High growth in the electromagnetic compatibility (EMC) field makes antennas absolutely important part in EMC test. There are two reasons below:

1) EMC test need: Antennas are used in many key EMC tests, antenna factors (AF) determined is a major work in making field strength measurements accurate for EMC compliance. For example, in the electromagnetic radiation disturbance test, we use antennas to receive the interference noise and transfer it to EMI receiver through coaxial cable.

2) Site validation need: Antennas are used to validate the performance of test sites such as Open Area Test Sites (OATS) and anechoic chambers, by measuring the parameter of normalized site attenuation (NSA) with ANSI C63.4 standard.

We can find out that the performance of the antennas affects the test results, accurate measurement and traceability of antennas are both important to EMC test. The measurement methods are prescribed in standards such as those published by the ANSI and CISPR [1]-[6]. Three commonly used measurement methods used to calibrate the AF are Reference Antenna Method (RAM), Equivalent Capacitance Substitute Method (ECSM) and Standard Site Method (SSM). These methods’ advantages and limitations are illustrated clearly in many papers.

ECSM is used to calibrate AF of monopole antennas frequency ranges from 9 kHz to 30 MHz according to standard ANSI C63.5-2003. RAM provide an AF measurement method based on the use of dipole with well?matched balun in 30 kHz to 1000 MHz. SSM is the most used method to measure AF parameter from 30 MHz to 1000 MHz. This paper provides the setup of SSM test system on OATS in Zhejiang Province of China.

In this paper, we adopt SSM method for the calibration of antennas obtained by the C63.5 procedure. To satisfy our own calibration condition, we establish a test system, which was shown in Figure 2. It consists of computer, frequency method, signal generator, spectrum analyzer, antenna amounting bracket and its controller. Here we choose discrete frequencies.

For AF measurement experiments, standard biconical antennas and log-pe- riodic antennas of Shanghai Institution of Measurement and Testing Technology (SIMT) are achieved on OATS in Zhejiang province as Figure 2 shows. Their specifications are shown in Table 1 and Table 2. To ensure the accuracy and reliability of the test, the AF data of standard antennas are also calibrated by the National institute of metrology on OATS.

We get the insertion loss (IL) of different frequency points with spectrum analyzer, and calculate AF data according to Formula (1) to Formula (3). By analyzing the above equations, AF is related to site attenuation and the maxi-

mum received field at separation distance. Figure 3 presents the AF curves of two biconical antennas measured by SIMT and National Institute of Metrology (NIM) and Figure 4 presents the AF curves of two log periodic antennas measured by SIMT and NIM separately.

Both curves obtained by SIMT and NIM keep in consistent at desired frequency ranges. Now we focus on the comparison between two testing organizations for two different antennas respectively.

Figure 3 indicates that there are certain differences of data measured by SIMT and NIM, especially in the frequency range from 80 MHz to 90 MHz. One of the most important reasons is that as a representative EMC antenna, biconical antenna has lower frequencies (range from 30 MHz to 300 MHz), so it is difficult to eliminate the influence of the OATS’ ground, so we should make sure that the floor is well conductive before test.

Figure 4 shows apparently that AF data measured by SIMT is higher than that of NIM. We should consider the mutual coupling between two antennas at 10 m distance. Besides of that, antennas are affected by different measurement system and equipments, OATS is an important influencing factor that cannot be neglected.

In the process of calibrated AF parameter, we should consider the uncertainty caused by measurement instruments and the factors of antenna coupling and OATS.

In metrology, measurement uncertainty is a parameter characterizing the dispersion of the values attributed to a measured quantity [7]. For antenna calibration, the measurement must include the uncertainty components, taking into consideration the environment in which the antenna is to be used for the testing.

The AF measurement process includes a spectrum analyzer, signal generator, amplifier, cables, attenuators, OATS. All these issues should be taken into ac-

count in the uncertainties determination. The contribution to measurement uncertainty from the system can be determined by using type A evaluation and other contribution can be determined by using type B evaluation.

Considering the uncertainty of AF measurement for log periodic antennas, here we discuss standard uncertainty components to each factor respectively.

1) Spectrum analyzers factor: Spectrum analyzers have two primary sources of uncertainty that shall be considered with respect to the calibration procedure. These sources are frequency accuracy, and amplitude accuracy. The influence of frequency accuracy can be ignored with respect to amplitude accuracy. So in the frequency range from 200 MHz to 1000 MHz, we get the uncertainty component:

2) Signal generator factor: The amplitude uncertainty contribution based on the amplitude stability is:

3) Impedance mismatch of Spectrum analyzer and antenna: In the frequency band of 200 MHz - 1000 MHz, VSWR of antenna (connected with 6dB attenuator) is 1.2 and that of Spectrum analyzer is 1.07, which are impedance mismatch, so the uncertainty component caused by the receive end is:

4) Impedance mismatch of Signal source and antenna: In the frequency band of 200 MHz - 1000 MHz, VSWR of antenna (connected with 6 dB attenuator) is 1.2 and that of signal generator is 1.5, which are impedance mismatch, so the uncertainty component caused by the transmitting end is:

5) Site shortage: Antenna factors can vary within the scope of ±1 dB due to mutual coupling with the ground plane. In every frequency band, we can look up table and get the maximum uncertainty value:

6) Repeatability of system: Measure NSA (unit: dBμV) for 10 times in the frequency point 100 MHz by using Log Periodic Antenna (Type: HK223, No: 100183). The measured data are: 25.1, 24.3, 24.6, 24.1, 24.2, 25.3, 24.2, 24.1, 25.3, 24.6, 24.1.

7) Change of measurement distance: We can find value from the Uncertainty calculation worksheet (Table 3):

Now we can get the AF combined uncertainty of Log Periodic Antenna:

In the end, we get AF expanded uncertainty of Log Periodic Antenna:

To same Log Periodic Antenna measured, we choose two data measured by SIMT and NIM respectively, which have maximum difference in same frequency point. As discussed above, we trace the standard antennas to NIM, which is the highest traceability institutions and has lower uncertainties than SIMT.

Table 4 shows, there exists maximum difference in the 300 MHz frequency point, the AF data is 14.7 (dB/m) measured by SIMT and AF data is 12.6 (dB/m) by NIM. So we get:

The results above show that:

So, the conclusion is that measured results on OATS by SIMT meet the requirements in all frequency points.

This paper reviews aspects of the standard site method in the American National Standards Institute’s ANSI C63.5 frequency band of 30 MHz to 1000 MHz is which calibrated using the standard method on OATS. Then it analyzes the issues affecting the measurement uncertainty and evaluates the expanded uncertainty of log-periodic antenna. In the end, it compares the data measured by SIMT and NIM, and verifies that our AF calibration method satisfies requirement.

Chen, C.C., Shi, H., Miao, Y. and Sang, Y. (2017) The Measurement and Uncertainty Analysis of Antenna Factor of Microwave Antennas Based on Standard Site Method. Int. J. Communications, Network and System Sciences, 10, 138-145. https://doi.org/10.4236/ijcns.2017.105B013