938

IEEE ANTENNAS AND WIRELESS PROPAGATION LETTERS, VOL. 9, 2010

Multiple Human Effects in Body Area Networks

George Koutitas, Member, IEEE

AbstractThis letter investigates the channel variations in

wireless body area networks (WBANs). For the purpose of the investigation, a ray solution for multiple closed surfaces is proposed
that utilizes a modified slope uniform theory of diffraction (UTD)
technique and can be applied to model scattering, radiation, coupling effects, or a combination of the three in a single formulation.
The solution considers all possible ray paths and is valid at the
transition boundaries of the scenario. The human bodies are
represented as closed cylindrical surfaces with constitutive parameters based on the Muscle model. The proposed algorithm
predicts channel variations in a ray format that requires less CPU
and is characterized by lower complexity compared to full-wave
techniques. Body-to-access point (BAP) and body-to-body networks are investigated.
Index TermsBody-to-access point (BAP) communications,
multiple closed surfaces, ray tracing, uniform theory of diffraction, wireless body area networks (WBANs).

I. INTRODUCTION
SER-CENTRIC and personal networks incorporate
sensors and actuators on a body that can communicate
with an access point, forming an off-body or body-to-access
point (BAP) network, or they can communicate between different bodies or parts of the same body forming a body-to-body
or an on-body network [1], [2]. Channel modeling in a wireless
body area network (WBAN) is usually obtained with extensive measurement campaigns that yield empirical estimations,
or with advanced electromagnetic codes that consider field
strength calculations, angle of arrival, and delay spreads.
Empirical channel models for on-body multiple-inputmultiple-output (MIMO) and ultrawideband (UWB) systems are
presented in [1][3]. It is concluded that human body shadowing is a prominent factor for short-range body area networks.
Deterministic techniques can provide accurate results and
avoid extensive measurement campaigns. The most suitable
deterministic electromagnetic algorithms are the ray formulation of UTD and the finite-difference time domain (FDTD).
The human body is modeled as a circular cylinder for the
cylindrical UTD ray formulation, or it can be modeled with advanced geometrical configurations to represent human anatomy
(FDTD solution). Based on FDTD analysis, the authors in
[4] and [5] derive path-loss formulations. In [6], a cylindrical
UTD ray theory that utilizes the Fock coupling functions is
compared to a FDTD solution and measurements for on-body
channel characterization. For the purpose of the investigation,
the authors apply a mechanistic multiplication of the UTD

Manuscript received June 12, 2010; revised August 23, 2010; accepted
September 25, 2010. Date of publication September 30, 2010; date of current
version October 11, 2010.
The author is with the School of Science and Technology, International Hellenic University, 57001 Thermi, Greece (e-mail: [email protected]).
Digital Object Identifier 10.1109/LAWP.2010.2082485

coupling coefficients of perfectly electrical conducting (PEC)

surfaces [7] to provide field predictions on a human body that is
modeled by three cylindrical surfaces, each one representing the
torso and the arms, respectively. The field predictions assumed
only the creeping waves around the human body, neglecting
cross-propagating rays over the cylinders, and this bounds
the applicability of the solution around the transition zones
of the scenario [8]. For the case of body-to-access point and
body-to-body networks and in the presence of multiple humans,
diffraction mechanisms occur in the transition boundaries of
the scenario, and slope terms are of major importance.
This letter presents a UTD ray solution for a cascade of
closed cylindrical surfaces that can be applied to multiple
scattering, radiation, coupling, and a combination of those
scenarios in a single formulation. In order to achieve this goal,
the UTD scattering coefficients of [7] are modified to model
radiation and coupling mechanisms (surface-mounted antennas
over the surface), according to [9]. The
with height
modified scattering coefficients are then incorporated within
the multiple slope cylindrical UTD formulation of [8]. Finally,
the algorithm is extended to incorporate cross-propagating
rays around the cylinders to provide field continuity in all the
transition boundaries of the scenario. The proposed solution is
applied to WBAN channel predictions, and simulation results
are validated with the method of moments (MoM) and measurements in the anechoic chamber of the Aristotle University
of Thessaloniki, Greece.
II. RADIATION FROM HUMAN BODY
The presence of a human body in a propagation channel can
be modeled as a circular cylinder with a typical radius of curcm, and UTD is proven to be accurate
vature
compared to real measurements [10]. At high frequencies, above
500 MHz, penetration of radio waves through the human tissue
is of minor importance, and creeping waves are the dominant
field components [5], [10]. For a body-to-access point WBAN
scenario, the transmitter or the receiver is surface-mounted (or
placed very close 15 cm) on the human body, resulting in
a radiation problem. In this section, the UTD scattering coefficient of [7] is modified according to [9] in order to approximate
surface-mounted antennas on a body. The solution is validated
with the UTD radiation case, the MoM, and measurements in
an anechoic chamber.
The problem under investigation is presented in Fig. 1. A reaway from the center
ceiver (RX) is placed at a distance
of the body, where
. For the scattering case, the field can
be modeled according to [7]

1536-1225/$26.00 2010 IEEE

(1)

KOUTITAS: MULTIPLE HUMAN EFFECTS IN BODY AREA NETWORKS

= 10
=02

939

Fig. 1. Proposed modified UTD scattering solution (h

= ) compared to
the UTD radiation case, the MoM for cylinder of radius R
: m, "
: ,

: S/m (based on the Muscle model) and measurements in an anechoic
chamber with a human body.

=47

= 48 2

where is an ONOFF parameter indicating if there is line of

sight (LOS) or non-line of sight (NLOS) between the transmitter
is the LOS field;
and
and the receiver, respectively.
represent the incident field at the reflection and diffraction
points, respectively; parameter represent the spreading factor
(that depends on distance ); and operators , represent the
left- and right-side field reaching the receiver around the closed
and
are the diffraction and reflection coeffisurface.
cients for the soft and hard case, respectively

(2)

(3)
where
, and
is the transition function. For
the diffraction case, is the external angle of the diffracting and
,
,
incident ray,
, is the distance parameter that depends on the distances ( ) of incident and diffracted or reflected rays, and is
the radius of curvature of the cylindrical surface. For the reflec,
tion case, is the angle of reflection,
and
. The terms
,
in (2) and
(3) are the so-called Fock scattering functions. They depend on
the polarization and the electrical characteristics of the curved
obstacle. For any surface impedance material, they are related
to the surface field function [11], [12]
(4)
where
and
represent the Fock-type Airy functions,
and depends on the impedance of the surface and the local
radius of curvature. For the radiation problem of an impedance

boundary cylinder, the Fock radiation functions are presented

in [13].
In [9], a heuristic modification of the diffraction and reflection
coefficients is presented, capable to approximate radiation or
coupling phenomena, caused by surface mounted antennas, with
the scattering coefficients of (2) and (3), providing acceptable
accuracy for engineering applications. For the case of antenna
(Fig. 1) and radius of curvature
, the
heights
scattering coefficients are accurate enough. The inaccuracies are
or
and for
observed for antenna heights
a receiver placed into the deep shadow region. This is because
the UTD scattering formulas do not reduce to Kellers form in
in (2) does
the deep shadow since the transition function
not approach unity fast enough. Based on this observation, the
when
for the hard case, and
term
when
for the soft case, representing field points that are
sufficiently into the deep shadow region. By employing these
modifications, the errors never exceeded 1.5 dB [9] for the case
and
. These conditions satof
isfy accurately enough surface-mounted antennas on a human
body in the frequency range of 900 MHz and above.
The modified UTD solution is compared to the case of the
, presented in [7]. For the radiUTD radiation problem
ation case, the UTD coefficients incorporate the Fock radiation
functions. The comparisons for a PEC surface of radius of curm, frequency 2.4 GHz, and the hard case is
vature
presented in Fig. 1. It can be observed that there is an excellent
agreement. Similar observations were observed for the soft case
as well. The case of an impedance boundary condition cylinder
with constitutive parameters based on the Muscle model [14]
and for a frequency of 5 GHz is compared to the MoM solution that employed a line source with rectangular pulses as basis
functions and a point-matching technique with points spaced at
. In addition, measurements of a surface-mounted antenna
on a real human body in the anechoic chamber show an acceptable agreement. The values were obtained as the mean of three
measurements of the field strength, from 0 to 180 with a step
of 10 . A signal generator and a spectrum analyzer were utilized. The antenna used is a commercial UWB antenna with
an efficiency of 85% at 5 GHz, a return loss of 15 dB, and
swept gain of 4 dBi. In order to provide minimum return loss
9 dB), the antenna was placed on an
from the antenna (
anechoic foam material of size 10 10 1 cm to model the
ground plane. A similar setup is also presented in [15]. One can
use an inverse Fourier transformation to model a line source excitation to a point source, similar to [15]. For comparison reasons, the known antenna pattern was extracted from the measured values. The agreement of the results validates the accuracy of the proposed single UTD solution to model humans as
cylinders that incorporate surface-mounted antennas.
III. RAY SOLUTION OF MULTIPLE EFFECTS
This section presents a ray solution based on UTD that incorporates a single formulation to model the radiation, scattering, coupling, and combination of the three phenomena in a
multiple-cylinder configuration. The superiority of the ray solution compared to analytical models [16] is that it provides
channel predictions, like the field strength, angles of arrival,

940

IEEE ANTENNAS AND WIRELESS PROPAGATION LETTERS, VOL. 9, 2010

Fig. 2. Ray representation of the field and shadow boundaries for the case of a
body-to-access point WBAN with two humans modeled as closed cylinders.
Fig. 3. Received field for the configuration shown in Fig. 2.

and delay spreads with less CPU and complexity. Furthermore,

it can be easily converted in the time domain, to model UWB
with Laplace transformations. The scenario under investigation
is presented in Fig. 2.
Two closed cylindrical surfaces represent the human bodies.
One antenna is mounted on the surface of body #2, and body #1
constitutes the obstruction to the transmitted by the access
point (TX) signals. For the multiple-cylinder scenario, the field
cannot be modeled only as the sum of the fields approaching the
receiver from the left and right side of the body, but cross-propagating rays are of major importance to achieve field continuity
at the shadow boundaries. Compared to the case of a single
body (Fig. 1), where the scenario incorporates two shadow
and
), the multiple-cylinders case
boundaries (
shadow boundaries, as shown in Fig. 2. The
incorporates
indices 0, 1, 2, 3 represent the TX, body #1, body #2, and RX,
respectively.
Based on the multiple-cylinder UTD solution, from open
for the case of
structures [8], the received field at position
multiple closed cylinders incorporating slope diffraction terms
can be modeled as

(5)
In (5), the first two terms characterize the received field from the
left and right side of the scenario if the cylinders were assumed
as open structures. The last term represents the cross-propagating rays due to the
cross-shadow boundaries. In this
case, the operator represents the cross-propagating rays and
can be either or , in any order, according to the nature of the

propagating ray. The amplitude diffraction or reflection coefficients, defined as a common parameter , in (5) are the coefficients presented in (2) and (3), taking into account the modifications for the transition function. The slope diffraction terms
are calculated in [8]. The distance parameters that are included
in the diffraction and reflection terms are computed according
to the continuity equations of [8] and are given by
(6)

(7)

IV. RESULTS
Fig. 3 presents the importance of cross-propagating rays to
uniform field predictions. The scenario under investigation is
m) is placed at a
the same as in Fig. 2. Body #1 (
distance
m and
m. Body #1 is moving on a
direction normal to the line between TX and body #2 (
m) with
m
m relative to axis
.
) at
The receiver is surface-mounted on body #2 (
. The difference between the case where all possible rays
are considered and the case where the cross-propagating rays are
not taken into account is presented. It is obvious that without the
cross-propagating field terms, the results are not uniform when
and
coincide with
the transition boundaries
and
, respectively.
Scenario 1 of Fig. 4 describes the case of an access point
m from body #1, which is placed
placed at a distance
at a distance
m from body #2. Body #2 has a sur), and for the measurements, it
face-mounted antenna (
is rotated with a step angle of 10 around its axis in the anechoic
chamber. The proposed solution is compared to measurements
at the frequency of 5 GHz for the hard case. There is an acceptable accuracy between theory and measurements. The observed
are due to the
ripples of the measured field at
multipaths created by the arms of body #1 that are in LOS with
), the
the RX. At the shadow region of body #2 (

KOUTITAS: MULTIPLE HUMAN EFFECTS IN BODY AREA NETWORKS

941

area-network channel estimations and was proven to be uniform

and accurate compared to MoM and measurements in an anechoic chamber. The cross-propagating rays around the bodies
of the scenario were proven to be of significant importance in
obtaining uniform results.
ACKNOWLEDGMENT
The author would like to thank Prof. D. Chrissoulidis,
Assoc. Prof. T. Youltsis, Dr. A. Goulianos, and Ch. Liontas
for their important contributions to the measurements at the
anechoic chamber of Aristotle University of Thessaloniki,
Greece.
Fig. 4. Case of a body-to-access point and body-to-body WBAN with two
: ,
: S/m.
human bodies modeled as cylinders "

= 48 2 = 4 7

Fig. 5. Relative power delay profile for the scenario of Fig. 2 with body #1
center moving from 1.2 to 0 m with equal steps, creating seven position IDs.

measured field is more stable as the multipaths from the arms of

body #1 are attenuated due to surface diffraction over body #2.
Scenario 2 of Fig. 3 represents the case of a body-to-body communication where the transmitting antenna is surface-mounted
) on body #1 at
. The observation is that
(
both scenarios have uniform field predictions. This is because
all possible ray paths, including slope field terms and continuity
equations, are considered in the simulations.
Fig. 5 presents the simulated relative delay spread of the multipaths for the scenario of Fig. 2. It can be observed that six dominant rays describe the field variations, and these are represented
depending on their orientation. At
, where the
as
two bodies are aligned, the relative delays are minimized due to
the similarity of the distance traveled by the multipaths.
V. CONCLUSION
A uniform ray solution for the case of multiple closed cylindrical bodies was presented. The proposed algorithm utilizes
slope UTD approximation and describes field variations due to
radiation, coupling, scattering, and a combination of the three,
with a single formulation. The algorithm was applied to body-

REFERENCES
[1] Y. Wang, I. B. Bonev, J. Nielsen, I. Z. Kovacs, and G. Pedersen, Characterization of the indoor multiantenna body-to-body radio channel,
IEEE Trans. Antennas Propag., vol. 57, no. 4, pp. 972979, Apr. 2009.
[2] A. A. Goulianos, T. C. Brown, B. G. Evans, and S. Stavrou, Wideband
power modeling and time dispersion analysis for UWB indoor off-body
communications, IEEE Trans. Antennas Propag., vol. 57, no. 7, pp.
21622171, Jul. 2009.
[3] A. Fort, J. Ryckaert, C. Desset, P. De Doncker, P. Wambacq, and L. Van
Biesen, Ultra-wideband channel model for communication around the
human body, IEEE J. Sel. Areas Commun., vol. 24, no. 4, pt. 1, pp.
927933, Apr. 2006.
[4] J. Ryckaert, P. De Doncker, R. Meys, A. De le Hoye, and S. Donnay,
Channel model for wireless communication around human body,
Electron. Lett., vol. 40, no. 9, pp. 543544, Apr. 2004.
[5] E. Reusens, W. Joseph, B. Latre, B. Braem, G. Vermeeren, E. Tanghe,
L. Martens, I. Moerman, and C. Blondia, Characterization of on-body
communication channel and energy efficient topology design for wireless body area networks, IEEE Trans. Inf. Technol. Biomed., vol. 13,
no. 6, pp. 933945, Nov. 2009.
[6] Y. Zhao, Y. Hao, A. Alomainy, and C. Parini, UWB on-body radio
channel modeling using ray theory and subband FDTD method, IEEE
Trans. Microw. Theory Tech., vol. 54, no. 4, pp. 18271825, Apr. 2006.
[7] D. A. Mcnamara, C. W. Pisturius, and J. A. Malherbe, Introduction to
the Uniform Geometrical Theory of Diffraction. Boston, MA: Artech
House, 1990.
[8] G. Koutitas and C. Tzaras, A UTD solution for multiple rounded surfaces, IEEE Trans. Antennas Propag., vol. 54, no. 4, pp. 12771283,
Apr. 2006.
[9] R. Paknys, On the accuracy of the UTD for the scattering by a
cylinder, IEEE Trans. Antennas Propag., vol. 42, no. 5, pp. 757760,
May 1994.
[10] M. Ghaddar, L. Talbi, T. A. Denidni, and A. Sebak, A conducting
cylinder for modelling human body presence in indoor propagation channel, IEEE Trans. Antennas Propag., vol. 55, no. 11, pp.
30993103, Nov. 2007.
[11] M. Abramowitz and I. A. Stegun, Handbook of Mathematical Functions. New York: Dover, 1970.
[12] L. W. Pearson, A scheme for automatic computation of Fock-type
integrals, IEEE Trans. Antennas Propag., vol. AP-35, no. 10, pp.
11111118, Oct. 1987.
[13] C. Tokgoz and R. J. Marhefka, A UTD based asymptotic solution for
the surface magnetic field on a source excited circular cylinder with an
impedance boundary condition, IEEE Trans. Antennas Propag., vol.
54, no. 6, pp. 17571757, Jun. 2006.
[14] S. Gabriel, R. W. Lau, and C. Gabriel, The dielectric properties of
biological tissues: III. Parametric models for the dielectric spectrum of
tissues, Phys. Med. Biol., vol. 41, pp. 227193, 1996.
[15] A. Fort, F. Keshmiri, G. R. Crusats, C. Craeye, and C. Oestges, A
body area propagation model derived from fundamental principles:
Analytical analysis and comparison with measurements, IEEE Trans.
Antennas Propag., vol. 58, no. 2, pp. 503514, Feb. 2010.
[16] S.-C. Lee, Dependent scattering of an oblique incident plane wave by
a collection of parallel cylinders, J. Appl. Phys., vol. 68, no. 10, pp.
49524957, Nov. 1990.