Metamaterial antennas have progressed considerably in the last few years. Kymeta demonstrated transmission to and receive from satellites both in the Ku and Ka-Bands using flat antennas, which use a tunable diffractive metasurface. Echodyne and Xerox PARC have developed metamaterial arrays for radar. The Army Research Laboratory funded the development of a metamaterial 250 to 505 MHz antenna with a λ/20 thickness. Complementing this, a conventional tightly coupled dipole antenna (TCDA) has been developed which provides a 20:1 bandwidth with a λ/40 thickness. Target cloaking has been demonstrated at microwaves using metamaterials. Stealthing, by absorption using a thin flexible and stretchable metamaterial sheet, has been shown to provide 6 dB absorption over an 8 to 10 GHz band, with greater absorption over a narrower range. Metamaterial has been used in cell phones to provide antennas that are 5× smaller (1/10th λ) having 700 MHz to 2.7 GHz bandwidth. It has also provided isolation equivalent to 1 m separation in antennas with 2.5 cm separation and is used for phased array wide angle impedance matching (WAIM).

Metamaterials are man-made materials in which an array of structures less than a wavelength in size are embedded (see Figure 1).1 These materials have properties not found in nature, such as a negative index of refraction. For one form of metamaterial the permittivity (ε) and permeability (µ) are both negative.  When this happens the index of refraction n =√µε is negative, the negative sign being used for the square root.2, 3 Actual materials have complex-valued ε and µ. The real parts of both ε and µ do not have to be negative to display negative refraction.3

Figure 1

Figure 1 A metamaterial is a synthetic material with properties based on the sub-wavelength repeated structure.1

Kymeta Figure 2

Figure 2 Kymeta electronically steered metamaterial communications antenna.7, 8

A material with negative n is rarely found in nature.  However, it can be produced by forming an array of metal split rings and rods (short parallel wires).  The split ring resonators produce a permeability µ that is negative while the rods produce a permittivity ε that is negative.  The dimensions of the metallic rings and rods must be smaller than a wavelength, but larger than an atomic dimension, to obtain a negative index of refraction.  With metamaterial it is possible to achieve imaging beyond Abbe’s diffraction limit which for modern optics is about λ/2. For regular materials, subwavelength imaging is hard to achieve because the evanescent waves containing the subwave-length information decay exponentially with distance, making them effectively nonexistent at the image plane.4

Figure 3

Figure 3 O3b constellation.

Figure 4

Figure 4 Waveguide row showing slot and resonator locations.11

Purdue University has shown through simulation that metamaterial can provide imaging beyond the diffraction limit for visible light having a wavelength of 0.7 µm using two layers of anisotropic material.4 The University of Illinois experimentally attained a 1/12th λ resolution at 0.38 µm by using thin layers of silver, germanium and chromium.5 The silver provided a negative permittivity which was sufficient for focusing beyond the diffraction limit, while the germanium enabled the silver film to be smooth.  Using what is called a resonant metalens, Institut Langevin, ESPCI Paris Tech & CNRs achieved a resolution of λ/80 in the far field at microwave frequencies.6 They believe that metalenses can be built at visible wavelengths using nanoparticles or nanowires as resonators. Applying imaging beyond the diffraction limit to integrated circuit lithography helps to advance Moore’s Law.

Figure 5

Figure 5 Three waveguide rows of metamaterial resonators.11

Figure 6

Figure 6 Metamaterial RLC resonator.7

The definition of metamaterial has been extended to include material having any combination of positive and/or negative ε and µ. It includes electromagnetic band gap (EGB) material (also called photonic crystals).2  For some in the RF community it includes frequency sensitive surfaces (FSS).2 Included here, also, are fractal frequency selective surfaces.

Figure 7

Figure 7 Generic parallel RLC resonator filter.


Kymeta Array

Kymeta is developing a metamaterial antenna for communications via satellites (see Figure 2).7-11 They are commercializing a product that operates in the Ku-Band (10 to 15 GHz). Overall data rates for the antennas depend on a number of factors like the size and operating frequency of the antenna. The RF radiated power is on the order of a few watts. Transmission from the ground to the satellites and back has been demonstrated. Kymeta originally received about $65 million in funding, mostly from Intellectual Ventures, about $10 million of which is from Bill Gates. They were working with O3b in the past for satellite communication; see Figure 3. They are now working with Honeywell Aerospace, Inmarsat and Panasonic Avionics Co.; see Kymeta website:

 Explanations of how the antenna might scan based on published material are now given for two architectures. For the first architecture, the array is formed from several rows or traveling wave feeds which could be a leaky waveguide over which a slotted metal cover is placed (see Figure 4).7,11 Think of it as a slotted waveguide. The antenna consists of rows of these slotted waveguides which are end fed (see Figure 5). Assume that it is desired to radiate in a specified direction. One then determines at which slots the signals have the desired phase shift to form a beam in that direction. Then only from these slots is the signal allowed to radiate. The signals from the other slots are blocked. The switch is a bandpass filter resonator placed over each slot that controls whether the signal is, or is not, radiated from the slot. When the resonator center frequency is at the frequency of the signal coming out of the slot, the signal passes through the resonator to radiate. If frequency of the resonator is shifted away from the signal frequency, the signal from that slot is blocked by the resonator and does not radiate. The resonators use liquid crystals whose dielectric constants can be controlled by bias voltages to shift their resonant frequencies.7 The spacing between the slots is much less than the conventional ~λ/2 in order to have a large number of slots with the desired phase shift.

Figure 8

Figure 8 Close-up of Kymeta antenna (from Kymeta website).

Figure 9

Figure 9 Center fed circular array architecture (from Intellectual Ventures website, Copyright©Intellectual Ventures Management, LLC [IV]).

This leads to a thinned array wherein the thinning is different for each different scanning angle. There are phase and amplitude errors for the radiated signals because the signals at a slot will not always have the exact phase and amplitude needed for that slot position, and the resonator center frequency will not always be at the signal frequency. The thinning and errors lead to antenna side lobes, which is common among finite antennas and only becomes an issue if the side lobes interfere with other satellites or appreciably reduce the receive capabilities of the antenna. Kymeta’s antenna has completed successful demonstrations without incurring complaints of adjacent satellite interference, and has described a feedback technique for lowering the side lobes. Since the Kymeta solution has numerous elements, the beam is formed by tuning some of the elements in the array while leaving others detuned. Phase shifting is not required when using this technology. Through using many elements at less than λ/2 spacing, an accurate hologram with cells at the right phase and amplitude can be achieved. Kymeta’s website shows such a thinning of the radiating elements.

Figure 10

Figure 10 Echodyne metamaterial array radars: MESA-D-DEV K-Band radar (a) and MESA-DAA K-Band radar (b).15

Table 1

Figure 6 shows an actual resonator,7 a circuit consisting of an etched capacitor and inductor in parallel. The etched circuit is placed over a liquid crystal dielectric (previously described) which is placed on top of a ground plane. The liquid crystal’s permittivity can be changed by applying a bias voltage -- between the etched RLC circuit and the ground plane. This bias voltage allows control of the resonator center frequency, placing it at the frequency of the signal when it is to be radiated and away from the signal frequency when it is to be blocked. The amplitude and phase shift of a generic resonator is shown in Figure 7. A close-up of the antenna face with its closely spaced elements is shown in Figure 8. Instead of a leaky waveguide with slotted cover one can use a microstrip, coplanar waveguide, parallel plate waveguide, dielectric slab or lossy waveguide.11 Because there are no active components, the cost of building this antenna with many slots, or elements, is low.

Table 2

Table 3

Whether or not to consider a metasurface of this variety a metamaterial was once hotly debated. It has, however, gained wide acceptance in the last several years as evidenced by the inclusion of a “metasurfaces” track at Metamaterials 2015 and 2016 as well as META 2016. The generally accepted notion of a metamaterial is that it is an array of subwavelength structures (not found in a natural material) designed for a specific and controlled electromagnetic response. The Kymeta solution uses the properties of each resonator independently to act as an on-off switch, yet it is the collective behavior of all the elements, whether off or on, in a holographic pattern that permits the Kymeta metasurface to function as a high-gain directional antenna. It is a clever and novel concept wherein the need for phase shifting (as in traditional phased array solutions) and active phase shifters is eliminated. The resonators were originally developed to create a metamaterial with a negative permittivity.14

Figure 11

Figure 11 Xerox PARC metamaterial array for car radar.18

Figure 12

Figure 12 Extremely low profile 250 to 505 MHz magnetic metamaterial antenna.20

Kymeta mTenna technology uses a thin structure with tunable metamaterial elements. The tunable elements scatter RF energy when activated. Software activates a pattern of tunable elements to generate a beam. To change the beam direction, the software changes the pattern of activated elements. Kymeta’s technology is heavily protected by a portfolio of global patents and trade secrets. A potential competing technology to the Kymeta approach is to use a conventional AESA built using low cost extreme MMIC technology.15, 16

Figure 13

Figure 13 Extremely low thickness wideband antenna array using tightly coupled dipole antennas.24, 25

Echodyne Array

A second company, Echodyne, has developed metamaterial arrays for radar using these antennas (see Figure 10 and Tables 1-3).17 Echodyne, like Kymeta, is funded by Intellectual Ventures and Bill Gates. Switching times needed for the intended radar applications, e.g., 1 µs, are much shorter than needed for communications.

Xerox PARC (see Figure 11) is developing a car radar that uses a metamaterial array antenna.18 It illuminates a wide angle on transmission and uses digital beam forming (DBF) to form many simultaneous beams and, in turn, an image of its surroundings. It has a 120° field of view and is intended for self-driving cars. Remember, Xerox PARC gave us the PC mouse as we know it as well as laser printing.


Army Low Profile VHF

A metamaterial with negative ε produces what is called an artificial magnetic ground plane or a magnetic dielectric.  Such a material would allow a dipole antenna (which ordinarily needs to be a ¼ wavelength above a metallic ground plane) to be flush with the artificial magnetic ground plane.  This is possible because the electric field in the artificial magnetic ground plane can be equal and parallel to that in the dipole just above it. This is in contrast to a conducting ground plane where the electric field would be opposite in the ground plane and thus cancel out the electric field. The promise is that it would allow the construction of conformal dipole arrays. Such an antenna could replace the highly visible (by the enemy) feet-high whip antennas that are mounted vertically on the side of a HMMWV, leading to greater survivability.19,20 The Army Research Laboratory has funded the development of a VHF/UHF metamaterial antenna (see Figure 12).20,21 Magnetic dielectrics having very wide bandwidths should be achievable in the band from 50 MHz to 20 GHz.22

Figure 14

Figure 14 Low cost, two-panel array with EBG enhancement for wide angle scan.26

Very Wideband

Thales has demonstrated the placement of a conformal spiral antenna on a metamaterial with this antenna having a bandwidth from 2 to 8 GHz.23  While discussing low profile wideband metamaterial antennas it is worth mentioning that a low thickness wideband antenna can be built without metamaterials using tightly coupled dipole antennas (see Figure 13).24,25

Figure 15

Figure 15 EBG used to achieve isolation between and transmit and receive antennas27 (courtesy of Professor K. Sarabandi, University of Michigan).

Figure 16

Figure 16 Invisibility cloak concept (a) and Duke University microwave metamaterial cloaking device using metamaterial with split rings (b).30

Isolation and WAIM

For their S-Band Digital Array Radar (DAR), Purdue University has used EBG material between patch radiating elements to reduce mutual coupling, resulting in a wider scan angle (see Figure 14).26 It serves to provide a wide angle impedance match (WAIM). In a program funded by the Army Research Lab, the University of Michigan used an EBG between the transmit and receive antennas separated by about 3 cm on a transponder operating at 2.72 GHz. They achieved 42 dB isolation, which is 24 dB above what would have been realized without the EBG (see Figure 15). This is the isolation one would have realized, conventionally, for 1 m separation.27

Figure 17

Figure 17 Human invisibility cloak.31, 32

Commercial Wireless

Metamaterial is used commercially in the wireless dual-band (2.4, 5 GHz) NDR3300 router.28 Here eight antennas are placed on a RAYSPAN® metamaterial which allows the antennas to be smaller with better isolation. Metamaterial antennas are also used in our cell phones for the same purpose. They are 2D antennas less than 10 mm × 50 mm in area and paper thin.29 Typically they are at least five times smaller than conventional antennas, i.e., 1/10th λ in size. Metamaterial antennas can be made broadband to support multiband operation such as 700 MHz to 2.7 GHz or GPS, Bluetooth, Wi-Fi and WiMax within one antenna array. It is claimed that they can be developed in a short time (two weeks to a month), are inexpensive to build, and provide low RF exposure to the user.29


Target cloaking was first demonstrated at Duke University using metamaterials at microwaves. With cloaking, the electromagnetic wave transmitted by a radar goes around the target, making it invisible (see Figure 16a). The Duke University microwave metamaterial cloaking device shown in Figure 16b uses concentric 1 cm wide rings. On each ring are etched split ring resonators that produce a negative index of refraction to guide the microwave signal around a 5 cm center region that contains the object being stealthed. The outer diameter is 13 cm (~5 inches). Cloaking is achieved only over a narrow bandwidth.

Cloaking has more recently been demonstrated using fractals by Fractal Antenna Systems (see Figure 17).31,32 An engineer at the company was first placed in the path between a transmitter and receiver, blocking the signal such that it was reduced by 6 to 15 dB over the band from 750 to 1250 MHz. After being “cloaked” within a cylinder with a fractal coating around its surface and, again, placed in the path between the transmitter and receiver, the signal was no longer blocked. It was attenuated by only a fraction of a dB over the same 50 percent bandwidth. Figure 18 shows both fractal and split ring resonator surfaces.

Another way to hide a target is for it to absorb the incident radar signal. Such stealthing has been demonstrated by simulation using a fractal frequency selective coating that is < 1 mm thick (see Figure 19).33 Absorption of 90 percent was achieved from 2 to 20 GHz and about 99 percent from about 10 to 15 GHz. Good absorption was achieved for all incident angles and polarizations. Iowa State University recently demonstrated stealthing with a stretchable, flexible metamaterial sheet consisting of silicon embedded with split ring resonators containing liquid metal alloy galinstan made of gallium, indium and tin. It achieved a 6 dB target cross section reduction from 8 to 10 GHz with higher absorption over narrower bands (see Figure 20).34 It should be possible to apply this material conformally over the object to be “stealthed.”


Metamaterials became an area of great interest as a result of a seminar paper by J. Pendry of the University of Cambridge.35,36 There are now over a dozen books on the subject.  One of these books (by Professor Munk),37 questions whether one can actually produce material with a negative index of refraction. Dr. Munk claims that results obtained with what are called negative index of refraction materials can be achieved with non-negative index of refraction material.  No matter what the explanation, it has been shown that it is possible, through the use of these materials, to achieve focusing beyond diffraction limit, cloaking and stealthing at microwave frequencies, conformal antennas at VHF/UHF, better isolation, electronic scanning arrays and reduced size antennas.

Figure 18

Figure 18 Images of cloaking fractal (a) and split ring resonator (b) surfaces.31


Figure 19

Figure 19 Stealth by absorption: < 1 mm thick fractal metamaterial coating (a) and performance (b).33

  1. S. Govind, “Realizing the Potential of Metamaterials,” Raytheon Technology Today Magazine, Issue 1, 2012, pp. 24–27.
  2. J.S. Derov, E.E. Crisman and A.J. Drehman, “Metamaterials and Their RF Properties,” Proceedings of the 2008 Antenna Applications Symposium, Vol. 2, September 2008, pp. 176–189.
  3. refraction.ogg.
  4. H. Liu and K.J. Webb, “NANOIMAGING: Bilayer Metamaterial Lens Breaks the Diffraction Limit,” Laser Focus World, September 2009, pg. 35.
  5. “Smooth Superlens Images at 1/12 Wave Resolution,” Laser Focus World, March 2010.
  6. G. Lerosey, “Resonant Metalens Resolves to Lambda/80,” Laser Focus World, Vol. 46, No. 7, July 2010.
  7. N. Kundtz, “Next Generation Communications for Next Generation Satellites,” Microwave Journal, Vol. 57, No. 8, August 2014.
  8. “MSA-T,” Intellectual Ventures,
  9. M.C. Johnson, S.L. Brunton, J.N. Kutz and N.B. Kundtz, “Sidelobe Cancelling for Optimization of Reconfigurable Holographic Metamaterial Antenna,” IEEE Transactions on Antennas and Propagation, Vol. 63, No. 4, April 2015, pp. 1881–1886.
  10. J.B. Pendry, “Metamaterials,” International Radar Conference, Keynote, October 2014.
  11. A. Bily, J. Dallas, R.J. Hannigan, N. Kundtz, D.R. Nash and R.A. Stevenson, US Patent No. 2014/0266946 A1, September 18, 2014.
  12. E. Brookner, “Practical Phased Array Antenna Systems,” Section 2.1, Artech House, 1991.
  13. S. Weiss (Army Research Lab) first alerted me to this.
  14. D. Schurig, J.J. Mock and D.R. Smith, “Electric-Field-Coupled Resonators for Negative Permittivity Metamaterials,” Applied Physics Letters, Vol. 88, No. 4, January 2006, pg. 041109.
  15. E. Brookner, “Radar and Phased Array Breakthroughs,” Microwave Journal, Vol. 58, No. 11, November 2015.
  16. D.W. Corman, P. Moosbrugger and G.M. Rebeiz, “5G/Massive MIMO Channel, The Industry’s Next Tipping Point,” Microwave Journal, Vol. 57, No. 5, May 2014.
  17. “MESA—Radars & Subsystems,”
  18. B. Casse, “Self-Driving Cars Need Better ‘Digital Eyes’ to Detect Pedestrians,” PARC blog, October 2015,
  19. R. Shahidain, Military Antennas, 2009.
  20. S.D. Keller et al., “Low Profile and Platform Specific Antennas,” Army Research Laboratory, Adelphi, MD.
  21. G. Mitchell and S. Weiss, Army Research Laboratory, Adelphi, MD; see also G. Mitchell and S. Weiss, “An Overview of ARL’s Low Profile Antenna Work Utilizing Anisotropic Metaferrites,” ARRAY-2016, October 2016.
  22. J. S. Derov, private communication.
  23. C. Renard and M. Soiron, “Wideband Multifunction Airborne Antennas,” International Radar Conference, October 2009, pp. 1–3.
  24. M. Sarcione, P. Hull, C. Whelan, D. Tonomura, T.V. Sikina, J. Wilson and R.E. Desrochers II, “Raytheon AESA Research: Past, Present and Future,” Raytheon Technology Today Magazine, 2014, Issue 1, pp. 8–13.
  25. J.A. Kasemodel, C.C. Chen and J.L. Volakis, “Broadband Planar Wide-Scan Array Employing Tightly Coupled Elements and Integrated Balun,” IEEE International Symposium on Phased Array Systems and Technology, October 2010, pp. 467–472.
  26. C. Fulton, Digital Array Radar, Ph.D. Thesis, Purdue University, 2010; see also C. Fulton and W. Chappell, “Low-Cost, Panelized Digital Array Radar Antennas,” IEEE International Conference on Microwaves, “Communications, Antennas and Electronic Systems,” May 2008, pp. 1–10.
  27. K. Sarabandi and Y.J. Song, “Subwavelength Radio Repeater System Utilizing Miniaturized Antennas and Meta-Material Channel Isolator,” IEEE Transactions on Antennas and Propagation, Vol. 59, No. 7, July, 2011, pp. 2683–2690; also see Joseph Mait, Army Research Lab, Adelphi, MD.
  28. T. Higgins, “What Do Metamaterials Really Do for Antennas?” SmallNetBuilder, January 2008,
  29. G. Poilasne, “RAYSPAN® Proprietary Metamaterial Antennas, A Proven SAR Reduction Solution,” RAYSPAN, 2009, pp. 1–4.
  30. D. Schurig, J.J. Mock, B.J. Justice, S.A. Cummer, J.B. Pendry, A.F. Starr and D. R. Smith, “Metamaterial Electromagnetic Cloak at Microwave Frequencies,” Sciencexpress, October 2006, pp. 1133628-1-4.
  31. N. Cohen, “Fractals,” Vol. 20, Nos. 3 & 4, pp. 227–232, World Scientific Publishing, 2012.
  32. N. Cohen, “Wideband Omnidirectional Microwave Cloaking,” Microwave Journal, Vol. 15, No. 1, January 2015.
  33. F. Yue-Nong and G. Rong-Zhou, “An Ultrathin Wide-Band Planar Metamaterial Absorber Based on Fractal FSS and Resistive Film,” Chinese Physics B, Vol. 22, No. 6, February 2013, pp. 067801-1-6.
  34. S. Yang, P. Liu1, M. Yang, Q. Wang and J. Song, “From Flexible and Stretchable Meta-Atom to Metamaterial: A Wearable Microwave Meta-Skin with Tunable Frequency Selective and Cloaking Effects,” Scientific Reports, Vol. 6, No. 21921, February 2016.
  35. J.B. Pendry, “Negative Refraction Makes a Perfect Lens,” Physical Review Letters, Vol. 85, No. 18, October 2000, pp. 3966–3969.
  36. R.R. Marqués, F. Martín and M. Sorolla, “Metamaterials with Negative, Parameters: Theory, Design, and Microwave Applications,” Wiley, 2008.
  37. B.A. Munk, “Metamaterials: Critique and Alternatives,” Wiley, 2009.

Figure 20

Figure 20 Stretchable, flexible metamaterial absorber (a) and absorption for different meta-skin stretchings (b).33

Dr. Eli Brookner received his MEE and DrSc from Columbia University in 1955 and 1962, respectively; and he received his BEE from CCNY in 1953. From 1962 to 2014, as a principal engineering fellow with Raytheon Co., he was involved radar design for air traffic control, military and defence, space and navigation. These included ASDE-X, ASTOR RADARSAT II, AGBR, major space-based radar programs, NAVSPASUR, COBRA DANE, PAVE PAWS, MSR, COBRA JUDY Replacement, THAAD, SIVAM, SPY-3, Patriot, BMEWS, UEWR, SRP, Pathfinder, Upgrade for >70 ARSRs, AMDR, Space Fence and 3DELRR. Before Raytheon, he was with Columbia University Electronics Research Lab (now RRI), Nicolet and Rome Air Force Labs.

Brookner received the IEEE 2006 Dennis J. Picard Medal for radar technology of application, the IEEE 2003 Warren White Award, the Journal of the Franklin Institute Premium Award for best paper (1966) and the IEEE Wheeler Prize for best applications paper (1998). He is a fellow of the IEEE, AIAA and MSS.

Brookner has nine patents and is the author of four books: “Tracking and Kalman Filtering Made Easy,” 1998, Wiley; “Practical Phased Array Antenna Systems,” 1991; “Aspects of Modern Radar,” 1988; and “Radar Technology; 1977,” Artech House. He has taught courses on radar, phased arrays and tracking worldwide (25 countries), with over 10,000 total attendees. He has authored over 230 papers, talks and correspondences (greater than 100 invited), contributed chapters to three books and has made many appearances as a banquet/keynote speaker. Six of his papers are reproduced in books of reprints (one in two books).