Microstrip patch antennas (MPA)1–5 and dielectric resonator antennas (DRA)6–7 have been studied extensively. They share the advantages of low profile, lightweight, ease of fabrication and are easy to excite by different methods. Both antennas are viable candidates for numerous applications, either as individual elements or in an array environment. However, metallic and dielectric losses are usually associated with most MPAs when they are printed on dielectric substrates.3 On the other hand, there is no metallic loss for the DRAs. Hence, they have the inherited advantage of higher radiation efficiency at high frequencies.7 A comparison between DRAs and microstrip patches was performed numerically.8 The advantages of the DRA over the MPA were clear. Measurement of the radiation efficiency could be a means to compare the losses in the antennas. The conduction losses, dielectric losses and surface wave losses all contribute to lowering the radiation efficiency, which can be accurately measured by the Wheeler cap method.9–11

In this article, the performance of a cylindrical DRA5,12–14 and a circular MPA15–18 at X-band are studied experimentally. Both the DRA and MPA are excited by a coaxial probe and resonate at 9.9 GHz. The locations of the feeds are carefully chosen in order to excite the broadside radiation modes. The MPA is excited in the TM11 mode and the DRA in the HEM11? mode. In order to have a fair comparison, the MPA is printed on the same dielectric material that is used to fabricate the DRA. The bandwidths for the DRA and MPA are 10.3 and 1.3 percent, respectively, with a dielectric constant of 10.2 used for both antennas. The antenna gain of the DRA is 1 dB higher than for the MPA. The Wheeler cap method is employed to measure the radiation efficiency. It is shown that the radiation efficiency for the MPA is 80 percent and 95 percent for the DRA. As expected, the conduction losses of the MPA reduced the radiation efficiency and the gain of the antenna in X-band.

Antenna Geometries and Measurement Setup

Figure 1 shows the geometries of the cylindrical DRA and the circular MPA. Both antennas are designed to have a good match at approximately 9.9 GHz. The circular-shaped MPA has a radius r = 2.45 mm (0.082 ?0) and is printed on a dielectric substrate with a dielectric constant ?r = 10.2 and a thickness t = 0.64 mm (0.021 ?0), where ?0 is the free space wavelength at 9.9 GHz. The TM11 mode is excited by an off center coaxial probe. The probe has a radius rf = 0.3 mm and is off-centered in the MPA by x = 0.4 ?m. The cylindrical DRA has the following dimensions: r = 3.3 mm (0.11 ?0), h = 5.08 mm (0.17 ?0), ?r = 10.2, rf = 0.4 mm, x = 2 mm and l = 2.54 mm. Both antennas are mounted on a square ground plane 30 by 30 mm (1 ?0 × 1 ?0). A two-port HP8510C vector network analyzer (VNA) is used to measure the transmission coefficients. The measurement setup is shown in Figure 2. Port 1 of the VNA is connected to a reference MPA, which is identical to the MPA mentioned earlier, and Port 2 is connected to the DRA or MPA for comparison. The antennas, connected to Ports 1 and 2, are separated by a certain distance d between the two ground planes. The measurement is taken when both antennas are facing each other. Since broadside modes are excited for both antennas, the maximum radiation is normal to the ground plane.

Measurement Comparisons

Transmission Coefficients and Antenna Gains

The measurement is made using the setup described in the previous section. The MPA is connected to Port 1, as a common transmitting antenna. The receiving port (Port 2 of the VNA) is connected to the antenna under test (AUT). First, the return losses of all the antennas are measured to ensure a good match at the desired frequency of 9.9 GHz. Figure 3 shows the return losses of the MPA and DRA. The resonant frequency for the DRA is 10.2 GHz. It has a matched frequency range from 9.75 to 10.81 GHz, with a 10.3 percent impedance bandwidth (S11 < –10 dB). The MPA has a minimum return loss of –25 dB at 9.9 GHz. Its impedance bandwidth is 1.3 percent with a matched frequency range from 9.87 to 9.99 GHz. The comparison is performed at 9.9 GHz, which shows a good match (S11 < –13 dB) for both antennas.

The transmission coefficients were measured with both antennas separated by a distance d = 27 mm (0.9 ?0), 43 mm (1.43 ?0) and 74 mm (2.47 ?0). These distances were chosen to have the antennas placed within both the Fresnel region (18.6 mm < d < 60 mm) and in the far field region (d > 60 mm). The measured transmission coefficients S21 are shown in Figure 4. The transmission coefficients for the DRA are found to be –11.8 dB, –15.7 dB and –20.4 dB when d = 27 mm, 43 mm and 74 mm, respectively. When the MPA is connected, the transmission coefficients are found to be –12.8 dB, –16.6 dB and –21.5 dB at d = 27 mm, 43 mm and 74 mm, respectively. The power received by the DRA is always 1 dB higher than by the MPA, at 9.9 GHz. In addition, the Q-factor of the DRA is found to be lower than the MPA, which leads to the transmission coefficient levels of the DRA dropping at a slower rate than for the MPA. The percentage increase of the power received by the DRA compared to the MPA is shown in Figure 5. The 1 dB differences between the DRA and MPA is equivalent to about 25 percent more power received/transmitted by the DRA than by the MPA, with the same input power. The matching of the MPA deteriorates quickly due to its narrow bandwidth, when the MPA will no longer receive or radiate the signal efficiently. It is noted that the percentage differences of the power received by the DRA and MPA grows quickly when the frequency is below or above 9.9 GHz. The same characteristics are found for different distances d. The percentage increase in the receiving power by the DRA is mainly due to the mismatch of the MPA.

A similar test was also performed in the far field range, inside an anechoic chamber. A standard gain horn from Narda Microwave was used as the transmitting antenna. The AUT was located two meters away from the transmitting horn. The average gains of the DRA and MPA, measured in the broadside direction, were 7.7 and 6.7 dBi, respectively.

Radiation Pattern

The radiation patterns of both the DRA and MPA were measured at 9.9 GHz. With the off-center probe excitation, the HEM11d and TM11 broadside modes are excited in the DRA and MPA, respectively. The co-polar radiation patterns of both antennas are depicted in Figure 6. The DRA has a 3 dB beamwidth of 86° and 90° in the E- and H-plane, respectively. The MPA 3 dB beamwidth is 88° in the E-plane and 90° in the H-plane. It is noted that both antennas are radiating with similar beamwidth and front-to-back ratio.

Radiation Efficiency

The radiation efficiencies of both the DRA and MPA are measured by the Wheeler cap method. For accurate results, it is required that the antenna be enclosed by a conducting cavity, the Wheeler cap.9 A conducting cylindrical cap is used to shield the antenna. Since the antenna is prevented from radiation by the cap, the measured input resistance is equivalent to the loss resistance.11 The radiation efficiency is defined by the ratio between the radiated power and the total input power. By computing the input resistance with and without the cap, the radiation efficiency can be found as


Prad, Rrad = radiated power and resistance
Ptot, Rtot = total input power and resistance
Rin = input resistance without the cap
Rcap = input resistance with the cap

The measurement setup is shown in Figure 7. Using Equation 1, the radiation efficiency of the DRA and MPA at 9.9 GHz are measured to be approximately 95 and 80 percent, respectively. Both antennas have similar amounts of dielectric loss. The lower radiation efficiency of the MPA is mainly caused by the conduction losses. A second way18 to measure the antenna efficiency by the Wheeler cap method is to determine the unloaded Q factor of both the radiating antenna, Q0rad, and the covered antenna, Q0cov. The radiation efficiency at the resonant frequency is then computed as

For the two antennas discussed here, the results obtained by the second method are also provided in Table 1. It can be seen that the results of both methods lead to the following conclusion: the DRA radiation efficiency is 93 percent or higher, while the MPA radiation efficiency is 81 percent or lower.


In this section, the performances of both the DRA and MPA are summarized. The MPA occupies a smaller volume than the DRA. The MPA has a 50 percent smaller volume than the DRA, as well as a lower profile. The height of the DRA is eight times greater than the MPA. The matched bandwidth of the MPA is only 1.3 percent, but the DRA bandwidth is nearly eight times greater (10.3 percent). Both the DRA and MPA are excited with broadside radiating modes and the radiation patterns are nearly the same. By comparing the gain and the radiation efficiencies, it is concluded that the conduction losses of the MPA causes the gain to be 1 dB lower than the DRA and the MPA has a radiation efficiency of only 80 percent.


The performance of a cylindrical dielectric resonator antenna (DRA) and a circular microstrip patch antenna (MPA) were compared at X-band. The DRA, without conduction losses, was showing higher radiation efficiency than the MPA. The DRA had a 1 dB gain higher than the MPA. In addition, the matched bandwidth of the broadside-radiating mode of the DRA is 10 percent and only 1.3 percent for the MPA. Both antennas showed similar radiation patterns with similar beamwidth. However, the advantage of using MPA was its smaller volume.


This work was partially supported by The Army Research Office under Grant No. DAAD19-02-1-0074. The authors would like to thank Rogers Corp. for providing the microwave substrate materials through the University Program.


1. G.A. Deschamps, “Microstrip Microwave Antennas,” 3rd USAF Symposium on Antennas, 1953.

2. J.Q. Howell, “Microstrip Antennas,” 1972 IEEE International Antennas and Propagation Symposium Digest, pp. 177–180.

3. J.R. James and P.S. Hall, Handbook of Microstrip Antennas, Peter Peregrinus Ltd., London, UK, 1989.

4. R. Garg, P. Bhartia, I. Bahl and A. Ittipiboon, Microstrip Antenna Design Handbook, Artech House Inc., Norwood, MA, 2001.

5. I. Bahl and P. Bhartia, Microstrip Antennas, Artech House Inc., Norwood, MA, 1980.

6. S.A. Long, M.W. McAllister and L.C. Shen, “The Resonant Cylindrical Dielectric Cavity Antenna,” IEEE Transactions on Antennas and Propagation, Vol. 31, No. 5, May 1983, pp. 406–412.

7. K.M. Luk and K.W. Leung, Dielectric Resonator Antennas, Research Studies Press Ltd., Hertfordshire, UK, 2003.

8. A.A. Kishk, M. Zunoubi and D. Kajfez, “A Numerical Study of a Dielectric Disk Antenna Above Grounded Dielectric Substrate,” IEEE Transactions on Antennas and Propagation, Vol. 41, No. 6, June 1993, pp. 813–821.

9. H.A. Wheeler, “The Radian Sphere Around a Small Antenna,” Proceedings of the IRE, Vol. 47, No. 8, 1959, pp. 1325–1331.

10. E.H. Newman, P. Bohley and C.H. Walter, “Two Methods for the Measurement of Antenna Efficiency,” IEEE Transactions on Antennas and Propagation, Vol. 23, No. 4, April 1975, pp. 457–461.

11. R. Chair, K.M. Luk and K.F. Lee, “Radiation Efficiency Analysis on Small Antenna by Wheeler Cap Method,” Microwave and Optical Technology Letters, No. 2, April 2002, pp. 112–113.

12. G.P. Junker, A.A. Kishk and A.W. Glisson, “Input Impedance of Dielectric Resonator Antennas Excited by a Coaxial Probe,” IEEE Transactions on Antennas and Propagation, Vol. 42, No. 7, July 1994, pp. 960–966.

13. A.A. Kishk, H.A. Auda and B.C. Ahn, “Radiation Characteristics of Cylindrical Dielectric Resonator Antennas with New Applications,” IEEE Antennas and Propagation Society Newsletter, Vol. 31, 1989, pp. 7–16.

14. S.A. Long, L.C. Shen and P.B. Morel, “Theory of the Circular Disk Printed-circuit Antenna,” Proceedings of the IEEE, Vol. 125, 1978, pp. 925–928.

15. A.G. Derneryd, “Analysis of the Microstrip Disk Antenna Element,” IEEE Transactions on Antennas and Propagation, Vol. 27, 1979, pp. 660–664.

16. S. Yano and A. Ishimaru, “A Theoretical Study of the Input Impedance of a Circular Microstrip Disk Antenna,” IEEE Transactions on Antennas and Propagation, Vol. 29, 1981, pp. 77–83.

17. W.C. Chew, J.A. Kong and L.C. Shen, “Radiation Characteristics of a Circular Microstrip Antenna,” Journal of Applied Physics, Vol. 41, 1980, pp. 3907–3915.

18. R.K. Mongia, A. Ittipiboon and M. Cuhaci, “Measurement of Radiation Efficiency of Dielectric Resonator Antennas,” IEEE Microwave and Guided Waves Letters, Vol. 4, No. 3, March 1994, pp. 80–82.