OWC-5

CSNDSP 2010

537

A MA TLAB-based simulation program for

indoor visible light communication system
H.Q. Nguyenl, J.-H. Choi\ M. Kang2, Z. Ghassem lool, D. H. Kim4,
T.-G.

Kang4, and C. G. Lee'

S.-K. Lim4,

IDepatment of Electronic Engineering, Chosun University, South Korea

2School of Computer Engineering, Chosun University, South Korea

3School of Computing, Engineering and Information Sciences. Northumbria University, United Kingdom
4LED Communication Research Team, ETRI, South Korea
*

Abstract-we
visible

light

MATLAB

report a simulation program for indoor

communication

and

E-mail: clee@chosun.ac.kr

Simulink.

The

environment
program

based

on

considers

the

positions of the transmitters and the relections at each wall.

For

visible

light

communication

environment,

the

illumination light-emitting diode is used not only as a

lighting device, but also as a communication device. Using
the simulation program, the distributions of illuminance
and root-mean-square delay spread are analyzed at bottom
surface.

I.

INTRODUCTION

Due to fast-growing demand of high-speed data

rnsport nd mobiliy, wireless commnication is vey
important technology in our lives. Recently, the indoor
wireless commnication technology using light-emitting
diodes (LEDs) is a new research ield [1-3]. Using LEDs,
there re a lot of advantages in viewpoints of lighting and
wireless commnication:
The lighting equipment wih LEDs is easy to install,
nice looking and safe for humn eye.
The LED lighting can be equipped with vrious shapes
which were diicult with raditional lightings such as
incndescent bulbs and luorescent lamps.
he low cost, low electric power consmption and
long life expectncy re he considerable advantages.
he possibility of high-speed data communication is
the indispensable chracteristic for using LED
inrasucture for lighting as communication devices.
From these advantages, it is expected that the indoor
optical wireless communication using LEDs will be used
widely in the near uture.
The undamental physical characteristics and problems
of wireless communication using LEDs are studied and
researched by many labs, for example, impulse response
[3,4], optical power distribution [S,6], bit error ratio
(BER), signal-to-noise ratio (SNR), nd shadowing [8],
relection [9], feasibility of communication link [10-12],
the way to enhance date rate [13,14], modeling a cluster
of LEDs as a directional point source according to he
LED radiation patten, arangement of LEDs, and the
number of LEDs, the far-ield condition and the precise
optical model of LED radiation have been reported in
[IS].

978-1-86135-369-6/101$25.00 2010 IEEE

n his paper, we report a simulation progrm

developed based on MATLAB and Simulik which
calculates the illumination distribution, RMS delay spread,
and received signal wavefom considering the positions of
the rnsmitters nd the relections on walls.

This paper is organized as follows. In Section II, the

modeling of physical structre is described. Section III
describes the illuminance performance and Section IV
describes the root mean squre (RMS) delay spread for
our system model. Finally, our conclusion is given in he
section V.
II.

MODELING OF OPTICAL CHANEL

We assume he physical prmeters for developing he

sinmulation program. The size of he oice room size is
Sm x Sm x 3m and the LEDs are installed on the ceiling;
the height of desk is 0.8S m nd the receiver is placed on
the working plne (the bottom surface in the room model
shown in Fig. 1). The other simulation prmeters are
listed in Table 1.
SIMULATION PARAMETERS.

TBLE 1.

30 [deg]
0.73 [cd]
3600 (60x60)
50 [deg]
0.8

Semi-angle at half power

Center luminous intensiy

Number of LED each group

Field of view
Relection coeicient

- - - -

t
I

---+....--.

I
I
I

I
I
HI
I

___

-+---+ --

. _,.--________-:,:
,----------------------r
w

,,

E_

Fig. I. Indoor visible light communication environment.

The distribution of illuminnce at a working plane is

discussed. It is assumed that he sorce of emission and

OWC-5

CSNDSP 2010

538

he relected points on wall have a Lmbertin radiation

patten [1, 6, 7]. he lambertin emission means hat the
light intensity emitted rom he source has a cosine
dependence on the ngle of emission with respect to the
srface nomal. Following the nction for an optical lik
[1, 5-10], the luminous intensiy in angle is given by:
/()

shows he disribution for 4 ransmitters wih he

semingle of 30 degree. he perfomance of LEOs with
narow ield of view is shown in here. It opens a chance
for improvement LEOs with the wide ield for illuminnce.
The value is in he rnge rom 62.80 to 803.91 x. The
average value is 371.53 x.

(1)

/(O) cosm()

, .. , '
oo

Where /(0) is the center luminous intensity of the group

LEOs, is the ngle of irradiance, m is the order of
Lambertian emission and is given by the semi-angle at
half illuminance of the LEO 112 as:
m

- In 2 / In( cos 1I2 )

(2)

A horizontal illuminance Ehor at a point (x,y,z) on the

working plane is given by :
Ehor (x,y, z)

y(m)

.(m)

Fig. 3. Distribution of illuminance in case of one transmitter.

Mximum value= 768.10 Ix.

/(O) cosm()
Dd2 cos(! )

(3)

800

Where Dd is the distance between rnsmitter nd receiver,

700

/ is the angle of incidence.

600

To survey the illuminance distribution of LEOs

system, we assume two conigrations for LEO position
on the ceiling. n case of one ransmitter, the position is
he center of the ceiling, nd for four transmitters the
ransmitters re located at the position like Fig. 2.
5---,

5-

300

200

(3.75,3.75)

y(m)

(2.5,2.5)
(1.25,1.25)
5

00

________

(1.25,3.75)
Position

500

x(m)

Fig. 4. Illuminance distribution incase of 4 transmitters with the LED

semiangle 000 degree. Mx= 803.91 Ix; Min= 62.80 Ix; Avrge=
371.53 Ix.

(3.75,1.25)

100
o

Fig. 2. The position of transmitters on the ceiling.

We used MATLAB2007R program to develop the

simulation program. First, the progrm calculates he
direct illumination, and the relected illumination for one
or plral transmitters with Lambertian radiation patten
(the results are given by mny previous researches [6, 7]).
Also, the program calculates he MS delay spread for
his effect.

1300
1200
1100
1000

900

800

700
600

III.

500

ILLUMINNCE PERFORMNCE

he distribution of illuminance of or system is shown

in igure below:
Fig. 3 shows the illuminnce with 1 ransmitter, he
semiangle at half power is 30 degree. he maximum value
of luminous lux in the center is 768.10 x. Fig. 4 below

400
0

x(m)

OWC-5

CSNDSP 2010

539

Fig. 5. Distribution of illuminance with 4 trnsmitters with the LED

semiangle of 70 degree. Max= 1342.50Ix; Min=315.90 Ix;
Average=958.l1Ix.

Fig. 5 is given for comparing our study with previous

reserches [6, 7]; the numerical calculation in illuminance
disribution of or system at the semiangle of 70 degree is
shown in here. All of prmeters re same.
According to the standardized by Intenational
Orgnization for Standrdization (ISO), the illuminance of
his system is rom 300 to 1500 x), t is suicient for
ofice work.
In our case, we assume hat the relections at the
relected point have Lmbertian pattens. Ater calculation
(including direction, relection wih Lambertin patten)
nd comparing with the illminance without Lmbertin
patten at relected points, the value is larger than 10.8 %
(in the sme conditions) in case of four ransmitters.

. HAO)

relections

L dHre/O)

(4)

, is the optical power transmitted rom an LED,

H d (0) is he channel DC gain on directed pahs, which is

given in [5-9] as

{(m+

I): cosm ()T//)g(/) cos(/),

2fDd

/> P c

O</<'c

/ > Pc

(7)
Where D I is the distance between an LED and a
relective point, D2 is the distance between a relective
point and a receiver, p is the relectance factor, .al is a
relective rea of small region, a is the angle of incidence
to a relective point, J is the angle of irradince to the
receiver.
With M direct pahs rom rnsmitters to a speciic
receiver and N relection paths to the same receiver, the
total power of he received optical signals is calculated as
M

he term of MS delay spread is he stndrd deviation

(or root-mean-squre) value of the delay of relections.
Considering boh he direct pah and the irst order
relected path, the received optical power at a point cn be
calculated by dividing the relection walls into points of
relections as follows:

LDSL{

cos (p) cos(a)...

COs(3), (/)g(/) COS(/),

ROOT EN SQUARE DELAY SPREAD

IV.

HAO)=

(m+l)A

2 -pAwal
-'2 ""
2iDI D 2

Pr LPdi + L.,j
j

(8)

Where j,i is the received optical power of a direct

light at the th point and pr,j mens the received optical

.

power of a relected light at he /h point. M denotes the

nmber of components for direct light nd N denotes the
number of components for relected light.
The MS delay spread provides n estimate for a kind
of nomalized delay time due to multiple relections.
Therefore, the MS delay spread will be a critical
performance criterion for the upper bound of the data
transmission rate.
The mean excess delay is defmed [16] to be

(9)

0 < / < Pc
(5)

,(/) is the gain of an optical ilter, and g(/) is he

gain of optical concentrator. / is the ngle of incidence.
Pc denotes the width of a ield of vision at a receiver.

he optical concentrator g(/) can be given as

Id,i is the propagation time for the th direct light Pd,i

and Ir,j is the propagation time of he /h relected light
,j'
The MS delay spread

TS

is given by:
(10)

(6)
where
n

denotes the reractive index.

dHref(O) is he channel DC gain on relection poits [5-9]

nd is given as:

It is noted that the MS delay spread depends on the

relative levels of optical power components wihin

Pr .

Fig. 6 shows the distribution of MS for one

transmitter. he maximum value is 3.2 ns (nanosecond),
the minimum is 0.86 ps. Fig. 7 shows the distribution of

OWC-5

540

RMS for one ransmitter. he mximum value is 3.57 ns

he minimum is 0.17 ns.
t is accepted that the maximum bit rate that cn be
ransmitted through the chnnel without needing n
equalizer will be limited as follows [11, 16]

CSNDSP 2010

TBLE II.
SAMPLE POINTS, FOLLOWING THE CUTTING PLANE
(2.5,Y,Z) ON THE CARTESIAN'S COORDINATE.
point

Co-ordinate

(25,05)

(25,1.0)

(25,1.5)

(25,2.0)

5(centre)

(25,25)

(12)
herefore, it is estimated that he maximum data rate
rom Eq. (9) will be limited to 28 Mb/s considering whole
bottom surface.
-"

(25,3.0)

(25,35)

(25,4.0)

(25,45)

x10
3

3.50E-09

2.5
4

g2
8
$ 1

..4 transmitters

3.0E-09
2.50E-09

_3

1 transmitter

2.0E-09

1.5

1.50E-09
1.oE09
5.0E-10

0.5

O.OOE+OO

-__=o__-__"'__

Fig. 8. RMS delay spread at several sample receiver positions.

Fig. 6. Distribution ofRMS delay spread for one trnsmitter. The

position of the trnsmitter is (2.5, 2.5) with the LED semingle of 30
degree.

source signal
O

-"
x10
3.5
3
4

"
g2
8
$ 1

2
1.5

0
5

OA

0.6

0.8

1
signal

1.2

1A

1.6

1.8

1.2

1A

1.6

1.8

noise

2.5

0.2

0.2

OA

0.6

0.8

receivered Signal conSidering reflection

O

0.2

OA

0.6

0.8

1.2

1.4

1.6

1.8

1.2

1A

1.6

1.8

1A

1.6

1.8

ater filter

0.5
o

X(m)

0.2

OA

0.6

0.8

recovered signal

Fig.7. Distribution ofRMS delay spread for four trnsmitters. The

positions of the trnsmitters are (1.25,1.25),(1.25,3.75),(3.75,1.25),
(3.75,3.75) with the LED semingle of 30 degree.

Fig. 8 shows the RMS delay spread performance at

some sample points. he RMS delay spread of system at
some speciic points in case 4 transmitters usually larger
than he value in case 1 transmitter. The main reason is
the multipaths of light for four transmitters.

0.2

OA

0.6

0.8

1.2

x 10-'

Fig. 9. Waveforms rom the simultion using MATLB/Simulink.

The trnsmitted signal is in the NZ-OOK modulation format. The bit
rate is set to be 20Mbps.

Fig. 9 shows the simulation for Z-OOK modulation

format using ATLB/Simulik. he noise components
rom ambient light sources are implemented based on the

OWC-5

541

measurements in [17]. The wavefom calculation

considering relections with Lambertian patten at the
relection points are displayed. The simulation program
needs to be upgraded by implementing the various
physical parmeters including the size of active area, the
responsivity, the noise components in the photodetectors,
the asymmetric relections on walls, and various
advnced modulation fomats.
V.

CONCLUSION

n this paper, we have reported the simulation prorm

for indoor visible light communication environment based
on ATLAB and Simulink. he progrm considered
irst-order relections at each wall. Using the simulation
program, he distributions of illuminnce nd MS delay
spread were analyzed at bottom surface. Also, he
wavefoms for NZ-OOK have been demonsrated. t is
expected to be upgraded with more realistic physical
parameters nd various advnced modulation fomats.

ACKNOWLEDGEENT
his work was supported by the IT R&D proram of
KEIKEIT [KIOOI930, Reserch of 380-780 nanometer
visible light RGB wireless communications for IT
illumination communication usion].
REFENCES
[1]

J. M. Kahn and 1. R Brry,"Wireless Inrared Communications,"

Proc. IEEE 85 (2),265-298 (1997).

[2]

D. O'Brien and M. Ktz, "Optical wireless communications within

fourth-generation wireless systems," 1. Opt. Network 4, 312-322
(2005).

[3]

M. Kavehrad and S. Jivkova, "Indoor broadband optical wireless

communications: optical subsystems designs and their impact on
channel chracteristics," IEEE Wireless Commun. 10, 30- 35
(2003).
1. B. Carruthers, S. M. Carroll, "Statistical impulse response
models for indoor optical wireless channels," Int. 1. Commun.
Syst. 18,267-284 (2005).

[4]

[5]

A. Sivabaln nd 1. John, "Improved power distribution in diuse

Indoor Optical Wireless systems employing multiple trnsmitter
conigurations," Opt. Quantum Electron. 38,711-725 (2006).

[6]

T. Komine, M. Nkagawa "Fundamental Analysis for Visible

Light communication system using LED light," IEEE Trns.
Consum. Electron. 50,100-107 (2004).
T. Komine and M. Nkgawa, "Performance evaluation of
Visible- Light Wireless Communication System using White LED
Lighting," in Proceedings of the Ninth IEEE Symposium on
Computers and Communications, pp.258-263 (2004).

[7]

[8]

T. Komine, S. Haruyama nd M. Nakagawa, "A Study of

Shadowing on Indoor Visible-Light Wireless Communication
Utilizing Plural White LED Lightings," Wireless Pers. Commun.
34,211-225 (2005).

[9]

K. Fan, T. Komine, Y. Tanka, M. Nakagawa, "The Effect of

Relection on Indoor Visible-Light Communication System
utilizing White LEDs," in Proceedings of the 5h Int. Symposium
on Wireless Personal Multimedia Communications (WPMC2002),

pp. 611-615 (2002).

[10] T. Komine and M. Nakagawa, "Integrated system of white LED
visible light communication and power-line communication,"
IEEE Trns. Consum. Electron. 49,71-79 (2003).
[11] Y. Tanaka, T. Komine, S. Haruyama and M. Nakagawa, "Indoor
Visible Light Data Trnsmission System Utilizing White LED
Lights," IEICE Trns. Commun. E86-B,2440-2454 (2003).

CSNDSP 2010

[12] C. G. Lee, C. S. Park, 1.-H. Kim, and D.-H. Kim, "Experimental

veriication of optical wireless communication link using high
brightness illumination light-emitting diodes," Opt. Eng. 46,
125005 (2007).
[13] D. C. O'Brien, "High-speed integrated trnsceivers for optical
wireless," IEEE Commun. Mag. 41,58-62 (2003).
[14] H. L. Minh,D. C. O'Brien,G. Faulkner,L. Zeng,K. Lee,D. Jung,
Y. Oh, "High-Speed Visible Light Communications Using
Multiple-Resonant Equalization," IEEE Photon. Technol. Lett. 20,
1243-1245 (2008).
[15] I. Moreno, c.-C. Sun, R Ivnov, "Fr-ield condition for light
emitting diode arrays," Appl. Opt.. 48,1190-1197 (2009).
[16] T. S. Rappaport, Wireless Communications (Prentice-Hall,2002).
[17] AJ.C. Moreira,RT.Valadas, A.M. de Oliveira Duarte, "Optical
interference produced by rtiicial light," Wireless Net., vol. 3,pp.
131-140,1997.