sensors Article
Non-Uniform Microstrip Antenna Array for DSRC in Single-Lane Structures Tiago Varum 1,2, *, João N. Matos 1,2 and Pedro Pinho 1,3 1 2 3
*
Instituto de Telecomunicações, Campus Universitário de Santiago, 3810-135 Aveiro, Portugal;
[email protected] (J.N.M.);
[email protected] (P.P.) Universidade de Aveiro, Campus Universitário de Santiago, 3810-135 Aveiro, Portugal Instituto Superior de Engenharia de Lisboa, Rua Conselheiro Emílio Navarro, 1959-009 Lisboa, Portugal Correspondence:
[email protected]; Tel.: +351-234-377-900
Academic Editor: Paolo Bellavista Received: 6 September 2016; Accepted: 6 December 2016; Published: 11 December 2016
Abstract: Vehicular communications have been subject to a great development in recent years, with multiple applications, such as electronic payments, improving the convenience and comfort of drivers. Its communication network is supported by dedicated short range communications (DSRC), a system composed of onboard units (OBU) and roadside units (RSU). A recently conceived different set-up for the tolling infrastructures consists of placing them in highway access roads, allowing a number of benefits over common gateway infrastructures, divided into several lanes and using complex systems. This paper presents an antenna array whose characteristics are according to the DSRC standards. Additionally, the array holds an innovative radiation pattern adjusted to the new approach requirements, with an almost uniform wide beamwidth along the road width, negligible side lobes, and operating in a significant bandwidth. Keywords: dedicated short range communications; antenna arrays; microstrip antennas
1. Introduction The evolution of wireless communications has been reflected in several areas, including transport networks, wherein the development of vehicular communications has become evident in recent years, offering several functionalities to the drivers and other travelers, allowing the improvement of important safety issues, reducing traffic jams, or even accidents. Dedicated short range communications (DSRC) is the technology enabling wireless communication among onboard units (OBUs) or between OBUs and roadside units (RSUs), allowing vehicle-to-vehicle and vehicle-to-infrastructure communications, being regulated by a set of standards. Electronic payments for tolls, parking, or gas stations are an application with increasing development in vehicular communications, due to the convenience and comfort that it provides to users. The electronic toll collection applied to highways is usually made based on input and output road logs, or using several gateways along the highway. These structures cover various lanes and allow quick access and the fluency of traffic. However, they involve at least one system covering each lane, and also require great care to overcome potential interferences between the various RSU’s, or with the vehicles passing in adjacent lanes. A recent approach suggests that the RSU’s gateways could be placed in the input and output access roads of the highways (typically single lane) as shown in Figure 1. This concept allows the use of only one RSU system, avoiding problems of interference between lanes and a simpler system, since in these lanes—commonly with curved shape—the speeds are much lower than in the highways.
Sensors 2016, 16, 2101; doi:10.3390/s16122101
www.mdpi.com/journal/sensors
Sensors 2016, 16, 2101 Sensors 2016, 16, 2101
2 of 11 2 of 8
Figure 1. Highway tolling system in access lane.
The CEN CEN EN12253 The EN12253 [1] [1] standard standard provides provides aa set set of of constraints constraints to to the the physical physical medium medium of of DSRC, DSRC, including for the RSU antenna. It mentions that the antenna must operate at least in a 20 MHz band including for the RSU antenna. It mentions that the antenna must operate at least in a 20 MHz band around 5.8 GHz, with a maximum equivalent isotropic radiated power of +33 dBm up to an angle of around 5.8 GHz, with a maximum equivalent isotropic radiated power of +33 dBm up to an angle ◦ with 70°70 with thethe vertical (outside thisthis zone maymay not not exceed +18 dBm). This This implies antenna sidelobes with of vertical (outside zone exceed +18 dBm). implies antenna sidelobes a level at least 15 dB below the maximum. In terms of polarization, it should be left hand circular with a level at least 15 dB below the maximum. In terms of polarization, it should be left hand circular (LHCP) with with aa rejection rejection of of 15 15 dB dB in in boresight, boresight, and and 10 10 dB dB in in the the zone zone where wherethe thegain gaindrops drops33dB. dB. (LHCP) Furthermore, since be wider wider than than the the highway highway lanes, lanes, the specific Furthermore, since the the access access roads roads tend tend to to be the specific application imposes a radiation pattern on the road plane (horizontal) larger than the usual DSRC application imposes a radiation pattern on the road plane (horizontal) larger than the usual DSRC tolls tolls applications, and awith a uniform gain across the to width tocommunication avoid communication at the applications, and with uniform gain across the width avoid failuresfailures at the ends or ends or passing without detection. passing without detection. Some antenna antenna arrays to RSU RSU of of DSRC DSRC systems systems are are proposed proposed [2–8]. [2–8]. It presented aa Some arrays targeted targeted to It is is presented 4 ××4 4right circularly polarized antenna array, with microstrip elements having an H-shape with 4 righthand hand circularly polarized antenna array, with microstrip elements having an H-shape a T-slot [2]. This array was designed for 5.8 GHz, showing a bandwidth of 109 MHz, sidelobes below with a T-slot [2]. This array was designed for 5.8 GHz, showing a bandwidth of 109 MHz, sidelobes −25 dB,−and about beamwidth. However, the design using ausing multi-layer structure insertsinserts some below 25 dB, and22 about 22◦ beamwidth. However, the design a multi-layer structure complexity in theindesign, in the manufacture, andand increases thethe cost. some complexity the design, in the manufacture, increases cost.AA44××44array array with with square square 22 structure size applies the microstrip elements is presented [3]. This array with a 500 × 500 mm microstrip elements is presented [3]. This array with a 500 × 500 mm structure size applies the sequential rotation rotation technique technique to to produce produce circular circular polarization. polarization. In In terms sequential terms of of sidelobes, sidelobes, these these are are higher higher than desired (between −10 and −15 dB), whereby the authors use an absorbent Korean paper with than desired (between −10 and −15 dB), whereby the authors use an absorbent Korean paper with Chinese ink, after the design to reach values on the order of −20 dB. The complexity of the final Chinese ink, after the design to reach values on the order of −20 dB. The complexity of the final structure making making it it possible possible to to obtain obtain reasonable reasonable results results concordant concordant with with the the DSRC DSRC standards is the the structure standards is main disadvantage of this solution. main disadvantage of this solution. An array with four four double double loop loop monopoles monopoles for for aa DSRC DSRC system system [4]. [4]. This An array is is presented presented with This antenna antenna has has ◦ 600 MHz bandwidth, beamwidth of 25, and sidelobes below −19 dB. However, it has a large ground 600 MHz bandwidth, beamwidth of 25 , and sidelobes below −19 dB. However, it has a large ground 2 plane (90 thethe absence of circular polarization. For aFor free-flow system, a microstrip patch plane (90 ××9090cm cm) 2and ) and absence of circular polarization. a free-flow system, a microstrip superimposed by a metamaterial layer designed using square split-ring resonator elements, which patch superimposed by a metamaterial layer designed using square split-ring resonator elements, increase the directivity of the single patch [5]. The authors propose a four-patch configuration to which increase the directivity of the single patch [5]. The authors propose a four-patch configuration ◦ obtain a radiation pattern with a 36 beamwidth and a directivity of 16.38 dBi. The complexity and to obtain a radiation pattern with a 36 beamwidth and a directivity of 16.38 dBi. The complexity volume make this this structure unattractive. In In addition, forfor the and volume make structure unattractive. addition, theinfrastructure infrastructureofofaa DSRC DSRC system, system, reference [6] presents a Fabry–Perot resonator antenna consisting of a circularly polarized reference [6] presents a Fabry–Perot resonator antenna consisting of a circularly polarized squaresquare patch patcha partially and a partially reflective superstrate. The characteristics main characteristics of the antenna are of a gain of and reflective superstrate. The main of the antenna are a gain 18 dBi, 18 dBi, sidelobe levelof(SLL) of −16 21 beamwidth. However, its complex and its sidelobe level (SLL) −16 dB, anddB, 21◦and beamwidth. However, its complex structurestructure and its volume volume are the main disadvantages. In reference [7], a 2 × 4 microstrip antenna array is presented are the main disadvantages. In reference [7], a 2 × 4 microstrip antenna array is presented which which combines complexity This its main feature a widebandwidth operating combines reducedreduced complexity and size.and Thissize. array hasarray as its has mainasfeature a wide operating bandwidth covering the DSRC bands, in addition to a radiation diagram complying with the covering the DSRC bands, in addition to a radiation diagram complying with the standards. However, standards. However, it has sidelobes that are slightly higher than desired. it has sidelobes that are slightly higher than desired. Finally, in reference [8], an antenna that overcomes the main problems revealed by the other reported antennas is shown, largely fulfilling the constraints of DSRC standards. Despite its design complexity, this 12 element array has a single-layer structure, which allows for simplicity of
Sensors 2016, 16, 2101
3 of 11
Finally, in reference [8], an antenna that overcomes the main problems revealed by the other reported is shown, largely fulfilling the constraints of DSRC standards. Despite Sensors 2016,antennas 16, 2101 3 ofits 8 design complexity, this 12 element array has a single-layer structure, which allows for simplicity manufacture andand reduced cost, as as well as as low weight and volume. In In addition, thethe array presents a of manufacture reduced cost, well low weight and volume. addition, array presents wide bandwidth (greater a wide bandwidth (greaterthan than1 1GHz GHzband) band)covering coveringthe theDSRC DSRCbands, bands,reduced reduced sidelobes, sidelobes, and and aa radiation radiation pattern pattern with with aa beamwidth beamwidth around around 22. 22◦ . All All of of the the proposed proposed antennas antennas in in the the literature literature focus focus on on the the common common toll toll gates gates and and free free flow flow configurations, configurations, showing showing radiation radiation patterns patterns with with narrow narrow half half power beam widths (HPBW) in the horizontal plane, in addition to highly complex structures. power Taking theDSRC DSRC and application requirements into account, this paper an presents an Taking the and the the application requirements into account, this paper presents innovative innovative whose characteristics fittolling the needs for tolling application in access including roads of array whosearray characteristics fit the needs for application in access roads of highways, highways, includingpattern, its widera radiation pattern, a uniform across thewith roadreduced width, and with its wider radiation uniform gain across the roadgain width, and sidelobes. reduced sidelobes. Furthermore, the designed entire array was microstrip designed using microstrip with the Furthermore, the entire array was using structures, withstructures, the advantages of advantages of beingsimple, light weight, simple, being light weight, and low cost. and low cost. This paperis divided is divided into sections. four sections. begins with an to introduction to vehicular This paper into four It beginsItwith an introduction vehicular communications communications (focusing on tolling and stating the objectives. The second (focusing on tolling applications), and applications), stating the objectives. The second section describes thesection design describes the design of the proposed antenna array, and Section 3 shows the simulation and of the proposed antenna array, and Section 3 shows the simulation and measurement results. Finally, measurement theconclusions. last section reports the main conclusions. the last sectionresults. reportsFinally, the main 2. Antenna Antenna Array Array Design Design 2. Due to (such as their simplicity and versatility), microstrip antennasantennas [9] are frequently Due totheir theirbenefits benefits (such as their simplicity and versatility), microstrip [9] are used in many wireless applications, achieving low profile low antennas with a simple anda inexpensive frequently used in many wireless applications, achieving profile antennas with simple and manufacture.manufacture. Moreover, they enablethey the design of array feed networks using microstrip inexpensive Moreover, enable the design of array feed(AFN) networks (AFN) using transmission lines in thelines planeinofthe theplane arrayofelements, structures. structures. microstrip transmission the arrayallowing elements,single-layer allowing single-layer 2.1. Antenna Antenna Array Array Element Element 2.1. The antenna antenna element element used used consists consists of of aa circular circular microstrip microstrip patch patch (as (as shown shown in in Figure Figure 2), 2), whose whose The resonance frequency frequency varies varies with with the the radius radius r. Additionally, two ∆rr)) were were created created for for resonance r. Additionally, two side side slits slits (∆ (Δyy × ×Δ LHCP generation [9]. To adapt the antenna input impedance to the impedance of AFN transmission LHCP generation [9]. To adapt the antenna input impedance to the impedance of AFN transmission lines, aa double double quarter quarter wavelength wavelength transformer transformer was was inserted inserted (l (lt1t1 and and llt2t2).). lines,
(a)
(b)
Figure Figure 2. 2. Microstrip Microstrip patch patch element: element: (a) (a)structure structure and and dimensions; dimensions; (b) (b) photograph photograph of of the the prototype. prototype.
The dimensions of the microstrip element designed for 5.8 GHz are: r = 9.1 mm, Δy = 2 mm, The dimensions of the microstrip element designed for 5.8 GHz are: r = 9.1 mm, ∆ = 2 mm, Δr = 1.1 mm, lt1 = 10.2 mm, wt1 = 0.45 mm, lt2 = 9 mm, and wt2 = 1.25 mm. The circular patchy element ∆r = 1.1 mm, lt1 = 10.2 mm, wt1 = 0.45 mm, lt2 = 9 mm, and wt2 = 1.25 mm. The circular patch element was manufactured, and the prototype is shown in Figure 2b. It was measured in terms of its main parameters, and results were compared with those obtained by simulation (presented below). Figure 3 shows a comparison between the simulated and measured results, both of the S11 and axial radio of the manufactured antenna. It is possible to observe a good agreement between simulated and measured curves. In terms of S11—assuming the usual criterion for a good impedance matching an S11 < −10 dB—the patch has a measured bandwidth of 210 MHz (5.68–5.89 GHz).
Sensors 2016, 16, 2101
4 of 11
was manufactured, and the prototype is shown in Figure 2b. It was measured in terms of its main parameters, and results were compared with those obtained by simulation (presented below). Figure 3 shows a comparison between the simulated and measured results, both of the S11 and axial radio of the manufactured antenna. It is possible to observe a good agreement between simulated Sensors2016, 2016,16, 16,2101 2101 of88 Sensors 44of and measured curves. In terms of S11 —assuming the usual criterion for a good impedance matching an S11 < −10as dB—the patch has apolarized measured of 210 MHzratio (5.68–5.89 GHz). Considered as Considered goodcircularly circularly anbandwidth antennawith with anaxial axial lowerthan than dB, thecircular circular Considered as aagood polarized an antenna an ratio lower 33dB, the apatch goodhas circularly polarized an antenna with an axial ratio lower than 3 dB, the circular patch has a measured50 50MHz MHzbandwidth bandwidthof ofcircular circularpolarization. polarization. patch has aameasured measured 50 MHz bandwidth of circular polarization.
Figure3.3.Simulated Simulatedand andmeasured measuredSS1111and and axial axialratio ratio(AR) (AR)of ofthe thedesigned designedcircular circularpatch patchantenna. antenna. Simulated ratio (AR) of the designed circular patch antenna. Figure 11 andaxial
The Figure44shows shows a comparisonbetween between thesimulated simulated andmeasured measured radiationpattern pattern ofthe the The The Figure Figure 4 shows aa comparison comparison between the the simulated and and measured radiation radiation pattern of of the antenna in terms terms of of EELHCP LHCP and ERHCP components, for the radiation plane φ = 0. A good agreement antenna components, for the radiation plane φ = 0.◦ A good agreement antenna in in terms of ELHCP and and EERHCP RHCP components, for the radiation plane ϕ = 0 . A good agreement between simulation and measurements is visible.As As expected,the the patchhas has a wideradiation radiation pattern, between between simulation simulation and and measurements measurements is is visible. visible. As expected, expected, the patch patch hasaawide wide radiationpattern, pattern, with a HPBW of 80°. with with aa HPBW HPBW of of 80°. 80◦ .
Figure 4. Simulated and measured radiation pattern of the designed circular patch antenna. Figure4. 4.Simulated Simulatedand andmeasured measuredradiation radiationpattern patternof ofthe thedesigned designedcircular circularpatch patchantenna. antenna. Figure
There isis aaa complete complete relationship between thethe axial ratio (AR) and and the Ethe ELHCP LHCPELHCP and EERHCP RHCP There completerelationship relationshipbetween between axial ratio (AR) and There the axial ratio (AR) and the and Ecomponents. RHCP components. components. ERHCP | ELHCP+ | E E AR(dB) = 20log (1) LHCP ERHCP RHCP AR((dB dB)) 20 20log log E|LHCP ELHCP − ERHCP (1) | AR (1)
EELHCP EERHCP
LHCP from RHCP given 3 (shown for θ = 0◦ ), respectively, Considering the simulated and measured AR Figure Considering the simulated and measured AR given from Figure (shownfor forθθ==0°), 0°),respectively, respectively, Considering the simulated ARmeas = 3 dB and AR dB. measured AR given from Figure 33(shown sim = 0.81and AR meas = 3 dB and ARsim = 0.81 dB. ARmeas = 3 dB and ARsim = 0.81 dB.
ARmeas 33 EELHCP 55..84 84 EERHCP 20 20log( log(EELHCP)) 20 20log( log( 5..84 84))20 20log( log(EERHCP)) AR 5 meas LHCP RHCP LHCP RHCP 15.3dB 1115.3dB
ARsim 00..81 81 EELHCP 21 21 EERHCP 20 20log( log(EELHCP)) 20 20log( log(21 21))20 20log( log(EERHCP)) AR sim LHCP RHCP LHCP RHCP 26.4dB 2226.4dB
These values values of of ΔΔ11 and and ΔΔ22 are are consistent consistent with with the the rejection rejection between between simulated simulated and and measured measured These components (shown in Figure 4), and the difference is justified by the slight deviation in the frequency components (shown in Figure 4), and the difference is justified by the slight deviation in the frequency of the the axial axial ratio ratio between between simulated simulated and and measured measured values. values. However, However, aa good good quality quality of of circular circular of
Sensors 2016, 16, 2101
5 of 11
ARmeas = 3 ⇔ ELHCP ∼ = 5.84 ERHCP ⇔ 20log( ELHCP ) ∼ = 20log(5.84) + 20log( ERHCP ) | {z } ∆1 =15.3dB
ARsim = 0.81 ⇔ ELHCP
∼ = 21 ERHCP ⇔ 20log( ELHCP ) ∼ = 20log(21) + 20log( ERHCP ) | {z } ∆2 =26.4dB
These values of ∆1 and ∆2 are consistent with the rejection between simulated and measured components (shown in Figure 4), and the difference is justified by the slight deviation in the frequency of the axial ratio between simulated and measured values. However, a good quality of circular polarization is ensured. 2.2. Antenna Array Structure Since the access lanes are typically large, the antenna for the RSU must have a wide radiation 5 of 8 pattern in the horizontal plane, with a HPBW around 60◦ . With uniform antenna arrays, it is not get anStructure HPBW of 60◦ with a uniform gain across the lane width while ensuring low sidelobes, 2.2.possible AntennatoArray so non-uniform excitation techniques must be used. Since accesstransform lanes aremethod typically large,[9] the antenna thearray, RSU since mustthis have a wideallows radiation Thethe Fourier (FTM) was used infor this method a feed pattern in the horizontal plane, with a HPBW around 60. With uniform antenna arrays, it ispattern. not distribution to be found for a set of array elements, so as to achieve a particular radiation possible getnumber an HPBW of 60 with uniform lane width while ensuring low by For antoodd of elements (N =a 2M + 1), gain each across elementthe feed coefficient can be calculated sidelobes, so non-uniform excitation techniques must be used. the following: The Fourier transform method (FTM) [9] was used in this array, since this method allows a feed T/2 π to achieve a particular radiation pattern. For distribution to be found w a set of array elements, 1so was 1 for jmψ − jmψ AFfeed dψ −can M be ≤m ≤M a = AF ψ e dψ = (ψ) e − m elements (N(= )2M + 1), each element an odd number of coefficient calculated by the(2) T 2π −π − T/2 following: Sensors 2016, 16, 2101
T/ 2
π
1 where AF(ψ) represents the desired array 1 factor and ψ = kdcos θ + β. a AFψ e jmψ dψ AFψ e jmψ dψ M m M (2) Figurem5 shows the desired array factor for this application, which consists of a uniform (maximum) T T/ 2 2π π amplitude over the lane width ∆θ, and zero outside this zone. To provide some margin, since the where AF() represents the desired array factor and = kdcos θ + β. HPBW is lower than the width between nulls of radiation pattern, the calculations were performed to Figure 5 shows the desired array factor for this application, which consists of a uniform ∆θ = 70◦ . The integral of Equation (2) can be decomposed by (maximum) amplitude over the lane width Δθ, and zero outside this zone. To provide some margin, since the HPBW is lower pattern, the calculations were of radiation ∆θ/2 ∆θ/2 w than the width between 1nulls 1 1 − jmψ − jmψ performed to Δθ of Equation (2) can−be decomposed by − M ≤ m ≤ M am ==70°. The integral (3) 1e dψ ⇒ am = e T 2π jmΔθ/2 −∆θ/2 Δθ/2
−∆θ/2 1 1 1 jmψ 1 e jmψ dψ am M m M (3) jm e T following 2πallows Δθ/2 to be calculated for each m-element of Δθ/2 resulting in the expression that the weights
am
resulting in thearray. following expression that allows the weights to be calculated for each m-element of this specific sin(0.57πm) this specific array. am = 0.57 −M≤m≤ M (4) 0.57πm sin ( 0.57 πm) am 0.57 M m M (4)
0.57πm
Figure 5. Desired array factor. Figure 5. Desired array factor.
Figure 6 shows the theoretical estimation of the array factor, using distance between elements of d1 = 0.5λ, and with the corresponding weights calculated via FTM applied to each element by varying the number of elements N.
Sensors 2016, 16, 2101
Figure 5. Desired array factor.
6 of 11
Figure 6 shows the theoretical estimation of the array factor, using distance between elements of Figure 6 shows the theoretical estimation of the array factor, using distance between elements of d1 = 0.5λ, and with the corresponding weights calculated via FTM applied to each element by varying d1 = 0.5λ, and with the corresponding weights calculated via FTM applied to each element by varying the number of elements N. the number of elements N.
Figure 6. array response with the number of elements of the of array, Figure 6. Variation Variationofoftheoretical theoretical array response with the number of elements the using array,Fourier using 6 of 8 transform methodmethod (FTM) coefficients. Fourier transform (FTM) coefficients.
Sensors 2016, 16, 2101
According to Figure 6, it is possible to observe that by increasing N, the radiation pattern clearly According Figure 6, it iswith possible to main observe that by increasing N,itthe radiation pattern clearly becomes better to defined, and lower lobe ripple; however, dramatically increases the becomes better defined, and with lower main lobe ripple; however, it dramatically increases the complexity of the array and the necessary AFN. The N = 9 elements was chosen, since it is the best complexity of the array and theradiation necessarycharacteristics, AFN. The N =the 9 elements was chosen,and since is the trade-off between the desired good performance, theit design best trade-off between the desired radiation characteristics, the good performance, and the design complexity of the array. The array of feed distribution (weight to apply to each element) for N = 9 complexity the array. The array elements is of presented in Figure 7. of feed distribution (weight to apply to each element) for N = 9 elements is presented in Figure 7.
Figure 7. Antenna array weights.
Figure 7. Antenna array weights.
Using a linear array of nine elements arranged along the horizontal plane (1 × 9), the radiation Using a linear array of has ninea elements arranged along the plane (1 × 9), the radiation pattern in the vertical plane wide HPBW, not fulfilling thehorizontal DSRC standard, which establishes a ◦ pattern in the vertical plane has a wide HPBW, not fulfilling the DSRC standard, which establishes maximum of 70 in this plane. To meet that requirement, another line element was inserted, placed ata of 70° in this plane. To meet that requirement, another line element was inserted, placed dmaximum 2 = 0.6λ and with equal feed distribution, so that the final antenna array consists of a planar structure at d 2 = 0.6λ and with equal feed distribution, so that the final antenna array consists of a planar of 2 × 9 elements. structure of 2 × 9 elements. 2.3. Array Feed Network 2.3. Array Feed Network The AFN has a great importance in the array design, implementing the feed distribution shown The AFN has case, a great importance into theprovide array design, implementing theand feedease distribution shown in Figure 7. In this it was designed simplicity, a low profile, of manufacture in Figure In this case, it was designed to provide simplicity, a low profile, and ease of manufacture on a single7.dielectric layer structure placed in the plane of the radiating elements. Figure 8 shows the on a single structure in the plane the radiating elements. Figure 8 shows the structure ofdielectric designedlayer array, with theplaced AFN created usingofmicrostrip lines (with 100 Ω characteristic structure of designed the AFN created using microstrip lines (with 100 Ω characteristic impedance), T-junctionarray, powerwith dividers (PDs), and quarter-wavelength impedance transformers. impedance), power dividers (PDs),with and two quarter-wavelength transformers. The AFNT-junction has a symmetrical arrangement branches (fed by impedance PDin divider), compensating ◦ the physical rotation between the two lines of nine elements by a 180 phase delay between the two branches. The amplitude distribution was accomplished using unequal power dividers (PD1, PD2, PD3, and PD4) with different ratios. Additionally, each patch was fed by different length of lines, to individually adjust the feed of all the microstrip elements, P1 ... P18 , with the desired phases.
The AFN has a great importance in the array design, implementing the feed distribution shown in Figure 7. In this case, it was designed to provide simplicity, a low profile, and ease of manufacture on a single dielectric layer structure placed in the plane of the radiating elements. Figure 8 shows the structure of designed array, with the AFN created using microstrip lines (with 100 Ω characteristic Sensors 2016, 16, 2101 impedance), T-junction power dividers (PDs), and quarter-wavelength impedance transformers. 7 of 11
Sensors 2016, 16, 2101
7 of 8
length of lines, to individually adjust the 8. feed of all the microstrip Figure Antenna array structure.elements, P1... P18, with the desired Figure 8. Antenna array structure. phases. The array was designed andsimulated simulated using Computer Technology’s The array was designed and using the thewith Computer Simulation Technology’s Microwave The AFN has a symmetrical arrangement two Simulation branches (fed by PDinMicrowave divider), Studio (CST MWS) electromagnetic simulator [10]. Figure 9 shows the simulation results the Studio (CST MWS) electromagnetic simulator [10]. Figure 9 shows the simulation results of theof signals compensating the physical rotation between the two lines of nine elements by a 180° phase delay signals arriving at each patch of one branch (identical in the other branch) in terms of amplitude and between the two branches. The amplitude accomplished using unequaland power arriving at each patch of one branch (identicaldistribution in the otherwas branch) in terms of amplitude phase. phase. For the frequency, 5.8 GHz frequency, these values are summarized 1.patch It is possible to different conclude dividers (PD1, PD2, PD3, and PD4) withare different ratios. Additionally, was fed by For the 5.8 GHz these values summarized in Table in 1. Table It each is possible to conclude that both that both results—amplitude and phases—are consistent with the desired theoretical relationships results—amplitude and phases—are consistent with the desired theoretical relationships expressed in expressed in Figure 7. Figure 7.
Figure9.9.Amplitude Amplitudeand and phase phase distribution distribution of Figure of the the array arrayfeed feednetworks networks(AFNs). (AFNs). Table1.1.Array ArrayFeed FeedNetwork–Simulated Network–Simulated Amplitude Table Amplitude and andPhase Phase(5.8 (5.8GHz). GHz). Patch 1 2 3 4 5 6 7 8 9
Patch 1 2 3 4 5 6 7 8 9
Amplitude Phase Amplitude 0.073 −50 0.073 0.07 134 0.07 0.073 131 0.073 0.37 −45 0.37 0.77 −47 0.77 0.39 −45 0.39 0.08 128 0.08 0.076 130 0.076 0.08 −54 0.08
Phase
−50◦ 134◦ 131◦ −45◦ −47◦ −45◦ 128◦ 130◦ −54◦
3. Results The antenna was designed, simulated, and manufactured using the Rogers RO4725JXR substrate, whose main characteristics are: dielectric constant εr = 2.55, thickness h = 0.78 mm, and loss tangent tgδ = 0.0026. The prototype is shown in Figure 10, and presents global dimensions
Sensors 2016, 16, 2101
8 of 11
3. Results The antenna was designed, simulated, and manufactured using the Rogers RO4725JXR substrate, whose main characteristics are: dielectric constant εr = 2.55, thickness h = 0.78 mm, and loss tangent tgδ = 0.0026. The prototype is shown in Figure 10, and presents global dimensions Sensors 2016, 16, 2101 8 of 8 Sensors 2016, 16, 2101 8 of 8 2 W g × Lg = 286 × 218 mm .
Figure 10. 10. Photograph Photograph of of the the antenna antenna array. array. Figure
3.1. Reflection Coefficient 3.1. Reflection Coefficient The prototype was measured, measured, and and a comparison comparison between between the the simulated simulated and and measured measured S11 11 of the The prototype was measured, and aa comparison between the simulated and measured SS11 of the antenna is presented in Figure Figure 11. 11. antenna is presented in Figure 11.
Figure 11. 11. Simulated Simulated and and measured measured reflection reflection coefficient coefficient(S (S11 11)) of of the the antenna. antenna. Figure the antenna. Figure 11. Simulated and measured reflection coefficient (S 11) of
The first first impression impression that that can can be be taken taken is is aa slight slight frequency frequency deviation; deviation; however, however, it it does does not not affect affect The The first impression that can be taken is a slight frequency deviation; however, it does not affect the the array’s performance, showing a good agreement between simulated and measured results. the array’s performance, showing a good agreement between simulated and measured results. array’s performance, showing a good agreement between simulated and measured results. Assuming Assuming as as aa common common criterion criterion for for an an impedance impedance matching matching an an SS11 11 < −10 dB, the antenna has a Assuming < −10 dB, the antenna has a as a common criterion for an impedance matching an S 11 < −10 dB, the antenna has a bandwidth of bandwidth of 358 MHz. bandwidth of 358 MHz. 358 MHz. 3.2. Polarization Polarization 3.2. The Axial Axial Ratio Ratio (AR) (AR) is is aa parameter parameter that that allows allows the the quality quality of of the the circular circular polarization polarization of of an an The antenna to to be be characterized. characterized. Figure Figure 12 12 illustrates illustrates the the simulated simulated and and measured measured ARs ARs of of the the designed designed antenna antenna array. It is possible to observe an acceptable agreement between the two results, with only aa array. It is possible to observe an acceptable agreement between the two results, with only antenna array. small deviation. small deviation. Assuming an an antenna antenna with with the the AR AR
