This article has been accepted for publication in IEEE Transactions on Antennas and Propagation.

This is the author's version which has not been fully edited and
content may change prior to final publication. Citation information: DOI 10.1109/TAP.2022.3209163

Communication
SAR Reduction with Antenna Cluster Technique
Harri Varheenmaa, Pasi Ylä-Oijala, Anu Lehtovuori, and Ville Viikari, Senior Member, IEEE

Abstract—An efficient and straightforward antenna design method various design cases to optimize performance of an antenna structure
is presented that maximizes the ratio of total efficiency and specific both in near and far-field to achieve the desired behavior both on
absorption rate (SAR). This goal is achieved with a multi-port antenna
SAR and efficiency point of view. This is attained by utilizing an
cluster technique where several ports are excited collectively with
appropriate feeding weights. These weights are found as an eigenvalue antenna cluster technique with collectively excited multiple feeding
solution formulated in terms of the radiated and near field power of an ports. The problem of finding optimal feeding weights is formulated
antenna. The method relies on both the near-field (SAR) and far-field as a Rayleigh quotient and the optimal weights are found as a solution
(radiation) physics of an antenna and it can be applied in various design of a generalized eigenvalue equation. Eigenvalue formulations and
cases. The importance of the maximization of total-efficiency-SAR ratio,
and the feasibility of the proposed approach, are demonstrated with a Rayleigh quotients [20] have been previously used to solve optimal
simple multiple-dipole antenna example and with a more realistic antenna multiport feeding signals with various goals, such as maximization
design where a metal-rimmed antenna is held in user’s hand. of impedance matching [21] or radiation efficiency [22]. To our
Index Terms—Antenna cluster, mobile antenna, Rayleigh quotient, knowledge this is the first time when these approaches are utilized
specific absorption rate. in SAR reduction.
To measure the performance of an antenna from both radiation
and SAR point of view, we define a figure-of-merit (FoM) – the
I. I NTRODUCTION
ratio of total efficiency and maximum SAR. Since this FoM cannot
The number of antennas used in close proximity of the user is be formulated as a Rayleigh quotient, we need to define another
continuously increasing emphasizing the importance of exposure- quantity that models it as well as possible. To this end, we consider
aware antenna design. Safety limits are determined with a parameter the ratio of radiated far-field power and surface near-field power.
called the specific absorption rate (SAR). This parameter measures Since this ratio can be formulated as a Rayleigh quotient, it leads
the absorbed electromagnetic power per unit mass of tissue. Both the to a computationally extremely efficient method requiring solution of
American [1] and the European standards [2] have defined limits for an eigenvalue equation which dimension agrees with the number of
maximum SAR values. These values have to be taken into account in the ports. Another benefit of the method is that it relies on both the
the design of mobile devices [3]–[6] as well as in different wearable near-field and far-field physics of an antenna, and thus is available
antenna solutions [7]–[9]. in various design scenarios.
Several detailed studies have been performed in the past to predict Numerical experiments with a simple multiple-dipole antenna are
SAR of an antenna see, e.g., [10], [11]. In these studies, the focus used to illustrate the importance of the choice of the optimization
is more on the estimation of the SAR than reducing it. In addition, goal. We also show that solely optimizing standard antenna parame-
SAR computation is typically performed as a post-processing step ters such as total or radiation efficiency, or matching, may not produce
once the antenna has been designed [12], rather than taking SAR optimal solution in the SAR point of view. The feasibility of the
minimization as a design goal. proposed approach using the ratio of far-field and near-field powers
The easiest way to achieve the SAR safety limits is to lower the as an optimization goal is demonstrated with a more realistic antenna
transmit power, but reduction in transmit power leads to weaker example where a metal-rimmed mobile antenna is hold in user’s hand.
transmitted fields and reduced coverage. Another straightforward
way to reduce SAR is to locate antennas as far as possible from
II. SAR O PTIMIZATION
the user, e.g., to the bottom of the device. This approach is not
suitable for modern multi-antenna mobile devices since all antennas In the design and analysis of modern complicated antenna systems
can not locate at the bottom of the device. Other widely used it is seldom sufficient to study or optimize only a single design
methods are insulating the antenna with wave-absorbing material, or parameter, rather several factors have to be treated simultaneously.
shielding it with conductive material [13]–[15]. Downside of these This is also the case with SAR optimization. The particular challenge
methods is that they typically affect negatively the gain, efficiency, with the minimization of SAR is that it easily leads to minimizing
and bandwidth of the antenna [4]. Metamaterial structures [16]–[18] radiation and matching efficiency, too.
and electromagnetic band gaps [19] have also been studied in SAR
reduction. In practice, however, implementation of these structures A. Definition of SAR and FoM
into modern thin and small ground clearance devices is difficult due
to their large size [3]. Let us consider a situation where a perfectly conducting lossless
The previously presented methods for reducing SAR typically antenna is in close proximity to human tissue. The human tissue is
require additional materials or sophisticated design of antenna struc- modelled by a homogeneous, dispersive, and lossy dielectric object
tures. In addition, these methods may have a negative effect on the with complex permittivity.
radiation properties of an antenna. In this paper, we introduce a novel SAR is defined as the power absorbed by the unit mass of tissue
and straightforward computational approach which can be applied to [23], [24]. The dielectric body is divided into small cubes Di with
mass m, and at a fixed time-harmonic frequency, SAR is associated
The authors are with the Department of Electronics and Nanoengineering, to the cube center point ci
Aalto University, Espoo, P.O. Box 15500, 00076 Aalto, Finland, (e-mail:
σ(r)∥E(r)∥2
Z
@aalto.fi). 1
Manuscript received XXXX XX, 2022; revised YYYY YY, 2022. SARi (ci ) = dV. (1)
2Vi Di ρ(r)

This work is licensed under a Creative Commons Attribution 4.0 License. For more information, see https://creativecommons.org/licenses/by/4.0/
 This article has been accepted for publication in IEEE Transactions on Antennas and Propagation. This is the author's version which has not been fully edited and
content may change prior to final publication. Citation information: DOI 10.1109/TAP.2022.3209163

Here Vi is the volume of Di , σ and ρ are the conductivity and density Here S∞ is a spherical surface in the far-field region, Ep is the
of the tissue, and E is the electric field. electric field due to unit input current at port p, and (·)∗ denotes
As mentioned above, it may not be sufficient to study solely SAR. complex conjugate. Using F far , we have
Rather, for a realistic antenna, e.g., handset or wearable, both the 1 H far
total efficiency and SAR must be taken into account during the Prad = I F I, (6)
2
design process. We define a FoM that describes the ratio of the total
efficiency, ηtot , and the maximum SAR where (·)H denotes Hermitian transpose. Analogously, we define a
near-field matrix with elements
ηtot Prad
FoM = ξ , ηtot = . (2)
Z
1
max(SARi ) Pin near
Fpq = Ep∗ (r) · Eq (r) dS, p, q = 1, . . . , P, (7)
η0 S
Here Prad and Pin are the radiated and input power of an antenna,
and ξ = Pin /m is a normalization factor. Mass m is either 1 g or and express the near-field surface power as
10 g depending on the used SAR standard. 1 H near
The benefit of the low SAR is that it enables higher input power Pnear = I F I. (8)
2
without exceeding SAR safety limits. Naturally, higher input power
Obviously both matrices F far and F near are positive definite, and
means higher transferred power. Since the FoM takes account both
the ratio of the far-field radiated power, and the near-field surface
the SAR and the total efficiency, the antenna with high FoM can
power, can be formulated as an RQ. Consequently, the generalized
transfer more power while SAR value is still below the safety limits.
eigenvalue equation to be solved reads

B. Rayleigh Quotient and Near-Field Power F far In = λn F near In (9)

From the numerical optimization point of view, the definition of and the optimal solution (port input currents) is the eigenvector In
maximum SAR poses two challenges. Firstly, SAR depends on the corresponding to the largest eigenvalue λn .
position and the optimization process can be sensitive to the choice
of the cube center ci . Secondly, the definition of a cube where SAR
is evaluated is rather complicated as the cube is near the surface of D. Numerical Implementation
the body [25]. The far-field and near-field matrices, needed in the eigenvalue
Our goal is to formulate the optimization problem as a Rayleigh equation (9), are implemented using an in-house MoM code for
quotient (RQ). The optimal solution of the problem can then be found combined PEC and lossy dielectric bodies [27], [28] with multiple
by solving an eigenvalue problem feeding ports. For given inputs, the program finds the equivalent
electric surface current on the antenna and the equivalent electric
A x n = λn B x n (3)
and magnetic surface currents on the surface of the dielectric block.
and time-consuming numerical optimization can be avoided. The Matrices F near and F far are constructed by using unit inputs at the
wanted optimal solution is the eigenvector xn corresponding to the ports, one by one, and by computing the corresponding electric field
maximum (or minimum) eigenvalue λn . Obviously FoM defined in in the near and far field regions.
(2) cannot be expressed as an RQ. Therefore, we need to find another While the assembly of the far-field matrix requires numerical
quantity that describes FoM as well as possible and can be expressed integration, the near-field matrix F near can be obtained directly from
as an RQ. the MoM solution. Let J = n × H and M = −n × E denote the
In [23] it is demonstrated that the maximum electric field appears equivalent electric and magnetic currents on the surface S of the
on the surface of the body block. This means that in minimizing dielectric body. The surface electric field at point r ∈ S is given by
maximum SAR it is sufficient to study the maximum electric field n(r)
on the body surface and to minimize it. However, using a field E(r) = n(r) × M (r) − ∇s · J (r). (10)
jωε
value defined at a single point is usually numerically unstable and
sensitive even to very small numerical errors. To avoid this problem, Here ∇s · J is the surface divergence of J and n is the exterior
we consider the near-field surface power defined here as unit normal vector of S. In the numerical solution, integral (7) is
Z evaluated at the center points of the triangles of the surface mesh
1
Pnear = ∥E(r)∥2 dS, (4) and the integration weights are the areas of the triangles. Thus, the
2η0 S elements of the near-field matrix can be obtained as
where the electric field is integrated over a surface S and η0 = N
p 1 X
µ0 /ε0 is the free space wave impedance. near
Fpq ≈ An Ep∗ (rn ) · Eq (rn ) (11)
The optimization goal is then formulated as the ratio of the radiated η0 n=1
(far-field) power and the surface (near-field) power. In the next where An and rn are the area and the center point of a triangle Tn
section, it is shown that this ratio can be expressed as an RQ in on S. The electric field is evaluated using (10).
terms of port input currents.

E. Other RQ-based Optimization Goals

C. Formulation with Port Input Currents
Above we showed how the ratio of far-field power and near-
Let us next assume that the antenna structure includes P feeding
field power can be expressed as an RQ. To compare this approach
ports and let I = [I1 , . . . , IP ]T denote the complex valued input
with other design goals, next we shortly review formulation of other
currents of the ports. These port input currents are the unknowns
antenna design parameters as RQs. Let S be the S-parameter matrix
to be optimized. First, the radiated power is expressed utilizing the
of a multiport antenna. Total Active Reflection Coefficient (TARC),
far-field matrix F far [26] with elements
Z defined as r
1 aH (S H S)a
far
Fpq = Ep∗ (r) · Eq (r) dS, p, q = 1, . . . , P. (5) TARC = , (12)
η0 S∞ aH a

This work is licensed under a Creative Commons Attribution 4.0 License. For more information, see https://creativecommons.org/licenses/by/4.0/
 This article has been accepted for publication in IEEE Transactions on Antennas and Propagation. This is the author's version which has not been fully edited and
content may change prior to final publication. Citation information: DOI 10.1109/TAP.2022.3209163

can be formulated as an eigenvalue problem [20]

1
λn = min eig(S H S), (13) 46 2
3
4 8 30
11
11 50
and the ”TARC-optimal” solution is the eigenvector corresponding to
the eigenvalue λn . Also radiation efficiency
72
Prad I H F far I 40
ηrad = = (14)
Prad + Pdiss I H RI
and total efficiency Fig. 1. Illustration of a three dipole cluster on top of a fat block. Dimensions
are in mm. The distance between the dipoles and the block is 5 mm. Discrete
Prad I H F far I ports, at the centers of the dipoles, are indicated with red arrows.
ηtot = = (15)
Pin aH a
can be formulated as RQs, since all involved matrices are Hermitian
and positive definite. Here a = (Z0 )−1/2 Z +Z0 I, with a diagonal

matrix Z0 including the characteristic impedances of the ports, and

Pdiss is the dissipated power due to dielectric losses. The dissipated
power can be expressed with the real part of the port input impedance
matrix, Z = R + jX, and the far-field matrix as
1 H
Pdiss = I (R − F far )I. (16)
2 (a) (b)
If the antenna is lossy, radiation efficiency should include contribution Fig. 2. (a) TARC, (b) FoM of the three dipoles with different feeding weights.
of metallic losses, too. In most practical situations dielectric losses
due to a human tissue dominate over metallic losses, and assumption
of a lossless antenna is reasonable. The FoM, defined as in (2), with 1 g SAR and with different
excitation weights is illustrated in Fig. 2(b). From this figure, it can
III. E XAMPLES be noticed that both ηrad and Prad /Pnear optimized weights performed
In this section, we demonstrate the feasibility of the proposed clearly the best and the FoM with TARC optimized weights is the
method in maximizing FoM and investigate solutions with different lowest in the entire frequency band. FoM with ηrad and Prad /Pnear
input weight optimization goals. The first example considers a cluster feeding weights is in average about 50 % higher than that of ηtot or
of three dipole antennas on top of a dielectric (fat) block. The second Equal feeding weights. Only, at a single frequency (2.85 GHz) ηtot
one involves a more practical metal-rimmed mobile antenna device weights achieves equally good FoM as ηrad and Prad /Pnear weights.
in a human hand. The optimal feeding weights for the multi-port The improvement in FoM can be analyzed by studying separately
antenna are solved with in-house codes, and the corresponding SAR, total efficiency and SAR. As Fig. 3(a) shows, the differences in total
S-parameter, and efficiency results are computed with CST Studio efficiency obtained with different feeding weights are rather small,
Suite. expect in the TARC optimal weights which lead to significantly lower
efficiency than others. Fig. 3(b) represents the 1-g maximum SAR
when input power is 22.4 dBm (175 mW). Defining the input weights
A. Three dipoles so that TARC is optimized produces clearly the highest SAR, almost
The goal of this section is to compare different input weight three times as high as with ηrad and Prad /Pnear weights. Also Equal
optimization goals to show that traditional approaches may lead to and ηtot weights produce about 30–80 % higher SAR than ηrad and
prohibitively high SAR values. We consider a cluster of three dipoles Prad /Pnear weights.
of lengths (46 mm, 48 mm, and 50 mm) placed above a fat block of SAR with both TARC and ηtot input weights varies rather much as
dimensions 40 mm × 72 mm × 30 mm. Due to the dissenting lengths a function of frequency while the other optimization methods achieve
of the dipoles, the resonance frequencies of the dipoles are different. less frequency-variant SAR. Both the high SAR, and its frequency-
This enables effective utilization of antenna cluster technique and to dependent behaviour, can be explained with SAR patterns shown in
achieve wider bandwidth. The simulation setup is shown in Fig. 1. Fig. 4. A high SAR value is obtained when the SAR maximum is
The distance between the dipoles is 11 mm, and the distance between concentrated on small area as shown in Fig. 4(b) with TARC weights.
the dipoles and the fat block is 5 mm. The size of the dielectric block Equal feeding achieves lower SAR than TARC-optimized weights
is chosen large enough to model an antenna close to a human body since its SAR has distributed on wider area near to the center dipole.
so that edges do not have a significant effect on the results. In this
example, the parameters of the fat block (εr = 10.6, δ = 0.389 S/m,
and ρ = 911 kg/m3 ) are assumed frequency-independent.
The feeding weights of the dipoles are found by using four different
optimization goals: TARC, ηrad , ηtot , and Prad /Pnear . The optimal
solutions are based on the RQ formulations introduced in the previous
section. In addition, we show results also for a non-optimized case,
i.e., when all three dipoles have equal feeding weights (Equal in
legends).
Fig. 2(a) shows TARC with different feeding weights. Naturally,
the TARC optimized weights give the best TARC. Differences (a) (b)
between other optimization goals are rather small but ηtot and Equal Fig. 3. (a) Total efficiency and (b) maximum 1 g SAR of the three dipoles
weights lead to slightly better matching than Prad /Pnear or ηtot . with different feeding weights. The total input power is 22.4 dBm (175 mW).

This work is licensed under a Creative Commons Attribution 4.0 License. For more information, see https://creativecommons.org/licenses/by/4.0/
 This article has been accepted for publication in IEEE Transactions on Antennas and Propagation. This is the author's version which has not been fully edited and
content may change prior to final publication. Citation information: DOI 10.1109/TAP.2022.3209163

Accordingly, the SAR of both ηrad and Prad /Pnear weights obtain the
lowest values since their patterns do not exhibit hot spots, like TARC,
but spread more constantly as shown in Fig. 4(c). SAR pattern with
ηtot weights behaves similarly as the one of TARC weight expect
around 2.85 GHz, where it obtains a local minimum. Around that
local minimum, the SAR distribution is similar as with ηrad and
Prad /Pnear weights.

2
4
8

6
W/kg

3.
6.
4.

1.
0
8

1 1 1
2 2 2
3 3 3

(a) (b) (c)

(a) (b)
Fig. 4. 1 g SAR patterns of the three dipoles at 2.8 GHz with Pin =
22.4 dBm. (a) Equal, (b) TARC, and (c) Prad /Pnear input weights. The
feeding ports, 1, 2, and 3 are marked with red arrows.

Fig. 5 shows the amplitudes and phases of the TARC, Prad /Pnear ,
ηrad , and ηtot optimized input weights. The input weights obtained
with Prad /Pnear and ηrad optimization goals are similar especially
in the middle of considered frequency band. Therefore, their FoM
and other results are similar. Clearly, Prad /Pnear weights have only
moderate frequency dependence while TARC and ηtot weights are
strongly frequency dependent. Since the radiation properties of an
antenna cluster are not unique, different feeding weights can give the
same far-field result. This also explains why at 2.85 GHz the results
with ηtot , ηrad , and Prad /Pnear input weights are almost identical.

B. Metal-rimmed Mobile Antenna Design (c) (d)

As a more realistic antenna design we consider a metal-rimmed Fig. 5. Magnitude (top) and phase (bottom) of (a) Prad /Pnear , (b) TARC,
handset antenna held in a human hand as shown in Fig. 6. The (c) ηrad , and (d) ηtot optimized input weights for the three dipoles.
overall dimensions of the device are 71 mm×150 mm×5 mm and
the ground clearance is 3 mm all-around the device. The minimum
distance between the rim and the hand is 1 mm. We compare two
antenna designs. In the first one, a 30-mm long element is placed on
2
the rim (port 2 in Fig. 6). In the second one, we have a ”three-element 5
71

stack” of equal length 30-mm elements, (ports 1, 2, and 3 in Fig. 6). 1

The elements are overlapping so that element 2 locates on the rim 30
3 4 1
(3 mm from the ground plane), elements 3 and 1 are at the distance 21 1
2
3
of 1 mm and 2 mm from the rim, respectively. The excitation for 5 150
G

2 4
N

element 2 is located at the center of the element, element 1 is excited

D

from the right-end, and element 3 from the left-end. The heights of 1
the elements 1 and 3 are 4 mm and the height of the element 2 is 3
5 mm. The equal length of the elements enables compact design and
Fig. 6. A mobile antenna with a metal rim and three feeding elements. The
asymmetrical location of the feeding ports are favorable for the cluster hand model is adopted from [29].
design.
The same feeding weight optimization methods are used as in the
case of three dipoles: TARC, ηrad , ηtot , and Prad /Pnear . The optimiza-
tion results are compared with the non-optimized case (Equal) and
with the one element case (1-Element).
Fig. 7 shows TARC and total efficiency for the metal-rimmed
antenna with different optimization methods. The 5G frequency band
n77 (3.3–4.2 GHz) has been marked with black dashed line to the
figure. Similar behavior as in the case of three dipoles can be noticed
and all optimization methods obtain adequate matching and total
efficiency in n77 band. (a) (b)
Similarly as in the 3-dipole case, a clear advantage of multi-port-
Fig. 7. (a) TARC and (b) total efficiency of the metal-rimmed mobile antenna
feeding with properly optimized weights, either Prad /Pnear or ηrad
with various feeding weights. 1-Element stands for the single element antenna
case, can be seen in the maximum SAR shown in Fig. 8. In the design. n77 band is marked with black dashed line.
considered frequency band, maximum SAR of the 1-port design is

This work is licensed under a Creative Commons Attribution 4.0 License. For more information, see https://creativecommons.org/licenses/by/4.0/
 This article has been accepted for publication in IEEE Transactions on Antennas and Propagation. This is the author's version which has not been fully edited and
content may change prior to final publication. Citation information: DOI 10.1109/TAP.2022.3209163

two to four times higher than that of the best three port case. We
may also observe that the choice of the optimization goal has a
significant effect on the results. For example, optimizing TARC of
the 3-port design can lead to even higher SAR than that in the 1-port
design. In addition, the relative differences between the results of the
optimization goals are larger in the case of 1-g SAR than in the case
of 10-g SAR. SAR for a limb is defined for 10 g tissue volume [1],
[2] but we also show 1-g SAR for comparison. Obviously, 1-g SAR
the tissue volumes are much smaller, since the maximum SAR values
are significantly higher and more focused on the surface of the body
than for 10-g SAR.

(a) (b)
Fig. 11. Magnitude (top) and phase (bottom) of (a) TARC and (b)
Prad /Pnear optimized input weights for the metal-rimmed antenna. n77 band
is marked with black dashed line.

(a) (b) 25 20 15 10 5 0
A/m
Fig. 8. (a) Maximum 1 g SAR and (b) maximum 10 g SAR of the mobile
antenna with various feeding weights. n77 band is marked with black dashed
line. The input power is 22.4 dBm (175 mW).
2 2
The SAR patterns with TARC and Prad /Pnear weights are shown
in Fig. 9. Similar behaviour as in the case of three dipoles can be
observed. The SAR pattern with TARC weights has clearly stronger 1 1
local maxima than with Prad /Pnear weights. Actually, one element,
and all other considered feeding methods, expect Prad /Pnear and
ηrad , show similar high local maxima as TARC weights. Hence, 3 3
Prad /Pnear and ηrad weights lead to clearly the lowest SAR values. (a) (b)
Fig. 12. Real part of the surface currents on the feeding elements of the metal-
rimmed mobile antenna with (a) TARC and (b) Prad /Pnear input weights at
W/kg 3.7 GHz. Locations of the discrete ports are marked with red arrows.
5
4
3 2 1 3 3 2 1
2
1 optimized input weights. These results show that on the n77 band,
0
the input weights of both methods are nearly constants, and their
(a) (b)
amplitudes are almost equal. The main difference appears in the phase
Fig. 9. 10 g SAR patterns of the metal-rimmed mobile antenna at 3.7 GHz difference of the feeding signal of port 3 compared to the other ports.
with Pin = 22.4 dBm. (a) TARC and (b) Prad /Pnear optimal inputs.
In the Prad /Pnear weights this difference is about -60◦ , while in the
Fig. 10 shows the FoM with various input weights. In computing TARC weights it is about +20◦ .
FoM, both 1-g and 10-g maximum SAR are used. The results In Fig. 12, we display the real part of the surface current on
demonstrate that by defining the input weights so that the far-near the antenna elements calculated with TARC and Prad /Pnear input
field power ratio is maximized, leads to the best efficiency-SAR ratio weights at 3.7 GHz. The currents on the two outer most elements,
on a wide frequency range. elements 1 and 2, are akin due to similar amplitude and phase,

(a) (b)
Fig. 10. FoM of the metal-rimmed mobile antenna calculated with (a) 1 g
SAR and (b) with 10 g SAR. n77 band is marked with black dashed line.

To get more insight into the function of the multi-port antenna

feeding design with different optimization goals in SAR reduction, (a) (b)
we next study both the input weights and the surface currents. Fig. 11 Fig. 13. Realized gain of the metal-rimmed mobile antenna with TARC and
shows the amplitudes and phases of the Prad /Pnear and the TARC Prad /Pnear input weights at 3.7 GHz. (a) yz and (b) xy planes.

This work is licensed under a Creative Commons Attribution 4.0 License. For more information, see https://creativecommons.org/licenses/by/4.0/
 This article has been accepted for publication in IEEE Transactions on Antennas and Propagation. This is the author's version which has not been fully edited and
content may change prior to final publication. Citation information: DOI 10.1109/TAP.2022.3209163

while significant differences on the currents on the element 3 can be [11] A. Hadjem, D. Lautru, C. Dale, M. F. Wong, V. Hanna, and J. Wiart,
observed. Since with Prad /Pnear weights the currents on the element “Study of specific absorption rate (SAR) induced in two child head
models and in adult heads using mobile phones,” IEEE Trans. Microw.
furthest from the hand (element 3) are stronger and the currents on
Theory Techn., vol. 53, no. 1, pp. 4–11, 2005.
the nearest element (element 2) are smaller than with TARC weights, [12] S. S. Zhekov, A. Tatomirescu, E. Foroozanfard, and G. F. Pedersen,
SAR is lower with Prad /Pnear input weights. “Experimental investigation on the effect of user’s hand proximity on a
Fig. 13 illustrates the realized gain in xy and yz planes with TARC compact ultrawideband MIMO antenna array,” IET Microw. Antennas &
and Prad /Pnear input weights at 3.7 GHz. Clearly, these patterns Propag., vol. 10, no. 13, pp. 1402–1410, 2016.
[13] M. I. Kitra, C. J. Panagamuwa, P. McEvoy, J. C. Vardaxoglou, and J. R.
are nearly identical and thus the antenna radiation pattern is almost James, “Low SAR ferrite handset antenna design,” IEEE Trans. Antennas
independent on the used input weights. This result further verifies that Propag., vol. 55, no. 4, pp. 1155–1164, April 2007.
with the proposed method low SAR values can be obtained without [14] K. H. Chan, K. M. Chow, L. C. Fung, and S. W. Leung,
“Effects of using conductive materials for SAR reduction
sacrificing antenna radiation.
in mobile phones,” Microwave and Optical Technology Lett.,
vol. 44, no. 2, pp. 140–144, 2005. [Online]. Available:
IV. C ONCLUSION https://onlinelibrary.wiley.com/doi/abs/10.1002/mop.20569
[15] A. Hirata, T. Adachi, and T. Shiozawa, “Folded-loop antenna with
A novel antenna design method based on the multi-port cluster a reflector for mobile handsets at 2.0 GHz,” Microwave and Optical
technique that simultaneously provides high efficiency and low SAR Technology Lett., vol. 40, no. 4, pp. 272–275, 2004. [Online]. Available:
is introduced. The optimal weights of the feeding ports are found https://onlinelibrary.wiley.com/doi/abs/10.1002/mop.11350
as a solution of a generalized eigenvalue problem expressed with [16] M. T. Islam, N. Misran, T. S. Ling, and M. R. I. Faruque, “Reduction
of specific absorption rate (SAR) in the human head with materials
the radiated and near-field power port matrices of the antenna. The
and metamaterial,” in 2009 International Conference on Electrical
feasibility of the proposed approach is demonstrated with a metal- Engineering and Informatics, vol. 02, Aug 2009, pp. 707–710.
rimmed mobile antenna held in user’s hand. More than 50% reduction [17] R. Ikeuchi and A. Hirata, “Dipole antenna above EBG substrate for
in maximum SAR values can be obtained compared to a conven- local SAR reduction,” IEEE Antennas Wirel. Propag. Lett., vol. 10, pp.
tional single-port antenna design, without significantly sacrificing 904–906, 2011.
[18] J.-N. Hwang and F.-C. Chen, “Reduction of the peak SAR in the human
the efficiency. It is also shown that by solely maximizing matching head with metamaterials,” IEEE Trans. Antennas Propag., vol. 54,
(minimizing TARC) can lead to prohibitively high SAR values. no. 12, pp. 3763–3770, Dec 2006.
[19] R. Das and H. Yoo, “Application of a compact electromagnetic bandgap
R EFERENCES array in a phone case for suppression of mobile phone radiation
exposure,” IEEE Trans. Microw. Theory Techn., vol. 66, no. 5, pp. 2363–
[1] R. E. Fields, “Evaluating compliance with FCC guidelines for human ex- 2372, 2018.
posure to radiofrequency electromagnetic fields,” OET bulletin, vol. 65, [20] C. Volmer, J. Weber, R. Stephan, K. Blau, and M. A. Hein, “An
no. 10, 1997. eigen-analysis of compact antenna arrays and its application to port
[2] “IEEE Standard for Safety Levels with Respect to Human Exposure to decoupling,” IEEE Trans. Antennas Propag., vol. 56, no. 2, pp. 360–
Electric, Magnetic, and Electromagnetic Fields, 0 Hz to 300 GHz,” IEEE 370, Feb. 2008.
Std C95.1-2019 (Revision of IEEE Std C95.1-2005/ Incorporates IEEE [21] J.-M. Hannula, J. Holopainen, and V. Viikari, “Concept for frequency-
Std C95.1-2019/Cor 1-2019), pp. 1–312, Oct 2019. reconfigurable antenna based on distributed transceivers,” IEEE Anten-
[3] H. H. Zhang, G. G. Yu, Y. Liu, Y. X. Fang, G. Shi, and S. Wang, nas Wireless Propag. Lett., vol. 16, pp. 764–767, 2017.
“Design of low-SAR mobile phone antenna: Theory and applications,” [22] R. Kormilainen, J. Hannula, T. Saarinen, A. Lehtovuori, and V. Viikari,
IEEE Trans. Antennas Propag., vol. 69, no. 2, pp. 698–707, Feb 2021. “Realizing optimal current distributions for radiation efficiency in prac-
[4] B. Lu, B. Pang, W. Hu, and W. Jiang, “Low-SAR antenna design and tical antennas,” IEEE Antennas Wirel. Propag. Lett., 2020.
implementation for mobile phone applications,” IEEE Access, vol. 9, pp. [23] Z. Zhang, Antenna Design for Mobile Devices, 2nd ed. Solaris South
96 444–96 452, 2021. Tower, Singapore: John Wiley & Sons, Incorporated, 2017.
[5] H. H. Zhang, G. G. Yu, X. Z. Liu, G. S. Cheng, Y. X. Xu, Y. Liu, and [24] B. Xu, M. Gustafsson, S. Shi, K. Zhao, Z. Ying, and S. He, “Radio fre-
G. M. Shi, “Low-SAR MIMO antenna array design using characteristic quency exposure compliance of multiple antennas for cellular equipment
modes for 5G mobile phones,” IEEE Trans. Antennas Propag., vol. 70, based on semidefinite relaxation,” IEEE Trans. Electromagn. Compat.,
no. 4, pp. 3052–3057, 2022. vol. 61, no. 2, pp. 327–336, April 2019.
[6] H. H. Zhang, X. Z. Liu, G. S. Cheng, Y. Liu, G. M. Shi, and K. Li, [25] “IEC/IEEE International Standard – Determining the peak spatial-
“Low-SAR four-antenna MIMO array for 5G mobile phones based on average specific absorption rate (SAR) in the human body from wireless
the theory of characteristic modes of composite PEC-lossy dielectric communications devices, 30 MHz to 6 GHz - Part 1: General require-
structures,” IEEE Trans. Antennas Propag., vol. 70, no. 3, pp. 1623– ments for using the finite-difference time-domain (FDTD) method for
1631, March 2022. SAR calculations,” IEC/IEEE 62704-1:2017, pp. 1–86, Oct 2017.
[7] C.-T. Liao, Z.-K. Yang, and H.-M. Chen, “Multiple integrated antennas [26] A. Stjernman, “Relationship between radiation patter correlation and
for wearable fifth-generation communication and internet of things scattering matrix of lossless and lossy antennas,” Electronics Letters,
applications,” IEEE Access, vol. 9, pp. 120 328–120 346, 2021. vol. 41, no. 12, pp. 678–680, June 2005.
[8] T. T. Le and T.-Y. Yun, “Wearable dual-band high-gain low-SAR antenna [27] P. Ylä-Oijala, M. Taskinen, and J. Sarvas, “Surface integral equation
for off-body communication,” IEEE Antennas Wirel. Propag. Lett., method for general composite metallic and dielectric structures with
vol. 20, no. 7, pp. 1175–1179, 2021. junctions,” Progress in Electromagnetic Research, vol. 52, pp. 81–108,
[9] M. El Atrash, M. A. Abdalla, and H. M. Elhennawy, “A wearable dual- 2005.
band low profile high gain low SAR antenna AMC-backed for WBAN [28] P. Ylä-Oijala and M. Taskinen, “Calculation of CFIE impedance matrix
applications,” IEEE Trans. Antennas Propag., vol. 67, no. 10, pp. 6378– elements with RWG and nxRWG functions,” IEEE Trans. Antennas
6388, 2019. Propag., vol. 51, no. 8, pp. 1837–1846, 2003.
[10] P. Bernardi, M. Cavagnaro, S. Pisa, and E. Piuzzi, “Specific absorption [29] R. Luomaniemi, P. Ylä-Oijala, A. Lehtovuori, and V. Viikari, “Designing
rate and temperature increases in the head of a cellular-phone user,” IEEE hand-immune handset antennas with adaptive excitation and character-
Trans. Microw. Theory Techn., vol. 48, no. 7, pp. 1118–1126, 2000. istic modes,” IEEE Trans. Antennas Propag., vol. 69, no. 7, pp. 3829–
3839, July 2021.

This work is licensed under a Creative Commons Attribution 4.0 License. For more information, see https://creativecommons.org/licenses/by/4.0/