Double-Dielectric Microstrip Ultrahigh-Frequency Antenna for Digital Terrestrial Television

applied sciences

Article Double-Dielectric Microstrip Ultrahigh-Frequency Antenna for Digital Terrestrial Television

Antonio Navarro *, Pedro Mostardinha, Tiago Varum , Joao Matos and Stanislav Maslovski Instituto de Telecomunicações and the Department of Electronics, Telecommunications and Informatics, University of Aveiro, P-3810-193 Aveiro, Portugal; [email protected] (P.M.); [email protected] (T.V.); [email protected] (J.M.); [email protected] (S.M.) * Correspondence: [email protected]; Tel.:+351-234-377900

Received: 7 November 2020; Accepted: 26 November 2020; Published: 3 December 2020

Featured Application: Although Yagi–Uda antenna is more than 80 years old and is still globally the most used antenna for TV reception, we believe that there is still room for improvement. This article discusses a new antenna design for application in terrestrial TV broadcasting. With regard to the antenna’s performance, our results are not limited to simulations. We present a prototype of a practical antenna supported by measurement results. This paper contributes to motivating and increasing research in the domain of antennas for digital terrestrial TV bands.

Abstract: In many countries, terrestrial broadcasting is the main delivery medium for television. In this paper, we propose a small-volume practical receiving antenna. Our design consists of a linear array of three vertically placed patch antennas which increase antenna gain. The antenna has a double-dielectric substrate (FR4 + air) in order to increase eﬃciency and bandwidth. In this paper, we also discuss simulations and practical results, and demonstrate that the proposed double-dielectric microstrip antenna is a viable design choice for digital terrestrial TV (DTT) reception. The designed antenna reached a gain of 10.5 dBi at the desired central frequency of 754 MHz.

Keywords: TV broadcasting; terrestrial television; microstrip antenna; UHF

1. Introduction In the modern world, switching from analog to digital television offers the user several advantages, such as high-definition (HD) video, six-channel surround audio (5.1), and data-broadcasting services. To make full use of the new audiovisual experience, the reception antenna must fulfill some requirements. Therefore, the design is very challenging for printed channel-bandwidth receiving antennas operating in the ultrahigh frequency band (UHF) [1]. Another well-known and important aspect of this challenge is that wireless communications suffer from interference that affects performance in terms of network capacity. This interference may arise, for example, from frequencies assigned to mobile networks (GSM, 3G, LTE, etc.) amplified by the input low-noise amplifier following the antenna. The upper limit of the digital television (DTV) band is next to either the GSM or LTE band [2]. Unlike other existing broadband antenna solutions, our novel proposed antenna avoids that interference because it has bandwidth close to a UHF IV and V band TV channel. Printed antennas have several advantages, such as ease of manufacturing, reduced cost, and low volume. They are also thin in one dimension, making them suitable to operate in indoor enviroments and placed on the wall like a photograph or art frame. This makes such antennas a promising choice for DTV reception. However, a better solution in this application is to design an antenna with adjacent-channel rejection. Thus, the goal is to have the necessary bandwidth centered on the reception

Appl. Sci. 2020, 10, 8640; doi:10.3390/app10238640 www.mdpi.com/journal/applsci Appl. Sci. 2020, 10, 8640 2 of 9 frequency while rejecting adjacent signals. In this framework, we designed an antenna that is intended for digital-terrestrial-television (DTT) reception. The next section presents the state of the art. In Section3, the problem of bandwidth is addressed. In Section4, the antenna design is discussed, and its structure and dimensions are presented. In Section5, antenna performance is substantively discussed, making use of several simulation results and practical measurements. Lastly, Section6 concludes the paper.

2. Related Work The Yagi–Uda antenna, commonly referred to as the Yagi antenna, is the most used antenna for TV reception. It was invented in 1926, so it is time to replace it with a novel design because of discoveries in new materials and structures since then. We need to move to a new type of antenna, but the research is tremendously challenging since the Yagi antenna’s performance is very good. Few antenna designs for DTV applications are presented in the literature. The antenna presented in [3] combines a planar inverted-F antenna (PIFA) and a loop antenna to achieve wideband operation in the UHF band. However, the operational bandwidth of the proposed PIFA–loop antenna is 460–870 MHz. Another interesting wideband antenna structure is a low-profile sinuous slot antenna constructed on a Rogers RO3206 substrate [4], where the gain varied from 2.5 to 6 dBi in the frequency range of 0.53–4.86 MHz. As discussed above, these antennas require filtering due to the nearby frequencies/channels of the LTE band presented in this range, which can cause interference to the desired existing allocated services. Another limitation is that this PIFA–loop antenna is targeted to be used only indoors. In [5], a monopole antenna is proposed for DTV applications with a wide operational bandwidth (451–912 MHz). Likewise, this design is not suitable for the main requirement, which is the reduction of interference from the LTE band. The monopole antenna [5] has a gain of 1.7 dBi, which is very low compared to that of other DTV antennas. In order to increase the gain, a solution was proposed in [6] that was based on a unidirectional array of bowtie antennas with an incision gap. The measured gain for this antenna array was 13.4–16.1 dBi. The design demonstrated good half-power beam width (HPBW) in the vertical plane of 15◦ from 470 to 862 MHz. Besides these features, the antenna also has return loss (RL) greater than 10 dB over a frequency range of 450–1014 MHz. A good example of a microstrip channel tuned antenna for Integrated Services Digital Broadcasting-Terrestrial (ISDB-T) was proposed in [7], but the gain was very low, around 2.16 dBi. − Another challenging problem is how to reach the appropriate bandwidth of a microstrip antenna. For example, as was shown in [8], a bandwidth increase of about 13% could be obtained by proximity coupling. However, this method has great complexity of design and construction. The simplest and most direct method is to increase the thickness of the substrate, that is, to use a thicker dielectric with a lower dielectric constant. This enables an increase in bandwidth and efficiency; however, the possibility of surface-wave formation may decrease antenna radiation performance [9]. Extending this principle in order to achieve broader bandwidth, we propose a double-dielectric array antenna that adds another dielectric layer between patch substrate and ground plane. The second dielectric should have low permittivity, as discussed in Section3. Alternatively, low-permittivity composite materials and metamaterials with a magnetic response, could be used to improve the bandwidth of microstrip antennas [10]. In this paper, we propose a new antenna that is well-matched at the frequency of interest and has the appropriate bandwidth, thus rejecting adjacent frequencies/channels such as the ones from the LTE band. The antenna was designed with linear polarization. Regarding the radiation pattern of the antenna, the main lobe had to be narrow enough to distinguish signals coming from different directions. Thus, by placing such antennas in a horizontal-array configuration, we could point the array radiation pattern maximum in the direction of the most favorable DTV broadcast signal (in terms of received power and modulation error rate), and simultaneously point the radiation pattern nulls in the direction of interfering signals. In this paper, we open a new path for motivating the research community to develop new antennas for TV reception. This paper takes a significant step towards Appl. Sci. 2020, 10, 8640 3 of 9 that goal by proposing a new antenna for digital TV reception. Our design makes use of a microstrip patch antenna as a radiating element with applications at relatively low frequencies (in the DTV–UHF band). The advantage of the proposed design is that it can easily be integrated into printed circuit boards (PCBs). It is also versatile in adjusting the antenna resonant frequency by simply changing the patch length. However, antennas of this type have a relatively narrow bandwidth, which can result in performance/efficiency degradation. Nevertheless, such antennas are expected to have enough bandwidth to cover at least a single 8 MHz channel with a suitable resonant frequency (754 MHz, single-frequency network in Portugal).

3. Bandwidth of Double-Dielectric Patch Antenna As discussed in Section2, in order to fulﬁll the bandwidth requirements, we chose air as the second dielectric in the double-dielectric substrate of the patch antenna. We aimed at obtaining antenna bandwidth of at least 8 MHz operating at the center frequency of 754 MHz. The bandwidth of a patch antenna above a dielectric substrate depends on the dimensions of the patch and substrate permittivity. A number of approximate formulas for antenna bandwidth are available from the literature. Here, we based our bandwidth estimation on the approach developed in [10]. Considering a resonant half-wavelength patch antenna with moderate patch width Wp satisfying 0.35 < Wp/λ0 < 2, where λ0 is the free space wavelength, the relative antenna bandwidth can be estimated [10] as 1 4h BW , (1) ≈ Qrad ≈ Wp 2 Wp √εr,eq 2 λ0 − π where Qrad is the radiation quality factor of the antenna, εr,eq is the equivalent relative substrate permittivity, and h is the substrate thickness. From this formula, increasing h and decreasing εr,eq resulted in a larger bandwidth for an antenna with a given patch width. As the air has a lower dielectric constant (εr = 1) compared to the average dielectric constant of the patch substrate (FR4), combining an air layer with a dielectric layer resulted in a reduction in the eﬀective dielectric constant of the double-dielectric substrate (FR4 + air). Such a compound substrate is characterized by an equivalent relative permittivity εr,eq that can be estimated by the following formula:

1 + r εr eq = , (2) , 1 + r εr where εr is the relative permittivity of the FR4 substrate, and r is a parameter determined by the hs thicknesses of the substrate and air layers hs and ha. In plate-capacitor approximation, r . A more ≈ ha accurate result could be obtained with the conformal mapping approach [10], in which

d r = s (3) da

K(ks) K(ka) πWp πWp where, ds = ( ) , da = ( ) ds, with ks = 1/ cosh , ka = 1/ cosh ( + ) , and K(x) is the K0 ks K0 ka − 4hs 4 hs ha complete elliptic integral of the ﬁrst kind [11]. On the one side, when hs ha, r 0 and the equivalent → permittivity approaches the permittivity of air. On the other side, when hs ha, r + and the → ∞ equivalent permittivity approaches the permittivity of the substrate. In our case, as shown in Section4, air height ha reached 3.2 mm.

4. Conﬁguration of Proposed Double-Dielectric Antenna The geometry of the proposed microstrip antenna is shown in Figure1. It was expected to achieve 3 dB more gain by narrowing the beamwidth in the elevation direction as long as we doubled the number of patches. The antenna, consisting of three patches placed above a double-layer dielectric Appl. Sci. 2020, 10, 8640 4 of 9

Appl. Sci. 2020, 10, x FOR PEER REVIEW 4 of 9 substrate, resulted in a good trade-oﬀ between size and gain. The three units increased the antenna Ωgain SMA by connector. about 4.7 dB.Series-fed It was fedcircuits bya were microstrip used to transmission drive the three line consecutive through a 50patches.Ω SMA The connector. distance betweenSeries-fed the circuits centers were of usedthe patches to drive in the the three array consecutive equaled 199.0 patches. mm The(~λ0/2 distance at the betweenoperation the frequency). centers of Allthe patchespatches inwere the arraydriven equaled in phase 199.0 because mm (~ theλ0/2 radiation at the operation amplitude frequency). and phase All patchesat each werepatch driven were determinedin phase because by the the cumulative radiation amplitude transmission and phasecharac atteristics each patch of the were preceding determined microstrip by the cumulative lines and patches.transmission The characteristicsconnecting lines of thehad precedingthe characteristic microstrip impedance lines and of patches. 100 Ω Theand connectingno influence lines on hadthe valuethe characteristic of the central impedance frequency of [12]. 100 Ω and no inﬂuence on the value of the central frequency [12].

z y

Figure 1. Geometry of double-dielectric microstrip antenna. Figure 1. Geometry of double-dielectric microstrip antenna. We followed a three-step approach. First, we theoretically designed the proposed antenna; second,We we followed simulated a three-step the antenna approach. behavior withFirst, the we CST theoretically Microwave designed Studio electromagnetic the proposed simulator;antenna; second,third, we we constructed simulated and the validated antenna the behavior antenna wi withth somethe CST measurements. Microwave Studio electromagnetic simulator; third, we constructed and validated the antenna with some measurements. On the basis of (1) and (2), we determined hs and ha. The upper dielectric was composed of FR4 On the basis of (1) and (2), we determined ℎ and ℎ . The upper dielectric was composed of with thickness hs = 1.6 mm, relative permittivityεr = 4.1 , and loss tangent tanδ = 0.015@1 GHz. FR4 with thickness ℎ =1.6 mm, relative permittivity 𝜀 =4.1, and loss tangent tan δ = The lower dielectric was composed of air with thickness ha = 3.2mm and relative permittivity equal to 0.015unity. @1GHz Section.3 The outlines lower how dielectric the height was of composed the antenna of inﬂuencedair with thickness bandwidth. ℎ =3.2 mm and relative permittivity equal to unity. Section 3 outlines how the height of the antenna influenced bandwidth. From the desired central frequency fc = 754 MHz, we estimated each patch dimension Wp and LP usingFrom the following the desired equations central [ 13frequency]: 𝑓 = 754 MHz, we estimated each patch dimension 𝑊 and 𝐿 using the following equations [13]: √2 W = √2 (4) 𝑊p = p 2 fc √ε0µ0 εr,eq + 1 (4) 2𝑓𝜀𝜇𝜀, +1 1 L𝐿P== −2∆𝐿,2∆L, (5) (5) 22f𝑓c√ε𝜀0µ𝜇0√𝜀εre f f − where ∆L is given by where ΔL is given by + Wp + 0.412h εre f f 0.3 𝑊h 0.264 =0.412ℎ(𝜀 +0.3) + 0.264 ∆L ℎ , (6) ∆𝐿 = Wp + , (6) εre f f 0.258 𝑊h 0.8 (𝜀 −− 0.258) +0.8 ℎ and h is the total height of the antenna, equal to hs + ha. All geometric parameters are listed in Table1. and h is the total height of the antenna, equal to ℎ +ℎ. All geometric parameters are listed in Table 1. Table 1. Geometric parameters of proposed double-dielectric microstrip antenna.

Frequency W L W L L h ε p p s a (MHz) r,eq (mm) (mm) (mm) (mm) (mm) (mm) 754 1.7 270 673 180 161.68 1.6 3.2 Appl. Sci. 2020, 10, x FOR PEER REVIEW 5 of 9

Table 1. Geometric parameters of proposed double-dielectric microstrip antenna.

Frequency W L 푾풑 푳풑 푳풔 풉풂 휺풓,풆풒 (MHz) (퐦퐦) (퐦퐦) (퐦퐦) (퐦퐦) (퐦퐦) (퐦퐦)

Appl. Sci. 2020, 10, x FOR PEER REVIEW754 1.7 270 673 180 161.68 1.6 3.2 5 of 9

2 Each of theseTable 1.three Geometric patches parameters had a size of pofroposed 푊푝 × d퐿ouble푝 = 180-dielectric× 161 .m7 icrostripmm . The antenna. overall size of the antenna was 푊 × 퐿 = 27.0 × 67.3 cm2. Frequency W L 푾풑 푳풑 푳풔 풉풂 Figure 2 shows simulated reflection휺풓,풆풒 coefficient 푆11 of the antenna. Assuming a return loss (RL) (MHz) (퐦퐦) (퐦퐦) (퐦퐦) (퐦퐦) (퐦퐦) (퐦퐦) greater than 10 dB as the well-accepted criterion of good impedance matching, the proposed antenna Appl.was Sci.well2020-matched, 10, 8640 at the 754central frequency1.7 270 of673 754 180 MHz 161.68 and in 1.6 channel 3.2 56 [750–758] MHz. 5 The of 9 simulated bandwidth was 17 MHz, obtained at 10 dB return loss. 2 Each of these three patches had a size of 푊푝 × 퐿푝 = 180 × 161.7 mm . The overall size of the The line length connecting patches had high impedance, and its length2 was equal to 164.0 mm. antennaEach was of these푊 × 퐿 three= 27. patches0 × 67.3 hadcm2 a. size of Wp Lp = 180 161.7 mm . The overall size of the In total, the electrical length of the line2 between each× consecutive× patch was equal to the wavelength antenna was W L = 27.0 67.3 cm . 푆 dividedFigure by 22, shows λ×/2. Therefore, simulated× thereflection input impedancecoefficient of11 theof thelinear antenna. array Assumingis equivalent a return to the loss parallel (RL) Figure2 shows simulated reﬂection coe ﬃcient S11 of the antenna. Assuming a return loss (RL) greaterconnection than of10 threedB as patchthe well antennas.-accepted A criterion λ/4 transformer of good impedance was then matching, used to convert the proposed antenna antenna input greater than 10 dB as the well-accepted criterion of good impedance matching, the proposed antenna wasimpedance well-matched into 50 Ω. at Thethe λcentral/4 transformer frequency line of had 754 a MHzlength and of 73 in mm channel and a 56width [750 of–758 24] mm. MHz. The The 50 was well-matched at the central frequency of 754 MHz and in channel 56 [750–758] MHz. The simulated simulatedΩ microstrip bandwidth line attached was 17to theMHz SMA, obtained connector at 10 had dB areturn width loss. of 17.5 mm. bandwidthThe line was length 17 MHz, connecting obtained patches at 10 dBhad return high impedance loss. , and its length was equal to 164.0 mm. In total, the electrical length of the line between each consecutive patch was equal to the wavelength divided by 2, λ/2. Therefore, the input impedance of the linear array is equivalent to the parallel connection of three patch antennas. A λ/4 transformer was then used to convert antenna input impedance into 50 Ω. The λ/4 transformer line had a length of 73 mm and a width of 24 mm. The 50 Ω microstrip line attached to the SMA connector had a width of 17.5 mm.

Figure 2. Simulated reﬂectionreflection coecoefficientﬃcient (S(S1111)) of of proposed proposed array antenna.

TheA gain line value length of connecting11.3 dBi was patches obtained had in high the impedance, direction of and the itsmaxim lengthal wasradiation, equal which to 164.0 was mm. a Inreasonable total, the valueelectrical since length the of linear the line array between was composed each consecutive of the patchthree wassingle equal patches. to the The wavelength gain in dividedfunction byof frequency 2, λ/2. Therefore, is shown the in Figure input impedance3. The proposed of the antenna linear array gain d isecrease equivalentd above to thechannel parallel 56, connectionas Figure 3 ofshows three. A patchrray efficiency antennas. was A λalso/4 transformersimulated, achieving was then 92% used at to the convert central antenna frequency, input as impedanceshown in Figure into 50 4.FigureΩ The. The 2.efficiency Simulatedλ/4 transformer peak reflection occurred line coefficient had at a754 length (S MHz11) of of proposedand 73 mm decrease and array ad widthantenna. at adjacent of 24 mm. channels. The 50 Ω microstrip line attached to the SMA connector had a width of 17.5 mm. A gain gain value value of of 11.3 11.3 dBi dBi was was obtained obtained in in the the direction direction of of the the maxim maximalal radiation, radiation, which which was was a reasonablea reasonable value value since since the the linear linear array array was was composed composed of the of three the three single single patches. patches. The gain The in function gain in functionof frequency of frequency is shown is in shown Figure 3in. TheFigure proposed 3. The antennaproposed gain antenna decreased gain abovedecrease channeld above 56, channel as Figure 563 , asshows. Figure Array 3 shows eﬃ.ciency Array wasefficiency also simulated, was also simulated achieving, achieving 92% at the 92% central at the frequency, central frequency, as shown as in shownFigure4 in. TheFigure e ﬃ 4.ciency The efficiency peak occurred peak atoccurred 754 MHz at and754 MHz decreased and decrease at adjacentd at channels. adjacent channels.

Figure 3. Simulated antenna gain over frequency.

Appl. Sci. 2020, 10, x FOR PEER REVIEW 6 of 9 FigureFigure 3. 3. SimulatedSimulated antenna antenna gain gain over over frequency. frequency.

FigureFigure 4. 4. SimulatedSimulated antenna antenna efficiency. eﬃciency.

5. Experimental Measurements The last step consisted of constructing the proposed antenna in order to validate our design. Thus, we checked all obtained simulation results and found the real antenna performance. The proposed double-dielectric linear array of three patch elements is illustrated in Figure 5. The bottom of Figure 5 shows the input SMA connector. This female connector was soldered to a 50 Ω microstrip line. The other parts that are also visible in the figure are the λ/4 transformer, the three radiating patches, and the connecting lines between patches. The base of substrate FR4 was raised with plastic M3 screws at a distance of 3.2 mm from the ground plane made of aluminum foil with 2.0 mm thickness. Experimental results of reflection coefficient S11 were obtained by measuring with a vector network analyzer (VNA), as shown in Figure 6. The center antenna frequency was clearly very close to 754 MHz, and at this frequency, S11 was less than –20 dB. We observed good agreement between the simulations and the measurements. However, from the practical measurements, we obtained a slightly greater bandwidth of about 25 MHz. The simulated and experimental radiation patterns obtained, in the anechoic chamber, of our proposed antenna are presented in Figure 7. The experimental and simulation diagrams were very similar. Nevertheless, there was a small difference of 7º in the radiation diagrams in the vertical plane. This very small rotation in the radiation diagram was caused by a misalignment between the transmitting and receiving antennas.

Figure 5. Construction of double-dielectric linear array.

The simulated and measured values of the gain were very similar, 11.3 and 10.5 dBi, respectively. Appl. Sci. 2020, 10, x FOR PEER REVIEW 6 of 9 Total Efficiency (%) Efficiency Total

Figure 4. Simulated antenna efficiency.

5. Experimental Measurements The last step consisted of constructing the proposed antenna in order to validate our design. Thus, we checked all obtained simulation results and found the real antenna performance. The proposed double-dielectric linear array of three patch elements is illustrated in Figure 5. The bottom of Figure 5 shows the input SMA connector. This female connector was soldered to a 50 Ω microstrip line. The other parts that are also visible in the figure are the λ/4 transformer, the three radiating patches, and the connecting lines between patches. The base of substrate FR4 was raised with plastic M3 screws at a distance of 3.2 mm from the Appl.ground Sci. 2020plane, 10 , 8640made of aluminum foil with 2.0 mm thickness. Experimental results of reflection6 of 9 coefficient S11 were obtained by measuring with a vector network analyzer (VNA), as shown in Figure5. Experimental 6. The center Measurements antenna frequency was clearly very close to 754 MHz, and at this frequency, S11 was less than –20 dB. We observed good agreement between the simulations and the measurements. However,The last from step the consisted practical ofmeasurements, constructing thewe proposedobtained a antenna slightly ingreater order bandwidth to validate of our about design. 25 Thus,MHz. weThe checked simulated all obtained and experimental simulation radiation results and patte foundrns the obtained, real antenna in the performance. anechoic chamber, The proposed of our proposeddouble-dielectric antenna linear are presented array of three in Figure patch 7. elements The experimental is illustrated and in Figuresimulation5. The diagrams bottom of were Figure very5 similar.shows the Nevertheless, input SMA connector. there was This a small female difference connector of was 7º solderedin the radiation to a 50 Ω microstripdiagrams in line. the The vertical other partsplane. that This are very also small visible rotation in the in ﬁgure the radiation are the λdiagram/4 transformer, was caused the three by a radiatingmisalignment patches, between and the transmittingconnecting lines and betweenreceiving patches. antennas.

Figure 5. ConstructionConstruction of double-dielectric linear array.

The basesimulated of substrate and measured FR4 was raisedvalues withof the plastic gain M3were screws very at similar, a distance 11.3 of and 3.2 mm10.5 fromdBi, respectively.the ground plane made of aluminum foil with 2.0 mm thickness. Experimental results of reflection coefficient S11 were obtained by measuring with a vector network analyzer (VNA), as shown in Figure6. The center antenna frequency was clearly very close to 754 MHz, and at this frequency, S was less than 20 dB. We observed good agreement between the simulations and the measurements. 11 − However, from the practical measurements, we obtained a slightly greater bandwidth of about 25 MHz. The simulated and experimental radiation patterns obtained, in the anechoic chamber, of our proposed antenna are presented in Figure7. The experimental and simulation diagrams were very similar. Nevertheless, there was a small difference of 7◦ in the radiation diagrams in the vertical plane. This very small rotation in the radiation diagram was caused by a misalignment between the transmitting and Appl. Sci. 2020, 10, x FOR PEER REVIEW 7 of 9 receiving antennas.

Figure 6. Figure 6. MeasuredMeasured and and simulated simulated S S1111 ofof proposed double-dielectric double-dielectric antenna. antenna.

(a) (b)

Figure 7. Simulated and measured radiation patterns. (a) 휙 = 90°; (b) 휙 = 0°.

The vertical/elevation plane (휙 = 0°) had half-power bandwidth (HPBW) of 30° and the horizontal/azimuth plane (휙 = 90°) had a HPBW of 65°. These results proved that the proposed antenna had fan-beam radiation characteristics. We also measured antenna input impedance through its input SMA connector, which was equal to ZA = 55.9–j10.4 Ω at the central frequency, as shown in the Smith chart in Figure 8. The Smith chart was drawn from 700 to 800 MHz. There was clear mismatching at frequencies away from the desired central frequency in the simulation and experimental curves.

Figure 8. S11 measurements in Smith chart.

Appl. Sci. 2020, 10, x FOR PEER REVIEW 7 of 9 (dB) 11 S

Appl. Sci. 2020, 10, x FOR PEER REVIEW 7 of 9

Figure 6. Measured and simulated S11 of proposed double-dielectric antenna. (dB) 11 S

Appl. Sci. 2020, 10, 8640 7 of 9 Figure 6. Measured and simulated S11 of proposed double-dielectric antenna.

(a) (b)

Figure 7. Simulated and measured radiation patterns. (a) 𝜙 = 90°; (b) 𝜙=0°.

The vertical/elevation (planea) (𝜙=0°) had half-power bandwidth(b) (HPBW) of 30° and the horizontal/azimuth planeFigureFigure (𝜙 7.7. Simulated = 90°) had and measureda HPBW radiationradiation of 65°. patterns.patterns. These (a) φ𝜙= results90 = ◦;(90°; b(b))φ 𝜙=0°=proved0◦. . that the proposed antenna had fan-beamTheThe simulated vertical/elevation radiation and measured characteristics. plane values (𝜙=0° of the) had gain half-power were very similar,bandwidth 11.3 and(HPBW) 10.5 dBi, of respectively.30° and the We alsohorizontal/azimuth measuredThe vertical /antennaelevation plane ( plane𝜙input = (90°φ ) impedancehad= 0 ◦a) HPBW had half-power ofthrough 65°. These bandwidth its results input proved (HPBW) SMA that ofconnector, the 30◦ proposedand the which was horizontalantenna had/azimuth fan-beam plane radiation (φ = 90◦) characteristics. had a HPBW of 65◦. These results proved that the proposed antenna equal to ZA = 55.9–j10.4 Ω at the central frequency, as shown in the Smith chart in Figure 8. The Smith had fan-beamWe also radiationmeasured characteristics. antenna input impedance through its input SMA connector, which was chart was drawn from 700 to 800 MHz. There was clear mismatching at frequencies away from the equalWe to also ZA = measured 55.9–j10.4 antenna Ω at the inputcentral impedance frequency, through as shown its in input the SMASmith connector, chart in Figure which 8.was The equalSmith desired centraltochart ZA frequency= was55.9–j10.4 drawn Ωfrominat the the 700 centralsimulation to 800 frequency, MHz. Thereand as shownexperimentalwas clear in the mismatching Smith curves. chart at in frequencies Figure8. The away Smith from chart the wasdesired drawn central from frequency 700 to 800 in MHz. the simulation There was and clear experimental mismatching curves. at frequencies away from the desired central frequency in the simulation and experimental curves.

Figure 8. S11 measurements in Smith chart.

6. Conclusions and FutureFigure Work 8. S11 measurements in Smith chart. This paper presented and discussed a novel practical digital TV antenna that is suitable for indoor environments. The antenna can be placed on a wall, for example, since it is very thin—about 6.8 mm including the aluminum foil. The upper dielectric is composed of FR4 with thickness hs = 1.6 mm, relative permittivity εr = 4.1, and tanδ = 0.015@1 GHz. The lower dielectric is composed of air, with thickness ha = 3.2 mm and relative permittivity equal to unity. The designed antenna has gain of 10.5 dBi, eﬃciency of 92%, and bandwidth of 25 MHz at the 10 dB return loss level. The proposed antenna overcomes the narrowband characteristics of common microstrip antennas, is mechanically robust, and can be connected to a coaxial cable without requiring any extra electronic components or devices such as baluns. Appl. Sci. 2020, 10, 8640 8 of 9

The proposed antenna has huge potential to be replicated and rearranged in an array. By adding electronic control to such an array, it is possible to perform smart beamforming in the azimuth plane, enabling new functionalities for DTV. An array of such antennas is quite useful for reducing self-interference caused by secondary transmitters in a single-frequency network (SFN) configuration. The antenna can also be spatially replicated to cover other applications in 5G broadcasting environments [14]. Another advantage of our proposed antenna is its bandwidth of 25 MHz, avoiding interference in the frequency domain. Although the antenna dimensions may seem large (67.3 by 27.0 cm), the experimental demonstration presented here proved the main concept. Further reduction in the dimensions can be achieved by using reflectors or metamaterials with magnetic response, which is reserved for future work. This antenna can be used to receive any modern digital TV signals such as those specified in DVB-T/T2, ISDB-T, or ATSC standards.

Author Contributions: Antenna research and design, A.N., P.M., T.V., and J.M.; simulation, P.M. and T.V.; theoretical formulation, J.M. and S.M.; laboratory and field tests, A.N. and P.M.; methodology, A.N. and J.M.; writing, A.N., T.V., J.M., and S.M.; writing—review and editing, A.N. and J.M.; coordination, A.N. All authors have read and agreed to the published version of the manuscript. Acknowledgments: This work is funded by FCT/MCTES through national funds and when applicable cofunded EU funds under the project UIDB/50008/2020-UIDP/50008/2020. Conflicts of Interest: The authors declare no conflict of interest.

References

1. Yeh, J.-T.; Liao, W.-J.; Chang, S.-H. Compact internal antenna for handheld devices with comprehensive DTV band coverage. IEEE Trans. Antennas Propag. 2014, 62, 3998–4007. [CrossRef] 2. Polak, L.; Kaller, O.; Klozar, L.; Prokopec, J. Influence of mobile network interfering products on DVB-T/H broadcasting services. In Proceedings of the IFIP Wireless Days (WD), Dublin, Ireland, 21–23 November 2012. 3. Van Trinh, T.; Kim, G.; Kim, J.; Jung, C.W. Wideband internal PIFA-loop antenna designed on the bezel of digital television applications for UHF band. IET Electron. Lett. 2018, 54, 1260–1262. [CrossRef] 4. Gamec, J.; Repko, M.; Gamcová, M.; Gladišová, I.; Kurdel, P.; Nekrasov, A.; Fidge, C. Low profile sinuous Slot antenna for UWB sensor networks. Electronics 2018, 8, 127. [CrossRef] 5. Huang, C.Y.; Jeng, B.M.; Yang, C.F. Wideband monopole antenna for DVB-T applications. IET Electron. Lett. 2008, 44, 1448–1450. [CrossRef] 6. Luadang, B.; Phongcharoenpanich, C. Unidirectional bowtie array antenna with incision gap for digital video broadcasting-T2 base station. IET Microw. Antennas 2015, 9, 1087–1095. [CrossRef] 7. Ballado, A.H.; Cruz, J.D.; Sejera, M.P.; Ingco, W.M.; Juan, C.M.; Nanao, D.N.; Navarro, J.A. Design of indoor microstrip TV antenna for integrated services digital broadcast—Terrestrial (ISDB-T) signal reception at 677 MHz. In Proceedings of the International Conference on Humanoid, Nanotechnology, Information Technology, Communication and Control, Environment and Management (HNICEM), Puerto Princesa Palawan, Philippines, 12–16 November 2014. 8. Pozar, D.M.; Kaufman, B. Increasing the bandwidth of a microstrip antenna by proximity coupling. IET Electron. Lett. 1987, 23, 368–369. [CrossRef] 9. Lau, N.U.; Sloan, R. Broadband millimeter-wave suspended microstrip antenna array. In Proceedings of the 2003 IEEE Antenna and Propagation Society International Symposium, Columbus, OH, USA, 22–27 June 2003. 10. Ikonen, P.M.; Maslovski, S.I.; Simovski, C.R.; Tretyakov, S.A. On artificial magneto-dielectric loading for improving the impedance bandwidth properties of microstrip antennas. IEEE Trans. Antennas Propag. 2006, 54, 1654–1662. [CrossRef] 11. Yoon, Y.J.; Kim, B. A New Formula for Effective Dielectric Constant in Multi-Dielectric Layer Microstrip Structure. In Proceedings of the IEEE 9th Topical Meeting on Electrical Performance of Electronic Packaging (Cat. No.00TH8524), Scottsdale, AZ, USA, 6 August 2000. 12. Yuan, T.; Yuan, N.; Li, L.W. A novel series-fed taper antenna array design. IEEE Antennas Wirel. Propag. Lett. 2008, 58, 362–365. [CrossRef] Appl. Sci. 2020, 10, 8640 9 of 9

13. Balanis, C.A. Antenna Theory, Analysis and Design: From Theory to Practice, 4th ed.; Wiley: Hoboken, NJ, USA, 2016. 14. Miao, W.; Luo, C.; Zhao, Z. Lightweight 3-D Beamforming Design in 5G UAV Broadcasting Communications. IEEE Trans. Broadcast. 2020, 66, 515–524. [CrossRef]

Publisher’s Note: MDPI stays neutral with regard to jurisdictional claims in published maps and institutional aﬃliations.

© 2020 by the authors. Licensee MDPI, Basel, Switzerland. This article is an open access article distributed under the terms and conditions of the Creative Commons Attribution (CC BY) license (http://creativecommons.org/licenses/by/4.0/).