Measurement of UHF RFID Tag Antenna Impedance

Measurement of UHF RFID Tag Antenna Impedance Xianming Qing, Chean Khan Goh, and Zhi Ning Chen Institute for Infocomm Research, 1 Fusionopolis Way, #21-01 Connexis, South Tower, Singapore 138632 Email: { qingxm, ckgoh, chenzn}@i2r.a-star.edu.sg ABSTRACT: A methodology to measure UHF RFID tag antenna impedance using network approach is presented. The balanced UHF RFID tag antenna is considered as a two-port network and the impedance of the antenna is characterized by network parameters. In the measurement, the balanced antenna is connected to the two ports of a two-port vector network analyzer (VNA) through a test fixture. The influence of the test fixture is de-embedded by using a portextension technique and then the antenna impedance is extracted directly from the measured S-parameters. The proposed method is validated by comparing the measured and the simulated impedance of an asymmetrical dipole antenna and a symmetrical meander line dipole antenna both for UHF RFID tags. INTRODUCTION The impedance of balanced antennas cannot be measured directly using most of measurement instruments since these instruments are terminated with un-balanced ports such as coaxial ports. When a balanced antenna is connected to an un-balanced test port, the currents fed to the two radiators of the antenna are unequal and thus the impedance of the balanced antenna cannot be characterized correctly. Meys and Janssens [1] developed a way of measuring the symmetrical balanced antennas using two ports S-parameters in 1998. Two 50-Ω microstrip lines, which were etched onto a printed circuit board (PCB) and mounted back-to-back, were used as a test fixture to connect a two-port vector network analyzer (VNA) to the antenna under test. The testing system was calibrated at the end of the microstrip lines by using a set of special-purpose standards. This method was able to characterize the impedance of a symmetrical dipole antenna up to 1 GHz. After that, Palmer and Rooyen [2], in 2006, proposed a method by using coaxial cables to replace the microstrip lines with the standard calibration. This technique allowed broadband impedance measurement. However, this technique could not extract the antenna impedance directly from the measured S-parameters. Instead, tedious computation, such as conversion of S-parameters to ABCD-parameters and Y-parameters, together with pre- and post-multiplication of ABCD-parameters, must be used to de-embed the effect of the coaxial cables. The method to characterize the impedance of asymmetrical/symmetrical balanced antennas is necessary in design of UHF RFID tag antennas because the majority of UHF RFID tag antennas are balanced designs while the VNAs very often used to measure the input impedance of the antennas are unbalanced. In this paper, the methodology of characterizing the impedance of asymmetrical balanced antennas is presented for UHF RFID tag antenna design. The impedance of the balanced antenna is expressed by using S-parameters of the equivalent network. Together with the coaxial fixture and the port-extension technique, the impedance of the balanced antennas can be extracted directly from the measured S-parameters. By the proposed method, the impedance of RFID tag antenna can be designed with desired value for matching to the specific integrated circuit (ASIC), which is vital for achieving maximum transfer between them and therefore achieving largest reading range. S-PARAMETER-BASED METHOD Fig. 1 shows the equivalent network of a typical asymmetrical dipole antenna [3]. The differential impedance of the antenna can be expressed as Zd =

V d V1 − V 2 = I I

(1)

Based on the definition of Z-parameters, the voltages of the ports can be expressed as V1 = Z 11 I 1 + Z 12 I 2

(2)

V 2 = Z 21 I 1 + Z 22 I 2 .

(3)

978-1-4244-4396-3/09/$25.00 ©2009 IEEE

Authorized licensed use limited to: ASTAR. Downloaded on May 19, 2009 at 05:04 from IEEE Xplore. Restrictions apply.

Fig. 1 Network representation of an asymmetrical dipole antenna and schematic diagram. The differential impedance of the antenna can be found as follows Zd

V d V1 − V 2 = I I

=

= (

(4)

Z 11 − Z 21 − Z 12 + Z 22 )

Transforming Z-parameters to S-parameters, the differential impedance can be expressed by using S-parameters as follows Zd

=

2 Z 0 (1 − S11 S 22 + S12 S 21 − S12 − S 21 ) (1 − S11 )(1 − S 22 ) − S 21 S12

.

(5)

For symmetrical balanced antenna, the equivalent network is symmetrical as well, namely S11 = S22, and S12 = S21, Equation (5) can be simplified to Zd

=

2 Z 0 (1 − S 112 + S 212 − 2 S 12 ) (1 − S 11 ) 2 − S 212

(6)

MEASUREMENT METHODOLOGY The setup of measurement is illustrated in Fig. 2 (a). The measurement is conducted using a two-port VNA (Agilent N5230A) and a test fixture. The test fixture is constructed with two coaxial cables which are soldered together on their outer conductors. One end of the fixture is with two SMA connectors and is connected to the ports of the VNA through the test cables. Another end of the fixture is open with small extensions of inner conductors to form the tips to connect to the input of antenna under test. In the measurement, the fixture prototype is constructed using two semi-ridge coaxial cables with length of 100 mm and the diameter of outer conductor of 2.2 mm.

(a)

(b)

Fig. 2 (a) photo of the impedance measurement setup in a chamber; (b) schematic demonstration for portextension.


The VNA offers high measurement accuracy with standard calibration. However, as shown in Fig. 2(b), the standard calibration can only be done at the ends of the test cables instead of the input port of the antenna under test. The fixture suffers frequency-dependent phase delay and loss, and therefore affects the accuracy of S-parameter measurement. To characterize the antenna impedance precisely, the fixture effect must be eliminated from the measurement. This can be done by using the port-extension function of the VNA to shift the calibration plane electrically from the ends of the test cables to the tip of the fixture where the antenna is connected. Fig. 3 illustrates how the port-extension technique eliminates the influence of the text fixture. The measured S11 (or S22) of the short-circuited test fixture shows multiple turn response in a Smith chart (Fig. 3(a)). The response is introduced by the test fixture and is dependent on the length of the test fixture and the measured frequency range. After performing the port-extension, the trace is concentrated on the most left side of a Smith chart (Fig. 3(b)), which indicates that the short-circuited test fixture is calibrated properly at the end of the test fixture.

(a)

(b)

Fig. 3 Measured impedance with short-circuited test fixture over 0.5−6 GHz; (a) before port-extension, (b) after port-extension. The procedures for the measurement are as follows: 1. Calibrate the VNA with standard calibration kit over the desired frequency range. 2. Connect the test fixture to the ports of the VNA and perform the port-extension to move the calibration plane from the ends of the test cables to the tips of the test fixture. 3. Connect the antenna to the test fixture and carry out the S-parameters measurement. 4. Extract the impedance using Equation (5) or (6). RESULTS AND DISCUSSIONS To validate the proposed method, an asymmetrical dipole antenna and a symmetrical meander line dipole antenna, which have been commonly used in UHF RFID antenna design were fabricated and measured. The measured impedance was then compared with the simulated impedance by IE3D [4]. Fig. 4 shows the configurations and dimensions of the measured antenna prototypes. The antenna was prototyped on an FR4 PCB with a thickness of 0.508 mm, dielectric constant of 4.4 and loss tangent of 0.027. Fig. 5(a) shows the comparison of the measured and simulated impedance of the asymmetrical dipole antenna. The measurement was done over the frequencies of 0.5–1.5 GHz. Good agreement between the simulated and measured results are observed. Fig. 5(b) exhibits the measured and simulated impedance of a meander line tag antenna over 0.8– 1.0 GHz. The measured impedance of small resistance and large reactance agree well with that of the simulation.


(a)

(b)

Fig. 4 Configurations of the antennas under test; (a) asymmetrical dipole; (b) meander line dipole.

R (ohm)

350 300

100

100

90

50

80

0

70

-50

250 -100

200

-150

150 100

-200

50

-250

0 0.5

Measured R Simulated R Measured X Simulated X

600 500 400

60 300

50 40

X (ohm)

400

150

R (ohm)

Measured R Simulated R Measured X Simulated X

450

X (ohm)

500

200

30

100

20

0.6

0.7

0.8

0.9

1.0

1.1

1.2

Frequency (GHz)

(a)

1.3

1.4

-300 1.5

0

10 0.80

0.85

0.90

0.95

1.00

Frequency (GHz)

(b)

Fig. 5 Comparison of the measured and simulated impedance of the antennas; (a) asymmetrical dipole antenna, (b) symmetrical meander line dipole antenna.

CONCLUSION An experimental methodology for characterizing the impedance of balanced UHF RFID tag antennas has been investigated. Using a two-port vector network analyzer, together with port extension technique, the impedance of the measured antenna can be extracted directly from the measured S-parameters. The methodology has been validated by measuring the impedance of an asymmetrical dipole antenna, and a meander line dipole antenna. The measured results have shown a good agreement with the simulated ones. REFERENCES [1] [2] [3] [4]

R. Meyers and F. Janssens, “Measuring the impedance of balanced antennas by an S-Parameter method,” IEEE Antennas and Propag. Mag., Dec. 1998, vol.40, pp. 62-66. K. D. Palmer and M. W. V. Rooyen, “Simple broadband measurements of balanced loads using a network analyzer,” IEEE Trans. Inst, and Meas., Feb. 2006, vol. 55, pp. 266- 272. X. Qing, C. K. Goh, and Z. N. Chen, “Impedance Characterization of Asymmetrical Balanced Antennas and Application in RFID Tag Design,” IEEE trans. Microwave theory and technique, to be published. IE3D version 11.01, “Zeland Software Incorporation,” Fremont, Calif, USA.
