By continuing to browse this site you agree to us using cookies as described in About Cookies
Notice: Please be advised that we experienced an unexpected issue that occurred on Saturday and Sunday January 20th and 21st that caused the site to be down for an extended period of time and affected the ability of users to access content on Wiley Online Library. This issue has now been fully resolved. We apologize for any inconvenience this may have caused and are working to ensure that we can alert you immediately of any unplanned periods of downtime or disruption in the future.
 The International Special Committee on Radio Interference (CISPR) (CISPR 16-1-4) measured the site voltage standing wave ratio (SVSWR) to determine the site imperfections of a radiated emission measurement site above 1 GHz. However, the technical background of the SVSWR measurement parameters is not included in this standard. In this paper, we first propose a simple propagation model to evaluate the relation between the SVSWR parameters and the measurement values. The calculated SVSWR values of the total displacement defined in CISPR 16-1-4 (0.4 m) were almost the same as those obtained when the transmitting antenna was moved up to 2.0 m from the reference point. In addition, the SVSWR values obtained with a 50 MHz frequency resolution, as defined in CISPR 16-1-4 were almost the same as those values obtained with a 1 MHz resolution. From these results, we confirmed that the CISPR measurement parameters are reasonable and effective.
 The International Special Committee on Radio Interference (CISPR) has standardized the method for measuring electromagnetic disturbance in the frequency range of 1–18 GHz (CISPR 16-2-3 [CISPR, 2006a]. CISPR also standardized the limit of interference due to radiation from information technology equipment in the 1–6 GHz range (CISPR 22 [CISPR, 2006b]). Measurement uncertainty for emission measurements above 1 GHz have been discussed (CISPR 16-4-2 [CISPR, 2011]).
 Measurement uncertainty is affected by site imperfections (expressed by the site voltage standing wave ratio (SVSWR)) (CISPR 16-1-4 [CISPR, 2007]), the material of the setup table, receiver characteristics, and other factors. To confirm the relevance of the CISPR uncertainty estimate (CISPR 16-4-2 [CISPR, 2011]), we determined these quantities with a round robin test [Tosaka and Yamanaka, 2009]. Then, we measured the SVSWR and the field strength radiating from the equipment under test (EUT) at 13 measurement sites. However, we were unable to determine a clear relationship between the SVSWR of a specific site and measured field strength of EUT on this site due to data dispersion in measurement results that were introduced by other factors such as the setup table's material and the receiver.
 We thus measured the SVSWR and field strength radiated from an EUT by changing only the area covered by the RF absorber [Tosaka and Yamanaka, 2010]. The results show a clear relationship between SVSWR and the measured field strength, and we can use the measured SVSWR of the test site to estimate the data dispersion of the field strength. We can therefore determine the measurement uncertainty of the measured field strength due to the test site. The technical background of the SVSWR measurement parameters like, the total displacement of the transmitting antenna and frequency step size are not well documented.
 In this paper, we first introduce a simple propagation model to evaluate the relationship between the SVSWR values and the measurement parameters, and then evaluate the SVSWR measurement positions from the measurement.
2. SVSWR Measurement Method
 To evaluate the effect of the reflected wave from defects or surrounding objects of a test site, SVSWR is measured based on CISPR 16-1-4 [CISPR, 2007], where this parameter can be expressed in dB as
where Vmax,dB and Vmin,dB are the measured levels of maximum and minimum values in dB, respectively, when the position of the transmitting antenna is changed according to the standard. These levels are calculated by normalizing the measured levels at any position other than the reference position to the reference position by assuming free space propagation, as follows:
where Vmeas.(n) represents the measured level at the nth (n = 1 to 6) position; D the distance between the transmitting antenna at a reference position and the receiving antenna; and d(n) the moving distance from the reference position.
Figure 1 shows the setup for a SVSWR measurement. The transmitting and receiving antennas are connected to a network analyzer. To measure the SVSWR, an omni-directional dipole antenna is used as the transmitting antenna. S21 is the S parameter measured with RF absorbers are placed between the transmitting and receiving antennas.
 The SVSWR values are measured at four reference positions (F6, C6, R6, and L6), indicated by gray symbols in Figure 1. The distance from the receiving antenna to the reference position F6 is basically 3 m. However, the distance of C6, R6, and L6 depends on the size of the test volume. Then, S21 was measured when the transmitting antenna was moved to distances of 2, 10, 18, 30, and 40 cm from the reference positions, and SVSWR values were calculated using equation (1). SVSWR must be measured at 50 MHz frequency increments in the frequency range of interest (for example, 1–6 GHz per CISPR 22 [CISPR, 2006a]). Using the maximum value of the SVSWR in that frequency band, the suitability of the emission measurement site is determined. The SVSWR values have to be 6 dB or less in order to compliant with CISPR 16-1-4 [CISPR, 2007].
3. Calculation Model of SVSWR
 In the SVSWR measurement, the standing wave ratio between the direct wave (ED) and reflected wave from surrounding objects is measured by changing the distance between a transmitting and a receiving antenna, and SVSWR is mainly affected by the reflected wave from the ground plane (EG), the left side (EL), the right side (ER), the ceiling (EC), and the behind of transmitting antenna (EB) in an anechoic chamber due to the reflection from the absorbing material or exposed metal surfaces of the ground plane. Figure 2 shows our propagation model. Figure 3 shows our calculation model reflected from ground plane. It can be assumed that the equivalent reflection plane exist on the top of the pyramidal absorber at GHz range [Shimizu and Nishimura, 1987]. The received field strength E at Rx can be expressed as the combination of the vector between ED, EG, EL, ER, EC, and EB, and is expressed as
where k is the wave number, R1, R2 and R3 are the reflection coefficients of the absorbers on the ground plane, that on the left and right panels, and that on the ceiling and the panel behind the transmitting antenna. rg, rl, rr, rc, and rb are the propagation path length for the reflected waves at ground, left panel, right panel, ceiling, and the panel behind the transmitting antenna, respectively. E, ED, EG, EL, ER, EC, EB, E0, R1, R2, and R3 are complex values. To simplify the calculation, we assumed E0 as 1.
 SVSWR can be calculated in dB changing the distance d(n) in Figure 3 as
where Emax,dB and Emin,dB show the field strength of the maximum and minimum values in dB, respectively. These field strengths are corrected to the values at the reference position by assuming the free space propagation, as follows:
where E(n) represents the field strength at the nth (n = 1 to 6 in CISPR measurement) position and d(n) is the moving distance from the reference position.
4. SVSWR Calculation
 The SVSWR of a test site is determined from the maximum value in the frequency range to be evaluated. We first calculated the SVSWR by a 1 MHz step using our model at three types of measurement site model, model 1: h = 1.0 m and t = 0.3 m; model 2: h = 1.3 m and t = 0.3 m; and model 3: h = 2.0 m and t = 0.3 m, respectively. To determine the SVSWR, we first calculate the field strength from equation (5). As an example, we calculate the field strength at model 1 when R1 = −0.3, R2 = −0.01, and R3 = −0.01. Basically, R1, R2 and R3 depend on the polarization and the incident angle to the wave absorber. In the actual SVSWR measurement defined by CISPR, a dipole antenna is used for transmitting antenna. Thus, SVSWR is mainly affected by the ground plane reflection and SVSWR for horizontal polarization is larger than that for vertical polarization. So, we consider only horizontal polarization that the direction of electric field was only existed for x direction in Figure 3. Thus, the incident wave on an absorber can be assumed oblique incidence for TE polarized wave. The changes of the incident angles are negligible since the moving distance is much smaller than the measurement distance. Accordingly, we assume that R1, R2, and R3 are constant.
 The result of the calculated field En,dB according to equation (5) is shown in Figure 4. The SVSWR is calculated by using the values from equation (4) as a function of frequency when R1 was changed from −0.3 to −0.5, since SVSWR is almost 6 dB, which is required by the CISPR, if these numbers are used for R1 when the antennas are placed at the heights of 1 m to 2 m above the ground plane. Figures 5, 6, and 7 show the calculated SVSWR when varying R1 from −0.3 to −0.5 and R2 and R3 were −0.01. From Figures 5, 6, and 7, it can be determined that the 6 dB acceptance criterion for SVSWR specified by CISPR corresponds to R1 = −0.355 at model 1, R1 = −0.382 at model 2, and R1 = −0.468 at model 3. These values are used in the following calculation.
5. Effect of Parameters on SVSWR Values
 To measure the SVSWR, the positions of a transmitting antenna and the frequency step size are specified in CISPR 16-1-4 [CISPR, 2007]. The rationale for the selection of these parameters is not described in CISPR 16-1-4 [CISPR, 2007]. In this section, we discuss the relationships between the SVSWR values and these parameters.
5.1. Frequency Step
 To evaluate the effect of the frequency step size, we calculated the SVSWR at each site model with two frequency steps of 1 and 50 MHz. In this calculation, the transmitting antenna is moved to the locations specified in CISPR 16-1-4 [CISPR, 2007].
 The calculated SVSWR values are shown in Table 1. The difference in SVSWR values measured with a step size of 1 MHz (CISPR1MHz in Table 1) and 50 MHz (CISPR50MHz in Table 1) was negligible and less than 0.07 dB (maximum difference was R1 = −0.5 at model 3). This means that the SVSWR measurement using a 50 MHz step is reasonable and the evaluation time is much shorter.
Table 1. Comparison Between the SVSWR's by CISPR Frequency Step (50 MHz) and by a 1 MHz Step at Three Different Site Modelsa
Here 50 MHz is the frequency step defined by CISPR. SVSWR's are shown in dB.
Model 1 (1 MHz)
Model 1 (50 MHz)
Model 2 (1 MHz)
Model 2 (50 MHz)
Model 3 (1 MHz)
Model 3 (50 MHz)
 We also calculated the SVSWR values for frequency step sizes of 2, 5, 10, 20, 100, 200, 500, and 1000 MHz, and then we determined the difference between the values achieved with 1 MHz and other frequency steps at each measurement site model, as shown in Figure 8. The 6 dB values at models 1, 2, and 3 are R1 = −0.355, −0.382, and −0.468, respectively. These values indicate that increased frequency step size lead to larger differences. In the case of 50 MHz specified in CISPR 16-1-4 [CISPR, 2007], the difference was less than 0.1 dB at each model. If the 0.2 dB difference is permitted in the SVSWR measurement, a 200 MHz step may be used.
5.2. Total Displacement of Transmitting Antenna
 In a CISPR-based measurement, the typical measurement distance is 3 m in the front position and the SVSWR is measured when the transmitting antenna is moved to 3.4 m. However, the total displacement of the transmitting antenna is not clearly evaluated in the CISPR document, to determine if a value of 0.4 m is sufficient. Therefore, we calculated the SVSWR when the transmitting antenna is moved up to 2.0 m with an increment of 1 cm. We assumed the diameter of the test volume to be 2.0 m, and we calculated the SVSWR using a 50 MHz frequency step size. Figures 9, 10, and 11 show the relationship between SVSWR and the total displacement of the transmitting antenna at each test site model. Each result shows that SVSWR increases rapidly with the distance first, and then gradually increases. The differences between the SVSWR's at 3.4 m and at 5.0 m are 0.49 dB in model 1, 0.76 dB in model 2, and 1.24 dB in model 3, respectively. Those results indicate that if h becomes larger then SVSWR converges to a lower number.
 At lower height of the antennas the path length difference between direct and reflected waves is not changed by the varying the position of the transmitting antenna from 3.4 m to 5.0 m. The SVSWR differs more significantly at higher antenna heights since the path length difference becomes larger and ED and EG may be in anti-phase. However, in an actual SVSWR measurement, EG is the vector sum of many reflected waves having different amplitudes originating from various surrounding object at the test site; therefore, perfect distractive summation seldom occurs.
 From the above consideration, it may be concluded that the variation distance of 0.4 m is reasonable for the total displacement of the transmitting antenna for small EUTs. For larger or taller EUTs longer displacement distances may be required.
5.3. Transmitting Antenna Position
 We found that the max SVSWR mainly depends on the total displacement of the transmitting antenna, i.e. reference and end positions, and does not clearly on the number of positions. Accordingly, we evaluated the effect of the transmitting antenna positions on the max SVSWR from measurement.
Figure 1 shows the general measurement setup. h is 1.0 m and the diameter of the test volume is 1.2 m. We consider three configurations of the RF absorber on the floor, as shown in Figure 12. The placement areas of the RF absorber at the measurement site are shown in small gray squares depicting absorber squares with dimensions 0.6 × 0.6 × 0.45 m. The dimensions of the emission measurement site (semi-anechoic chamber) were 6.0 × 8.0 × 5.5 m (W × L × H).
 The SVSWR value strongly depends on the placement of the RF absorbers. Thus, we changed the SVSWR by controlling the placement as shown in Figure 12. Figures 13 and 14 show the measured SVSWR value for the horizontal polarization at position F, and C of the test volume. The maximum SVSWR values in configuration a, b, and c were shown in Table 2 at all positions of the test volume, and these values were less than the CISPR acceptance criterion of 6 dB.
Table 2. Maximum SVSWR at Each Position
 To find the relation between the number of positions of transmitting antenna and the SVSWR, we determine the max value of the SVSWR for less number of positions shown in Table 3. For examples, Figures 15 and 16 show the results for sites a and c at position F. As a whole, the SVSWR becomes larger as the number of positions increase. The dependency of the max SVSWR on the number of positions are summarized in Figures 17 and 18 at positions F and C. From Figures 15 and 16, we found that if the value of SVSWR become smaller, the same max SVSWR is obtained by a smaller number of positions. Thus, we can evaluate the site having a small number of SVSWR using a smaller number of positions, extremely by two positions. On the other hand, as the SVSWR become larger, at least five or six positions are necessary for correctly determining the SVSWR value through measurement.
Table 3. Transmitting Antenna Positions in the SVSWR Measurement
Number of Positions
Transmitting Antenna Positions
3.0 and 3.4 m
3.0, 3.02, and 3.4 m
3.0, 3.02, 3.1 and 3.4 m
3.0, 3.02, 3.1, 3.18, and 3.4 m
3.0, 3.02, 3.1, 3.18, 3.3, and 3.4 m
 From the above measurement results, we can conclude that the six-position measurement specified by CISPR is considered to be reasonable and proper.
 In CISPR 16-1-4 [CISPR, 2007], the site voltage standing wave ratio (SVSWR) is measured to evaluate the site imperfection of a radiated emission measurement site above 1 GHz. However, some of the related measurement parameter are not explained in detail in this document.
 In this paper, we first proposed a simple propagation model to evaluate the relationship between the SVSWR values and the measurement parameters. The calculated SVSWR values of the total displacement defined in CISPR 16-1-4 [CISPR, 2007] (0.4 m) were almost the same as those obtained when the transmitting antenna was moved up to 2.0 m from the reference point. The similar SVSWR results were obtained when using a 50 MHz frequency step size as defined in CISPR 16-1-4 [CISPR, 2007] and when using a 1 MHz frequency step size. We then evaluated the effect of the transmitting antenna positions on the SVSWR through measurement. We found that if the site has better characteristics, then the same max SVSWR is obtained by a smaller number of positions. On the other hand, if the site becomes worse, five or six positions are necessary at least for correctly determining SVSWR value.
 From these results, we confirmed that CISPR measurement parameters are reasonable and effective. In addition, we also found that a larger measurement frequency step size can be used for the SVSWR measurements and that smaller numbers of antenna positions may be sufficient for the measurement of smaller SVSWR values.