Bioelectromagnetics 36:517^526 (2015)

SAR Compliance Assessment of PMR 446 and

FRSWalkie-Talkies
G€
unterVermeeren,* Wout Joseph, and Luc Martens
Department of InformationTechnology, iMinds/UGent, Ghent, Belgium
The vast amount of studies on radiofrequency dosimetry deal with exposure due to mobile devices
and base station antennas for cellular communication systems. This study investigates compliance
of walkie-talkies to exposure guidelines established by the International Commission on Non-
Ionizing Radiation Protection and the Federal Communications Committee. The generic walkie-
talkie consisted of a helical antenna and a ground plane and was derived by reverse engineering of
a commercial walkie-talkie. Measured and simulated values of antenna characteristics and
electromagnetic near fields of the generic walkie-talkie were within 2% and 8%, respectively. We
also validated normalized electromagnetic near fields of the generic walkie-talkie against a
commercial device and observed a very good agreement (deviation <6%). We showed that peak
localized specific absorption rate (SAR) induced in the oval flat phantom by the generic walkie-
talkie is in agreement with four commercial devices if input power of the generic walkie-talkie is
rescaled based on magnetic near field. Finally, we found that SAR of commercial devices is within
current SAR limits for general public exposure for a worst-case duty cycle of 100%, that is, about 3
times and 6 times lower than the limit on the 1 g SAR (1.6 W/kg) and 10 g SAR (2 W/kg),
respectively. But, an effective radiated power as specified by the Private Mobile Radio at 446 MHz
(PMR 446) radio standard can cause localized SAR exceeding SAR limits for 1 g of tissue.
Bioelectromagnetics. 36:517–526, 2015. © 2015 Wiley Periodicals, Inc.

Key words: modeling; exposure; two-way radio transceiver; absorption assessment; helical
antenna

INTRODUCTION In recent years, walkie-talkies are more fre-

quently encountered as a short-range communication
Although a vast amount of studies on dosimetry
device (e.g., adults give a walkie-talkie to their
deal with exposure due to mobile devices and base
children when they go to play outside their house;
station antennas for cellular communication systems,
parking lot attendants employ a walkie-talkie to direct
the mobile phone is not the only device operated in
the traffic inside a parking lot to assist drivers in
proximity of the human head. Using the case study of
finding a free parking spot, etc.) Advantages of
the walkie-talkie, we draw attention to a device that
walkie-talkies with respect to cellular phones are their
gained interest as an alternative to cellular phones for
ease-of-use (push-to-talk), their inexpensiveness, and
short-range communications. Walkie-talkies, or two-
free-of-charge communications.
way radios, operate close to the human head as is the
In Europe, a walkie-talkie operates according to
case for mobile phones. Typical operating positions of
the Private Mobile Radio at 446 MHz (PMR 446)
a walkie-talkie are in front of the face, whereas a
standard; in the United States, Family Radio Service
mobile phone is typically operated next to the ear.
(FRS) is authorized for license-free short-range voice
When operating a walkie-talkie, its antenna could be
communication [ERC, 1998; ETSI, 2001a, 2015;
just in front of the eye. The International Commission
on Non-Ionizing Radiation Protection (ICNIRP)
restricts local-averaged specific absorption rate (SAR) Conflicts of interest: None.
in 10 g (applicable in Europe) based on cataract in the *Correspondence to: G€unter Vermeeren, iMinds/UGent-INTEC,
eye of a rabbit [ICNIRP, 1998]. As opposed to mobile Gaston Crommenlaan 8 box 201, Ghent B-9050, Belgium.
phones, no or little attention is paid to walkie-talkies E-mail: [email protected]
in dosimetry, although compliance tests also apply to
these devices. Cecil et al. [2014] and Dimbylow et al. Received for review 21 August 2014; Accepted 13 August 2015
[2003] investigated numerically peak 10 g localized DOI: 10.1002/bem.21935
SAR induced by Terrestrial Trunked Radio (TETRA) Published online 6 September 2015 in Wiley Online Library
transmitters positioned around the human body. (wileyonlinelibrary.com).

2015 Wiley Periodicals, Inc.

518 Vermeeren et al.

FCC, 2015]. FRS uses 14 channels within frequency radio, that is, the COBRA MT500 (Cobra Electronics,
bands 462.5625–462.7125 MHz and 467.5625– Chicago, IL) shown in Figure 1a. We dismantled the
467.7125 MHz. PMR 446 specifies eight channels COBRA MT500 and developed a generic walkie-
within frequency range 446.0–446.1 MHz. Each chan- talkie (Fig. 1b) based on physical dimensions of the
nel has a bandwidth of 25 and 12.5 kHz for FRS and antenna structure (including coating around antenna)
PMR 446, respectively. Maximum allowed Effective and Printed Circuit Board (PCB). The model consisted
Radiated Power (ERP) is 500 mW for both systems. of a helical antenna mounted on a rectangular ground
Frequency modulation (FM) has been adopted as plane. Terminals of the source connected the helical
modulation scheme. Relative low frequency, effective antenna and ground plane of the model. To limit
radiated power of 500 mW, use of walkie-talkies by complexity of the model, the PCB of the walkie-talkie
children, position in front of the face, and limited was replaced by a perfectly conducting ground plane
number of exposure studies for walkie-talkies and the case was removed. The helical antenna fits in
attracted our attention to these devices. a dielectric cover. This cover around the helical
The objective of this study was to develop and antenna influences antenna characteristics of the
evaluate an accurate helical antenna of a generic model. Dielectric properties of this cover were
walkie-talkie for compliance assessment with ICNIRP unknown and we derived properties by tuning simu-
[1998] and FCC [2001] guidelines. We designed a lated antenna characteristics and near-field distribu-
generic walkie-talkie and evaluated it in free space tions to measured results.
and below a flat phantom against a real device. This
validation between model and real device will allow SAR Assessment
us to use the walkie-talkie model in future studies We tested SAR compliance for the generic
(e.g., to evaluate absorption in anatomical human walkie-talkie as well as for four commercially avail-
body models.) To our knowledge, compliance of able walkie-talkies (Table 1) according to IEC stand-
walkie-talkies operating according to PMR 446 and ard 62209-2 [IEC, 2005]. We placed the walkie-talkie
FRS standards under worst-case conditions has not at a distance (d) below an oval flat phantom as shown
yet been investigated in literature. In addition, com- in Figure 2. We did not take the user’s hand into
pliance of four commercial walkie-talkies was eval- account. IEC standard 62209-2 [IEC, 2005] for com-
uated. The model is obtained by reverse engineering pliance testing does not specify hand position because
and is validated by reflection, near-field, and far-field dosimetric studies [Balzano et al., 1995; Kuster et al.,
measurements. The use of helical monopole antennas 1997a, b; Meyer et al., 2001] suggest that excluding
as a model for portable handheld devices have been the hand in modelling constitutes a conservative case
mainly discussed for mobile-phone technologies scenario for SAR in head.
[Lazzi and Gandhi, 1998; Koulouridis and Nikita,
2004] operating in frequency bands around 900 MHz Measurement Setup
and 1800 MHz. In this study, a helical monopole Measurements of reflection and input impedance
antenna model of a walkie-talkie operating at were performed in an anechoic room using a vector
446 MHz was designed to study electromagnetic network analyzer (VNA) (type HP8710, Agilent
fields induced in the human head by walkie-talkies. A Technologies [formerly Hewlett Packard], Palo Alto,
model is preferred over a real walkie-talkie device CA). The far-field characteristics in terms of Total
mainly for two reasons: firstly, antenna characteristics Radiated Power (TRP) and the Effective Radiated
(e.g., reflection at input terminals of a model can be Power (ERP) were measured. TRP measurements
measured accurately); secondly, input power is easily were performed in a reverberation chamber according
adjustable as one can feed using a signal generator to TCO’01 Certification of Mobile Phones [TCO
contrary to a real device. The model also allows study Development, 2008]. Measurement uncertainty was
of influence of dimensions and material parameters on 1 dB. ERP measurement was performed in an
antenna characteristics and absorption or SAR assess- anechoic room according to European Telecommuni-
ment in realistic human head models. cations Standards Institute (ETSI) standard EN 300
296-2:2001-03 [ETSI, 2001b]. Measurement uncer-
tainty was 3.3 dB, which is within the uncertainty
MATERIALS AND METHODS boundary of 6 dB specified by the ETSI standard. This
large uncertainty is due to simplifications built into
Generic Walkie-Talkie Design measurement methodology to reduce time and costs
We developed a generic walkie-talkie by reverse of tests. For walkie-talkies, ERP was only measured
engineering of a commercially available PMR 446 in eight different directions in the azimuth plane
Bioelectromagnetics
 SAR Compliance of Walkie-Talkies 519

Fig. 1. (a) COBRA MT500 (Cobra Electronics); (b) back of derived model with helical antenna
enclosed by a dielectric cover (bazooka balun is not shown).

resulting in an uncertainty of 3.3 dB. Near fields were generator (HP8647A, Agilent), a network analyzer
measured in an indoor environment using a robot and (HP8753E, Agilent), and the flat phantom (ELI4,
DASY3 measurement system with the following SPEAG) filled with head simulating liquid (HSL450,
probes: E-field probe ER3DV6 and H-field probe SPEAG). The DASY4 measurement system consisted
H3DV6 (SPEAG, Zurich, Switzerland). Reflections of data acquisition electronics (DAE3 from SPEAG)
of the environment and robot arm were minimized by and a dosimetric probe (ET3DV6 from SPEAG).
placing absorbers. Figure 2 shows setup for performing dosimetric
Measurement setup for SAR compliance testing measurements. The dosimetric system (probe and data
consisted of a robot (Staubli Type Rx90B L, Staubli, acquisition electronics) was attached to a robot. Robot
Pf€affikon, Switzerland), a DASY4 measurement sys- and data acquisition electronics were connected to a
tem (SPEAG), a power meter (Agilent E4419B), server and managed through the graphical user inter-
power sensors (8482H, Agilent), a directional coupler face of a laptop or desktop computer. Walkie-talkies
(HP775D Dual Directional Coupler 450-940 MC, and the generic walkie-talkie were placed at short
Agilent), an RF termination (Meca 480-1, Agilent), a

TABLE 1. Four Commercially Available Walkie-Talkies

Walkie-talkie (manufacturer)
COBRA MT 500 (Cobra Electronics, Chicago, IL)
TwinTalker 3300 (Topcom, Tilburg, The Netherlands)
TwinTalker 1300 (Topcom, Tilburg, The Netherlands)
Alecto FR-20 (Alecto Electronics, ’s-Hertogenbosch,
The Netherlands)
Fig. 2. Drawing of setup for dosimetric measurements.
Bioelectromagnetics
520 Vermeeren et al.

distance (d) below oval flat phantom. Oval flat set by the FDTD solver to obtain a selected efficiency
phantom was filled with head simulating liquid of 99.9%. Padding (minimum distance between
(HSL450 from SPEAG) as suggested by IEC 62209 absorbing boundaries [UPML] and the bounding box
[IEC, 2001]. Dielectric properties of the liquid at around the walkie-talkie or combination of walkie-
450 MHz were s ¼ 0.87 S/m and er ¼ 43.5. Separa- talkie and the flat phantom) was a quarter of a
tions between liquid and walkie-talkie were 2 mm wavelength. Grid step for the helical antenna equaled
(touch position), 15, 30, and 50 mm. For the generic 0.5 mm (133 times smaller than a tenth of the
walkie-talkie we did not measure in touch position wavelength in free space at 450 MHz), and maximum
(we did not want the ground plane of the generic grid step in the flat phantom was 2 mm (5 times
walkie-talkie to make contact with the shell of the smaller than a tenth of the wavelength in the tissue
phantom) and at 50 mm (due to antenna’s low simulating liquid at 450 MHz). Based on these
efficiency, we could not measure at this distance). In settings, we estimated uncertainty on peak local-
the case of the generic walkie-talkie, it was connected averaged SAR from the study of Bakker et al. [2010,
to a radiofrequency (RF) generator. Input power of 2011]. Expanded uncertainty U (k ¼ 2) on SAR10g in
antennas was measured by a power meter via a the flat phantom filled with tissue-simulating liquid
directional coupler. equaled 11.9% (there is no uncertainty on dielectric
The worst-case uncertainty of the DASY4 meas- properties of tissue-simulating liquid because the
urement system was mentioned in the manual values are specified by standards).
[SPEAG, 2008] provided with the system. Combined
standard uncertainty on the local-averaged SAR in 1 g
RESULTS
and 10 g equals 10.9% and 10.7%, respectively;
expanded standard uncertainty on local-averaged SAR Free-Space Evaluation of the Generic
in 1 g and 10 g equals 21.9% and 21.4%, respectively. Walkie-Talkie
This worst-case uncertainty budget for DASY4 was
Figure 3 shows dimensions of the derived
assessed according to IEEE 1528 [IEEE, 2003]. For
generic walkie-talkie operating at 446 MHz and the
specific tests and configurations, uncertainty can be
coordinate system used in this study. Model dimen-
considerably smaller [SPEAG, 2008].
sions and dielectric parameters are listed in Table 2.
The commercially available walkie-talkies were
These values can be modified to specific dimensions
placed below the flat phantom with their push-to-talk
of different types of walkie-talkies. The helical
button fixed during measurement in order to transmit
antenna fits in a dielectric cover with a relative
continuously. Before every measurement, fully-
permittivity (er) of 3.9 and a conductivity (s) of
charged batteries were placed in walkie-talkies to
12 mS/m.
ensure transmission at maximum power. A walkie-
talkie uses half-duplex communication resulting in a
varying duty cycle from 0% to 100%. A continuously
transmitting device has a duty-cycle of 100%. Hence,
a worst-case duty cycle of 100% was selected for
compliance testing.
Besides four commercially available walkie-
talkies, we also assessed peak local-averaged SAR in
1 g and 10 g for the hand-made generic walkie-talkie.
Measured peak local-averaged SAR values were
compared with simulations.
Numerical Methods
We employed the finite-difference time-domain
(FDTD) solver available in the three-dimensional full-
wave electromagnetic software package SEMCAD-X
(SPEAG) for performing numerical analysis in free
space and below the oval flat phantom. In FDTD
calculations, simulation domain is finite and boundary
conditions are used to mimic free space. We applied Fig. 3. Generic walkie-talkie in (a) xz-plane and (b) yz-plane.
uni-axial perfectly matched layers (UPML) at the Measurement line (y-axis) and plane (y ¼ 20 mm) for near
boundaries. The number of layers was automatically fields are also shown.
Bioelectromagnetics
 SAR Compliance of Walkie-Talkies 521

TABLE 2. Dimensions of a Walkie-Talkie Model Radiation efficiency (h) of an antenna is defined

as ratio of radiated power (Prad) and input antenna
Helical antenna Ground plane
power (Pin), or h ¼ Prad/Pin [Balanis, 1982]. Radiation
a 4.2 mm l 80 mm efficiency was calculated using FDTD simulations.
b 9.8 mm w 45 mm For the generic walkie-talkie, we obtained an effi-
c 10 mm t 0.5 mm ciency of 53% at 446 MHz. This low efficiency was
e 5 mm Antenna housing
due to the reflection at input terminals of the helical
f 2 mm ri 3.5 mm antenna and ohmic losses in housing around the
g 8.54 mm ro 7 mm helical antenna introduced by the conductivity (s) of
h 42 mm er 3.9
k 7.2 mm s 12 mS/m
12 mS/m.
nturns 21 hi 51 mm Far-field behavior of the generic walkie-talkie
rh 2.4 mm ho 53 mm has been investigated numerically in terms of gain
rw 0.5 mm m 1 mm (G). Using FDTD, a gain of 1.07 for simulation model
was obtained. This value is lower than the standard
gain of a half-wave dipole antenna (Gd), mainly due
to losses in the dielectric cover around the helical
We realized a balanced feed current at terminals antenna of the model. ERP can be calculated as
of the measurement model with a bazooka or a sleeve follows [FCC, 1997]:
balun [Balanis, 1982]. Operating frequency of the
bazooka balun was 446 MHz. (The bazooka balun is G
ERP ¼ Pin ð1Þ
not shown in Fig. 1b). Gd
Return loss jS11 j in dB of the model with respect Effective radiated power was measured for the
to 50 V is shown in Figure 4. A very good agreement generic walkie-talkie with coating. For an input power
is observed between simulations and measurements. of 10 mW, measured maximum ERP was 4.9 mW
Resonance frequency fres calculated by the FDTD tool (measurement uncertainty was 3.3 dB). This agreed
is 436.5 MHz and deviates only 0.3% of the measured well with simulated ERP of 4.8 mW (relative error
value fres, that is, 435.2 MHz. Simulated input impe- was 2%). Based on simulated ERP for an input power
dance at resonance Zin,res equaled 20.8 V, measured of 10 mW, we calculated that input power of the
input impedance equaled 24.5 V. At operating fre- helical antenna (or output power of generator) must
quency of 446 MHz, input impedance of model Zin be set to 1042 mW to obtain maximum allowed ERP
was 28.6 þ i30.4 V and 23.1 þ i29 V for measurement of 500 mW as specified by the PMR 446 [ERC,
and FDTD simulation, respectively. 1998] and FRS standard [FCC, 2015]. This input
We already mentioned that dielectric parameters power is used below to determine compliance of the
of the cover (relative permittivity of 3.9, conductivity model with safety guidelines [ICNIRP, 1998; FCC,
of 12 mS/m.) around the antenna were derived by 2001].
tuning these parameters until simulations agreed with Near fields of the model were simulated and
measurements in term of resonance frequency, return measured in the y-plane at 20 mm from the model
loss, and near fields. (y ¼ 20 mm) according to the setup shown in Figure 3.
This plane was parallel to the ground plane of the
generic walkie-talkie (or PCB of real devices) at a
0
distance that might be considered a typical separation
distance value when a walkie-talkie is operated in
−2 front of the face. ERP was set to 500 mW. Figure 5
shows simulated ([a] and [b]) and measured ([c] and
[d]) RMS electric Erms ([a] and [c]) and RMS
|S11| (dB)

−4
magnetic Hrms ([b] and [d]) near fields of the model at
−6 y ¼ 20 mm. One observes that distribution of the
electric and magnetic field as well as absolute field
−8 values of the simulations agree well with measure-
model (meas)
model (FDTD) ments. Dots in Figure 5 show the position of the
−10 maximum field values. The position of the maximum
400 420 440 460 480 500
Frequency (MHz) RMS E-field and H-field was situated near the helical
antenna and terminals of the model, respectively. In
Fig. 4. Return loss jS11 j of model. the plane y ¼ 20 mm, simulated Erms,max (Fig. 5a)
Bioelectromagnetics
522 Vermeeren et al.

Erms Hrms
(V/m) (mA/m)
50 50
600 900
750

x (mm)

x (mm)
• 500
0 400 0 • 600
450
300 300
200 150
−50 −50
−100 0 100 −100 0 100
z (mm) z (mm)
(a) (b)

Erms Hrms
(V/m) (mA/m)
50 50
600 900
750
x (mm)

x (mm)
500
0 • 400 0 • 600
450
300 300
200 150
−50 −50
−100 0 100 −100 0 100
z (mm) z (mm)
(c) (d)

Fig. 5. (a) and (b) simulated and (c) and (d) measured electromagnetic near fields at y ¼
20 mm of generic walkie-talkie.

equals 762 V/m in the point (x ¼ 10 mm, z ¼ 55 mm) radiated power varied from 51 mW to 138 mW. This
and deviates only 1.6% from measured field value, is about 9.8–3.6 times below specification of allowed
that is, Erms,max of 750 V/m in point (x ¼ 5 mm, ERP for PMR 446 radios.
z ¼ 50 mm). In the same plane y ¼ 20 mm, simulated Near fields of COBRA MT500 were measured
Hrms,max is 1080 mA/m in (x ¼ 5 mm, z ¼ 15 mm) and and compared to the model of the walkie-talkie
deviates 7.7% from measured Hrms,max of 1003 mA/m (which has dimensions based on dimensions of the
in (x ¼ 5 mm, z ¼ 15 mm). COBRA MT500). We positioned the walkie-talkie in
the same way as the model (see Fig. 3) such that the
Free-Space Evaluation of Four Commercially feed point coincided with the origin of the coordinate
Available Walkie-Talkies system. We normalized fields to the maximum field
Besides the COBRA MT500, three other PMR value in the plane y ¼ 20 mm, because we were not
446 radios were acquired. In this section, the radiation able to determine input power for the real walkie-
characteristics of all the walkie-talkies (Table 1) and talkie. Figure 6 shows normalized electric and mag-
near-field behavior of the COBRA MT500 are inves- netic field in a plane for the real walkie-talkie (only
tigated. measurements). These distributions agree well with
Radiation characteristics of real devices were simulated (see Fig. 5a and c) and the measured (see
measured at SP Technical Research Institute of Fig. 5b and d) near field distributions of the generic
Sweden (Borås, Sweden). Table 3 lists TRP, maxi- walkie-talkie. Compared to the real walkie-talkie,
mum ERP, and averaged ERP. Maximum effective maximum relative error for the measured and simu-
lated model on the normalized electric field were only
1.2% and 5.2%, respectively. For normalized mag-
TABLE 3. Total Radiated Power and Effective Radiated netic near field, maximum relative error for the
Power of Four Commercial Walkie-Talkies measured and simulated model was 5.2% and 5.5%,
TRP Max. ERP Avg. ERP respectively. These low deviations show that the
Walkie-talkie (mW) (mW) (mW) model behaves electromagnetically like a real walkie-
talkie and can be used for analysis of interaction with
COBRA MT500 105 138 117
TwinTalker 59 79 68
the human body.
3300
TwinTalker 51 69 63 SAR Assessment of Real Walkie-Talkies
1300 Figure 7 shows peak local-averaged SAR in
Alecto FR-20 44 51 47
1 g and (SAR1g) and 10 g (SAR10g) of the four
Bioelectromagnetics
 SAR Compliance of Walkie-Talkies 523

Erms/Erms,max Hrms/Hrms,max
50 50
0.75 0.75

x (mm)

x (mm)
0.6 0.6
0 • 0.45 0 • 0.45
0.3 0.3
0.15 0.15
−50 −50
−100 0 100 −100 0 100
z (mm) z (mm)
(a) (b)

Fig. 6. Normalized RMS (a) electric and (b) magnetic near field at y ¼ 20 mm of a real
walkie-talkie.

commercial walkie-talkies for a duty-cycle of with the SAR-limits for 1 g (i.e., 1.6 W/kg) and 10 g
100% (i.e., continuously speaking for at least (i.e., 2 W/kg), respectively, even for a worst-case duty
6 min.) The Alecto was not measured at a distance cycle of 100%.
of 15 mm between the device and the liquid inside Cecil et al. [2014] and Dimbylow et al.
the flat phantom. Except for the touch position, the [2003] reported peak 10 g SAR values up to about
COBRA MT500 showed a higher SAR in 1 g and 50% of ICNIRP limit for general public exposure for
10 g than the walkie-talkies from Topcom (Tilburg, TETRA transmitters. This is higher than the maxi-
The Netherlands) and Alecto (’s-Hertogenbosch, mum of 0.35 W/kg observed in our study. Besides the
The Netherlands). difference in phantom (human body model vs. flat)
Highest values of the SAR1g and SAR10g occurred and distance between the walkie-talkie and the
when the walkie-talkie touched the flat phantom: phantom, the difference was mainly due to the differ-
SAR1g ¼ 0.52 W/kg (Alecto) and SAR10g ¼ 0.35 W/kg ence in applied power: Cecil et al. as well as
(Alecto). These values for the touch position comply Dimbylow et al. used a transmitted power of 0.25 W
for the hand-held TETRA devices whereas in our
study, SAR values were for typical TRP-values (in
(a) free space) of walkie-talkies ranging from 44 mW to
Alecto FR20 105 mW (Table 3).
0.5 COBRA MT500
Topcom TT1300
SAR1g (W/kg)

0.4 Topcom TT3300

SAR Assessment of Generic Walkie-Talkie
0.3 The input power of the generic walkie-talkie
was set to 10 mW during measurement taking into
0.2
account reflection losses. We validated use of the
0.1 generic walkie-talkie as a substitute for real devices
by comparing peak local-averaged SAR of real
0
touch 15 20 30 50 devices with ones of the model for the same ERP
d (mm) of the real devices and for the same magnetic near
(b) field. Above, we discussed ERP values of the four
walkie-talkie devices. We observed that ERP of
Alecto FR20
0.5 COBRA MT500 considered devices was at least 3.6 times smaller
than maximum allowed ERP for PMR 446 radios.
SAR10g (W/kg)

Topcom TT1300
0.4 Topcom TT3300 We rescaled ERP of the generic walkie-talkie
0.3 (measured value of 4.9 mW and simulated ERP
value of 4.8 mW for an antenna input power of
0.2
10 mW) to values listed in Table 3. The antenna
0.1 input power needed to reach ERP of the walkie-
0
talkies was applied during dosimetric measurement.
touch 15 20 30 50 Figure 8 shows the SAR1g and SAR10g of the
d (mm)
generic walkie-talkie with the same ERP of the
Fig. 7. Peak local-averaged SAR in (a) 1g and (b) 10 g induced
COBRA MT500. We observed that the generic
in oval flat phantom by four commercially available walkie- walkie-talkie overestimated peak local-averaged
talkies with duty cycle of 100%. SAR of the real devices (Fig. 7). Rescaling input
Bioelectromagnetics
524 Vermeeren et al.

(a) walkie-talkie to the maximum value of the magnetic

1
M2 (meas)
field in the same plane above the real device. Peak-
0.8 M2 (sim) local averaged SAR of the generic walkie-talkie for
a rescaling based on magnetic field is shown in
SAR1g (W/kg)

0.6 Figure 9. We observed that maximum deviation on

the peak local-averaged SAR was less than 26%.
0.4 Thus, magnetic near field is a better quantity than
far-field parameter ERP for estimating peak local-
0.2
averaged SAR of a real device.
0 Finally, we estimated worst-case peak local-
touch 15 20 30 50 averaged SAR for the walkie-talkies. We deter-
d (mm)
mined peak local-averaged SAR in 1 g and 10 g for
(b) an input power that gives an ERP of 500 mW in
0.8
M2 (meas) free space (we assumed a duty cycle of 100% for
M2 (sim) worst-case evaluation) for the generic walkie-talkie
0.6
SAR10g (W/kg)

and taking into account a correction factor of two

(see above). Figure 10 shows that the generic
0.4 walkie-talkie is not compliant to the FCC limit of
1.6 W/kg in a cube of 1 g at 15 mm. The worst-case
0.2 peak local-averaged SAR in 1 g of tissue exceeded
the FCC limit by up to 7% (based on simulated
0 value). If we compared peak local-averaged SAR in
touch 15 20 30 50
d (mm) 10 g of tissue with ICNIRP basic restriction of
2 W/kg, then we observed that the walkie-talkies
Fig. 8. Peak local-averaged SAR in (a) 1g and (b) 10 g are compliant.
induced in flat phantom by generic walkie-talkie for same ERP
as COBRA MT500 (100% duty cycle).
(a)
0.4
power based on ERP (a far-field characteristic) was M2 (meas)
M2 (sim)
inadequate for assessment of SAR in human tissue 0.3
SAR1g (W/kg)

despite the use of the helical antenna with coating

from the COBRA MT500. The ratio for SAR1g and 0.2
SAR10g (for ERP equal to ERP of the devices) of
the model and device was 2.5 and 2.3, respectively.
0.1
The deviation originates from uncertainties on ERP
and SAR measurements, simplifications of the
0
model with respect to the real device and from touch 15 20 30 50
differences in coupling when the device or model d (mm)
operates at close distance to the phantom. Ratio for (b)
the other real walkie-talkies was maximum 2.8 0.4
(Topcom tt3300) and 2.5 (Topcom tt3300) for M2 (meas)
M2 (sim)
SAR1g and SAR10g, respectively. For further numer- 0.3
SAR10g (W/kg)

ical analysis of the SAR10g in the flat phantom

using the generic walkie-talkie and based on a 0.2
rescaling of ERP, we will take into account a
correction factor of two for SAR10g value. 0.1
A better quantity for validating the generic
walkie-talkie as a substitute for a real device for
0
assessing peak local-averaged SAR is the magnetic touch 15 20 30 50
near field. Kuster and Balzano [1992] showed that d (mm)
the SAR is mainly proportional to incident mag-
Fig. 9. Peak local-averaged SAR in (a) 1g and (b) 10 g
netic near field. We rescaled the maximum of the induced in the oval flat phantom by generic walkie-talkie for
magnetic near field in a plane at a distance of same magnetic field at 20 mm above ground plane as COBRA
20 mm of the ground plane of the generic MT500 (100% duty cycle).
Bioelectromagnetics
 SAR Compliance of Walkie-Talkies 525

(a) assessing peak local-averaged SAR is the magnetic

2
M2 (meas) near field with deviations of less than 26%.
M2 (sim) We investigated compliance to SAR safety limits
1.5
of a walkie-talkie. An effective radiated power as
SAR1g (W/kg)

specified by the PMR 446 radio standard can cause a

1 local-averaged SAR, which exceeds limits. However,
based on effective radiated power of four real walkie-
0.5 talkies, we showed that peak local-averaged SAR
values are unlikely to exceed current SAR limits.
0
touch 15 20 30 50
d (mm)
REFERENCES
(b)
1.5 Bakker JF, Paulides MM, Christ A, Kuster N, van Rhoon GC.
M2 (meas) 2010. Assessment of induced SAR in children exposed to
M2 (sim) electromagnetic plane waves between 10MHz and 5.6
GHz. Phys Med Biol 55:3115–3130.
SAR10g (W/kg)

1 Bakker JF, Paulides MM, Christ A, Kuster N, van Rhoon GC.

2011. Corrigendum for assessment of induced SAR in
children exposed to electromagnetic plane waves between
0.5 10 MHz and 5.6 GHz. Phys Med Biol 56:2883.
Balanis CA. 1982. Antenna theory, analysis and design. Hoboken,
NJ: John Wiley& Sons. pp 58, 368.
Balzano Q, Garay O, Manning T. 1995. Electromagnetic energy
0
touch 15 20 30 50 exposure of the users of portable cellular telephones. IEEE
d (mm) Trans Vehic Technol 44:390–403.
Cecil S, Neubauer G, Rauscha F, Stix G, M€uller W, Breithuber
Fig. 10. Worst-case assessment of peak local-averaged SAR Glanzer M. 2014. Possible risks due to exposure of workers
in (a) 1 g and (b) 10 g induced by generic walkie-talkie for and patients with implants by TETRA transmitters. Bio-
ERP ¼ 500 mW (100% duty cycle). electromagnetics 35:192–200.
Dimbylow P, Khalid M, Mann S. 2003. Assessment of specific
energy absorption rate (SAR) in the head from a TETRA
CONCLUSIONS handset. Phys Med Biol 48:3911–3926.
An accurate model for a walkie-talkie has been European Radio Communications Committee (ERC). 1998. ERC
decision of 23 November 1998 on the harmonised
developed and very good agreement reported from frequency band to be designated for PMR 446 (ERC/DEC/
simulations and measurements in terms of reflection (98) 25), Brussels, Belgium.
and transmission characteristics. Near fields of the European Telecommunications Standards Institute (ETSI). 2001a.
model have been validated with measured near fields ETSI European standard, electromagnetic compatibility and
of a real walkie-talkie. A relative error of less than radio spectrum matters (ERM); land mobile service; radio
equipment using integral antennas intended primarily for
5.5% has been observed on the near electric and analogue speech; part 1: Technical characteristics and
magnetic field indicating that the walkie-talkie model methods of measurement. ETSI EN 300 296-1, Sophia-
behaves electromagnetically as a real walkie-talkie. Antipolis, France.
Therefore, the model can be used to determine European Telecommunications Standards Institute (ETSI). 2001b.
electromagnetic interaction with the human body and ETSI European standard, electromagnetic compatibility and
radio spectrum matters (ERM); land mobile service; radio
to test compliance with safety limits. Total radiated equipment using integral antennas intended primarily for
power and effective radiated power were measured analogue speech; part 2: Harmonized EN covering essential
for four real or commercially available walkie-talkies requirements under article 3.2 of the R&TTE directive.
and a generic walkie-talkie. Measured effective radi- ETSI EN 300 296-2, Sophia-Antipolis, France.
ated power of the real walkie-talkies is about 3.6–9.8 Federal Communications Commission (FCC). 1997. Evaluating
compliance with FCC guidelines for human exposure to
times lower than allowed effective radiated power for radiofrequency electromagnetic fields. Tech. Rep. OET
PMR 446 radios. bulletin 65, Washington, DC.
We found that rescaling the input power of the Federal Communications Commission (FCC). 2001. Evaluating
generic walkie-talkie based on the ERP (a far-field compliance with FCC guidelines for human exposure to
characteristic) is inadequate for assessment of SAR in radiofrequency electromagnetic fields. Tech Rep Supple-
ment C to bulletin 65, Washington, DC.
human tissue: we observed a ratio for SAR1g and Federal Communications Commission (FCC). Undated. Family
SAR10g between the model and the device of 2.5 and radio service (FRS). Available from: http://www.fcc.gov/
2.3. A better quantity for validating the generic encyclopedia/family-radio-service-frs [Last accessed 11
walkie-talkie as a substitute for a real device for Aug 2015], Washington, DC.

Bioelectromagnetics
526 Vermeeren et al.

Institute of Electrical and Electronics Engineers (IEEE). 2003. Koulouridis S, Nikita KS. 2004. Study of the coupling between
IEEE recommended practice for determining the spatial- human head and cellular phone helical antennas. IEEE
peak specific absorption rate (SAR) in the human body Trans Electromagn Compat 46:62–70.
from wireless communications devices: Experimental tech- Kuster N, Balzano Q. 1992. Energy absorption mechanism by
niques. Std. 1528-2003, New York, NY. biological bodies in the near field of dipole antennas above
International Commission on Non-Ionizing Radiation Protection 300MHz. IEEE Trans Veh Technol 41:17–23.
(ICNIRP). 1998. Guidelines for limiting exposure to time- Kuster N, Balzano Q, Lin JC. 1997a. Mobile communications
varying electric, magnetic, and electromagnetic fields. safety. London, UK: Chapman & Hall. pp 21–22.
Health Phys 74:494–522. Kuster N, Kastle R, Schmid T. 1997b. Dosimetric evaluation of
International Electrotechnical Commission (IEC). 2001. Proce- hand-held mobile communications equipment with known
dure to measure the specific absorption rate (SAR) for precision. IEICE Trans Commun E80-B:645–652.
hand-held mobile wireless devices in the frequency range Lazzi G, Gandhi O. 1998. On modeling and personal dosimetry of
of 300 MHz to 3 GHz. IEC62209, Geneva, Switzerland. cellular telephone helical antennas with the FDTD code.
International Electrotechnical Commission (IEC). 2005. Human IEEE Trans Antennas Propag 46:525–530.
exposure to radio frequency fields from hand-held and Meyer F, Palmer K, Jakobus U. 2001. Investigation into the accuracy,
body-mounted wireless communication devices—human efficiency and applicability of the method of moments as
models, instrumentation, and procedures—part 2: Proce- numerical dosimetry tool for the head and hand of a mobile
dure to determine the specific absorption rate (SAR) in the phone user. Appl Comput Electromagn Soc J 16:114–125.
head and body for 30 MHz to 6 GHz handheld and body- SPEAG. 2008. DASY4 System Handbook. Zurich, Switzerland.
mounted devices used in close proximity to the body. TC/ TCO Development. 2008. TCO'01 Certification of Mobile Phones.
SC106/90/NP, Geneva, Switzerland. Stockholm, Sweden.

Bioelectromagnetics