Electrically Small Wideband Monopole Antenna Partially Loaded with Low Loss Magneto-Dielectric Material
←
→
Page content transcription
If your browser does not render page correctly, please read the page content below
Article Electrically Small Wideband Monopole Antenna Partially Loaded with Low Loss Magneto-Dielectric Material Aladdin Kabalan, Ala Sharaiha * and Anne-Claude Tarot Institut d’Electronique et des Technologies du numéRique (IETR), University of Rennes 1, 263 Avenue du Général Leclerc, 35700 Rennes, France; aladdin-kabalan@hotmail.com (A.K.); anne-claude.tarot@univ-rennes1.fr (A.-C.T.) * Correspondence: ala.sharaiha@univ-rennes1.fr Abstract: A miniaturized new topology of the planar monopole antenna using a Magneto-Dielectric Material (MDM) is proposed in this paper. The antenna element is realized by introducing slots partially covered by the MDM. We optimized and modified the MDM topology and dimensions to enhance the impact of this material on the planar monopole antenna, including slots in its structure. This new monopole shows a miniaturization rate of 60% of the antenna’s height (51 cm antenna’s height is miniaturized to 20 cm) by covering only 5% of the antenna surface by the MDM. The measured results show the antenna’s central working frequency of 130 MHz, while the bandwidth is 30% using a broadband matching circuit using the Real Frequency Technique (RFT). Keywords: magneto-dielectric materials; planar monopole antenna; miniaturization 1. Introduction Planar monopole antennas are widely used in many communication systems due to Citation: Kabalan, A.; Sharaiha, A.; their simple structure, compact size, ease of manufacture and implementation, especially Tarot, A.-C. Electrically Small in aeronautical applications [1–4] in the VHF and UHF frequency bands. The antenna size Wideband Monopole Antenna is relatively important in these bands, which causes an integration problem for most of the Partially Loaded with Low Loss applications in these bands. Therefore, a small antenna size is necessary and becomes a Magneto-Dielectric Material. critical problem in communication systems. Magnetism 2022, 2, 229–238. Among the various existing miniaturization techniques, the use of semi-massive https://doi.org/10.3390/ Magneto-Dielectric Materials (MDM) [5] is one of the most promising for reducing the magnetism2030017 antenna size. Instead of using a material with high permittivity, the same miniaturization Academic Editor: Paolo Baccarelli can be obtained by using MDM of the same refractive index n = (µr ε r )0.5 with moderate Received: 22 May 2022 permittivity and permeability avoiding highly concentrated field confinement, and the Accepted: 7 July 2022 medium is far less capacitive when compared to the dielectric-only high permittivity, which Published: 12 July 2022 results in low antenna efficiency and narrow-band operation [6,7]. On the other hand, antenna matching is difficult in a medium of high permittivity because of the high-quality Publisher’s Note: MDPI stays neutral factor value. with regard to jurisdictional claims in Since the work of Hansen and Burke [8], magneto-dielectric materials have been published maps and institutional affil- known to be more advantageous in antenna miniaturization and, in ultra-wideband antenna iations. applications, since the work of Volakis [9]. Many other papers have confirmed, at least in theory, that magneto-dielectric materials give larger bandwidths and also better efficiencies compared to dielectric ones [8,10–12]. Furthermore, the MDM should have sufficiently Copyright: © 2022 by the authors. low dielectric and magnetic losses tangents to ensure good performance of antennas made Licensee MDPI, Basel, Switzerland. of them. Recent papers have proposed a new hexaferrite structure [13] with a high FOM This article is an open access article (Figure of merit of the product µ0 , Q and the operating frequency f) which has a strong distributed under the terms and potential for low-loss high-frequency applications. conditions of the Creative Commons These last years, modern material manufacturing technology has made it possible to Attribution (CC BY) license (https:// design composite or nanocomposite substrates with different kinds of magnetic inclusions creativecommons.org/licenses/by/ mixed with dielectric host materials. Most of these MDM substrates were used to minia- 4.0/). turize planar antennas (patch, PIFA and IFA), with the ratio µr /ε r of 1.5 [12,14], 5.9 [15] Magnetism 2022, 2, 229–238. https://doi.org/10.3390/magnetism2030017 https://www.mdpi.com/journal/magnetism
inclusions mixed with dielectric host materials. Most of these MDM substrates were used to miniaturize planar antennas (patch, PIFA and IFA), with the ratio ⁄ of 1.5 [12,14], 5.9 [15] and 2.1 [16], and loss tangents (tanδε,μ ) less than 0.1 for a frequency band up to 700 MHz. Showing a better performance compared to commercial dielectric materials. Magnetism 2022, 2 In [17], we studied the effect of MDM composed of NiZnCo ferrite, which in the VHF 230 band has a higher permeability ( = 16) than the permittivity ( = 12), on the miniatur- ization of a planar monopole antenna in the VHF band 100–200 MHz. We achieved an optimum miniaturization rate of 15%, covering only 12% of the total antenna size where and 2.1 [16], and loss tangents (tan δε,µ ) less than 0.1 for a frequency band up to 700 MHz. the intensity of the surface current is maximum. In order to increase the magnetic impact Showing a better performance compared to commercial dielectric materials. of the MDM, we propose, in this paper, modifying the monopole topology by introducing In [17], we studied the effect of MDM composed of NiZnCo ferrite, which in the slots and by optimizing the size and the position of the MDM, as well as its geometry. VHF band has a higher permeability (µr = 16) than the permittivity (ε r = 12), on the Then the effect of the air gap, due to the sheet metal thickness when using Planar MDM miniaturization of a planar monopole antenna in the VHF band 100–200 MHz. We achieved (P-MDM), on the magnetic field distribution was investigated. an optimum miniaturization rate of 15%, covering only 12% of the total antenna size where The paper is organized in the following way: after briefly presenting the fabricated the intensity of the surface current is maximum. In order to increase the magnetic impact MDM and the reference antenna in Section 2, we presented the principal results of the of the MDM, we propose, in this paper, modifying the monopole topology by introducing antenna miniaturization using a small portion of P-MDM and introduced slots in Section slots and by optimizing the size and the position of the MDM, as well as its geometry. 4. Then in Section 5, we presented a parametrical study, where the antenna geometry and Then the effect of the air gap, due to the sheet metal thickness when using Planar MDM the dimensions (P-MDM), of magnetic on the the P-MDMfieldwere optimized, distribution wasand then in Section 6, a first prototype investigated. using The P-MDM was designed and validated by measurements. paper is organized in the following way: after briefly Finally, Section presenting the7fabricated presents some concluding remarks. MDM and the reference antenna in Section 2, we presented the principal results of the antenna miniaturization using a small portion of P-MDM and introduced slots in Section 4. 2.Then MDM Description in Section 5, we presented a parametrical study, where the antenna geometry and the An MDMofwas dimensions developed the P-MDM with were the objective optimized, andof obtaining then in Sectiona high 6, a permeability first prototypewith a using low loss inwas P-MDM thedesigned frequencyandband of 100–200 validated MHz. The MDM by measurements. basedSection Finally, on a ferrite Ni-Zn-Co 7 presents some was elaborated concluding by the Lab-STICC laboratory, and the method is presented in [5]. remarks. The relative permeability (μr) and permittivity (εr) spectrum, as well as the magnetic 2. MDM and Description dielectric loss tangent in the frequency band of 100–200 MHz, are shown in Figure 1a,b, respectively. An MDM was developed with the objective of obtaining a high permeability with a lowItloss wasinobserved that between the frequency band of100 and 200 100–200 MHz.MHz,Thethe MDM ′ / ′ on ratiobased is greater than a ferrite unity Ni-Zn-Co wasvaries and elaborated by the between 1.46Lab-STICC and 1.37. laboratory, and the Magnetic losses aremethod is presented low, between in [5]. 0.03 and 0.05. This propertyTheisrelative permeability related to (µr ) dimensions the nanometric and permittivity (εr ) that of grains spectrum, as wellduring are preserved as the heat mag- netic and dielectric loss tangent in the frequency band of 100–200 MHz, are shown treatment. in Figure 1a,b, respectively. (a) (b) Figure Figure1.1.Manufactured ManufacturedMDM MDMspectrum spectrumininthe theband band1010MHz–6 MHz–6GHz. GHz.(a)(a)Typical Typicalmeasured measuredrelative relative permeability and permittivity. (b) Magnetic and dielectric loss tangents. permeability and permittivity. (b) Magnetic and dielectric loss tangents. 3. Geometry of the Planar It was observed Monopole that between 100 Antenna and 200 MHz, the ratio µ0 /ε0 is greater than unity and varies Anbetween 1.46 and (UWB) ultra-wideband 1.37. Magnetic planar losses are low, monopole between antenna 0.03 and 0.05. is designed This property to operate in the is related VHF bandto(118–156) the nanometric MHz, dimensions of grains that with the dimensions are preserved shown in Figureduring heat treatment. 2a (antenna A1); the other dimensions W2 = 5 mm and = 40° are optimized for a larger bandwidth. Figure 2b 3. Geometry shows of the the antenna Planar input Monopole impedance Antenna with a bandwidth of 31.5% (for VSWR < 2.5) around An ultra-wideband (UWB) planar monopole its resonance frequency f0 = 130 MHz. We assumed in antenna is designed our study to operate that the ground in of plane the VHF band (118–156) MHz, with the dimensions the monopole antenna has infinite dimensions. shown in Figure 2a (antenna A1); the ◦ other dimensions W2 = 5 mm and θ = 40 are optimized for a larger bandwidth. Figure 2b shows the antenna input impedance with a bandwidth of 31.5% (for VSWR < 2.5) around its resonance frequency f 0 = 130 MHz. We assumed in our study that the ground plane of the monopole antenna has infinite dimensions.
The planar monopole antenna was chosen for our project of aircraft application, which is the suitable type of antenna for the VHF band. The scope statement goal of re- ducing the antenna height from 51 cm to 20 cm corresponds to a miniaturization rate of 60.7%. The monopole antenna with 20 cm of height called “ARef” is shown in Figure 2a, and its input impedance is represented on the Smith chart in Figure 2b. The size of the Magnetism 2022, 2 antenna A1 is in the order of ⁄4.5 at the frequency 130 MHz, while the reference antenna231 Aref must be miniaturized up to ⁄11.5. (a) (b) Figure Figure2. 2.(a)(a) Simple monopole Simple antenna monopole A1A1 antenna ofof the height the heightofof 510 mm, 510 mm,and the and thereference referencemonopole monopole antenna ARef of the height of 200 mm. (b) Smith chart of the antenna’s impedances in the band 118 antenna ARef of the height of 200 mm. (b) Smith chart of the antenna’s impedances in the band 118 to to 156 MHz. 156 MHz. 4. Miniaturization of the Monopole The planar monopole Antenna antenna was chosenUsing a P-MDM for our project of aircraft application, which In this is the section, suitable typewe of study antenna thefor effect the of VHFtheband. PlanarThe MDM (P-MDM) scope statement to shift goal the resonance of reducing the frequency of the reference antenna height from 51 cm antenna to 20 cmARefcorresponds , consideringtofor the moment thatrate a miniaturization the of metal thick- 60.7%. The ness monopole antenna with of the antenna is zero. 20 cm of height called “ARef ” is shown in Figure 2a, and its input We showed impedance in previouson is represented workthe [17] Smiththat it is in chart sufficient to cover Figure 2b. the regions The size close to A1 of the antenna theis source in thewhere order the intensity of λ/4.5 at theoffrequency the magnetic 130field MHz, is while the highest to obtain the reference an optimum antenna Aref mustmin-be iaturization miniaturized rate. upTherefore, to λ/11.5.by covering the reference antenna partially with the P-MDM (3 mm thick, 60 mm height) on both sides (see Figure 3a), antenna A2 is obtained. The 4. Miniaturization latter resonates at 176ofMHz the Monopole lowering the Antenna frequency Using bya only P-MDM about 35%. We should em- phasize Inthat thisour section, we study objective was tothe effect130 attain of MHz the Planar without MDM (P-MDM) increasing the to shift size the an- of the reso- nanceThe tenna. frequency P-MDMofthickness the reference antenna is limited to A 3 Ref mm , considering because of for the moment technical reasons,thatknowing the metal thickness that the material antenna t ahas of the thickness is zero. a direct effect on the effective medium (effective permea- We showed in previous bility and permittivity). In our application,work [17] that the itincrease is sufficient in theto cover the P-MDM regionsleads thickness closetoto the source where the intensity of the magnetic field is the highest an effective permeability more important. However, the objective is to have the desired to obtain an optimum miniaturization miniaturization rate.of Therefore, effect the P-MDMby covering with the minimumthe reference quantity antenna partially with the of the material. P-MDM In order(3 mm thick, 60high to achieve mm height) on both sides miniaturization rates,(see weFigure 3a), antenna increased the impact A2 isofobtained. the P- The latter resonates at 176 MHz lowering the frequency MDM. For this, we proposed the antenna A3 structure with the inclusion of two open- by only about 35%. We should emphasize that our objective was to attain 130 MHz without ended slots in this area (see Figure 3b). These two slots, with a length of 50 mm and increasing the size of the a antenna. The P-MDM thickness is limited to 3 mm because of thickness of 5 mm, serve to concentrate the surface currents in the area of intersection technical reasons, knowing that thethem between material thickness (Figure has a direct 3c), which effect on leads consequently the effective mediumin to an increase (effective permeability the magnetic field and permittivity). In our application, the increase in the P-MDM intensity in the zone covered by the P-MDM. Thus, an increase in the effective permeabil- thickness leads to an ityeffective permeability of the medium more providing is obtained, important.a higher However, the objectiverate. miniaturization is to have the desired miniaturization effect of the P-MDM with the minimum The effective medium can be calculated numerically by simulation, quantity of the material. by determining In order to achieve high miniaturization rates, we increased the shift of the resonance frequency caused by the material, using the following the impact of the P-MDM. equation: For this, we proposed the antenna A3 structure with the inclusion of two open-ended slots in this area (see Figure 3b). These two slots, with a length of 50 mm and a thickness of 5 mm, serve to concentrate the surface currents in the area of intersection between them (Figure 3c), which consequently leads to an increase in the magnetic field intensity in the zone covered by the P-MDM. Thus, an increase in the effective permeability of the medium is obtained, providing a higher miniaturization rate.
the permittivity of P-MDM does not have any effect on the resonant frequency of the an- tenna, and the effective permittivity is neglected. Magnetism 2022, 2, FOR PEER REVIEW For a of about 20, the effective permeability increases from about 2.28 for the an- 4 tenna without slots A2 to 4.6 for the antenna with the two slots A3, respectively (see Figure Magnetism 2022, 2 4). Consequently, the resonance frequency decreases from 175 MHz to 130 MHz, increas- 232 ing the miniaturization rate to 60.7% (see Figure 5). 0, 0, = (1) √ where 0, and 0, are the resonance frequencies of the antenna without and with MDM, respectively. are the effective permittivity and permeability, respectively. In the regions close to the source, the intensity of the electric field is the lowest; therefore, the permittivity of P-MDM does not have any effect on the resonant frequency of the an- tenna, and the effective permittivity is neglected. For a of about 20, the effective permeability increases from about 2.28 for the an- tenna without slots A2 to 4.6 for the antenna with the two slots A3, respectively (see Figure 4). Consequently, the resonance frequency decreases from 175 MHz to 130 MHz, increas- ing the miniaturization rate to 60.7%(b)(see Figure 5). (a) (c) Figure Figure 3. 3. Monopole Monopole antenna antenna miniaturized miniaturized byby thethe P−P−MDM: MDM: (a)(a)without withoutslots slots(A2) (A2)and and(b)(b) with with two two slots of 50 mm × 5 mm (A3). (c) Density of the surface currents (A/m), which corresponds to the slots of 50 × 5 mm (A3). (c) Density of the surface currents (A/m), which corresponds to the intensity intensity of the magnetic field. of the magnetic field. The effective medium can be calculated numerically by simulation, by determining the shift of the resonance frequency caused by the material, using the following equation: f 0,air f 0,MDM = √ (1) ε e f f µe f f where f 0,air and f 0,MDM are the resonance frequencies of the antenna without and with MDM, respectively. ε e f f µe f f are the effective permittivity and permeability, respectively. In the regions close to the source, the intensity of the electric field is the lowest; therefore, the (a) permittivity of P-MDM does not(b) have any effect on the resonant frequency (c) of the antenna, and the effective permittivity is neglected. Figure 3. For Monopole a µr ofantenna about 20,miniaturized the effectiveby permeability the P−MDM: (a) withoutfrom increases slotsabout (A2) and 2.28(b) forwith the two of 50 mm slotsantenna × 5 mm (A3). (c) Density of the surface currents (A/m), which corresponds without slots A2 to 4.6 for the antenna with the two slots A3, respectively (see to the intensity of Figure the Figure4). magnetic 4. Effective field. Consequently, mediumthe resonance in function frequency of the decreases from relative permeability 175miniaturized for both MHz to 130antennas: MHz, increasing the (A2) without slots miniaturization and with tworate slotsto 60.7% (see Figure 5). (A3). Figure Figure 4. Effective 4. Effective medium medium inin functionofofthe function the relative relative permeability permeabilityfor both for miniaturized both antennas: miniaturized antennas: without without slotsslots (A2)(A2) andand with with twoslots two slots(A3). (A3).
Magnetism 2022, 2, FOR PEER REVIEW Magnetism 2022, 2 233 Figure 5. Imaginary parts of the input impedance of 3 antennas: ARef, A2 and A3. 5. Parametrical Study Figure In this section, we studied the impact of the size of the P-MDM used (WMD 5.Imaginary Figure 5. Imaginary parts parts ofinput of the the input impedance impedance of 3 antennas: of 3 antennas: AA3. ARef , A2 and Ref, A2 and A3. HMDM ) as well as the slot length Lslot and height Hslot on the resonance frequency 5. Parametrical Study the 5. corresponding Parametrical Studyμeff (see Figure 6). In this section, we studied the impact of the size of the P-MDM used (WMDM and The initial parameters are the following: WMDM = 96 mm, hMDM = 20 mm, Ls HMDM In) as this section, well welength as the slot studied the height Lslot and impact ofonthe Hslot thesize of thefrequency resonance P-MDMf 0 used and (WMD mm, H slot = 5 mm. Hthe corresponding µeff (see MDM ) as well as the Figure slot 6). Lslot and height Hslot on the resonance frequency length the corresponding μeff (see Figure 6). The initial parameters are the following: WMDM = 96 mm, hMDM = 20 mm, Lsl mm, Hslot = 5 mm. Figure 6. Parametrical study of the Antenna A3 with the parameters: H MDM MDM slot , W slot,L and H . Figure 6. Parametrical study of the Antenna A3 with the parameters: HMDM , WMDM , Lslot an The initial parameters are the following: WMDM = 96 mm, hMDM = 20 mm, Lslot = 50 mm, 5.1. HslotOptimization = 5 mm. of the Position and the Dimensions of the P-MDM Figure Here, we considered H 5.1. Optimization of the Position 6. Parametrical study ofand MDM the as the height of the P-MDM. Several values of the theAntenna Dimensions A3ofwith the P-MDM the parameters: HMDM , WMDM , Lslot an wereHere, tested webetween considered 20 Hmm and 55 mm when the other dimensions were fixed wi MDM as the height of the P-MDM. Several values of the =H50 mm, 5.1.MDM wereH tested Optimization = 5 mm between and slot of the Position 20 mmWand = 96 andthe MDM 55 mmmm (Figure when Dimensions 7a). theofother the It is shown dimensions P-MDM in fixed were Figure 7b t with Lslot = rate reduction 50 mm,of fH0 slot as =a 5function mm and W ofMDM HMDM= 96decreased mm (Figureonly 7a). Itby is shown aboutin2%, Figure 7b correspondi Here, that the reduction rate of f0 as aH we considered MDM as function ofthe height HMDM of the decreased P-MDM. only by aboutSeveral values of the 2%, correspond- Δμeff of 0.35. were a ∆µ between ing totested of 0.35. 20 mm and 55 mm when the other dimensions were fixed wit Next,effwe varied the width WMDM from 8 mm to 96 mm with a constant height o = 50 mm, Hslot = 5 mm and WMDM = 96 mm (Figure 7a). It is shown in Figure 7b t = 20 mm (Figure 8a) and keeping L and Hslot the same. We can see in Figure reduction rate of f0 as a function of Hslot MDM decreased only by about 2%, correspondi at WMDM = 40 mm, the reduction rate of f0 is equal to 41% and converges slowly to 4 Δμeff of 0.35. the maximum width WMDM = 96 mm. Next, we varied the width WMDM from 8 mm to 96 mm with a constant height o = 20 mm (Figure 8a) and keeping Lslot and Hslot the same. We can see in Figure at WMDM = 40 mm, the reduction rate of f0 is equal to 41% and converges slowly to 4 the maximum width WMDM = 96 mm.
Magnetism 2022, 2, FOR PEER REVIEW 6 Magnetism 2022, 2 234 Magnetism 2022, 2, FOR PEER REVIEW 6 (a) (b) Figure 7. (a) Figure 7. Monopole antenna (a) Monopole of the height antenna of the200 mm with height 200 two mmslots withof two the width slots 50 of mm. the (b) Reduc- width 50 mm. tion rate of f0 and μeff in function of P-MDM height HMDM with a constant WMDM = 96 mm. (b) Reduction rate of f 0 and µ eff in function of P-MDM height HMDM with a constant WMDM = 96 mm. (a) Next, we varied the width WMDM from 8 mm to(b)96 mm with a constant height of HMDM Figure 7. (a) = 20 mm (Figure Monopole antenna 8a) and of the keeping height Lslot 200 mm andtwo with Hslot the slots of same. We50can the width mm. see(b)inReduc- Figure 8b thatofatf0W tion rate and MDMμeff= in 40function mm, the ofreduction rate H P-MDM height f 0 is ofMDM equal with to 41%WMDM a constant and converges = 96 mm. slowly to 49% for the maximum width WMDM = 96 mm. (a) (b) Figure 8. (a) Monopole antenna of the height 200 mm with two slots of the width 50 mm. (b) Reduc- tion rate of f0 and μeff in function of P-MDM width WMDM with a constant HMDM = 20 mm. 5.2. Slots Dimensions Effect on the Miniaturization Rate (a) Here, Lslot varies between 10 mm and 90 mm (Figure (b)9a) with a constant H slot of 5 mm. Figure The 8. (a) P-MDM Monopole size antenna Figure 8. (a) Monopole antenna ofMDM is fixed of the to W height 200 =mm4 cmwithand the height 200 mmMDM two H slots of= 2 the cm. width We 50 noticed mm. with two slots of the width 50 mm.(b) that Reduc- the reduction tion rate of f 0 rate and (b) Reduction rate increases μ in slowly function eff of f and µ of until P-MDM the slot width length W MDM of with 40a mm, constant where H MDM the= intersection 20 mm. be- 0 eff in function of P-MDM width WMDM with a constant HMDM = 20 mm. tween the two slots begins. For Lslot > 40 mm, the effective medium increased quickly to 5.2. Slots 5.2.the reach Dimensions SlotsvalueDimensions Effect of 10.4, on the Effect onMiniaturization corresponding to a 69% Rate the Miniaturization size Rate reduction for Lslot = 70 mm. We noticed thatHere, theHere, reduction Lslot Lvaries slot rate varies increases between between 10 more 10 mm mm andslowly and90 90 mm for mm L > 70,9a) (Figure (Figure slot where 9a) withthe with slots exceed aaconstant constant H Hslot the of 55P- slot of mm. MDM The border. P-MDM size mm. The P-MDM size is fixed toMDM is fixed to W WMDM = 4 cm and = 4 cm and H MDM HMDM = 2 cm. We noticed that the = 2 cm. We noticed that the reduction rate reduction Therefore, increases we rate increases obtained slowly slowly high until the until miniaturization slot length the slotof 40 mm, length rates 40 by of where mm, increasing the theLslot intersection where more than between intersection be-40two the Magnetism 2022, 2, FOR PEER REVIEW mm. tween slotsThis thebegins. is because two slots we For begins. increased Lslot > 40For mm, Lslotthe the surface effective > 40 mm, the area medium of effective the increasedintersection medium quickly between to reach increased the the value quickly to 7 of two slots, reach 10.4, and the therefore corresponding value of 10.4, thecorresponding tointensity a 69% size of the tomagnetic reduction a 69% for sizefield slot in Lreduction = 70 themm.P-MDM for We =was Lslotnoticed 70 mm. increased, that Wethe which reduction noticed allows that rate for the reduction modifying increases rate the more slowlyeffective increases Lmedium. formore slot > 70, slowly We where for found L the slot > a similar slots 70, exceed where effect the the when P-MDM slots we exceed varied border. the the P- H MDM between slot border. 1 and 10 mm with a constant L slot of 90 mm (Figure 9b), the μ eff increases until the slot height Therefore, of 10 mm, we obtained high where the slots exceed miniaturization ratesthe by P-MDM. increasing Lslot more than 40 mm. This Regarding is because thesewe results, increasedwe suggest the surface reducingarea theof the sizeintersection of the P-MDM betweento 4 cm the×two2 cm to cover slots, and only therefore the region of the intersection the intensity of the magnetic between fieldthein thetwoP-MDMslots. We obtained was increased, the antenna which A4 (see allows forFigure modifying 10a). theIn the realized effective antenna, medium. We we had aa problem found similar effect withwhenan airwe gapvaried of about the 1 Hslot between 1 and 10 mm with a constant Lslot of 90 mm (Figure 9b), the μeff increasesof mm between the MDM and the antenna surface. This air gap can reduce the intensity the magnetic until the slot height field of inside 10 mm, the where MDM,the andslots as aexceed consequence, the P-MDM. that reduces the miniaturiza- tionRegarding rate. Therefore, to compensate for the effect these results, we suggest reducing the size of the of the air gap onP-MDM the miniaturization to 4 cm × 2 cm rate, towe cover hadonlyto adjust the regionthe dimensions of the intersection of the slot to increase between the two theslots. intensity of the magnetic We obtained the antenna field A4inside (a) (see the Figure MDM. 10a). InInorder the to have realized the antenna,resonance we had at a 130 MHz, problem (b) we with had an to air increase gap of the about slot 1 dimensions mm between the MDM to L slot = 93 mm (instead of 50 and the antenna surface. This air gap mm) and H = 10 mm (instead slot can reduce the intensity of of 5 mm), Figure which 9. 9. (a)(a) increased Reductionthe μrate ofoff0 f and value μµeff This in in function of of the theslot’s length A5LLwith slot.. (b) (b) Reductionfre- rate of the Figure magnetic Reduction field inside effrate the MDM, 0to 4.34. and and effas gave us afunction consequence, antenna slot’s thatlength reduces slotthe aminiaturiza- resonant Reduction rate of f0 and μeff in function of the slot’s height Hslot . quency tion 0 andof frate. 130 Therefore, µeff inMHz to(see function ofFigure the slot’s compensate 10b). height for theHeffect slot . of the air gap on the miniaturization rate, we had to adjust the dimensions of the slot to increase the intensity of the magnetic field inside the MDM. In order to have the resonance at 130 MHz, we had to increase the slot dimensions to Lslot = 93 mm (instead of 50 mm) and Hslot = 10 mm (instead of 5 mm), which increased the μeff value to 4.34. This gave us the antenna A5 with a resonant fre-
Magnetism Magnetism 2022, 2022, 2, FOR2 PEER REVIEW 7 235 2022, 2, FOR PEER REVIEW 7 Therefore, we obtained high miniaturization rates by increasing Lslot more than 40 mm. This is because we increased the surface area of the intersection between the two slots, and therefore the intensity of the magnetic field in the P-MDM was increased, which allows for modifying the effective medium. We found a similar effect when we varied the Hslot between 1 and 10 mm with a constant Lslot of 90 mm (Figure 9b), the µeff increases until the slot height of 10 mm, where the slots exceed the P-MDM. Regarding these results, we suggest reducing the size of the P-MDM to 4 cm × 2 cm to cover only the region of the intersection between the two slots. We obtained the antenna A4 (see Figure 10a). In the realized antenna, we had a problem with an air gap of about 1 mm between the MDM and the antenna surface. This air gap can reduce the intensity of the (a) (b) magnetic field inside the MDM, and as a consequence, that reduces the miniaturization rate. Figure 9. (a) Reduction Therefore, rate of f0for to compensate andtheμeff in function effect of the air of gap the slot’s on the length Lslot . (b) Reduction miniaturization rate, werate had to of f0 adjust and μeff theindimensions function of theof slot’s height the slot Hslot . to increase the intensity of the magnetic field inside the (a) (b) MDM. In order to have the resonance at 130 MHz, we had to increase the slot dimensions to Lslot = rate Figure 9. (a) Reduction 93 mm of f(instead 0 and μof eff 50 inmm) and H function the=slot’s ofslot 10 mm (instead length Lslotof 5 mm), . (b) which rate Reduction increased of f0 and μeff the µeff value in function of to the4.34. This slot’s gaveHus height slotthe . antenna A5 with a resonant frequency of 130 MHz (see Figure 10b). (a) (b) Figure 10. (a) Miniaturized monopole antenna A4 by the P-MDM of dimensions 4 × 2 cm with two slots of length 51.3 mm. (b) Miniaturized monopole antenna A5 by the P-MDM with two slots of dimensions 93 × 10 mm. (a) (b) 6. Realization and Experimental Validation Figure 10. (a) Miniaturized Figure The antenna monopole antenna 10. (a) Miniaturized A5 was realized A4antenna monopole with 1by mmthethick P-MDM theof A4 bybrass dimensions P-MDM plate 4 × 2 cm of dimensions and measured × with 4using 2 cmatwo with two 1-m slots of length 51.3 slots mm. of (b) length Miniaturized 51.3 mm. (b) monopole Miniaturized antenna monopole A5 by the antenna P-MDM A5 by the with P-MDM diameter ground plane. The two P-MDM of the antenna A5 were integrated into two plas-two slots with two ofslots of dimensions 93 × 10 mm. dimensions 93 × 10 mm. tic supports to tighten the plates as much as possible on the antenna (see Figure 11). Figure 6. 12 shows Realization theExperimental and imaginary and real parts of the measured and simulated imped- Validation 6. Realization and ance of theExperimental Validation antenna A5. There is a good agreement between the results in the band 118– The antenna A5 was realized with 1 mm thick brass plate and measured using a 1-m 156 MHz.A5 The antenna The measured antenna 1resonates almost at plate the same resonance frequency found diameterwas realized ground plane.with The twomm thick P-MDM ofbrass the antenna and measured A5 were using integrated intoa two 1-mplastic in the diameter ground simulation, plane. which The two is around P-MDM 130 of MHz. the antenna A5 were integrated into two plas- supports to tighten the plates as much as possible on the antenna (see Figure 11). tic supports to tighten the plates as much as possible on the antenna (see Figure 11). Figure 12 shows the imaginary and real parts of the measured and simulated imped- ance of the antenna A5. There is a good agreement between the results in the band 118– 156 MHz. The measured antenna resonates almost at the same resonance frequency found in the simulation, which is around 130 MHz. (a) (b) Figure 11. (a)11. Figure planar monopole (a) planar A5 realized monopole with with A5 realized two P-MDM. (b) Slots two P-MDM. and two (b) Slots and P-MDM. two P-MDM. Figure 12 shows the imaginary and real parts of the measured and simulated impedance of the antenna A5. There is a good agreement between the results in the band 118–156 MHz. (a) (b) Figure 11. (a) planar monopole A5 realized with two P-MDM. (b) Slots and two P-MDM.
Magnetism 2022, 2 236 (a) (b) (a) (b) Magnetism 2022, 2, FOR PEER REVIEW 8 Figure 12. Imaginary (a) and real (b) parts of the measured and simulated input impedance of the Figure 12. Imaginary (a) and real (b) parts of the measured and simulated input impedance of the antenna A5. The measured antenna resonates almost at the same resonance frequency found in the antenna A5. simulation, which is around 130 MHz. Therefore, a 60.7% of size reduction can be obtained using P-MDM. However, the use Therefore, a 60.7% of size reduction can be obtained using P-MDM. However, the use of a matching circuit is required to tune the input reactance to zero. We matched the an- of a matching circuit is required to tune the input reactance to zero. We matched the an- tenna’s input impedance to cover 118–156 MHz in the VHF band for VSWR less than 3.5 tenna’s input impedance to cover 118–156 MHz in the VHF band for VSWR less than 3.5 (S11 < −5 dB). In order to achieve this goal, we proposed using the « Real Frequency Tech- (S11 < −5 dB). In order to achieve this goal, we proposed using the « Real Frequency Tech- nique (RFT) [18] » to design the broadband matching circuits shown in Figures 13 and 14. nique (RFT) [18] » to design the broadband matching circuits shown in Figures 13 and 14. The matching circuit of the antenna A5 was made on an FR4 substrate of a thickness of 1.6 The matching circuit of the antenna A5 was made on an FR4 substrate of a thickness of 1.6 mm and using SMD components of the Murata series [19]. mm and using SMD components of the Murata series [19]. Then, by connecting the antenna to port 2 of the matching circuit, the reflection coef- Then, by connecting the antenna to port 2 of the matching circuit, the reflection coef- ficient S11 was measured at the circuit input and compared with the simulation result in ficient (a) S11 was measured at the circuit input and compared (b) with the simulation result in Figure 15. As the individual component tolerances have a profound effect on the overall Figure 15. Figure12. As the individual 12.Imaginary Imaginary (a)the andreal component real(b) (b)parts parts tolerances have a profound effect onofthetheoverall circuit Figure performance, (a) and shaded ofofthe curves the measured show measuredthe andand simulated potential simulated input errors input impedance ofimpedance the reflection of the coeffi- circuit antenna performance, A5. the shaded curves show the potential errors of the reflection coeffi- cient S11 antenna A5.due to the component tolerances. Therefore, a smaller error between measured cient S11 due to the component tolerances. Therefore, a smaller error between measured and simulated Therefore,aresults60.7% can a 60.7% bereduction of size obtainedcan using components bebeobtained using with smaller P-MDM. tolerance However, usevalues. thethe and simulated Therefore, results can of be reduction size obtained can using components obtained using with P-MDM.smaller tolerance However, values. Inofaddition, a matching thecircuit realizedis gain oftothe required tunematched the input antenna reactance A5towaszero.validated We by the matched the measure- an- use of a matching circuit is required to tune the input reactance In addition, the realized gain of the matched antenna A5 was validated by the measure- to zero. We matched the ments tenna’s antenna’swith input a impedance input good agreement impedance totocover between cover 118–156 118–156 the MHzMHzresults, in in thethe VHFas VHF shown band bandforin for Figure VSWR VSWR 16. less This than less 3.5gained than ments with a good agreement between the results, as shown in Figure 16. This gained respect (S11 3.5 (S11 for < −5< − the dB). 5 scope In dB). order In statement to achieve order to goals achieve thisof goal, this goal, our we project. proposed we The proposedusingsimulated the «the using Real radiation « Frequency Real pattern Tech- of the Frequency respect for the scope statement goals of our project. The simulated radiation pattern of the matched nique (RFT) Technique antenna (RFT) »A5 [18][18] isdesign shown totodesign the inbroadband Figure 17. thebroadband We noticed matching matching circuitsthat circuits the highly shown miniaturized in Figures Figures 13 and 13 and14.14. mon- matched Thematching antenna matching A5 circuit is shown of the antenna in Figure A5 17. We noticed that the highly miniaturized mon- opole The antennacircuit A5 maintains of A5 waswasmade its omnidirectional madeon ananFR4 FR4substrate radiation on ofof performance substrate a thickness well. of of a thickness 1.6 opole mm 1.6 mm antenna andandusing using A5 SMD maintains SMD components components its omnidirectional of of thetheMurata Murata radiation series series[19]. [19]. performance well. Then, by connecting the antenna to port 2 of the matching circuit, the reflection coef- ficient S11 was measured at the circuit input and compared with the simulation result in Figure 15. As the individual component tolerances have a profound effect on the overall circuit performance, the shaded curves show the potential errors of the reflection coeffi- cient S11 due to the component tolerances. Therefore, a smaller error between measured and simulated results can be obtained using components with smaller tolerance values. In addition, the realized gain of the matched antenna A5 was validated by the measure- ments with a good agreement between the results, as shown in Figure 16. This gained Figure 13. Matching Figure 13. Matchingcircuit circuit ofof thethe miniaturized miniaturized monopole monopole antenna antenna A5 forA5 thefor the118–156 band band 118–156 MHz. MHz. Figure 13.for respect Matching circuit the scope of statementthe miniaturized goals monopole of our project. antenna The simulated A5 for the band radiation 118–156 pattern of the MHz. matched antenna A5 is shown in Figure 17. We noticed that the highly miniaturized mon- opole antenna A5 maintains its omnidirectional radiation performance well. Figure 14. Realized matching circuit of the antenna A5. Figure 14. Realized matching circuit of the antenna A5. Figure 14. Realized matching circuit of the antenna A5. Then, by connecting the antenna to port 2 of the matching circuit, the reflection coefficient S11 was measured Figure 13. Matching at miniaturized circuit of the the circuit input and compared monopole antenna A5with theband for the simulation 118–156 result MHz. in Figure 15. As the individual component tolerances have a profound effect on the overall circuit performance, the shaded curves show the potential errors of the reflection coefficient S11 due to the component tolerances. Therefore, a smaller error between measured and simulated results can be obtained using components with smaller tolerance values. In addition, the realized gain of the matched antenna A5 was validated by the measurements with a good agreement between the results, as shown in Figure 16. This gained respect for the scope statement goals of our project. The simulated radiation pattern of the matched antenna A5 Figure 14. is shown Realized in Figure matching circuit17. Weantenna of the noticedA5. that the highly miniaturized monopole antenna A5 maintains its omnidirectional radiation performance well.
Magnetism 2022, 2, FOR PEER REVIEW 9 Magnetism Magnetism 2022, 2022, 2,2, FORFOR PEER PEER REVIEW REVIEW 99 Magnetism 2022, 2 237 Figure 15. Measured and simulated reflection coefficient S11 of the matched antenna A5 including Figure Figure Figure the sim-ulated 15. 15. Measuredand Measured Measured 15.individual andsimulated and simulated simulated component reflection reflection reflection tolerances coefficient coefficient coefficient effects.. S11S11S11 of ofofmatched the thematched the matched antenna antenna antenna A5including A5 including A5 including the the sim-ulated sim-ulated individual individual component component tolerances tolerances effects.. effects.. the sim-ulated individual component tolerances effects.. Figure 16. Measured and simulated realized gain of the miniaturized and matched antenna A5. Figure Figure Figure 16. Measured 16.Measured and andsimulated Measuredand simulated realized simulated gain realized realized of of gain gain the miniaturized ofthe and matched theminiaturized miniaturized antenna andmatched and matched A5. antenna antenna A5. A5. (a) (b) (a)(a) (b) (b) FigureFigure 17. Simulated radiation 17. Simulated pattern radiation of the pattern miniaturized of the and miniaturized andmatched matched antenna A5atatthe antenna A5 thefrequency frequency Figure Figure 130 MHz. 17. 17. (a) 130 MHz. in Simulated Simulated azimuth (a) radiation radiation plane in azimuth inpattern pattern (b) (b) plane ofofthe elevation in elevation the miniaturizedand miniaturized plane. plane. andmatched matchedantenna antennaA5 A5atatthe thefrequency frequency 130MHz. 130 MHz.(a)(a)ininazimuth azimuthplaneplane(b)(b)ininelevation elevationplane. plane. 7. Conclusions 7. Conclusions 7.7.Conclusions Conclusions In this paper, we proposed a new geometry of planar monopole antenna using slots to Intake thisadvantage paper, we ofproposed the low a new loss Planar geometry of planar monopole Magneto-Dielectric Material antennainusing (P-MDM) slots orderusing to slots InInthis to takeachieve thispaper, advantage paper, of we the welowproposed proposed loss a anew Planar new geometryofofplanar geometry Magneto-Dielectric planar monopole monopole Material antenna antenna (P-MDM) inusing order slots very high miniaturization rates. By placing the slots between two small pieces of the tototake to achievetake advantage advantage very high ofofthe thelow lowloss miniaturization loss Planar Planar rates. Magneto-Dielectric ByMagneto-Dielectric placing the field Material Material slotsintensity between (P-MDM) (P-MDM) two ininorder small pieces order P-MDM, we increased the concentration of the magnetic and consequently toto achieve achieve of the obtained P-MDM,a we very very high high increased higher miniaturization miniaturization effectivethe concentration magnetic rates. rates. medium of ByBy placing placing andthe the magnetic then the slots slots a higher field between between two two intensity and miniaturization small small pieces pieces rate.conse- The ofof quently the the proposedP-MDM, P-MDM, obtained wewe increased increased miniaturization the the concentration concentration methodmagnetic a higher effective was validated medium ofof the the by theand magnetic magnetic measurements field field then a higher intensity intensity of the and conse- miniaturizedconse- and miniaturization quently quently antennas rate. The obtained obtained a a higher higher shownminiaturization proposed effective effective in this article. magnetic magnetic medium medium and and then then a a higher higher method was validated by the measurements of the miniaturization miniaturization rate. rate. The The proposed proposed miniaturized antennas shown miniaturization miniaturization methodwas method in this article. wasvalidated validatedbybythe themeasurements measurementsofofthe the miniaturizedantennas miniaturized antennasshown shownininthis thisarticle. article.
Magnetism 2022, 2 238 Author Contributions: Supervision, A.S.; Writing—original draft, A.K., A.S. and A.-C.T. All authors have read and agreed to the published version of the manuscript. Funding: This work was carried out in the framework of the MISTRAL project, which is funded by the French National Research Agency (ANR) under contract number ANR−15−CE24−0030. Conflicts of Interest: The authors declare no conflict of interest. References 1. Sairam, C.; Khumanthem, T.; Ahirwar, S.; Singh, S. Broadband blade antenna for airborne applications. In Proceedings of the 2011 Annual IEEE India Conference, Hyderabad, India, 16–18 December 2011; pp. 3–6. 2. Arand, B.A.; Shamsaee, R.; Yektakhah, B. Design and fabrication of a broadband blade monopole antenna operating in 30 MHz–600 MHz frequency band. In Proceedings of the 21st Iranian Conference on Electrical Engineering, ICEE, Mashhad, Iran, 14–16 May 2013; pp. 1–3. 3. Ahirwar, S.D.; Sairam, C.; Kumar, A. Broadband blade monopole antenna covering 100–2000 MHz frequency band. In Proceedings of the 2009 Applied Electromagnetics Conference (AEMC), Kolkata, India, 14–16 December 2009; pp. 1–4. 4. Rhee, C.Y.; Kim, J.H.; Jung, W.J.; Park, T.; Lee, B.; Jung, C.W. Frequency-reconfigurable antenna for broadband airborne applications. IEEE Antennas Wirel. Propag. Lett. 2014, 13, 189–192. [CrossRef] 5. Mattei, J.-L.; le Guen, E.; Chevalier, A. Dense and half-dense NiZnCo ferrite ceramics: Their respective relevance for antenna down- sizing, according to their dielectric and magnetic properties at microwave frequencies. J. Appl. Phys. 2015, 117, 084904. [CrossRef] 6. Hwang, Y.; Zhang, Y.P.; Zheng, G.X.; Lo, T.K.C. Planar inverted F antenna loaded with high permittivity material. Electron. Lett. 1995, 31, 1710–1712. [CrossRef] 7. Ittibipoon, A.; Mongia, R.K.; Cuhaci, M. Low profile dielectric resonator antennas using a very high permittivity material. Electron. Lett. 1994, 30, 1362–1363. 8. Hansen, R.C.; Burke, M. Antennas with magneto-dielectrics. Microw. Opt. Technol. Lett. 2000, 26, 75–78. [CrossRef] 9. Erkmen, F.; Chen, C.C.; Volakis, J.L. UWB magneto-dielectric ground plane for low-profile antenna applications. IEEE Antennas Propag. Mag. 2008, 50, 211–216. [CrossRef] 10. Rozanov, K.N.; Ikonen, P.M.T.; Tretyakov, S.A.; Osipov, A.V.; Alitalo, P. Magnetodielectric Substrates in Antenna Miniaturization: Potential and Limitations. IEEE Trans. Antennas Propag. 2006, 54, 3391–3399. 11. Mosallaei, H.; Sarabandi, K. Magneto-Dielectrics in Electromagnetics: Concepts and Applications. IEEE Trans. Antennas Propag. 2004, 52, 1558–1567. [CrossRef] 12. Niamien, M.A.C.; Collardey, S.; Sharaiha, A.; Mahdjoubi, K. Printed inverted-F antennas over lossy magneto-dielectric materials: Theoretical approach and validations. IET Microw. Antennas Propag. 2014, 8, 513–522. [CrossRef] 13. Li, Q.; Chen, Y.; Yu, C.; Young, L.; Spector, J.; Harris, V.G. Emerging magnetodielectric materials for 5G communications: 18H hexaferrites. Acta Mater. 2022, 231, 117854. [CrossRef] 14. Zaid, J.; Farahani, M.; Denidni, T.A. Magneto-dielectric substrate-based microstrip antenna for RFID applications. IET Microw. Antennas Propag. 2017, 11, 1389–1392. [CrossRef] 15. Lee, J.; Heo, J.; Lee, J.; Han, Y. Design of small antennas for mobile handsets using magneto-dielectric material. IEEE Trans. Antennas Propag. 2012, 60, 2080–2084. [CrossRef] 16. Karilainen, A.O.; Ikonen, P.M.T.; Simovski, C.; Tretyakov, S.A.; Lagarkov, A.N.; Maklakov, A.; Rozanov, K.N.; Starostenko, S.N. Experimental studies on antenna miniaturisation using magneto-dielectric and dielectric materials. IET Microw. Antennas Propag. 2011, 5, 495. [CrossRef] 17. Kabalan, A.; Tarot, A.-C.; Sharaiha, A. Miniaturization of a broadband monopole antenna using low loss magneto-dielectric materials in VHF band. In Proceedings of the Loughborough Antennas & Propagation Conference, Loughborough, UK, 13–14 November 2017. 18. Ramahi, O.M.; Mittra, R. Design of a Matching Network for an HF Antenna Using the Real Frequency Method. IEEE Trans. Antennas Propag. 1989, 37, 506–509. [CrossRef] 19. Murata Manufacturing Co., Ltd. Available online: https://www.murata.com/ (accessed on 21 June 2018).
You can also read