Electrically small metamaterial-based antennas - have we seen any real practical benefits?
aa r X i v : . [ c ond - m a t . m t r l - s c i ] F e b Electrically small metamaterial-based antennas –have we seen any real practical benefits?
Pekka Ikonen
Nokia Devices R&D Productization Technology ManagementP.O.Box 407, 00045 NOKIA GROUP, Finland [email protected]
Abstract — Electrically small metamaterial-based antennas arediscussed from the industrial point of view using mobile phonesas the application example. It appears, that despite the in-teresting theoretical findings, the commercial acceptability ofthese antennas is low. Some of the issues possibly leadingto this situation are addressed. Discussion topics range fromchallenging application environment, through the response offinite-size composite-material samples, all the way to the requiredconstructive criticism and acknowledgement of prior art. Selectedissues are discussed in more details, and proposals how topossibly improve the commercial acceptability of metamaterial-based antennas are made.
I. I
NTRODUCTION
The number of papers about electrically small metamaterial-based antennas is big and steadily growing, e.g. [1], [2],[3], [4], [5], [6], [7], [8], [9], [10], [11], [12], [13], and thereferences therein. Interesting theoretical discussions predictgreat advantages from metamaterials in small-antenna design.For example, resonant conditions for strongly subwavelengthpatch antennas, or possibilities to overcome the small-antenna Q − limit have been discussed.Undoubtedly, metamaterial-inspired theoretical ideas canoffer new points of view in the “traditional” small-antennadesign. Also, many of the theoretical works are already backedup with prototypes experimentally verifying the proposedideas. However, to push the proposed antennas in commer-cial applications (e.g., in mobile phones), it is necessary toproperly demonstrate the practical benefits of metamaterial-based antennas when compared with “traditional” referenceantennas for the same application. Unfortunately, at the timebeing, solid comparative demonstrations can hardly be foundin the literature.What is the root cause for the lack of these demonstrations?Understanding this is very important as convincing experi-mental demonstrators are essential to maintain the industrialinterest in small-antenna enhancement using metamaterials.Below we discuss some challenges for the utilization of meta-materials in mobile-phone antennas, and address some otherissues that might be hindering the commercial acceptabilityof these design schemes. Previously, some challenges relatedto metamaterials in small-antenna design have been discussed,e.g., in [14], [15]. II. M ETAMATERIALS IN MOBILE - PHONE ANTENNAS : SOMEGENERAL OBSERVATIONS
To understand better how we could possibly push moremetamaterial-based (-inspired) antennas to mobile phones, westart by listing down some related general challenges. Later,some of the items listed below are discussed in more details,and proposals for clarification practices are made. At theend, the goal is to ensure that always the most competitiveantenna finds its way to the target application.1. Mobile phone is a very challenging application environmentfor artificial composite materials. • Available volume in the vicinity of antennas is only asmall fraction of free-space wavelength. • Antenna manufacturing complexity should be kept low. • Spatial near fields exciting finite-size material samplesare highly complex. • Some of the antennas need to couple strongly enough tothe rest of phone mechanics to gain sufficient bandwidth.2. Resonant nature of typical metamaterials is challenging. • Interesting material phenomena tend to occur in thevicinity of the resonance. • Effect of dispersion and resonant losses can be strong. • Non-radiating resonances coupled with antenna reso-nances are unwanted.3. Metamaterial-based antennas are rarely fully benchmarkedagainst reference antennas. • Full experimental characterization in the real applicationenvironment is always needed. • Reference antenna should be targeted (desirably alreadybeing used) for the same application. • Proper figure-of-merit should be used: bandwidth com-parison is not enough if efficiency degrades.4. Occasionally self-driven constructive criticism towardsmetamaterial-based antennas is missing. • Why should the proposed antenna be actually used,instead of “traditional” antennas? • Possible drawbacks (increased weight, cost etc.) of theproposed designs should be openly stated. • The realized response of metamaterials is always re-stricted by fundamental physical laws → these restric-tions should always be stated even with theoretical works. Also “not-optimally-working” antenna designs might bevaluable, if the study increases physical understanding.5. Proper acknowledging of prior works and proper marketingis needed. • Artificial materials in microwave engineering have a longhistory. • Practical realizations of metamaterial-antennas often re-semble “traditional” antennas. When should one marketthe solution as “metamaterial-based antenna”? • Metamaterial-based antennas should be marketed as al-ternative solutions to traditional antennas, rather than theonly possible solution. • Consistent terminology derived from prior works shouldbe used → “traditional” antenna features should not behidden behind new terminology.III. M OBILE - PHONE ANTENNAS AS THE APPLICATION FORMETAMATERIALS
The largest dimensions of a mobile phone are roughly λ / ...λ over the commonly used communications frequen-cies ( λ is the free-space wavelength). The volume reserved,e.g., for the cellular antenna is therefore only a small fractionof the wavelength. In addition to this, the spatial near-field pro-file in the vicinity of the antenna is typically highly complexdue to the antenna pattern details, and closely located mechan-ics components (display, speakers, etc.) Thus, it is impossibleto create the ideal homogenization conditions assumed inmany theoretical works. Also, due to the very small volumereserved for the antenna, the whole phone is typically utilizedas a radiator in order to increase the obtainable bandwidth[16]. Apparently, the difference between the response (eventhe goal of the desired response) of a free-standing antennaelement, and the element mounted in a real mobile phone canbe significant. To promote the findings successfully from thecommercial point of view, it is therefore essential to make surethat the proposed antenna offers the best size-vs.-performancecharacteristics also in the real phone environment.To maintain low manufacturing complexity (and associ-ated costs), a big portion of mobile phone antennas is stillimplemented on planar surfaces. Even though 3D compositematerial covers (e.g. [1], [6], [10]) would allow (in theory) toobtain natural matching for a highly sub-wavelength antenna,it is difficult to envision the actual realization of such coversin low-volume and low-cost applications. At the end, the per-formance enhancement obtained even with planar substratesunder volumetric antenna elements (e.g., planar inverted F-antenna) should clearly outweigh the increased manufacturingprocess complexity (costs), increased weight, and implicationsof the reserved volume.IV. O N THE RESONANT NATURE OF TYPICALMETAMATERIALS
Typical realizations of metamaterials proposed forelectrically-small antennas are composite substrates orsuperstrates based on resonant inclusions. For example,Lorentz-type resonant magnetic behavior is achieved with a lattice of broken loops, and Drude-type artificial permittivitybehavior is achieved with a lattice of thin wires. Alternatively,transmission-line meshes can be used to create high − k ( k isthe propagation constant in the mesh) appearance for a waveoscillating in the mesh with the goal to obtain size reduction.Even when excluding the above described challenges relatedto the mobile-phone volume constraints, there remain somefundamental questions on other challenges. For example, ar-tificial magnetics have no natural magnetic polarization, thus,work has to be done to polarize the loops to obtain collectivemicrowave response. Moreover, the loops are electrically verysmall, thus, their contribution to total radiation is typicallynegligible. Rather, the loops tend to store energy in the nearfield around them. How could this kind of material helpboosting the performance of the main radiator whose mainloss mechanism should come through radiation?In general, typically the exotic, “metamaterial-like” phe-nomena occur in the vicinity of the material resonance, thus,such a material possesses strong dispersion and resonantlosses. Coupling this kind of materials with inherently ratherhigh − Q antennas creates some apparent challenges: strongdispersion further increases the antenna Q (most often un-desirable, example discussion in [17]), or a discrete collectionof inclusions acts more as a non-radiating parasitic resonatorthan a “true” material load [18]. In the latter case, it some-times becomes difficult to identify the differentiating advan-tage offered by metamaterial-based antenna implementationswhen compared to “traditional” solutions utilizing parasiticresonators to boost the bandwidth. Moreover, due to very tightsystem requirements for the radio performance, non-radiatingresonances only boosting the impedance bandwidth (and notthe radiation efficiency bandwidth) are typically unwanted.V. P ROPER EXPERIMENTAL ANTENNA PERFORMANCEBENCHMARKING IS ESSENTIAL
How to get new antenna concept adopted in commercialuse, e.g., in mobile phones? First, the benefits (smallervolume or improved performance with a fixed volume, etc.)stemming from the proposed solution should clearly enoughoutweigh the possibly associated challenges (increased weight,complexity and cost, etc.). Second, given the performance ofthe proposed antenna seems feasible, this performance shouldbe compared with the performance of a reference antennabeing used for the particular application. Below we list somegeneral issues that help to build a convincing demonstration.1. The reference antenna is properly chosen. • The reference antenna should preferably already be usedfor the proposed application. • Several different antennas are being used in mobilephones → it is most convincing to compare the proposedantenna to the most competitive available antenna.2. The antennas are experimentally characterized in properenvironment. • Antennas targeted to mobile phones should be character-ized over proper-size chassis.
Value of the results is increased if real mechanics com-ponents close to the antenna (battery, speakers, etc.) areincluded to the printed-wiring-board prototype. • Most convincing demonstrations are obtained with realphone mechanics (using existing phones).3. The antennas are completely characterized. • Only the absolute value of S − parameter is clearly anon-complete description of small-antenna performance. • Measured efficiency and input impedance behaviorshould be presented. • Value of the results is further increased by consideringalso the user effect on the antenna performance.4. Proper figure-of-merit is used in the performance compari-son. • For single-resonant antennas proper figure-of-merit de-scribing size-vs.-radiation bandwidth characteristics is theradiation quality factor. • For multi-band antennas, possibly accompanied with amatching circuit, determining a proper figure-of-meritbecomes more challenging. • Often in these cases performance has to be evaluatedas a compromise between required volume, impedancebehavior, total efficiency, tolerance effects of matchingcomponents, tolerance to user effects, and manufacturingcomplexity and cost.5. Both the benefits and drawbacks of the proposed solutionare fully reported.VI. S
ELF - DRIVEN CONSTRUCTIVE CRITICISM TOWARDSTHE PROPOSED SOLUTIONS
The world is full of differently seeming electrically smallantennas. Evidently, a lot of attention has been paid to theselection of certain antennas for the use in mobile phones.Some of the issues typically affecting this selection processhave been described above. Therefore, as metamaterial-basedantennas are being proposed for mobile phones, the proposalshould first clearly answer to the question: “Why should theproposed antenna be used over all the other alternatives?”Especially in the beginning of metamaterial research thesematerials were in many occasions advertised to provide char-acteristics not found in nature. Such advertisements, accom-panied with some first theoretical studies based on simplifiedmaterial models, have created a lot of expectations towardsmetamaterials also in the field of small antennas. It is apparent,however, that as we approach the experimental realizationof antennas utilizing these materials, inevitable performancerestrictions (e.g., dispersion and losses) are strongly limitingthe actual performance. Therefore it is important to understandand openly state the practical limitations even in the caseof (typically the first) most theoretical studies, not to createhypothetical expectations.For example, for some time artificial magnetic materialswere considered as a very good miniaturization technique formicrostrip antennas due to the low-loss nature of the corre-sponding microwave (artificial) magnetism (background for magnetic materials with microstrip antennas is available, e.g.,in [19]). However, the experimental demonstrations availablein the literature were incomplete, or failed to validate theobservations based on simplified analysis (see [17] and thereference therein for more discussion). When the inherentmaterial dispersion (coming as an inevitable side result of theexperimental realization) was included into the analysis, it wasrevealed that such materials can never outperform referenceantennas [17].VII. A
CKNOWLEDGING PRIOR WORKS AND PROPERMARKETING
The history of artificial materials in microwave engineeringis very long, especially when it comes to to the utilization ofartificial dielectrics (see, e.g., [20] for a collection of relatedearly references). Also, the transmission-line and resonatortheories have been well established for several decades ago.Thus, as already outlined above, some of the realizations ofmetamaterial-based antennas might bear strong resemblancewith the “traditional” solutions. However, still in this case theproposed antennas might offer some benefits not seen in theprior solutions. Nevertheless, when introducing the proposedantennas it is important to understand and respect the priorworks, to be able to clearly highlight the differentiating aspectsof the proposed solution.An illustrative example of a good practice is the case ofnegative permittivity resonator (sphere antenna) [6]. This an-tenna seems to bear a striking resemblance with the sphericalhelix resonator introduced by Wheeler 50 years ago [6], [21].Nevertheless, the example shows how a metamaterial-inspiredtheoretical idea materialized into an interesting antenna de-sign, and one of the contributors explicitly acknowledged theresemblance to the prior works [22].Other issues possibly helping to improve the commercial ac-ceptability of metamaterial-based antennas through better un-derstanding relate to using consistent terminology. Currently,confusion is created as occasionally non-standard evaluationmeasures are used (for related criticism see, e.g., [23]), orwidely studied structures are called differently in differentsources. For example, despite the different terminology beingused, all the structures considered in [13], [17], [24] physicallyboil down to a periodic array of broken loops (authors of[25] further call a principally similar substrate “magneticmetamaterial” substrate). A reader not experienced with theprogress in this field might have the illusion that differentstructures are studied in all of these papers.It is also common that many antenna structures available inthe recent literature are called “metamaterial-based antennas”or simply “metamaterial antennas”, even though the actualstructures do not contain anything that can be described as (ar-tificial) material according to general definition [26]. Examplesof such antennas include, e.g., microstrip antennas utilizingonly one discrete resonant grid as a superstrate, or antennasutilizing few discrete resonators (often broken loops) withinthe antenna volume. For many people having background inthe field of small antennas (but not necessarily in the fieldf metamaterials) the use of such terminology might createthe feeling of an attempt to hide traditional antenna featuresbehind newly established terminology.VIII. S
OME CONCLUDING REMARKS
Electrically small metamaterial-based antennas have beenbriefly discussed from the industrial point of view using mo-bile phones as the application example. We have listed downseveral issues possibly affecting the fact that, despite interest-ing theoretical findings, the commercial use of metamaterial-based antennas, e.g., in mobile phones is low. Some of theissues, like the challenging application environment, cannotbe affected. Other issues, like the proper experimental charac-terization of the proposed antennas, will have a clear impactwhen trying to push these antennas to commercial applications.Also, it has been highlighted that one has to be constructivelycritical towards the proposed antennas, as this, added to properexperimental characterization and acknowledgement of priorart, is the best way to ensure that at the end the mostcompetitive antenna finds its way to the target application.R
EFERENCES[1] R. Ziolkowski, “Application of double negative materials to increase thepower radiated by electrically small antennas,”
IEEE Trans. AntennasPropag. , vol. 51, no. 10, pp. 2626–2640, 2003.[2] S. F. Mahmoud, “A new miniaturized annular ring patch resonatorpartially loaded by a metamaterial ring with negative permeability andpermittivity,”
IEEE Ant. Wireless Propag. Lett. , vol. 3, pp. 19–22, 2004.[3] S. A. Tretyakov and M. Ermutlu, “Modeling of patch antennas partiallyloaded with dispersive backward-wave materials,”
IEEE Ant. WirelessPropag. Lett. , vol. 4, pp. 266–269, 2005.[4] F. Qureshi, M. A. Antoniades, and G. V. Eleftheriades, “A compact andlow-profile metamaterial ring antenna with vertical polarization,”
IEEEAnt. Wireless Propag. Lett. , vol. 4, pp. 333–336, 2005.[5] C.-J. Lee, K. M. K. H. Leong, and T. Itoh, “Composite right/left-handedtransmission line based compact resonant antennas for RF moduleintegration,”
IEEE Trans. Antennas Propag. , vol. 54, no. 8, pp. 2283–2291, 2006.[6] H. R. Stuart and A. Pidwerpedtsky, “Electrically small antenna elementsusing negative permittivity resonators,”
IEEE Trans. Antennas Propag. ,vol. 54, no. 6, pp. 1644–1653, 2006.[7] R. Ziolkowski and A. Erentok, “Metamaterial-based efficient electricallysmall antennas,”
IEEE Trans. Antennas Propag. , vol. 54, no. 7, pp. 2113–2130, 2006.[8] A. Alu, F. Bilotti, N. Engheta, and L. Vegni, “Subwavelength, com-pact, resonant patch antennas loaded with metamaterials,”
IEEE Trans.Antennas Propag. , vol. 55, no. 1, pp. 13–25, 2007.[9] R. W. Ziolkowski and A. Erentok, “At and below the Chu limit: passiveand active broad bandwidth metamaterial-based electrically small anten-nas,”
IET Proc. Microw. Antennas Propag. , vol. 1, no. 1, pp. 116–128,2007.[10] S. Chadarghadr, A. Ahmadi, and H. Mosallaei, “Negative permeability-based electrically small antennas,”
IEEE Ant. Wireless Propag. Lett. ,vol. 7, pp. 13–17, 2008.[11] A. Erentok and R. Ziolkowski, “Metamaterial-inspired efficient electri-cally small antennas,”
IEEE Trans. Antennas Propag. , vol. 56, no. 3, pp.691–707, 2008.[12] M. Hirvonen and J.-E. Sten, “Power and Q of a horizontal dipoleover a metamaterial coated conducting surface,”
IEEE Trans. AntennasPropag. , vol. 56, no. 3, pp. 684–690, 2008.[13] F. Bilotti, A. Alu, and N. Engheta, “Design of miniaturized metamaterialpatch antennas with µ -negative loading,” IEEE Trans. Antennas Propag. ,vol. 56, no. 6, pp. 1640–1647, 2008.[14] R. Mittra, “A critical look at metamaterials for antenna related applica-tions,”
Journal of Communications Technology and Electronics , vol. 52,no. 9, pp. 1051–1058, 2007. [15] P. Ikonen, “Some reflections on the appearance of metamaterials inmicrowave engineering,” presented at the
Young Scientist Meeting onMetamaterials , 2008.[16] P. Vainikainen, J. Ollikainen, O. Kivekas, and I. Kelander, “Resonator-based analysis of the combination of mobile handset antenna andchassis,”
IEEE Trans. Antennas Propag. , vol. 50, no. 10, pp. 1433–1444,2002.[17] P. M. T. Ikonen, S. I. Maslovski, C. R. Simovski, and S. A. Tretyakov,“On artificial magnetodielectric loading for improving the impedancebandwidth properties of microstrip antennas,”
IEEE Trans. AntennasPropag. , vol. 54, no. 6, pp. 1654–1662, 2006.[18] P. Ikonen, S. Maslovski, and S. Tretyakov, “PIFA loaded with artificialmagnetic material: practical example for two utilization strategies,”
Microw. Opt. Techn. Lett. , vol. 46, no. 3, pp. 205–210, 2005.[19] R. C. Hansen and M. Burke, “Antennas with magnetodielectrics,”
Microw. Opt. Techn. Lett. , vol. 26, no. 2, pp. 75–78, 2000.[20] J. Brown, “Artificial dielectrics,”
Progress in Dielectrics , vol. 2, pp.195–225, 1960.[21] H. A. Wheeler, “The spherical coild as an inductor, shield, or antenna,”
Proc. IRE , vol. 46, pp. 1595–1602, 1958.[22] H. R. Stuart, “An electromagnetic comparison of the tapered sphericalhelix and the negative permittivity sphere,” in
Proc. IEEE InternationalSymposium on Antennas and Propag. ’07 , Hawaii, USA, June 2007, pp.3472–3475.[23] P. S. Kildal, “Comments on “application of double negative materialsto increase the power radiated by electrically small antennas”,”
IEEETrans. Antennas Propag. , vol. 54, no. 2, p. 766, 2006 (Authors’ reply:pp. 766-767).[24] H. Mosallei and K. Sarabandi, “Design and modeling of patch antennaprinted on magneto-dielectric embedded-circuit metasubstrate,”
IEEETrans. Antennas Propag. , vol. 55, no. 1, pp. 45–52, 2007.[25] K. Buell, H. Mosallei, and K. Sarabandi, “A substrate for small patch an-tennas providing tunable miniaturization factors,”
IEEE Trans. Microw.Theory Tech. , vol. 54, no. 1, pp. 135–145, 2006.[26] S. A. Tretyakov,