over an Isotopic Radiation Source db measured over CP Isotropic Radiation Source Differential GPS Department of Defense Frequency Division Multiple Access Global Navigation Satellite System (Russian) Global Navigation Satellite System Global Positioning System High Zenith Antenna In-phase Institute of Electrical and Electronics Engineers Integrated Multipath Limiting Antenna

14 14 ION L1 L2 L5 LHCP Mcps MCS MEO MLA NAVSTAR NGA PCB PRN PSD QPSK RF RHCP RL S-parameters SV U U.S. USAF USCG VSWR Institute of Navigation GPS Link 1 frequency ( MHz) GPS Link 2 frequency ( MHz) GPS Link 5 Frequency ( MHz) Left Hand Circular Polarization Million chips per second Master Control Station Medium Earth Orbit Multipath Limiting Antenna Navigation System with Timing and Ranging National Geospatial-Intelligence Agency Printed Circuit Board Pseudo-Random Noise Power Spectrum Density Quadrature phase Shift Keying Radio Frequency Right-Handed Circular Polarization Return Loss Scattering Parameters Space Vehicle Undesired Signal United States US Air Force United States Coast Guard Voltage Standing Wave Ratio

15 15 1. INTRODUCTION The purpose of this research was to design and investigate the performance of a planar cross-dipole Global Navigation Satellite System (GNSS) receiver antenna to operate in GNSS L1 frequency band with multipath mitigation performance. This chapter provides a brief overview of various topics that will be discussed throughout the thesis and also lists the challenges and scope of this thesis Background: Owned by the United States (U.S.) Government, developed by the Department of Defense (DoD) and managed by the U.S. Air Force Space Command (AFSC), the Global Positioning System (GPS) is a popular, robust, and sophisticated satellite/space-based radio navigation system that works on the principle of trilateration to provide an all-weather worldwide position, navigation, and timing service [1] [2]. GPS satellites use Code Division Multiple Access (CDMA) technology for their continuous navigation signal broadcasts. Currently, GPS can operate at three center frequencies: Radio Frequency (RF) Link1 or L1 ( MHz), RF Link2 or L2 ( MHz), and RF Link5 or L5 ( MHz) [3] [4]. GPS satellites or Space Vehicle s (SV s) located in the Medium Earth Orbit (MEO) broadcast continuous Coarse Acquisition (C/A) code and Precision (Y) (P(Y)) code at L1 frequency to serve both military and civilians and the P(Y) code on L2 frequency, which is intended only for authorized use [3]. Block IIR-M, IIF, and later additions of SV s are broadcasting an additional Pseudorandom Noise (PRN) codes for the civilian use on L2 frequency which is designated as L2 Civil (L2 C) [3]. GNSS

16 16 use Right Hand Circular Polarization (RHCP) to help minimize the signal fluctuations due to the orientation mismatches between the transmitter and the receiver antennas. Other nations have either developed or are developing their own GNSS; the Russian GLONASS GNSS is presently operational. European Galileo and Chinese COMPASS or Beidou-2 (BD2) is currently under development. Few regional navigation systems like France s DORIS, Japan s Quasi-Zenith Satellite System (QZSS) and India s Indian Regional Radio Navigation Satellite System (IRNSS) are under the process of development to provide Position, Navigation, and Timing (PNT) services for a particular region GPS Fundamentals: Operated by the United States, GPS is the best-known satellite-based radio navigation and positioning system among the GNSS, which provides accurate and instantaneous Position, Velocity and Timing (PVT) services for an unlimited number of civilian and military users in all weather conditions. A GPS receiver needs at least four visible satellites to calculate the three-dimensional position and time solution [2]. Any GNSS including GPS system is mainly made up of three major segments: Space Segment (SS) or Satellite Segment. Control Segment (CS) or Operational Control Segment (OC). User Segment (US) Space Segment (SS) GPS consists of about 24 to 32 satellites at the MEO (20,200 Km above Earth s surface), where each SV is identified with a PRN code. Located in six orbital planes inclined at 55 degrees (four to six SV s in each plane), each SV circles the Earth

17 17 approximately every 12 hours and transmits continuous navigation signals in the L- band [5] [1]. SV transmission consists of navigation data (SV orbital and clock information, satellite health, ionosphere data, almanac data, ephemeris data etc.) and ranging data with which a military/civilian user can determine PVT when equipped with an appropriate receiver. During the time of this writing, there are 31 satellites in GPS space segment, which are classified as Block IIA, Block IIR, Block IIR-M, and Block IIF. Out of these, the first Block II satellite was launched in Feb 14, 1989 (no longer in service) and the most recent SV, Block IIF-2, was launched on Jul 16, 2011 [6]. Individual satellite history and operational status is given in [7]. Originally, Rockwell International developed the concept validation satellites with the Block I satellites and these Block I satellites were decommissioned in The GPS constellation is monitored by the GPS CS, and timing services are provided by USNO [8]. A legacy GPS satellite (i.e., Block IIA and IIR) will transmit signals on two L- band frequencies: L1 and L2 with PRN ranging codes in use: The coarse/acquisition (C/A) code, P(Y) code is used when anti spoofing (AS) mode is enabled Control Segment (CS) Monitoring the satellite orbits and SV health, maintaining the GPS time, predicting ephemeris and clock data, updating SV navigation messages and commanding small maneuvers are the functions of the CS [9] [10]. The GPS CS consists of: Master Control Station (MCS), located at Schriever AFB in Colorado.

18 18 Six Air Force monitoring stations (Hawaii, Kwajalein, Cape Canaveral, Ascension Island, Diego Garcia and Colorado Springs). Four ground antennas (Kwajalein, Cape Canaveral, Ascension Island and Diego Garcia). At the heart of CS is the MCS, which operates and controls the GPS satellite constellation by providing command and control functions [9]. At the MCS, ranging data obtained from the monitoring systems is processed 24/7. Six monitoring stations were setup so that each satellite must be viewed by at least one monitoring station. These six monitoring stations are unable to provide the complete global coverage; about four percent of the satellites were left unmonitored most of the time. So, in 2005, the GPS CS added six National Geospatial-Intelligence Agency (NGA) monitor stations all over the globe (United Kingdom, Argentina, Ecuador, Bahrain and Australia). This 12-station monitoring system enabled the operator to view each satellite from at least two monitoring stations at any given point of time. Furthermore, stations will be added by NGA so that each satellite is viewed from at least four monitoring stations at any given point of time [10] [9]. GPS and GLONASS have similarly functioning CS [11]. The four ground antennas, which transmit, control and command signals in S- band ( MHz), act as an interface between the MCS and the SVs. Once the MCS processed and update the data, at least one or two times a day data will be sent to the satellites via these ground antennas [7] [9] User Segment (US) The User Segment consists of an appropriate GNSS receiver with well-tuned antennas and signal processors to receive and track the satellite signals. This GNSS

19 19 receiver equipment must be capable of processing the signals from a minimum of four SV s simultaneously to obtain an accurate position (latitude, longitude and height), velocity, and time solution. Two major services provided by the GPS are: The Standard Positioning Service (SPS), which is a continuous worldwide positioning and timing service, which is available to the users at free of cost. SPS is available on GPS L1 frequency band. SPS consists of a Coarse-Acquisition or C/A code and a Navigation data or NAV data message. With this information SPS can provide 95% accurate predictable PVT solution. [9] [12] [3].Additionally, SPS includes the use of the L2 carrier signal for some users. The Precise Positioning System (PPS) is another continuous worldwide PVT Service offered by the GPS to authorized users, where accuracy and security is a primary concern. PPS is available only to the users authorized by the U.S. This service is unavailable to the commercial user by cryptography. PPS can be obtained from both L1 and L2 frequency bands with the P(Y) code. PPS can provide up to 95% or more predictable accuracy [9] [1] GLONASS Fundamentals: GLONASS is a Russian version of a GNSS that provides the three-dimensional PVT services across the globe in all weather conditions, similar to GPS made up of a; satellite Space Segment, monitoring Control Segment, and User Segment [11]. The Space Segment has eight, equally spaced satellites (a total of 24) in three orbital planes. Each satellite takes 11 hours, 15 minutes and 44 sec to complete an orbit with 64.8ᴼ inclination to each other [11]. GLONASS transmission frequency is

20 determined by its channel number, K and its frequency bands are calculated using the following expression. 20 ( ) (1.1) ( ) (1.2) Where, The Control Segment consists of a satellite tracking and command stations network located across Russia to monitor the satellites and provide the corrections to the orbital parameters and navigation data as needed [11]. The User Segment is equipped with receiver equipment to receive and process the GLONASS signals transmitted by the satellites to determine the user PVT solution [11]. GLONASS historically has used FDMA in similar frequency bands as GPS (i.e., L1, L2, L5), and is now developing a CDMA-based GLONASS GNSS Galileo Fundamentals: Galileo is the European GNSS, which is planned to provide highly accurate global positioning services to various users. Once deployed completely, the Galileo Space segment will consist of 27 operational satellites and three operational spare satellites (total 30) in three circular MEO orbit planes inclined at 56ᴼ and 29, km semi major axis. The Control Segment and User Segment are similar to GPS and GLONASS as discussed earlier [13].

21 Signal Characteristics: All modernized GNSS (expect GLONASS) satellites use CDMA technology for their continuous navigation signal transmissions. CDMA allows navigation data to be modulated by a PRN code, which allows the SV s to transmit at the same center frequency without interfering with other GNSS SV signals. The frequency allocations for the GPS, GLONASS, and GALILEO GNSS signals are shown in Figure 1.1 [10]; in which Figure 1.1 (a) illustrates the frequency band allocations of the carrier frequencies and bandwidths of three GNSS s discussed so far and Figure 1.1 (b) illustrates the normalized Power Spectral Densities (PSD s) in L1 band. Figure 1.1 (a) GNSS Frequency Allocation in Radio Navigation Bands [10]

22 22 Figure 1.1 (b) Normalized GNSS PSD s in L1 Band [14] Figure 1.1 GNSS L1 Signal Characteristics For a GNSS antenna with sufficient bandwidth it will receive multiple GNSS signals for the GNSS receiver. From this thesis point of view, when a GNSS antenna is designed at 1600 MHz operating frequency with 103 MHz minimum bandwidth, it will receive signals from the GPS L1, Galileo E1, GLONASS L1, and COMPASS B1 frequency bands. Table 1.1 summarizes the GPS, GLONASS, Galileo and COMPASS GNSS that are operational and under development around the world. Within the scope of this research, the signal characteristics of the various GNSS are summarized in the Table 1.2 [3] [11] [15] [13].

23 23 Table 1.1 Description and Operational Status of GNSS around the World GNSS (Nation) GPS (U.S) GLONASS (Russian) Galileo (European Union) BEIDOU/ COMPASS (China) Coding CDMA FDMA/ CDMA CDMA CDMA Orbital Height in Km & Time in Hr 20,200 & ~ ,100 & ~ ,222 & ~ ,150 & ~12.6 Number of satellites Operational Frequencies in MHz 24 L1 ( ) L2 ( ) L5 ( ) 24 (FDMA) 30 (CDMA) 30 (3 Spares) 5 (GEO) & 32 (Non- GEO) L1 ( ) L2 ( ) E2-L1-E1 ( ) E6 ( ) E5a and E5b ( and ) B1( ) B1-2 ( ) B3 ( ) Operational Yes Yes (FDMA), CDMA is under development 2 GIOV launched and System is under development 10 satellites launched; 35 satellites are planned

24 24 Table 1.2 Signal Characteristics of a GNSS (L1, E1, B1) bands GNSS Freq. GPS, L1 GLONASS, GALILEO, E1 BEIDOU, B1 Band L1 Signal CDMA FDMA CDMA CDMA transmission type Minimum dbw -161 dbw -157 dbw -163 dbw Signal Power Modulation BPSK BPSK BOC( 1,1) QPSK BOC(6,1) Polarization RHCP RHCP RHCP RHCP Center Frequency (MHz) Chip Rate (Mcps) (C/A) (P(Y)) C/A 5.11 P(Y) BOC(1,1) BOC (6,1) Code Length (Chips) 1,023 (C/A) e12 (P(Y)) 511 (C/A) 5.11e6 (P(Y)) 4092 BOC(1,1) BOC(6,1) S-parameter Analysis: At higher frequencies the electrical characteristic of an RF device can change drastically vs. frequency. So, Scattering parameters (S-parameters) can be used to characterize the performance of a 2-port network or 1-port device at higher frequencies. For a 1- port device (e.g., an antenna), S11 represents the ratio of reflected power at port 1 to the incident power at port 1 [16]. S11 will be used in this thesis to characterize the impedance match of the antenna.

25 Scope: The goal of this thesis is to design and simulate a crossed-dipole planar antenna for a GNSS receiver to operate in GNSS L1 frequency band and mitigate multipath. Simulations of multipath mitigation techniques and tuning of the planar dipole antennas were done in Agilent s Advanced Design System (ADS) Momentum update (ADS 2009U1) on a Rogers Corp s RO3010 substrate. Once the antenna was designed, then the performances of the various configurations were investigated and the best configuration was finalized from the software simulation results and then the antenna was placed on top of an appropriate CAN with a suitable ground plane. This term CAN is used to define the antenna enclosure because the CAN structure looks a coffee CAN. When an antenna is designed for the L1 frequency band with sufficient bandwidth, it can be used to receive multiple GNSS signals with their carrier frequencies within the bandwidth range. Once designed, an antenna shall receive the signals transmitted by the GNSS SV s in GPS L1 frequency band, GLONASS L1 frequency band, Compass B1 frequency band, and Galileo E1 frequency band. This document is organized as follows: Chapter 1 gives a brief introduction to the various GNSS and scope of this research. Chapter 2 provides an overview of GNSS antenna characteristics. Chapter 3 the printed - dipole antenna design basics, radiation mechanisms, implementation and feeding techniques. Chapter 4 explains ADS momentum software basics and provides the step-by-step design procedure in freespace, with and without concentric rings. Chapter 5 provides the design approach of the GNSS antenna prototype with infinite and finite ground planes and detailed analysis of the various combinations of rings and cavity backed CAN structure for all

26 26 the configurations. Chapter 6 provides the conclusions, followed by recommendations and references, and finally, the results and plots obtained from the free-space analysis and infinite ground plane analysis are presented in Appendix A and B.

27 27 2. GNSS ANTENNA In this chapter, GNSS antenna performance requirements and antenna properties that affect functionality and performance of a GNSS antenna are presented. The antenna is a crucial front-end component of a GNSS receiver system. In general, the function of any receiver antenna is to capture the electromagnetic signals from free-space and convert them into electrical signals in order to be processed by the receiver. At the receiver, the RF signal strength of a signal received from the GNSS satellite is very weak and these GNSS signals can arrive from any direction. The following are some of the important GNSS antenna properties that affect the performance and functionality of the GNSS antenna: Radiation characteristics at the center frequency in the GNSS frequency bands, Antenna polarization, Axial ratio, Standing Wave Ratio Bandwidth, Desired to undesired ratio (D/U ratio) (e.g., gain pattern shape), Antenna phase center, and Multipath mitigation GNSS Antenna Performance Requirements: Keeping GNSS application in mind, various characteristic properties of antenna design are defined briefly. Then additional performance requirements for the various GNSS application are explained.

28 Resonant Frequency ( ): The resonant frequency or operating frequency is a frequency at which capacitive and inductive reactance of an antenna cancel out each other (as shown in Figure 2.1). Usually, a resonant frequency of interest can be achieved by tuning an antenna to a particular frequency. In reality, there can be a shift in the resonant frequency of a microstrip type due to the antenna packaging, ground plane size and input feeds. Real and imaginary parts of the normalized impedance in a complex plane can be represented on a Smith Chart, where the real part varies from 0 to, and the imaginary part of impedance varies from - to. The end points located on the left and right side on horizontal axis of a Smith Chart refers to the short circuit and open circuit respectively. Similarly, the top and bottom most points on the vertical axis represents the inductive and capacitive nature of the circuits, respectively. In order to obtain a good impedance match at the antennas the input impedance should be at the center of Smith Chart. The point at which the antennas impedance intersects the horizontal axis and close to the center of the Smith Chart, indicates the The Smith Chart display is an integral part of Agilent s ADS Momentum software; Figure 2.1 shows a S11 magnitude plot illustrating the resonant frequency on the Smith Chart. This metric is used for fine tuning of the GNSS antenna in the L1 frequency band [17]. Figure 2.1 shows a S11 plot illustrating the on the Smith Chart.

29 29 Figure 2.1 Impedance Smith Chart for an Antenna Resonating at GHz Voltage Standing Wave Ratio (VSWR) and Return Loss (RL): VSWR is a scalar measurement that characterizes the amount of signal reflected from the antenna with respect to the signal incident, at the antenna terminal due to the impedance mismatch. VSWR corresponding to a perfect mismatch is infinity, and a perfect match is 1, but in reality a perfect match is difficult to achieve. The loss obtained due to the mismatch can also be described as a return loss (RL). VSWR is an important measure to characterize the GNSS antenna performance. Due to a very small time delay of any reflections VSWR measure less than 2:1, (corresponds to a RL of -9.5 db) is used for most GNSS applications. A lower VSWR may be suitable for certain high performance GNSS applications. Equations correspond to RL and VSWR are given in equations 2.1 and 2.2 [16]. VSWR plot (i.e., RL vs. Frequency) of an antenna resonating at is shown in Figure 2.2.

30 30 ( ) [( ) ] (2.1) ( ) ( ) (2.2) Where:, Characteristic Impedance, Load Impedance, is zero, infinity, and equal to ; for the short, open and matching circuits respectively. Figure 2.2 VSWR Plot for an Antenna Resonating at 1.59 GHz

31 VSWR Bandwidth (BW): BW of an antenna is defined as, the range of frequencies within which the performance of the antenna, with respect to some characteristic, conforms to a specified standard [18]. Based upon the application type, there are several measurements to define an antenna bandwidth [17], of which VSWR (e.g., < 2:1) and axial ratio (e.g., < 3 db) are two important BW definitions for the GNSS antenna application. The GNSS antenna designed here should have a minimum bandwidth of 103 MHz at VSWR < 2:1 (or RL < -9.5 db) around its center frequency (i.e., 1600 MHz) Antenna Polarization: Polarization of an electromagnetic field is defined as a curve traced by the tip of an instantaneous electric field vector as the wave is propagating away from the observation point [17] [19]. Elliptical polarization is a more common type of polarization with the linear and circular polarization as its extreme cases. The polarization of a GNSS antenna describes how the antenna is sensitive to the polarization of the wave incident upon it. In general, GNSS use RHCP to minimize the effect of polarization fading and Faraday rotation (due to the ionosphere) [20]. Moreover, Circular Polarization (CP) minimizes the signal fluctuations due to the orientation mismatches between the transmitter and the receiver antennas. Therefore, in order to obtain the maximum RF signal reception from SV s, the designed GNSS antenna should be RHCP Antenna Axial Ratio (AR): AR of the polarization ellipse is the ratio of its major to minor axes length [17] [19]. An AR close to one (0 db) indicates good circular polarization where, AR

32 32 greater than one indicates RHCP and less than one indicates LHCP, and an AR close to infinity indicates good linear polarization. Calculations for AR are shown in equation 2.3 [21]. For GNSS applications, the AR is usually specified at the antenna boresight because the AR will increase along with the increase in boresight angle (which decreases with the increase in elevation angle). Low AR is desirable for most of the elevation angles. At the upper hemisphere elevation angles ( ), the boresight AR should be less than 3 to 6 db for a high performance GNSS application [22]. AR should be between 0 db to 1 db for the high-end GNSS application antennas like choke ring and geodetic quality antennas [23]. [ ( ) ( ) ], (2.3) Where: ( ) ( ) ( ) ( )

33 Antenna Pattern: The upper hemisphere GNSS antenna radiation pattern should have sufficient and uniform gain/efficiency to effectively receive the GNSS SV signals at the various azimuth and elevation angles in the upper hemisphere [24]. For most GNSS applications, the antenna should have a uniform radiation pattern over the upper hemisphere with a sharp roll off at the lower elevations in order to reduce the multipath and lower hemisphere interference [24]. GPS receivers typically use a mask angle of at the lower elevation angles to minimize multipath and atmospheric effects [25]. The theoretical normalized radiation pattern of an antenna at the 0 degrees bore sight and a typical real world antennas are shown in Figure 2.3. Figure 2.3 Comparisons of Theoretical Radiation Pattern and the Typical Radiation Pattern of a GNSS Antenna

34 Desired to Undesired Ratio: The multipath rejection performance of an antenna can be investigated with the help of a desired (D) to undesired (U) signal ratio; in general, the multipath due to the ground reflections can be mitigated by passing the D signals at positive elevation angles into a reasonable gain GNSS antenna and by attenuating the U signals at the corresponding negative elevation angles with the help of an antenna radiation pattern. In this research, D/U is calculated by taking the antenna RHCP gain (in db) difference between the positive and negative elevation angles [26] Antenna Phase Reference: The antenna phase response will vary as a function of the azimuth and elevation angle. The antenna phase response can be measured with respect to an Antenna Reference Point (ARP) where the measured antenna phase response can then be used to compensate for these antenna phase variations, with respect to the ARP, depending upon the performance requirement [27] Choke Ring Antenna Concept: A choke ring antenna consists of concentric cylindrical structures around the radiating antenna element to minimize multipath. The primary advantage of these cylindrical structures is to prevent the antenna element from receiving the multipath signals caused due to ground reflections. Choke ring antennas are typically used in places where high precision is a primary requirement. The concept of the choke ring antenna is depicted in Figure 2.4 [28]. The antenna is located at the center and will receive signals from all the visible satellites and the signals from the lower elevations will be reflected from various obstacles (buildings, trees and the Earth s surface) and received with a delay. These reflected (multipath) signals will be diffracted at the

35 edge of the rings placed around the antenna element. Designs of choke-ring antennas are frequency specific [29]. 35 Figure 2.4 Principle of Choke Ring GNSS Antenna [28] 2.2. Challenges: For microstrip type antennas, the design challenges include the selection of an appropriate substrate material; fine tuning of dipole strips and feed locations to resonate at GNSS L1 frequency band to obtain good RL and minimal gain variation over the upper hemisphere radiation pattern; optimizing the multipath performance with the addition of concentric circular rings in various numbers and sizes with the inclusion of cavity backing; reproducing the results with and without CAN (antenna enclosure); and maintaining the alignment of the concentric rings on top and bottom layers of the substrate.

36 Objective: The objective of this research is to design, simulate and investigate the performance of a planar crossed-dipole GNSS antenna prototype, which is expected to operate at GNSS L1, E1, B1 frequency bands with a good multipath mitigation. A printed dipole design is chosen for this prototype because of the following merits: ease of fabrication and mounting, multipath performance, relatively low profile, easy integration with active components. In this thesis, concentric circular ring layouts were designed around the GPS crossed-dipole antenna on a high permittivity substrate along with a cavity backing in order to obtain good multipath mitigation performance at GPS L1 frequency band Performance Requirements: The microstrip type GNSS antenna designed for this research should operate at a center frequency of 1.6 GHz with BW is at least 103 MHz, i.e., the difference between the lower edge of Galileo and the upper edge of the GLONASS L1 along with their additional bandwidths added to accommodate multiple GNSS bands. As discussed earlier the polarization is RHCP. The AR should be between 0 db to 3 db at the boresight for the GNSS application. Gain variation should be less than 15 db from the zenith to 10ᴼ elevation angle. The D/U should be a minimum of 20 db at 45ᴼ elevation angle in order to obtain the good multipath mitigation performance. The performance metrics of this thesis design discussed so far are tabulated in Table 2.1.

37 37 Table 2.1 Performance Requirements of a Planar Crossed-Dipole L1 GNSS Antenna Prototype Parameter Requirement Resonant Frequency or Operational Frequency Bandwidth (VSWR < 2:1) Polarization 1600 MHz GPS L1, GLONASS L1, Galileo E1, COMPASS B1. At least 103 MHz on L1 band Right Handed Circular Polarization (RHCP) Axial Ratio Gain Variation 0 db to 3 db at boresight Less than 15 db from Zenith to 10ᴼ elevation angle Zenith Gain Desired to Undesired Ratio (D/U) Greater than 0 dbic (desired) Minimum 20 db at 45ᴼ Elevation angle

38 38 3. GNSS ANTENNA DESIGN This chapter covers some of the fundamentals of microstrip antennas, printed planer dipole antennas, radiation mechanisms, feeding techniques and brief overview of the design Antenna Selection Criteria: The GNSS antenna selection criterion is not only based on performance characteristics listed in Chapter 2 but also includes the following: Application type (e.g., aircraft, ground station), Frequency of operation (single or multiple), Profile size, Cost and performance, The system requirements (single or multiple elements). Microstrip patch antennas (MPA) are usually selected for the GNSS applications because of their size, mass and cost. In general, a thick substrate is used to increase the bandwidth by increasing the antenna height [30]. But the goal of this research is to design a printed dipole on a dielectric substrate and investigate its performance Advantages and Limitations of the Microstrip Antennas: The principle advantages of microstrip antennas over a conventional microwave antenna are: Thin profile, less weight, small volume. Ease of mass production and low fabrication cost. CP and linear polarizations are easy to obtain with simple feed.

39 39 Easy to make dual-polarization and dual-frequency antennas. Easy to integrate with the external circuit boards. Feeds and matching networks can be fabricated along with the antenna structures. On the other hand, the limitations of microstrip antennas when compared to the conventional microwave antenna structures are: Narrow BW, Lower gain (~6dB), Power handling capability is limited (~100 W), Surface wave excitations, Possibility of reduced gain, cross-polarization and mutual coupling in high frequency array environments, Additional radiations from junctions and feeds are possible, Difficult to achieve polarization purity, Use of thick high substrate results in poor efficiency. However, there are several techniques to overcome most of these limitations [30] Various Configurations: The microstrip type antenna can be designed with various geometries and dimensions to obtain the comparable performance of conventional microwave antennas. Microstrip antennas can be categorized into four groups: microstrip patch antennas, microstrip or printed dipoles, printed slots and microstrip travelling wave antennas. A comprehensive list of various patch antennas geometric shapes,

40 dimensions and configurations can be found in [31]. This document will discuss only the printed dipole antennas used as a basic for this research Overview: Microstrip type antennas can be classified into two categories based on their length to width ratio of the radiating strip, a narrow rectangular strip (typically.05 ) is called a microstrip dipole and one with wider strip (e. g., square shape) is called microstrip patch [30]. Apart from the similar radiation patterns due to longitudinal current distributions, the geometric shape, length to width ratio, radiation resistance, cross-polarization and beam-width of a printed dipole antenna are different from the regular rectangular patch antenna [31]. When compared to a rectangular patch antenna, a microstrip dipole antenna can occupy less area and produce more bandwidth. Microstrip planar dipole antenna design is similar to a patch antenna design where a thin dielectric substrate with a relative permittivity of is used with very thin conductive strip on top and/or bottom, typically copper [17]. A dipole with small length to width ratio on a high permeability dielectric substrate is considered for the proposed design, where the dipole length dimensions are fraction (N) of wavelength (λ). A 3-D view of a single-dipole antenna element of length L = Nλ (with N 0.5) and width W (W < =.05λ) is etched on one side of the substrate of height t, in freespace is shown in the Figure 3.1. A detailed discussion on the dipole antennas will be discussed later in this document.

41 41 Copper Strip W Substrate L = Nλ Z ε r t in mm X Y Figure D View of a (L x W) Planar Dipole Antenna Printed on a Dielectric Substrate 3.5. Multipath Mitigation and CAN: Multipath is the dominant error source in Differential Global Positioning System (DGPS) application. This research is focused on the mitigation of multipath signal by attenuating the multipath signals that enter the receiver by limiting the antenna gain in the lower hemisphere region [25]. Most GNSS antennas used for ground reference DGPS or geodetic applications are designed to have some multipath mitigation capability. One antenna, the High Zenith Antenna (HZA) used within the Integrated Multipath Limiting Antenna (IMLA) utilized a crossed-dipole design [24]. A similar idea is used in this thesis as follows, the positive conductive strips of a dipole are printed on top of the substrate and the negative conductive strips of the dipole are printed on bottom of the substrate. A microstrip dipole designed on a substrate, operated in free-space will produce radiation on the top and bottom side of the antenna. Considering, GNSS application a CAN with micro-wave absorbing material is designed with a depth of λ/4, where the λ here is within the CAN material. The idea behind this microwave absorbing material

42 42 in the cavity is to absorb the unwanted radiation at the bottom side of the antenna. Any remaining signal reflected from the bottom of the CAN will add in phase to the original radiation on top side of the antenna. Figure 3.2 shows the crossed-dipole on a RO3010 substrate with positive legs of the dipole on top of the substrate and the negative legs of the dipole on the bottom of the substrate, with a cavity backing structure. This cavity backing consists of a thin RGDS-124 microwave material placed inside the conductive CAN. The space between the substrate and the absorbing material is filled with a commercially available Styrofoam material. Detailed picture with the dimensions and feed will be shown in the later discussion. Styrofoam is filled in the gap because it has a dielectric constant, which is very close to free-space. Hence, the Styrofoam material is expected to provide stability to the antenna structure without causing any variation to the antenna pattern. In Figure 3.2, and corresponds to the height of the Styrofoam and height of the absorbing material respectively.

43 43 Figure 3.2 Cavity Backing of a Crossed-Dipole Printed on RO Feeding Technique and Feed Locations: There are many techniques to feed the microstrip dipole antenna, where most commonly used ones are: aperture coupled feed, microstrip transmission line feed and coaxial or probe feed [32]. In co-axial cable or probe feed, a coaxial cable is passed through the substrate in such a way that one end will make contact with the conductive radiating element, and the other end will be attached to a conductive ground. Typically, the feed location on the microstrip dipole is selected in such a way that the antenna impedance is nearly equal to 50 In a dipole design, the resonant frequency can be tuned by varying the dimensions of the dipole strips [30]. However, during the simulation process, it was noticed that the closer the feed ports (i.e., small feed gap), the better the performance. In reality, the closer coaxial feed structures are hard to design for cross-dipole antenna strips

44 44 because this cross-dipole design requires two coaxial cables to be fed from the bottom side of the antenna, which in turn requires a minimum separation between them such that the returns can be grounded. So, keeping this in mind, feed locations were specified at (-4, 0) mm, (4, 0) mm, (0, 4) mm, (0, -4) mm on the X and Y axis, respectively; the origin (0, 0) mm was selected at the middle of the antenna. With these feed locations, two 1 mm center conductor semi-rigid cables can be used as a coaxial probe feed for the proposed cross-dipole antenna. The feed to a single-dipole is shown in Figure 3.3 that illustrates a semi - rigid cable where the center conductor is feeding the positive conductive strip on top and the return from the negative conductive strip is attached to the outer metallic coating which is attached to the ground plane. Styrofoam and the microwave absorbing material are not shown in the Figure 3.3 for the clarity. Figure 3.3 Antenna Fed with the Semi Rigid Cable from the Bottom

45 Design Considerations for a Printed Dipole Antenna This section covers the design considerations of a printed dipole antenna, used in this research Resonant Frequency ( ) and Dipole Dimensions (L and W) The resonant frequency of a dipole antenna depends upon the electrical length (L) of the dipole arms and the relative permittivity ( ) of a dielectric substrate on which the dipole is designed. Calculation for the design value of L is given in equation 3.1 [33] [30] indicates that the is inversely proportional to the L of the dipole. Typically, the physical length of an antenna is smaller than the theoretical value due to the substrate effects [34]. However, the theoretical formula can be used for the calculation and provides a good approximation of the physical length. The final value of the dipole length can be determined by fine-tuning the antenna length with the help of simulation tools like ADS Momentum. The width (W) of a dipole strip will have a very minimal effect on and the radiation pattern. ( ) Where: ( ) ( ) is the free-space wavelength (meters) ( ) ( ) ( )

46 Dipole Bandwidth: Typically, the bandwidth of a thin cylindrical dipole antenna can be increased by keeping its length constant and increasing the diameter [17]. A similar concept is implemented here in the design to obtain a near optimum BW. In a printed dipole design BW depends upon the distance between the dipole strips (i.e., one on top and bottom). Bandwidth of a dipole can also be improved by stacking the dipoles (i.e., one on top and bottom), whereas this increment depends on the dipole separation, which turns out to be a substrate thickness in the design [35]. In this GNSS antenna design, a dipole is printed on a relatively thin (RO3010 ) with the positive leg on top of the substrate and the negative leg on the bottom of the substrate to help enhance the bandwidth Radiation Pattern: The far-field radiation pattern of a dipole antenna can be estimated by the calculating of the electric charge distribution on the dipole strips and the Green s function. The radiation pattern of an antenna is usually defined as a far-field directivity (or Gain) of an antenna plotted in units of db relative to an isotropic radiator at various points in terms of spherical coordinates ( ) for the propagation of interest. As discussed earlier the radiation will be in the upper hemisphere of the antenna with the CAN. The amount of the back lobes depends upon the dimension on the ground plane and CAN structure. Radiation on the bottom side of the antenna will be very low when an antenna is placed over a very larger conductive ground plane. In reality, if the antenna patterns have a minimum radiation levels on the backside of the radiation pattern, then it is considered as a good multipath mitigating antenna [26].

47 Polarization of Planar Dipole Antenna: As discussed earlier, the antenna designed for the GNNS application should be RHCP. For simplicity, in this single-layered cross-dipole antenna design, the positive dipole strips were fed by two equal amplitude (±1 volt) ports with a 90ᴼ phase difference to achieve the circular polarization. To physically implement this a 90ᴼ hybrid combiner would be used. For this design simulation the ADS Momentum software has a flexibility define phase and amplitude at each port individually Substrate Selection Criteria: Selection of the substrate material is a consideration in the printed antenna design as it can influence the antenna performance in terms of the gain/directivity pattern, efficiency, BW and resonant frequency. When a dipole is designed on a thick substrate, it can increase the BW up to a certain limit based on the placement of dipoles. However, a thick substrate will increase the resonant length. [30]. Due to the material availability the Rogers RO3010 which is considered relatively thin substrate (t = 1.27 mm) with a relatively high ( = 11.2) was used for this prototype antenna design Design of Concentric Rings: In order to help mitigate multipath, a concept similar to the choke ring GNSS antenna design (but not exactly the same) is implemented on a Printed Circuit Board (PCB) on which the dipole is printed. A multipath signal coming towards the antenna can be diffracted by adding at least one ring or the combination of two or more rings. In this GNSS antenna design, multiple rings ranging from one to 3 were placed around the crossed-dipole antenna element as shown in Figure 3.4.

48 48 Antenna s multipath mitigation performance is investigated in this prototype design with the help of concentric rings etched around the designed dipole where the smallest ring is at least λ/4 away from the outer edges of the dipole arms. In the end, 6, 7, and 8 inch diameter rings were placed on the top and bottom layers of the substrate surrounding the dipoles. Initially each ring was individually simulated, followed by combinations of two rings at a time and finally all three rings were simulated at once. Figure 3.4 Concentric Rings used for Multipath Mitigation Antenna Design Summary: A planar GNSS planar crossed-dipole antenna is printed along with the concentric rings is simulated on a Roger s RO3010 substrate. This substrate is placed on a commercially available aluminum CAN of quarter wavelength height. The quarter wavelength CAN is filled with a thin Cuming s C-RAM RGDS-124 microwave absorbing material in order to absorb the radiation on the bottom side of the antenna

49 49 substrate. The gap between the substrate material and the microwave absorbing material is filled with a simple commercially available Styrofoam. All simulations were performed in Agilent s ADS 2009U1 simulation software. Complete design overview is shown in Figure 3.5 below. Detailed explanation on simulation analysis and results will be discussed in the later chapters. Figure 3.5 Crossed-Dipole Antenna Prototype with Concentric Rings Designed Using Rogers RO3010 and Cavity Backing Note: grounded return is connected to the absorbing material (for clarity previously illustrated semi rigid cable type feed is not shown here)

50 4. SIMULATION SETUP AND PERFORMANCE ANALYSIS OF PLANAR CROSSED-DIPOLE ANTENNA CONFIGURATIONS IN FREE-SPACE ADS Momentum All of the simulations and measurements of a prototype antenna design were done in the Agilent ADS Momentum 2009U1, 3-dimentional planar Electromagnetic simulation software. This chapter provides an overview of ADS Momentum 2009U1 software followed by definition of substrate parameters, microwave absorbing material properties, the simulation setup and measurements of a planar cross-dipole in the free-space and the infinite ground plane Overview: Momentum is an integrated part of Advanced Design System 2009 Update1 (ADS 2009u1) electromagnetic (EM) simulation software that computes the S- parameters and radiation patterns of various planar circuits using the method of moments (MOM). In Momentum, the design can be done in the schematic (circuitbased designing) or layout (two-dimensional dimensional physical designing), where schematic and layout are the integrated part of the Momentum software [36]. The designed prototype antenna was simulated in the layout window as it is easier to create the custom shapes of the radiating elements for the physical design. Momentum can mainly be operated in two different modes: RF mode is used for geometrically complex, electrically small, and do not radiate designs and microwave Momentum mode for the full-wave EM simulations. In this thesis, Momentum mode is used for the planar cross-dipole antenna design.

51 ADS Layout: In ADS momentum designs can be done in Schematic (which is a circuit based design) or Layout (two-dimensional physical design). In this thesis antenna design is done in Layout mode because in layout window, we have the flexibility to fine tune the physical antenna s physical dimensions. A screenshot of the layout window and the various command functions available with the Momentum are shown in Figure 4.1. In the layout pane, one can define the substrate parameters, port or feed types, mesh parameters and box or waveguide to create a physical boundary for the antenna radiation. Also, the design can be analyzed with the S-parameter simulation and 3-D visualization. Figure 4.1 Screenshot of the Crossed-Dipole Antenna Layout in ADS 2009U1 Showing Various Commands Available in Momentum

52 Simulation Setup in ADS Momentum: The following are the steps for simulating a design in the ADS Momentum. Define the substrate parameters. Pre-compute the substrate for the desired frequency range. Define Layout Layers in the layout window. Draw the layout using the component palate. Define the port location and port types (internal mode, differential mode, common mode and ground reference). Set up and generate a customized mesh for all the layers in the design, where mesh is used to break down the design into the user-defined number of cells per wavelength for the purpose of effective analysis. The cell pattern shape (triangular or rectangular) is auto-generated by the ADS. Set up the simulation for desired frequency range and choose the sweep type (adaptive, linear, logarithmic and single) to compute the S-parameters. As a part of post processing, visualize the radiation characteristics in 2-D and 3-D. In microstrip type dipole antenna simulation setup, initially as a part of substrate definition, the substrate properties are entered. Once the substrate is defined, it is verified to operate in the 1 to 2 GHz frequency range, this is done by using mesh analysis with the help of the pre-processing option available in the ADS Momentum menu. The layout layers are defined for the Rogers RO3010, Styrofoam, Cuming microwave s RGDS-124, and the aluminum CAN as shown in Figure 4.2.

53 53 Figure 4.2 Defining the Substrate Layers and Layout Layers for the Infinite Ground Plane Simulations in ADS Momentum 2009U1

54 54 Once the valid substrate and the layout layers are defined for the specified frequency range, then the dipole strips were drawn and tuned with the appropriate ports/feed in the layout workspace. At this point, port properties were defined, positive legs were fed with an internal port and the negative port was fed with a ground reference with respect to the corresponding positive dipole strip. A differential port can also be defined here but while defining the differential, it is mandatory that both positive and return ports be on a same plane (along X-axis or Y-axis). Once the ports are defined and set up, then a customized mesh at 2 GHz frequency is generated (in general, the highest frequency of the simulation is chosen as the mesh frequency) [37]. In order to set up the S-parameter simulation, an adaptive sweep was performed over the frequency range 1-2 GHz. The simulation time depends on the number and shape of the mesh cells Substrate Material Properties: The Rogers RO3010 substrate material was chosen for this prototype design which is relatively thin and has a high relative permittivity dielectric material sandwiched between the two sheets of a 1 Oz., electron deposited copper. The properties of the RO3010 material as defined in the Rogers microwave impedance calculator MWI2010.exe [38] are shown in Table 4.1.

55 55 Table 4.1 Properties of Rogers RO3010 Substrate Material Substrate Relative Permittivity, 11.2 Loss tangent (tan ) Thickness Clad Type 1.27mm Copper on both sides Copper Clad Thickness 1 Oz = 33um Conductivity 5.813E+007 Siemens/m 4.3. Properties of the Absorbing Material and the Finite Ground Plane: In infinite and finite ground plane cases, the combination of the Cuming Microwave s C-RAM RGDS-124 microwave absorbing material [39] and Styrofoam is added between the ground plane and the dipole elements in order to absorb the unwanted back lobe radiation to a certain extent. Properties of the Styrofoam and Cuming microwave s C-RAM RGDS-124 and the aluminum CAN are shown in Table 4.2.

56 56 Table 4.2 Properties of the Styrofoam, Absorbing Material and Ground Plane Material Styrofoam Relative permittivity 1.03 (approx.) Loss tangent (tan ) Thickness 0 (approx.) 17.4 mm Cuming Microwave s C-RAM RGDS 124 Permittivity, Real 21.5 Imaginary -1 Permeability, Real 5.3 Imaginary -2.1 Thickness 12.7 mm (two sheets) Ground Plane Conductivity 3.5E+007 Siemens/m Thickness 2mm 4.4. Design Configurations: The complete design approach can be classified into four steps, as described below: In step one, on a simulated RO3010 substrate, a single-dipole antenna is tuned to operate at 1.6 GHz frequency in free-space and then the effect of adding the individual and various combinations of concentric rings around the tuned dipole is measured and analyzed. In step two, a crossed-dipole antenna on the RO3010 was simulated and tuned over an infinite ground plane generated by the ADS software with Styrofoam and C-RAM RGDS-124 microwave absorbing material between the antenna elements and the ground plane. Then the effects of adding

57 57 individual and various combinations of concentric rings were measured and analyzed. In step three, the infinite ground plane in step two is replaced by a userdefined finite ground plane. In step four, in order to analyze the performance of an antenna in a closed CAN structures (i.e., where the CAN has sides and a finite round ground plane) all the antenna configurations designed with the infinite and the finite ground plane were placed into 6, 7 or 8 aluminum CAN, respectively. In order to create a quick reference, all the design configurations are listed in Table 4.3. In this Table each antenna configuration is identified with a sequence of 3 digit alpha numeric numbers starting from 001 to 019C where, each configuration number is briefly explained in the design description column and a 2-D layout view is depicted for each configuration in Table 4.3. From Table 4.3: Antenna Configurations (001 to 011) Single-Dipole and Crossed-Dipole Antenna. Single 6, 7 or 8 inch concentric rings on top and bottom. Two rings at a time or all three rings at a time. Dipoles [012 (A, B,C) to 015 (A, B,C) ] Configurations with at least two concentric rings are simulated with a 6, 7, or 8 inch CAN with sides. Dipoles [016 (A, B,C) to 019(A, B,C) ] Configurations with only one concentric ring are simulated with a 6, 7, or 8 inch CAN with sides.

58 58 Table 4.3 Description of the Antenna Configurations Antenna Description of the Design Configuration Number 001 Single-Dipole Layouts Configurations in 2D 002 Crossed-Dipole strips on the top metal sheet of a substrate 003 Crossed-Dipole with positive strips on top and returns on bottom 004 Crossed-Dipole 003 with a 6 ring on top 005 Crossed-Dipole 003 with a 6 ring on top and bottom conductor 006 Crossed-Dipole 003 with a 7 ring on top and bottom conductor 007 Crossed-Dipole 003 with a 8 ring on top and bottom conductor 008 Crossed-Dipole 003 with a 6 and 8 rings on top and bottom conductor 009 Crossed-Dipole 003 with a 6 and 7 rings on top and bottom conductor

59 Crossed-Dipole 003 with a 7 and 8 rings on top and bottom conductor 011 Crossed-Dipole 003 with a 6, 7 and 8 rings on top and bottom conductor A Crossed-Dipole 008 with 6 CAN 12B Crossed-Dipole 008 with 7 CAN 12C Crossed-Dipole 008 with 8 CAN A Crossed-Dipole 009 with 6 CAN 13B Crossed-Dipole 009 with 7 CAN 13C Crossed-Dipole 009 with 8 CAN

60 A Crossed-Dipole 010 with 6 CAN 14B Crossed-Dipole 010 with 7 CAN 14C Crossed-Dipole 010 with 8 CAN A Crossed-Dipole 011 with 6 CAN 15B Crossed-Dipole 011 with 7 CAN 15C Crossed-Dipole 011 with 8 CAN A Crossed-Dipole 005 with 6 CAN

61 61 16B Crossed-Dipole 005 with 7 CAN 16C Crossed-Dipole 005 with 8 CAN A Crossed-Dipole 006 with 6 CAN 17B Crossed-Dipole 006 with 7 CAN 17C Crossed-Dipole 006 with 8 CAN A Crossed-Dipole 007 with 6 CAN 18B Crossed-Dipole 007 with 7 CAN 18C Crossed-Dipole 007 with 8 CAN

62 A Crossed-Dipole 003 with 6 CAN 19B Crossed-Dipole 003 with 7 CAN 19C Crossed-Dipole 003 with 8 CAN Notes: 1. Width/thickness of the rings and the dipole strips are the same in all of the designs. 2. CAN and ring dimensions are given in Table A, B, C corresponds to the 6, 7 or 8 CAN, respectively, in all the designs dipole simulations are valid for free-space, infinite and finite ground planes A-019C dipole simulations are valid for infinite and finite ground plane simulations. Table 4.4 Dimensions of the CAN used in the Design Size of the CAN Approx. thickness of the CAN walls (mm) Radius of the CAN (mm) Thickness on the ground plane Inside Outside (mm) Performance Analysis of Planar Dipole in Free-Space: As discussed earlier, the scope of this research is to design and investigate the multipath performance of a planar cross-dipole GNSS L1 Antenna with the concentric

63 63 planar rings structured around it. The antenna design is initially done in a free-space and then followed by the infinite and finite ground planes. This free-space simulation analysis is limited to the antenna configurations 001 to 011 listed in Table Design and Simulation of a Single Half-Wave Dipole in Free-Space: In order to generate the free-space simulations in ADS Momentum, the substrate stack is defined in such a way that it is a three-layered structure (freespace01, Rogers RO3010, Frespace02) in the substrate layers window and a copper conductor strip is defined on top of a Rogers RO3010 layer in the layout layers window. Once the substrate is defined, a single half-wave dipole is drawn in the layout window with the theoretical length of the dipole calculated from Equation 3.1, with the 1600 MHz,, and width of the dipole defined as 2.42 mm, which is twice the substrate thickness (t= 1.21 mm). Theoretical calculations yield the resonant frequency at 1.6 GHz but during the simulation process in the ADS Momentum software it is noticed that there is a difference of 300 MHz due to substrate effects. The length of the dipole strips is tuned in order to achieve the desired operating frequency and then the width is slightly tuned. In order to eliminate any possible inner edge radiation from the cross-dipole design, the inner edges were tested with various shapes as shown in Figure 4.3. Due to the smaller dimensions of the dipole strips, it is noticed that the inner edge shape has very minimal or no effect on the S-parameters and the radiation pattern in the designs as the outputs obtained from all three configurations were about the same. The rectangular inner edge configuration is considered throughout this thesis to ease the design process and analysis of cross-dipole antenna configurations. Considering a widely available 0.91mm semi-rigid cable center conductor diameter feed and

64 64 keeping the practical implementation possibilities in mind, the feeds of the dipole are separated by 8mm and the inner edges of the dipole are separated by a 6mm gap, as shown in Figure 4.3. Figure 4.3 Inner Edge Configuration of a Planar Dipole Strip The simulation of the substrate parameters and inner edge configuration remains constant for all the free-space dipole antenna configurations. The tuned values of the dipole lengths (from the center of the dipole to its outer edge) and the widths for the respective configurations are listed in Table 4.4. Figure 4.4 illustrates the 2-D view and 3-D isometric views of a single-dipole. Due to the limitation in the ADS Momentum 2009U1 software license, the lateral substrate dimensions are not finite in all of the designs. In reality, if we define a constant substrate size, the results may vary slightly from the obtained ones. However, we can always re-tune the dipole dimensions in order to obtain a desired operating frequency range.

65 65 Figure 4.4(a) Top View of a Single-Dipole Layout Figure 4.4(b) Snapshot of a 3-D Isometric View Figure 4.4 Geometry of a Single-Dipole Striped on RO3010 in Free-Space

66 Analysis of Single Half-Wave Dipole Results: Results obtained from the ADS Momentum simulations were mainly classified into two categories; S-parameters and Radiation Pattern, from which the S-parameter results are used to calculate the RL, resonant frequency and VSWR Bandwidth (at 10 & 20 db RL). The Radiation Pattern results were used to analyze radiation performance of an antenna at the designated center frequency of 1600 MHz. Screenshots of the S-parameter and radiation results obtained from the simulation process are presented in Figure 4.5, from which Figures 4.5 (a) & 4.5 (b) correspond to the S-parameter plots and Figure 4.5 (c) to a 3-D radiation pattern. In Figure 4.5 (a), marker m1 corresponds to the resonant frequency at which the RL is best; m2 and m3 are the markers placed on the plot at 10 db RL from which the VSWR 2:1 BW is calculated as a difference between m3 and m2. Similarly, markers m4 and m5 were placed at 20 db RL to calculate the VSWR 1.22:1 BW. Typically a 3-D doughnut shape radiation pattern is expected from a single-dipole antenna configuration in free-space but there is a sharp roll off near the substrate material because the dipole elements are thin virtually. This 3-D pattern obtained from the single free-space dipole simulation is depicted in Figure 4.5 (c).

67 67 Figure 4.5 (a) S11 RL vs. Frequency for Single Half-Wave Dipole Antenna Figure 4.5(b) S11 on Smith Chart for Single Half-Wave Dipole Antenna

68 68 Figure 4.5 (c) 3-D Radiation Pattern of a Single Half-Wave Dipole Antenna Figure 4.5 Simulation Results of a Single Half-Wave Dipole Antenna in Free-Space 4.6. Validation Single Half-Wave Dipole Antenna Simulation in Free-Space: Radiation characteristics of the half-wave dipole antenna designed on the substrate were validated with the help of an ideal half-wave dipole radiation pattern [17]. 2-D radiation patterns of an ideal half-wave dipole and the dipole printed on the substrate are presented in Figure 4.6 (a) and Figure 4.6 (b). From Figure 4.6 it is noticed that there is a sharp roll-off near the substrate for the half-wave dipole radiation characteristics due to the substrate material.

69 69 Z Figure 4.6 (a) Radiation in X-Y Plane

70 70 Figure 4.6 (b) Radiation in Z-Y Plane Figure 4.6 Validation of Half-Wave Dipole in Free-Space Tuned dimensions and results obtained for the half-wave dipole antenna configuration 001 are presented in the first row of Table A.1 in Appendix A, whereas the normalized gain elevation pattern obtained from the radiation pattern for this configuration is plotted in Figure A.2 along with the normalized gain elevation patterns of the various free-space dipole antenna configurations ( ) provided in Table 4.3.

71 Performance Analysis of the Planar Cross-Dipole in Free-Space: In this section, the single-dipole designed in the previous section was used as a reference to design and analyze the crossed-dipole antenna prototype. A crosseddipole shape is being chosen for the design to obtain the good RHCP at the receiver end Design and Simulation of a Crossed-Dipole in Free-Space An additional planar half-wave dipole antenna with similar dimensions is added perpendicular to the Single half-wave dipole antenna designed from the previous section. Now the newly designed crossed-dipole antenna has two perpendicular positive feeds on the top layer and negative feeds (returns) on the bottom layer of the substrate, respectively. Figure 4.7 illustrates the top view of a cross-dipole layout and the second row of Table A.1 shows the dimensions and S-parameter results obtained from the simulations. Figure 4.7 Top View of Cross-Dipole Antenna on the Top Substrate Layer in Free-Space

72 Crossed-Dipole Free-Space Configurations with Rings: As discussed earlier in Section 4.4, in order to make the practical design possible by reducing the number of feeds on the top layer, the cross-dipole antenna is designed in such a way that one conductive radiating element (i.e., half of the half-wave dipole) with positive feed was placed on the top layer and the other conductive radiating element (i.e., other half of the half-wave dipole) with the return or negative feed was placed on the bottom layer of the substrate. Figures A.1 (a) and A.1 (b) in Appendix A, illustrates the 2-D and 3-D layout of this design and Figure A.1 (c) illustrates the 3- D radiation pattern obtained from this design. RHCP is achieved by feeding the ports 1 & 2 with ±1 volt 90ᴼ out of phase and with the ports 3 & 4 connected to their respective grounded returns. Once the crossed-dipole was designed in free-space, the antenna configurations 004 to 011 were implemented by placing the 6-inch, 7-inch & 8-inch concentric conductive rings ranging from one to three on the top and bottom sides of the substrate. On each side, these conductive rings were placed on the top and bottom, and the width of each ring was set equal to the tuned width of the dipole in each design.. Figure A.1 (d) and A.1 (e) in Appendix A, illustrates the 2-D and 3-D views of a free-space dipole design with 3 concentric rings Analysis of Results Obtained From the Free-Space Simulations: The S-parameter results obtained from these free-space antenna configurations are shown in Table A.1 and the normalized gain pattern plots for all of the configurations are depicted in Figure A.2 in Appendix A. From the results shown in Table A.1, it is noted that the length and width of the first three configurations are the same, from which it is evident that the once the

73 73 dipole is tuned in free-space, it can be easily transformed into the cross-dipole antenna configuration without requiring any additional tuning. Moreover, the configuration 003 yields maximum VSWR 2:1 BW of approximately 195 MHz, and similar gain and directivity patterns are noticed when compared to the configuration 002. Considerable changes in dimensions are noticed by adding the larger ring/rings. It is also noticed that the larger the ring, the wider the dipole width for a good impedance match. In Table A.1, when tuning the dipole with an 8-inch ring (dipole #007), the width of the dipole elements are noticed to be 5.0mm and with slightly decreased length. It looks like the 8 inch concentric ring is interacting with the actual dipole design. Gain values in db at the respective theta angles and the 2-D radiation pattern data are extracted from radiation levels obtained by simulating crossed-dipole antenna prototype in the ADS Momentum in free-space. Extracted gain data from each individual configuration are normalized and plotted using the MATLAB at the respective angles of theta (from -180 to 180 ). In order to limit the document length, all the results are consolidated and plotted in the single plot in Figure A.2. All prototype antennas designed in the free-space are operable at approximately 1.6 GHz frequency range with a good RL of approximately 40 db and the VSWR bandwidth of approximately 185 MHz at 2:1 (10 db RL) and approximately 60 MHz at 1.22:1 (20 db RL). Moreover, from the S-parameter Table and normalized elevation patterns plotted in Figure A.2, it is evident that the addition of a single 6 ring or a larger ring with a 6 ring to a crossed-dipole yields better performance in terms of the bandwidth and uniform radiation pattern. At a elevation angle, the normalized gain of the antenna is greater than -7 db for the cross-dipole antenna

74 74 configurations without rings and is greater than -17 db for the crossed-dipole antenna configurations with rings. The main purpose of adding these concentric rings is to improve the antenna s performance by rejecting the multipath signals at the receiver but the results shows that the additions of rings definitely have an impact on the antennas pattern. Effects of adding rings are further investigated with an infinite ground plane and the finite ground plane.

75 75 5. SIMULATION AND MEASUREMENT OF L1 PLANAR CROSSED- DIPOLE ANTENNA ON INFINITE AND FINITE GROUND PLANES AND WITH A CAN In this chapter, simulations discussed at the end of the Chapter 4 will be repeated with a infinite and a finite ground planes. In addition, different combinations of 6-, 7-, & 8-inch concentric rings structures, a single 6-, 7- or 8-inch CAN structure with Styrofoam and microwave absorbing material is added to the bottom side of the antenna. Results obtained from the infinite and finite ground plane simulations with a CAN are presented and analyzed in the end Simulation Setup for the Planar Crossed-Dipole Design with an Infinite, Finite Ground Plane and a CAN and Configuration Descriptions: In order to generate the infinite ground plane simulations in ADS Momentum, FreeSpace02 layer defined in Section is replaced by the cavity backing (Styrofoam and Microwave absorbing material) and an infinite ground plane (defined by the ADS Momentum). Specifications of three additional layers used in the infinite ground plane case are: a mm thick Styrofoam, b mm thick (i.e., height) microwave absorbing material (two sheets of commercially available Cumming microwave s C-RAM RGDS 124 were used) and, c. An infinite ground plane defined by the Momentum simulation software. Properties of the absorbing material and the Styrofoam used in this are given in Table 4.2. The total thickness of the Styrofoam and absorbing material located

76 76 between the substrate and the ground plane is approximately equivalent to a quarterwavelength. Once the substrate is defined, steps followed in the free-space antenna configurations ( ) are repeated and the dipole dimensions and ring s thickness were tuned accordingly in order to achieve the resonant frequency, 1600 MHz. In addition to the configurations , the remaining configurations 012A through 019A shown in the configuration table are classified into two sections, where 12A through 15C are the designs in which a CAN was added to the cross-dipole antenna configurations having a combination of the rings (6 & 8, 6 & 7, 7 & 8 and 6, 7 & 8 ). Configurations 016A through 019C are the designs in which a single CAN was added to the cross-dipole antenna having an individual ring (6 or 7 or 8 ) or no rings. In all the designs, A, B, C corresponds to the 6, 7 or 8 CAN, respectively Performance of the Planar Crossed-Dipole Design Configurations with an Infinite Ground Plane: For this group of test antenna configurations were designed and implemented for a center frequency of 1600 MHz, there were 35 antenna configurations simulated with the infinite ground plane, out of which dipole 012B was randomly selected for this analysis purpose. Figures B.1 (a) and B.1 (b) in Appendix B, illustrate the twodimensional and three-dimensional views of the planar cross-dipole, respectively, and it is designed in such a way that there is one ring inside and one ring outside the CAN (6 and 8 rings with a 7 CAN).

77 77 S-parameter plots obtained from the ADS Momentum simulations for the selected antenna 012B configuration is illustrated in Appendix B Figure s B.2 (a) and B.2 (b), and a snapshot of the 3-D radiation pattern is presented in Figure B.2 (c) Analysis of the Results Obtained from Planar Crossed-Dipole Design Configuration with the Infinite Ground Plane Simulations: All the S-parameter results obtained from Momentum simulations were tabulated and are shown in Table B.1, from which it is noticed that the first three configurations ( ) require no additional tuning to go from the single half-wave dipole to the crossed half-wave dipole (Cross-dipole). But once the rings are added, fine tuning is needed to achieve the desired resonant frequency; also for these results, all the configurations have the VSWR BW of approximately 205 MHz at 2:1 (or 10 db RL) and 60 MHz at 1.22:1 (or 20 db RL). The average of the maximum RL from all the antenna configurations is greater than 40dB, which meets the GNSS antenna requirement. Unlike the free-space case, the elevation pattern plots obtained in the infinite ground plane case are grouped into three different sections and plotted in Figure B.3 in appendix B, where Figure B.3 (a) has the elevation patterns for the antenna configurations , Figure B.3 (b) has the plot for the antenna configurations 012A-015C, and Figure B.3 (c) has plots for the antenna configurations 016A-019C. From these consolidated normalized elevation pattern plots, it is noticed that the addition of rings and various size CAN do have a considerable effect on the antenna performance; but due to the infinitely larger ground plane, it is hard to analyze the multipath mitigation performance because the radiation on the back side of the antenna is completely absorbed by this ground plane. In this infinite ground plane

78 case, the total radiation underneath the antenna is zero. Also from Figure B.3, it is 78 noticed that, at a elevation angle, the normalized gain of an antenna in the upper hemisphere is greater than -15 db for the GNSS antenna configuration s , greater than db for the antenna configurations 012A-015C, and greater than db for the antenna configurations 016A-019C. From these data shown it is noticed that, for the antenna configurations with one or more concentric ring placed on top of a CAN the gain variation from zenith to an elevation angle of 10ᴼ is less than 13 db, which is acceptable for the GNSS application Finite Ground Plane Simulations This section provides the simulation setup and results analysis of the GNSS antenna designed on a finite ground plane. Just like the infinite ground plane dipole simulation case, all S-parameter results are tabulated and the normalized gain patterns over the spherical coordinate angle theta are presented in plots below. In addition to elevation pattern plots, desired to undesired ratio (D/U) plots are also included for the detailed analysis of the newly designed antenna prototypes Simulation Setup for the Planar Cross-Dipole Design with a Finite Ground Plane: Definition of the finite ground plane in the ADS Momentum simulation is similar to the infinite ground plane; except for the fact that the infinite ground plane is replaced by a finite round metal conductor (properties of this ground plane are given in Table 4.2). Firstly, in order to define this finite ground plane, the previously defined infinite ground plane was removed and then an additional free-space layer is added to the bottom side of substrate layer. Then an aluminum ground plane of 2mm

79 79 thick is placed on top of it. Now, a custom shape of the ground plane can be defined in the layout window during the design. In all the antenna configurations, the shape of the ground plane is circular with variable diameter equal to the outer diameter of the specific CAN design. For the CAN configurations (012A-019C), the diameter of the finite ground plane is equal to the diameter of the CAN; For the no CAN antenna configurations ( ), the finite ground plane diameter is equal to the diameter of the larger concentric ring used in the design; and finally, for the no CAN & no ring antenna configuration ( ), it is equal to the tuned length of the dipole from one end to another end Performance of the Planar Crossed-Dipole Antenna with a Finite Ground Plane: In order to analyze the radiation beneath the antenna and to simulate a reliable design configuration, a finite ground plane was defined in place of the infinite ground plane and the performances of 35 configurations were analyzed. Total numbers of antenna configurations designed for the infinite ground plane case were redesigned, simulated, and tuned on a finite ground plane. All S-parameter results obtained as a part of simulation process are shown in Table 5.2 and the normalized gain patterns obtained as a result of radiation from the antenna are plotted in three groups in Figure 5.2.3, similar to the infinite ground plane case. The same antenna configuration (012B) is selected for the infinite ground plane analysis is considered here to show the changes made to the design to convert it into the finite ground plane antenna configuration. Figure illustrates the two- and three-dimensional views of this antenna configuration; Figure illustrates its results, where Figure (a) and Figure (b) shows the S11 RL and Smith Chart

80 80 data plots, followed by a snapshot of the three-dimensional top and side views of the 012B radiation pattern in Figure (c) and Figure (d). These are the results at the tuned resonant frequency, illustrating the VSWR (2:1 & 1.22:1) bandwidth and the normalized gain patterns for this antenna configuration. Detailed list of the results obtained from these 35 configurations are presented in Table (a) 2-D top view of a 012B layout.

81 (b) Snapshot of 3-D isometric view of 012B antenna configuration. Figure D and 3-D layout views of an Antenna Configuration 012B with the Finite Ground Plane (a) S11 RL vs. Frequency for 012B Antenna Configuration with a Finite Ground Plane

82 (b) S11 on Smith Chart for 012B Antenna Configuration with a Finite Ground Plane (c) 3-D top View Radiation Pattern of 012B Antenna Configuration with the Finite Ground Plane.

83 (d) 3-D Isometric View of Radiation Pattern of 012B Antenna Configuration with the Finite Ground Plane. Figure Simulation Results of a Planar Crossed-Dipole GNNS Antenna Prototype with a Finite Ground Plane

84 84 Table 5.1 S-parameter Results of Planar GNSS L1 Prototype Antennas Simulated with Finite Ground Plane Antenna Configuration Index Dipole Strip Dimension (mm) Length, l Width, w Radius of the Ground Plane (mm) S-parameters (S11) Resonant Frequenc y (GHz) RL at Resonance (db) VSWR Bandwidth (MHz) 2:1 BW at 10. db RL 1.22:1 BW at 20. db RL A B C A B C A B C

85 A B C A B C A B C A B C A B C

86 Analysis of the Results Obtained from Planar Crossed-Dipole Design Configuration with the Finite Ground Plane Simulations: From the S-parameter results in Table 5.1, it is noticeable that the average RL for all the configurations is greater than 43 db at 1.6 GHz and the bandwidths at VSWR 2:1 & 1.22:1 (or 10 & 20 db RL) are about 205 & 62 MHz, respectively, these data tabulated in Table 5.2 indicates that these prototype antenna configurations meet the BW requirement established in Table 2.1. In the finite ground plane case, the elevation patterns for the configuration s 001 to 011 are shown in Figure (a), the blue color trace is drawn for the single halfwave dipole antenna configuration (001), the red and orange traces are the crossdipole antenna configurations (002 and 003), and the rest of the traces is for the crossed-dipole antenna configurations with one or more rings. From these data, it is evident that the addition of one or more concentric rings around the dipole suppresses the unwanted radiations on the bottom side of the antenna. From Figure (b), it is noticed that antenna configurations 012A-015C (two or three rings with a CAN) have a similar radiation pattern on the upper hemisphere region of the antenna but the suppression of radiation on the bottom side of an antenna depends on the diameter of rings and the CAN. From Figure (b), it is also noticed that the design configurations with the at least one ring located outside the CAN and the one ring inside the CAN has a better performance in terms of multipath suppression over the other configurations. The radiation patterns on the back side of the antenna for the Design Configurations 012B and 015B have a uniform suppression throughout the backside of the antenna. Finally, from the elevation patterns data shown in (c), it is noticeable that a crossed-dipole design without concentric rings and with a CAN of

87 87 different sizes provides worse multipath mitigation, when compared to the mitigation provided from the designs with two or more concentric rings and a CAN. Moreover, from these results, it is also noticed that a smaller ring (i.e., a 6 ring) with a medium sized CAN (i.e., 7 CAN) provides a good multipath mitigation when compared to a design simulated with larger ring (8 ring) and a larger CAN (i.e., 8 CAN). Figure (a) Elevation Patterns for Antenna Configurations with a Finite Ground Plane.

88 Figure 5.2.3(b) Elevation patterns for Antenna Configurations 012A 015C with a Finite Ground Plane. 88

89 89 Figure 5.2.3(c) Elevation Patterns for Antenna Configurations 016A 019C with a Finite Ground Plane. Figure Normalized Gain Patterns for the Various Planar Crossed-Dipole GNSS L1 Antenna Configurations with a Finite Ground Plane

90 Analysis of D/U Ratio Obtained from Antenna Configurations with a Finite Ground Plane: Apart from the S-parameter and radiation pattern analysis, an additional set of results obtained from the elevation pattern data were considered to investigate the multipath performance of the antenna prototypes. This analysis was done by calculation of the D/U ratio for all the gain patterns at the respective elevation angles. Figure illustrates the D/U ratios for all the antenna configurations with a finite ground plane. As discussed earlier (in section ), the D/U ratios are plotted for the entire elevation angle from 0ᴼ to 90ᴼ elevation angle ( ). Gain patterns THETA values over the 0ᴼ to 180ᴼ are used to calculate the elevation angle. (i.e., = 90ᴼ -THETA), For example, if THETA is equal to 50ᴼ then THETA_E will be 40ᴼ. D/U ratio plots shown in Figure 5.2.4, corresponds to the results obtained by subtracting the normalized RHCP gain values at positive elevation angles from the normalized RHCP gains form the corresponding negative elevation angles. From the D/U plots presented in Figure 5.2.4, it is observed that the configurations with two or three rings have minimum D/U ratios of at least 16 db at the 45ᴼ elevation angle which provides the good multipath mitigation in the GNSS antenna design. From Figure 5.2.4, it is also noticed that there is a sudden shift appeared in the D/U ratios at very lower elevation angles. This is due to the step size in ADS simulation software used to design the antenna prototypes. But in reality, this shift will not be sudden. When results shown in Figure are compared with the results in Figure 4.8 presented in [26], it is noticed that the performance of this antenna is better than the

91 91 commercially available choke ring antennas. However, the results presented here are ideal simulations, and the actual results from a fabricated antenna would likely be worse. From the finite ground plane analysis it is noticed that, the addition of concentric rings around the antenna will exhibit good multipath suppression on the backside of the antenna at a cost of reduced gain at the lower elevation angles. Figure (a) D/U ratio vs. Elevation Angle for the Antenna Configurations with a Finite Ground Plane.

92 92 Figure 5.2.4(b) D/U ratio vs. Elevation Angle for the Antenna Configurations 012A- 015C with a Finite Ground Plane. Note: 012B configuration has D/U ratio of 18.5 db at 45 degrees elevation angle

93 93 Figure 5.2.4(c) D/U ratio vs. Elevation Angle for the Antenna Configurations 016A- 019C with a Finite Ground Plane. Figure D/U Ratio Plots for various Antenna Configurations with a Finite Ground Plane.

94 94 6. CONCLUSION A total of 105 design configurations of the L1 GNSS prototype dipole based antenna were designed and simulated in free-space, with an infinite or finite ground plane with and without concentric rings, and with and without a CAN using the Agilent s ADS Momentum U simulation software. All antenna configurations simulated were finely tuned to have good RL of approximately 40 db at the resonance frequency (approximately 1.6 GHz) and a minimum BW of 205 MHz (finite ground Plane). The advantage of using the multiple concentric rings is clearly shown in Chapters 4 and 5 discussions, to reduce the radiation on the backside of the antenna to mitigate multipath. Bandwidth calculated from the S-parameter results is approximately 190 MHz at 10 db RL (VSWR 2:1) and 60 ± 5. MHz at 20 db RL (VSWR 1.22:1) for all the configurations. The antenna radiation pattern is RHCP with the less than 3 db boresight axial ratio for all the crossed-dipole antenna configurations. The variation of the gain over the pattern is less than 15 db from zenith to 10ᴼ elevation angle which is essential for good GNSS receiver performance. D/U ratio is between 17 db and 20 db at 45ᴼ elevation angle for all configurations which have more than one ring with a CAN and with a finite ground plane. The D/U is 18.5 db for the crossed-dipole antenna designed with a 6 and 8 ring with a 7 CAN (i.e., antenna configuration 012B). The performance of a planar crossed-dipole with no rings and a bigger CAN (019A-019C) has more radiation on the bottom side of the antenna when

95 95 compared to the crossed-dipole antenna configurations with at least one ring or more rings and a CAN (012 to 018). The performance of an antenna with two or more concentric rings, and a CAN is better than the antenna with one ring or no rings, and a CAN. Addition of one or more concentric rings, with a ground plane or CAN to an antenna will exhibit a good multipath suppression on the backside of the antenna, at a cost of reduced radiation of the gain at the lower elevation angles. Finally, the performance of the crossed-dipole antenna with two or more rings in a medium size CAN (012 to 015) are good choice for the practical designs as the performance of the multipath mitigation is better and consistent at the lower elevation angles and on the backside of the antenna. In conclusion, the configuration that performed the best, with respect to multipath mitigation for low elevation GNSS signals is a planar cross-dipole with the 6- & 8-inch rings and a 7-inch CAN where the simulated prototype antenna performance was consistent in both the upper and lower hemisphere of the antenna radiation pattern while meeting the various design requirements for the frequency, BW, polarization, and RL.

96 96 7. RECOMMENDATIONS While the scope of the research was limited to simulation analysis, it is recommended to build the actual GNSS L1 band antenna prototype based upon the simulated results obtained from the ADS Momentum software and test the antenna s performance. Additional research is could be performed to investigate the similar printed antenna configurations (stacked dipoles, triangular dipoles, spiral etc.), and to investigate the antenna performance by using various feed structures and dielectric substrates materials. It is also recommended to investigate the use of a low substrate in future designs to minimize losses in the antenna performance. Additional research would also be needed to expand this single GNSS L1 band antenna configuration to a multiband GNSS antenna (e.g., L1 and L2 or L1 and L5, etc.). Moreover, it is also recommended that further analysis has to be perused using a more robust 3-D electromagnetic simulation software for higher fidelity modeling of the 3D antenna structure.

97 97 REFERENCES [1] E. D. Kaplan and C. J. Hegarty, Understanding GPS: Principles and Applications, 2nd ed., E. D. Kaplan and C. J. Hegarty, Eds., Norwood, MA: ARTECH HOUSE, INC., [2] U.S. Air Force, 15 September [Online]. Available: [Accessed 9 October 2011]. [3] Navstar GPS Joint Program Office (SMC/GP), "Interface Specification IS-GPS- 200 Rev E, Navstar GPS Spsce Segment/Navigation User Interfaces", N. G. J. P. O. (SMC/GP), Ed., El Segundo, CA: Sciennce Applications International Corporation GPSW SE&I, [4] September [Online]. Available: [Accessed 09 February 2012]. [5] USAF Global Positioning System Directorate, 03 Novermber 2010 [Online]. Available: [Accessed 2012 June]. [6] U. S., Navy, "United States Naval Observatory (USNO)," March [Online]. Available: ftp://tycho.usno.navy.mil/pub/gps/gpsb2.txt [Accessed 25 May 2012]. [7] U. S. Navy, "United States Naval Observatory (USNO)," April [Online]. Available: ftp://tycho.usno.navy.mil/pub/gps/gpstd.txt [Accessed 05 April 2012]. [8] U. S., Navy, "United States Naval Observatory (USNO)," April [Online]. Available: [Accessed 25 September 2011]. [9] C. Dave, April [Online]. Available: sonairforceandnga.aspx [Accessed 2011 September 15]. [10] M. Pratap and P. Enge, Global Positioning System: Signals, Measurements, and Performance, 2nd ed., M. Pratap and P. Enge, Eds., Lincoln, Massachusetts: Ganga-Jamuna Press, ISBN , 2006.

98 98 [11] "GLONASS Interface Control Document (ICD)," [Online]. Available: [Accessed 03 July 2011]. [12] U.S. Coast Guard Navigation Center, Apr [Online]. Available: [Accessed 19 September 2011]. [13] European Union and European Space Agency, "European GNSS (Galileo) Open Service," September [Online]. Available: [Accessed 21 August 2011]. [14] M. S. Grewal, A. P. Andrews and C. G. Bartone, Global Navigation Satellite Systems, Inertial Navigation, and Integration, 3rd Edition ed., John Wiley, [15] China Satillite Navigation Office, December [Online]. Available: a36cc1d1363.pdf [Accessed 09 May 2012]. [16] D. M. Pozar, Microwave Engineering, 2nd ed., New York: John Wiley & Sons, INC., [17] C. A. Balanis, Antenna Theory: Analysis and Design, 3rd ed., I. : , Ed., Hoboken, New Jersey: John Wlley & Sons, Inc., [18] IEEE Antennas and Propagation Society, "IEEE Standard Definitions of terms for Antennas," New York, USA, [19] J. D. Kraus, Antennas, second ed., J. D. Kraus, Ed., New Delhi: Tata McGraw- Hill Publishing Company Limited, [20] J. C. Ostergaard, "Preliminary Investigation of Faraday Rotation Effects and Description of Polarization Measurements on AFGL High Latitude Meteor Scatter Test Bed," Hanscom AFB, Massachusetts, [21] W. Kunysz, "Effect of Antenna Performance on the GPS Signal Accuracy," in Proceedings of ION, National Technical Meeting, Denver, Colorado, January, [22] RTCA, Inc., "Minimum Operational Performance Standards for Global Navigation Satellite System Airborne Active Antenna Equipment for the L1 frequency band," December 13, 2006.

99 [23] G. J. K. Moernaut and D. Orban, "An Introduction to Bandwidth, Gain Pattern, Polarization, and All That," GPS World, 1 February [Online]. Available: [Accessed 11 November 2011]. [24] D. B. Thornberg, D. S. Thornberg, M. F. DiBenedetto, M. S. Braasch, v. G. Frank, and C. Bartone, "LAAS Integrated Multipath-Limiting Antenna," Summer [25] M. Braasch, "Optimum Antenna Design for the DGPS Ground Reference Stations," in ION GPS 94, Salt Lake City, UT, [26] I. M. Barton, "Antenna Performance Analysis For The Nationwide Defferential Global Positioning System," Athens, Ohio, [27] V. G. Frank, C. G. Bartone and T. Arthur, "GPS Antenna Phase and Group Delay Corrections," in ION, San Diego, California, January, [28] Javad, "Choke Ring Theory," [Online]. Available: ry.html [Accessed 10 June 2011]. [29] W. Kunysz, "Novatel Inc.," november [Online]. Available: [Accessed 22 August 2011]. [30] R. Garg, P. Bhartia, I. Bahl and A. Ittipiboon, MicroStrip Antenna Design Handbook, Boston: Artech House ISBN , [31] J. R. James and P. S. Hall, Hand Book of Microstrip Antennas, London: Peter Peregrinus, [32] W. L. Stutzman and G. A. Thiele, Antenna Theory and Design, 2nd ed., John Wiley & Sons, [33] R. E. Munson, "Conformal Microstrip Antennas and Microstrip Pahsed arrays," in IEEE Transactions on Antennas and Propagation, [34] M. Kirschning, R. Jansen and N. Koster, "Accurate Model for Open-End Effect of Microstrip Lines," in Electronics Letters, [35] P. B. Katehi and N. G. A. a. I. Y. Hsia, "A Bandwidth Enhancement Method for Microstrip Antennas," in IEEE transactions on Antennas and Propagation, 99

100 [36] Agilent Technologies, [Online]. Available: file:///c:/ads2009u1/doc/mom/momentum_basics.html [Accessed 06 June 2011]. [37] Agilent Technologies, Inc, "Advanced Design System 2009 Update 1 - Momentum," October [Online]. Available: [Accessed 09 April 2011]. [38] Rogers Corporation, 31 October [Online]. Available: [Accessed ]. [39] Cuming Microwave, "Dielectric Data," August [Online]. Available: [Accessed 23 January 2011].

101 101 APPENDIX A: FREE-SPACE SIMULATION RESULTS This section consists of the design and simulation results (tuned dimensions, S- parameters, radiation patterns, elevation pattern plots) of the planar GNSS L1 frequency band crossed-dipole antenna configurations in free-space. Figure A.1 (a) 2-D Top View of an Antenna Configuration 003 in Free-Space

102 102 Figure A.1 (b) 3-D View Cross-Dipole Layout in the Free-Space (003 Antenna Configuration) Figure A.1 (c) 3-D Isometric View of Crossed-Dipole Antenna Radiation Pattern (003 Antenna Configuration)

103 103 Figure A.1 (d) 2-D Layout view of Crossed-Dipole Antenna with 6, 7, & 8 concentric rings Figure A.1 (e) 3-D Layout view of Crossed-Dipole Antenna with 6, 7, & 8 Concentric Rings Figure A.1 Design and Radiation Pattern of a Crossed-Dipole Antenna Prototype with and without Rings in Free-Space

104 Table A.1. S-parameters of Planar GNSS L1 Prototype Antennas Simulated in Antenna Configuration Index Dipole strip dimension (mm) Length, l Width, w Free-Space S-parameters (S11) Resonant Frequency (GHz) RL at Resonance (db) 104 VSWR Bandwidth (MHz) 2:1 (BW at 10. db RL) 1.22:1 (BW at 20. db RL) ~ ~ ~ ~ ~ ~ ~ ~ ~ ~ ~ Notes: (a). Length is defined from the origin to the outer edge of the dipole as shown in Figure 4.4

105 105 Figure A.2 Normalized Gain Elevation Pattern for the Antenna Configurations ( ) in Free-Space

106 106 APPENDIX B: INFINITE GROUND PLANE SIMULATION RESULTS This section consists of the results (tuned dimensions, S-parameters, Radiation patterns, elevation pattern plots) of the planar crossed-dipole GNSS L1 frequency band dipole antenna configurations with cavity backing and infinite ground plane conductor. Figure B.1 (a) 2-D Top View of the 012B Layout with an Infinite Ground Plane

107 107 Figure B.1 (b) 3-D Isometric View of 012B Antenna Configuration with an Infinite Ground Plane Figure B.1 2-D and 3-D layout views of an Antenna configuration 012B with the infinite ground plane Figure B.2 (a) S11 RL vs. Frequency for 012B Antenna Configuration with an Infinite Ground Plane.

108 108 B.2 (b) S11 on Smith Chart for 012B Antenna Configuration with an Infinite Ground Plane Figure B.2 (c) 3-D Isometric View of 012B Antenna Radiation Pattern Figure B.2 S-parameter and 3-D Radiation Plots for the Antenna 012B with an Infinite Ground Plane

109 Table B.1 S-parameter Results of a Planar Cross-Dipole GNSS L1 Antenna Prototypes Simulated with Infinite Ground Plane 109 Antenna Configuration Index Dipole Strip Dimension(mm) Length, l Width, w S-parameters (S11) Resonant Frequency (GHz) RL at Resonance (db) VSWR Bandwidth (MHz) 2:1 BW at 10. db RL 1.22:1 BW at 20. db RL A B C A B C A B

110 110 14C A B C A B C A B C A B C A B C

111 Figure B.3 (a) Elevation patterns for antenna configurations with an Infinite Ground Plane. 111

112 Figure B.3 (b) Elevation patterns for antenna configurations 012A 015C with an Infinite Ground Plane. 112

113 113 Figure B.3 (c) Elevation patterns for antenna configurations 016A 019C with an infinite ground plane. Figure B.3 Normalized Gain Patterns for the Various Planar Crossed-Dipole GNSS L1 Antenna Configurations with an Infinite Ground Plane.

114 !!!!!!!!!!!!!!!!!!!!!!!!!!!!!!!!!!!!!!!! Thesis and Dissertation Services!