Design of a rectangular patch antenna.1.ECM 521 TECHNICAL REPORT 2011 Design of a Rectangular Microstrip Patch Antenna cost. The selection of microstrip antenna technology can fulfill these requirements 3. Abstract— Throughout this technical report, WLAN in the 2.4 GHz band(2.4-2.483 GHz) hasit will briefly describe the design of a made rapid progress and severalIEEE standardsrectangular patch antenna in free space as are available namely 802.11a, b, g and j1.required aided with special software. The Various design techniques using defected groundsoftware used is Computer Simulation structure (DGS) in the patch antenna have beenTechnology 2011(CST). Based on an ordinary suggested in previous publications 2-4. DGS ispatch antenna, the antenna has isolated realized by etching a defect in the ground plane oftriangle gaps and crossed strip-line gaps etched planar circuits and antennas. This defect disturbson the metal patch and ground plane, the shield current distribution in the ground planerespectively.
Demonstrated to have left-handed and modifies a transmission line such as linecharacteristics, the patterned metal patch and capacitance and inductance characteristics 6.finite ground plane form a coupled capacitive- Accordingly, a DGS is able to provide a wideinductive circuit of negative index band-stop characteristic in some frequency bandsmetamaterial. It is shown to have great impact with a reduced number of unit cells. Due to theiron the antenna performance enhancement in excellent pass and rejection frequency bandterms of the bandwidth significantly broadened characteristics 6, DGS circuits are widely usedfrom a few hundred MHz to a few GHz, and in various active and passive microwave andalso in terms of high efficiency, low loss and millimeter-wave devices 7. The purpose of thislow voltage standing wave ratio.Experimental work is to design and enhance a rectangulardata show a reasonably good agreement microstrip patch antenna using the parametersbetween the simulation and measured results. Given.This antenna has strong radiation in thehorizontal direction for some specificalapplications within the entire band.Thesimulation and results that were going to be II.
Design Equations Patch Antenna Reviews
METHODOLOGYdiscussed in this report was based on theparameter value given as such that the A. Review Stagefrequency is 3(GHz) with the material type ofRT/Duroid 5870 while the value of permittivity To begin with, the dimensions of the rectangularis 2.33. Besides that the given value of substrate patch were calculated using formulae as beingthickness(mm) and copper thickness(mm) is shown below.The width is critical in terms of0.508 and 0.035 respectively. Power efficiency, antenna impedance andbandwidth.
It is largely dependent on the operating frequency and the substrate Keywords—Patch antenna, microstrip, CST, dielectricconstant. The equation (1) below waspermittivity, fringing effect, return loss used to work out the width of the patch. Other widthscould have been used but if it is too small then radiator efficiency will suffer and if it istoo I. INTRODUCTION large higher order modes will be excited, resulting in field distortions 1, pg-57. Icrostrip patch antenna antenna used to send onboard parameters of article to the groundMwhile under operating conditions.
Such asWireless local area networks (WLAN), mobileand sets require lightweight, small size and low 1.ECM 521 TECHNICAL REPORT 2011 To determine the extension of length, we use this formula: (3)Figure 1 –Front view of the patch (4)Figure 2 –Structure of the patch (1) The actual length, L of the patch is using the following formula: (5)The effective dielectric constant due to the airdielectric boundary is given by: (2) 2.ECM 521 TECHNICAL REPORT 2011 B. Final Stage as calculated previously was used to calculate the value of conductance as shown below; (7) (6)where (6-a) 0 0 sin 3 (8) Where using the asymptotic methods 5 that it can be shown as following; So,The calculated is then being referred toAPPENDIX: Cosine and Sine Integrals 5. To find the value of, we use equation below; (9)Therefore, solve for equation (6); (10) 3.ECM 521 TECHNICAL REPORT 2011Using the CST2011 software, is thencalculated, and the value obtained is 1.5218. RESULT AND DISCUSSIONThe software used to design and simulate the Figure 4 –the S-parameter diagrammicrostrip patch antenna can be used to calculate Based on the figure above, the S-and plot S11 parameters, VSWR, radiation pattern parameter is not achieved -10dB which is the idealand others. The return loss for this figure is: The figures below were the results on 20 log10 S11 = -2.6designing the microstrip patch antenna with thegiven specifications. S11 = 0.7413 Figure 5 –Farfield pattern of the patchFigure 3 –Top view of the designated patch The patch’s radiation at the fringing fieldScattering parameter ( S- Parameter ) is used to results in a certain far-field radiation patterns.
Themodel N-Port linear electrical networks. For N = figure above shows that the antenna radiates more1, S-Matrix is S11 consists of single element. S11 power in the red area (z-axis) than another parameter also known as Reflection Coefficient. The directivity of this antenna isIt is a complex number that has magnitude and 7.057dBi.The IEEE standard is for an antenna tophase angle.
When magnitude of S11 is expressed have a -10dB S-parameter magnitude. Thein decibels, is known as return loss at the input scattering parameter shown above shows that theand it always in decibels.
It can be expressed by: curve falls at a frequency of 2.98GHz which isRLinput = 20 log10 S11 dB near enough to 3GHz as given in the specification. The nearest to its frequency, the better the performance as it has a sharper curve and a lower S-parameter magnitude in dB. A lower dB indicates the antenna having a greater directivity 4.ECM 521 TECHNICAL REPORT 2011and gain.A DGS unit cell is a defect in the ground Now, the S-Parameter achieved -10dB as theplane of a physical transmission line be it minimum ideal of parameter. By formula of returnmicrostrip, coplanar waveguide or whatever loss shows before, the return loss is calculated as:structure where a reference ground planeconductor exists, which is capable of producing a 20log S11 = -10.98dBtransmission zero in the response of the structure. S11 = 0.282The resonant frequency of this transmission zerodepends on the physical dimensions of the defect.
As we compared the value of the return lossThe main advantage of DGS, that it is easier to before and after done the DGS, the return lossmodel and therefore to use in more complex after is less than before doing the DGS. The first DGS structure is the well- smaller the value of the return loss, the smaller theknown dumbbell shaped DGS, published in 1998 reflected power due to we want the maximum8. The use of a DGS (slot in the ground plane) power transferred. So when there are less loss inallows the apparition of a stop band controlled by the antenna, it will achieve best performance oftuning the dimensions of the slot. The ground radiation and high gain as well.plate which has a strong E-Field is cut away inorder to get the better bandwidth. Figure 9 –Farfield pattern after DGS IV.
CONCLUSIONFigure 7 –E-field resulted after DGS From the analysis as above, it can be concluded that the greater substrate thickness will get a greater gain and directivity. The results demonstrated that the radiation properties of the antenna with DGS is better performance than the antenna without DGS as the S-parameter is measured to below than -10dB. It is also shown that DGS have an effectto the performance of the antenna characteristics.As the dielectric constant of the substrate increase, the antenna bandwidth decrease and therefore decrease the impedance bandwidth.Figure 8 –The S-parameter after DGS 5.ECM 521 TECHNICAL REPORT 2011 V. RECOMMENDATIONS Suppression For Microstrip Patch Antennas”, Microwave and Optical Technology Letters, pp. Microstrip patch antennas radiate primarily 103-105 Vol. 1, January 2007.because of the fringing fields between the patchedge and the ground plane.
For a good 3 Haiwen Liu, Zhengfan Li, Xiaowei Sun, andperformance of antenna, a thick dielectric Junfa, “Harmonic Suppression With Photonicsubstrate having a low dielectric constant is Bandgap and Defected Ground Structure for anecessary since it provides larger bandwidth, Microstrip Patch Antenna”, IEEE Microwave andbetter radiation and better efficiency. However, it Wireless Components Letters, VOL. 2,leads to a larger antenna size. In order to reducethe size of the Microstrip patch antenna, substrates Feb. 2005.with higher dielectric constants must be used 4 Y. Kim, and Y.-S.
Kim,which are less efficient and result in narrowbandwidth. Hence a trade-off must be realized “Harmonics Reduction With Defected Groundbetween the antenna performance and antenna Structure for a Microstrip Patch Antenna”, IEEEdimensions Antennas and Wireless Propagation Letters, VOL.
5 Constantine A. Balanis, “Antenna Theory Analysis And Design”, 3rdedition, page 811-842 VI. ACKNOWLEDGMENT 6 D. Qian, In the name of Allah s.w.t and He most and T.
Itoh, “A design of the low-pass filter using merciful with Salawat and salam to prophet Muhammad s.a.w, Alhamdulillah thanks to Him the novel microstrip defected ground structure,” with the help and His permission giving us the IEEE Trans. Microwave Theory Tech., vol. 49, idea and health, also the opportunity to pp. Successfully completed this project entitled “Design of A Rectangular Patch Antenna Using 7 C.
Lim, CST”for Computer Engineering System Design “A novel 1-D periodic defected ground structure subject within the given time duration. For planar circuits,” IEEE Microwave Guided Wave Lett., vol. 131–133, Apr.2000. Here, we would to express our great thankful wishes to all of those who have been 8 J. Very supporting and helpful to us especially our Antenna and Propagation lecturer, Pn.Nor Qian, D. Itoh,―Modeling of a Hasimah Baba for her guidance, advices,and photonic bandgap and its application for the low- monitoring us to complete this project.
Special pass filter design,‖ Proceedings of Asia Pacific thanks also to all classmates of EE2405A and Microw. 331–334, Singapore other friends who are involved directly or 1999. Indirectly to accomplish this task. The full and kindness of support, attention, time and advises gives a full memories to me. REFERENCES1 Bahl, I. J and Bhartia, P; “MicrostripAntennas”, Artech House, 1980.2 M.
Sanyal, and A.Chakrabarty, “An Improved Design Of Harmonic 6.
A typical circular microstrip antenna.Typical sizes of a microstrip patch range from λ/2 to λ/3, and the dielectric thickness for such antennas is usually in the range of 0.003 λ to 0.05 λ. The relative constant ε r is in the 2.5 to 3.2 range.The main disadvantage of the microstrip antennas lies in their quality factor Q.
Microstrip antennas have high Q ( Q 100 is common), which means that they have small bandwidths. The bandwidths can be increased by increasing the thickness of the dielectric, but as this is done, surface waves (instead of transverse waves) use more and more of the delivered power, which could be considered as a power loss. Furthermore, as the size of the microstrip increases, it can allow resonant frequencies of two or more resonant modes to exist, leading to instabilities.When a current is injected into a microstrip antenna, a charge distribution becomes present at the microstrip surface and ground plane.
For thin microstrips, most of the current resides at the bottom of the microstrip and on the top surface of the ground plane. Therefore, the component of the magnetic field which is tangential to the patch edge is small.The input impedance of the microstrip antenna has both reactive and resistive components.
Its resistive components account for the power radiated by the antennas. The presence of complex poles means that the imaginary parts of these poles account for the power loss by radiation and by dielectric and conduction losses.
The real part and antenna poles are dependent on the shapes of their modal distributions. The microstrip antenna is a relatively modern invention. It was invented to allow convenient integration of an antenna and other driving circuitry of a communication system on a common printed-circuit board or a semiconductor chip (Carver and Mink, 1981; Pozar, 1992). Besides other resulting advantages, the integrated-circuit technology for the antenna fabrication allowed high dimensional accuracy, which was otherwise difficult to achieve in traditional fabrication methods. The geometry of a microstrip antenna consists of a dielectric substrate of certain thickness d, having a complete metalization on one of its surfaces and of a metal “patch” on the other side. The substrate is usually thin ( d ≪ λ).
The metal patch on the front surface can have various shapes, although a rectangular shape, as shown in Figure 6.17, is commonly used. The antenna may be excited using various methods (Pozar, 1992; Pozar and Schawbert, 1995). One common approach is to feed from a microstrip line, connecting the microstrip antenna at the center of one of its edges. The microstrip line may be connected to a feeding circuitry or directly fed by connecting a signal source across the microstrip line and the ground plane. The microstrip antenna produces maximum radiation in the broadside (perpendicular to the substrate) direction and ideally no radiation in the end-fire (along the surface of the substrate) direction. The size of the antenna is usually designed such that the antenna resonates at the operating frequency, producing a real input impedance. For a rectangular microstrip antenna, this requires the length of the antenna, L, to be about half a wavelength in the dielectric medium.
The width of the antenna, W, on the other hand, determines the level of the input impedance. The microstrip antenna can be thought of as a rectangular cavity with open sidewalls. The fringing fields through the open sidewalls are responsible for the radiation. However, the structure is principally a resonant cavity, with only limited fringing radiation.
Therefore, the bandwidth of the radiation is poor compared to the bandwidth of antennas discussed earlier. The small bandwidth, however, is adequate in a large class of communication applications. Readers may refer to Balanis (1997) and Carver and Mink (1981) for some analytical modeling of a microstrip antenna. Simple and approximate expressions for the radiated electric field components of a microstrip antenna are given by Carver and Mink (1981). The authors of reference 142 review recent work in this field with some emphasis on work performed in mainland Europe. A variety of realizations yield f 1/ f 2 ratios ranging from 1.6 to 3; one antenna ensemble consisting of a cross resonant at f 2 and a subarray of four patches, shown in Figure 3.18a, designed to operate at f 1, yields an f 1/ f 2 ratio ranging from 2.9 to 3.2 142, 143. By following the design procedure outlined in the latter reference, the ensemble can be used to populate an aperture capable of operation at S- and X-bands simultaneously.
Instead of having an arrangement consisting of a cross and four patches, one could have two concentric circular patches designed to operate at widely different frequencies. Dual-band antennas. (a) Cross and subarray of square patches.
The low-frequency cross resonates at the low-frequency band, while the patches of the subarray resonate at the high-frequency band. (b) Dipole-based dual-band antenna; the concept can be extended to multiband antennas. (From 142; © 1997, IEEE.), (Adapted from 146; © 1995, IEEE.) Copyright © 1997The authors of reference 144 propose a multilayer aperture-coupled patch antenna that can be used for dual-polarization operation at C- and X-bands. Two design examples operating at these bands (5.3 and 9.6 GHz) are reported and exhibited a polarization isolation in excess of 28 dB and a port-to-port isolation of 25 dB. For SAR applications, high polarization purity and interport isolation are of considerable importance.In reference 145 it is demonstrated that the resonant frequency of a circular patch designed to resonate at the high frequency band centered at f 1, can be substantially lowered if a shorting pin is used; the position and dimensions of the shorting pin are critical and have been determined experimentally.
The f 1/ f 2 ratio can vary between 2.55 and 3.85 when the shorting pin is in place. The approach is experimental and details are given of the antenna realized when the high frequency, f 1, is at L-band. The attractors of this approach are:.A small patch designed to operate at f 1, can operate efficiently at a much lower frequency with the aid of a shorting pin.An aperture capable of operation at two bands can be realized if it is populated by circular patches of the same size; some of the patches utilize a shorting pin and the distribution of the two sets of patches on the aperture is random or aperiodic.While considerable work is required in this area, the approaches already reported are important and promising. 3.6.4.2 Multiband AntennasAn array capable of operation at multiple bands is an attractive proposition.
We have already reported that efficient antennas operating at different frequencies can be realized on the same multilayer arrangement consisting of several layers of alumina ceramic and polyimide 128. Two antennas operating at 10 and 20 GHz have been realized on the same multilayer substrate and the realization of other antennas operating at different frequencies is possible.Without any loss of generality, let us consider a linear array that operates at two frequency bands, proposed in 146 and illustrated in Figure 3.18b. The concept can be extended to planar arrays and to arrays operating at several bands selected on the basis of system requirements. The array consists of two dipoles, each of which resonates at the high-frequency band.
The band-pass filters Z T1, Z T2, and Z T3 can present either an open-circuit or short-circuit to ports 1, 2, and 3 respectively. The high-frequency band is obtained when: Z T2 = ∞ and ports 1 and 3 are matched to the high-frequency dipoles. Similarly, operation at the low frequency band is attained when both dipoles are short-circuited to form one dipole resonating at the low frequency. More specifically, the necessary conditions are that the bandpass filters Z T, = Z T3 = 0, V g1, = V g3 = 0 and Z T2 is matched to that of the effective dipole at the low-frequency band. 8The measurements obtained from an experimental array operating at the UHF, L- and S-bands agree with the analytical work undertaken 146. NICHOLAS FOURIKIS, in, 2000 1.1.1 Typical Phased ArraysAn active radar array consists of m 1 antenna elements that take the form of conventional dipoles, slotted waveguides, horns, or microstrip antennas tightly spaced and each antenna element is connected to a transceiver.
If the number of transceivers is n 1, n 1 = m 1 for an active array; by contrast, m 1 n 1 in a passive array. A circular/elliptical aperture populated with antenna elements or transceivers yields a pencil or fan beam, respectively; the latter beam is wide in the direction of the shortest dimension of the ellipse and narrow in the direction of its longer dimension.A typical or notional radar phased array operating at a wavelength λ has an aperture of 50λ × 50λ and yields a pencil beam of 1° 3. If costs are not constrained, the aperture is densely populated by 10 000 antenna elements, the maximum number of elements spaced λ/2 apart. This interelement spacing insures that the array grating lobes are eliminated—see section 2.2.1.1. If costs are constrained, the aperture can be populated by a smaller number of antenna elements commensurate with the available funds.
Let us assume that the number of antenna elements of the typical radar phased array is between 5000 and 10 000 with the understanding that some arrays will be larger or smaller.In another realization, the array antenna elements take the form of conventional reflectors that are either located on terra firma or are spaceborne. These are the classical radiotelescope realizations where a few tens of parabolic reflectors are scattered over very large areas. A stable and sensitive receiver selected from a suite of receivers usually follows each antenna and this arrangement enables the radiotelescope to receive radiation from celestial bodies of interest at the frequency band usually defined by the receiver.Compact radioastronomy arrays (or radiotelescopes) extend over some tens of kilometers and all antenna elements are connected by low-loss transmission lines such as coaxial cables, oversized waveguides, or optical fibers. The spatial resolution attained by these arrays is of the order of one arc second, comparable to the spatial resolution attained by Earth-bound optical telescopes.The spatial resolution of compact arrays can be increased if the antenna elements are further separated, in which case the connection between antennas is not implemented by low-loss transmission lines. The antennas of these arrays can be dispersed in one or more continents, forming a very long-baseline interferometer, or VLBI network.
The signals received by each antenna of the VLBI network are recorded together with exceptionally stable time signals derived from an atomic clock or hydrogen maser. After the observations are completed, the signals recorded by the network elements are coherently processed in one central processor.If all antennas used are Earth-bound, the spatial resolution attained is of the order one milli-arc second; when satellite-borne antennas are used in conjunction with a VLBI network of antennas, the attained resolution is even finer—see section 1.3.3. NICHOLAS FOURIKIS, in, 2000 4.3.1.4 The Multichip ModuleThe other design philosophy advocates that the module assembly contains the antenna and a small number of chips that perform the many module functions. We have already seen that several microstrip antennas afford the designer the freedom to integrate the antenna to the MMIC-based module.
Before the recent availability of reliable interconnects between chips 50, this approach lowered the module reliability; now multichip modules (MCMs), reliably interconnected, are common. This design philosophy allows designers:.to use a finite set of fully tested, ‘standard’ chips to realize a variety of systems 51, a very attractive proposition; and.to design, realize, and test T/R modules over short intervals of time.Typically, a prototype dual-band multichip receive module was designed, fabricated, and tested within a period of six months at a cost of US$950 51.
This design philosophy is particularly attractive for small research groups residing in universities or research institutes since it allows them to explore important aspects of research in phased arrays without any access to IC design and fabrication facilities. It is commonly acknowledged that diversity in research is a basic ingredient of success.While the average module for radar phased arrays consists of 5–8 chips, an EW module was reported to have 72 MMICs 52.Both design philosophies are valid and allow designers coming from diverse backgrounds to realize tailor-made modules. Several types of distinctive antennas are now in use. By choosing an appropriate shape, size, and configuration of ceramic and conductor, dielectric antennas with acceptable characteristics can be designed for desired working frequencies. Figure 16 shows a microstrip antenna (or patch antenna) for a global positioning system (GPS) receiver, using a high K dielectric substrate.
Dielectric ceramic loaded stripline antennas have several advantages, namely small size, fairly narrow but still wide enough frequency band, and good temperature stability. GPS receivers are required to receive multiple signals from three or four different satellites simultaneously. This antenna has axially symmetric gain characteristics around the vertical axis, so that this antenna is suitable as a multidirectional antenna. Apart from the ferrite films and ferrite bulk materials mentioned above, a composite with large and equal relative permittivity and relative permeability is prepared for miniaturization and impedance match to free space ( Petrov et al., 2008 ). This composite is made of nickel zinc ferrite and BST that is prepared by conventional ceramic processing.
The geometry of the half-wavelength resonant microstrip antenna on the composite substrate is shown in Figure 10.33. The diameter of the composite disc is 22 cm and the thickness is 0.85 cm. The length of the microstrip is 22 cm and the width is 0.65 cm.
Return loss and radiation pattern are measured with the microstrip antenna as the receiver and shown in Figures 10.34 and 10.35, respectively. A transmitter made of traditional dipole antenna is used and the transmitter–receiver distance is 42 m. The radiation pattern and the signal level are close to expected data. The antenna has a signal amplitude of 0 dB at 0°. The back lobe of the radiation is sufficiently small and is measured to be −4.8 dB. The antenna has a high quality factor Q that represents the losses associated with the stripline and the substrate. A large Q leads to narrow bandwidth and low efficiency.
However, Q can be reduced by increasing the thickness of the substrate. Hansen, in, 2002 Printed-Circuit AntennasPrinted-circuit antennas include stripline slots, printed-circuit-board dipoles, and microstrip patches. A stripline slot consists of a narrow rectangular slot cut in the top stripline ground plane, with the slot excited by proper positioning of the center conductor below the slot. A linear resonant array of collinear slots may be fed by a single center conductor, with this conductor centered under each slot at the proper angle. Boxed stripline is used to reduce higher modes.Dipoles may be printed for low-cost fabrication, either as an array of dipoles on a single dielectric substrate, with each dipole fed by a balun which is normal to the dipole array face, or with each dipole and balun on a separate dielectric substrate. The array comprises a stack of these sheets. The most widely used printed-circuit antenna is the microstrip patch, which in its simplest form is a rectangular or circular patch of metal fed by the microstrip upper conductor; see Fig.
Thus the element and feed line, and usually other elements, power dividers, etc., can all be prepared as a single etched PC board. A patch can also be fed from below by a coaxial probe, but this in large part removes the cost advantage. Early work on microstrip antennas was done by Deschamps and Sichak. † A review is given by Pozar. Increasing the substrate height does increase the bandwidth, but the coupling to surface wave modes and the cross-polarized radiation also increase, with undesirable pattern effects. Surface waves may be a problem for larger arrays or large substrates. An easy technique to obviate these is to make the dielectric anisotropic, by drilling holes perpendicular to the slab.
Such a slab is sometimes called a Photonic Band Gap material in analogy to quantum mechanics, but this nomenclature is at best misleading. Most techniques for extending the bandwidth of patches have done so at the expense of efficiency, i.e., through use of matching networks with high standing waves. However, the parasitic patch. avoids these problems by keeping the substrate thickness low, with a parasitic patch above the driven patch increasing the effective radiation height. Bandwidth can be doubled, with the parasitic patch dimensions and height above the driven patch adjusted to give a symmetrical impedance curve. 28 sketches the parasitic patch configuration. These may be arrayed as are ordinary patches.
In practice, the parasitic patches are printed on a thin dielectric substrate, with a foam layer used to support this substrate above the microstrip. Rectangular or circular patches, as described, are mostly linearly polarized radiators, with patch widths roughly a third of a wavelength. Accordingly, the patterns are between those of a short dipole and a half-wave dipole. Although a square patch could be fed on two adjacent sides with the proper phases to obtain circularly polarized radiation, simpler circularly polarized patches have been developed. ‡ The key is to modify dimensions to allow the two cross-polarized modes to be of equal amplitude and 90 degrees out of phase.
29 sketches four ways of accomplishing this. A simple analysis has been given by Lo and Richards.
§ Separate feeds provide wider bandwidth. A 38-GHz high-pulse repetition frequency (HPRF) mono-pulse Doppler-radar front-end was developed at the University of Michigan by Caekenberghe et al. 28, 29 for a long-range automotive radar application such as pre-crash detection and ACC.
The radar front-end is based on an RF MEMS electronically scanned array (ESA) design and low-loss MEMS time-delay units (TDU), which offer cost and performance advantages over the other technologies. A maximum radar range up to 150 m and a resolution of 1.5 m are obtained from this design. The scanning angle resolution for this car radar is ±11° in the azimuth plane with an angular deviation of ±2.5° from an amplitude mono-pulse tracking, as compared to the scanning angle.
This front-end was optimized for 38 GHz; however, it can be scaled to 77 GHz to match the ITU-Rs recommendations. 28 The front-end concept of the RF MEMS-based HPRF automotive radar is shown in Fig. As shown in Fig. 16.8, the radar front-end consists of a T/R module, an intermediate-frequency (IF) impulse processor and two passive sub-arrays which are individually connected to the T/R module. In the transmitting mode, a transmitted signal is fed to the transmitter module via an IF impulse processor, while the received signal is separated and fed to the sum and the azimuth-different channels to correct the angular errors.
The radiation patterns of the transmitting and receiving modules are shown in Fig. 16.9 a and b. Two passive sub-arrays, as the key units, are composed of 16-dB-gain 64 elements of microstrip antennas optimized at 38 GHz. The line antenna arrays fed through a slotted aperture from a microstrip corporate feed a network consisting of Wilkinson power dividers and grounded coplanar waveguide (GCPW) TDUs, which are analogue DMTLs, and each line is loaded with 60 elements of RF MEMS varactor.
This design offers a low-loss and wideband device with good power handling capability at 38 GHz. Radiation characteristic of the transmit (a) and receive (b) mode of the 38-GHz HPRF mono-pulse Doppler-radar. 28, 29The front-end module was fabricated on a 6-inch fused silica wafer and a Rogers TMM3 substrate using wafer-scale monolithic tile construction. The fused silica wafer contains two passive sub-arrays which are the MEMS TTD feed networks coupled to microstrip antenna arrays which are fabricated on the Rogers substrate. Apart from the microstrip antenna arrays, the other modules were fabricated in a class-100 cleanroom with a wafer-scale RF MEMS process with SiCr integrated thin-film resistors and high-resistivity bias lines.
16.3.2 Robert Bosch 77-GHz automotive radar front-end. Schoebel et al. At Robert Bosch GmbH developed two designs for 77-GHz analogue-beam-forming automotive radar front-end. The first approach was based on a MEMS single-pole–multi-throw switch, which is employed to select one of several beams of a planar Rotman lens and then feeding branches of a patch-antenna array. The second approach was a more conventional patch-antenna array beam-steering concept, whereby RF MEMS phase shifters, fed via a Wilkinson power divider, were used for configuring the phase distribution of the signals of the antenna array. 7, 9 These designs offer better performance, ease of design, and low manufacturing costs in comparison to the other millimetre-wave technologies. These two 77-GHz RF MEMS-based automotive radar concepts are shown in Fig.
Two 77-GHz RF MEMS automotive-radar front ends from Robert Bosch GmbH. 9 The designs are based on Rotman lens (left) and MEMS phase shifters (right).For the first design, an automotive radar front-end implemented with RF MEMS phase shifters, with less than three bits, and a patch-antenna array was investigated. The distance between each antenna element was half-wavelength, as compared to the wavelength in free space. Therefore, fixed scanning beams of 0°, ±14.5°, ±30°, and ±48.6° from the boresight axis are obtained from this design. The beam forming is implemented by constructing three stages of a two-port Wilkinson power divider with the employment of a 20-dB Chebyshev pattern.
RF MEMS stub-loaded-line phase shifters employed in this automotive radar design are based on MEMS capacitive shunt switches, offering 90° and 180° phase shift and less than 4° phase tolerance. This phase shifter exhibits a very wide bandwidth, which is very suitable for ultra-wideband (UWB) applications. The second alternative design consists of a Rotman lens and a single-pole–quadruple-throw (SP4T) switch. This design is very straightforward and offers a completely symmetrical device, exhibiting uniform loss for all scan beams.
The number of scan angles reduces to four, pointing to ±6° and ±18° from boresight axis. The Rotman lens was implemented by microstrip technology, and the body of the lens is designed as a parallel-plate waveguide with microstrip feeding. The Rotman lens with patch-antenna and waveguide interconnects is realized on a 5-mil Ro3003 substrate. The SP4T switch is designed based on the capacitive shunt single-pole–double-throw (SP2T) switch design.
The SP2T switch offers an insertion loss of 1.8 dB at 77 GHz. Figure 16.11 shows the fabricated MEMS automotive radar based on the Rotman lens and the SP4T switch. The radiation pattern of the design with the Rotman lens and the antenna array is shown in Fig.
16.12, in comparison to the three-stage Wilkinson power divider and MEMS phase shifters, with the same microstrip antenna array.