Sie sind auf Seite 1von 6


discussions, stats, and author profiles for this publication at:

Analyzing the non-linear effects at various

power levels and channel counts on the
performance of DWDM based optical fiber
communication system
DOI: 10.1109/ICET.2012.6375446




Iftikhar Rasheed

Muhammad Abdullah Sandhu

The Islamia University of Bahawalpur

The Islamia University of Bahawalpur





Mahwish Chaudhary

Available from: Iftikhar Rasheed

Retrieved on: 17 February 2016

Analyzing the Non-linear Effects at various Power Levels and Channel

Counts on the Performance of DWDM based Optical Fiber
Communication System
Iftikhar Rasheed, Muhammad Abdullah, Shahid Mehmood, Mahwish Chaudhary
Department of Telecommunication Engineering
The Islamia University of Bahawalpur
Bahawalpur, Pakistan,,,
Abstract In order to expand the capacity of optical fiber
communication system the most important phenomena is the
emergence of dense wavelength division multiplexing (DWDM).
Using DWDM, multiple channel of information can be
transmitted on single fiber. There are some limiting factors
related to data rate and capacity in DWDM optical fiber
communication system. These limiting factors can be linear or
non-linear. We can compensate the linear effect such as
dispersion and attenuation by using pulse having right shape
called soliton but the non-linear effect accumulate.
The most important non-linear effect occur in the fiber optics
communication system are Self phase modulation(SPM),
stimulated brillouin scattering (SBS), Cross phase modulation
(XPM), Four wave mixing ( FWM) and Stimulated Raman
scattering (SRS). These non linear effects are the most important
limiting mechanism to the capacity of optical fiber
communication system. Self phase modulation (SPM), Stimulated
brillouin scattering (SBS) occurs only in single channel fiber
optics communication system and Cross phase modulation
(XPM), Four wave mixing (FWM), Stimulated Raman scattering
(SRS) have impact on multichannel DWDM fiber communication
system. In this paper, we analyze the impact of cross phase
modulation (XPM), Four wave mixing (FWM), stimulated
Raman scattering (SRS) on DWDM communication system at
different power level & different number of data channels. We
also describe some novel technique to reduce the effect of these
non linearities (XPM, FWM, SRS) in DWDM fiber
communication system.
For simulation we use OptiSystem
Key word: DWDM, Cross phase modulation, Four wave
mixing, Stimulated Raman scattering

Non-linearitys in fiber arose as the number of data
channel, transmission length, data rate & power level increase.
Mainly non-linearity occurs in fiber because of dependence of
refractive index on power going through fiber.
This phenomena is shown by equation (1)
n = no + n2 (P/Aeff) [1]


Here no is core refractive index, n2 is non linear refractive index

coefficient, P is optical power in watt and Aeff is fiber cores
effective area [1].

978-1-4673-4451-7/12/$31.00 2012 IEEE

Practically change in refractive index with optical power is

small but this becomes effective when the length of fiber is
hundreds of kilometer.
The brief introduction of XPM, FWM & SRS is given below.
A. Cross phase modulation(XPM)
Cross-phase modulation (XPM) is an example of fiber
nonlinear effects which can limit the distance and the capacity
of DWDM optical fiber transmissions[2].
When a single optical pulse transmit through the optical
fiber, then because of refractive index dependence on power,
the leading edge of pulse cause increase in refractive index &
its trailing edge cause decrease in refractive index.[3] Leading
edge of pulse shift toward longer wavelength and trailing edge
toward shorter wavelength. This phenomenon called self phase
modulation, causes overall spreading of pulse.
In case of multiple pulse travel in the fiber as in DWDM
communication system, multiple pulses overlap to cause cross
phase modulation.
Distortion & pulse broadening is caused by cross phase
B. Four wave mixing(FWM)
Four wave mixing (WM) is the important factor having
catastrophic effect on DWDM communication system. This
non linear effect arises when two or more pulses transmit
through same fiber.
Generally FWM effect occur if the three light pulses,
having different wavelength & traveling through single fiber,
interact together to generate a new pulse.
If the three wavelength A, B & C are propagating through
single fiber, these wavelengths will interact according to
equation (2) to generate new D
D = A B C here




In general, the number of crossing product K, for M number of

input channel is given by equation (3)


Equation (3) shows that non linear effect FWM increase as
number of channel in DWDM system increase.

Interchange crosstalk effect is caused in DWDM system by

FWM non- linearity.
FWM effect can be reduced by using unequal channel
spacing & by dispersion management.
C. Stimulated Raman scattering(SRS)
Stimulated Raman scattering (SRS) is the ultimate
performance limiting phenomenon in multichannel optical
Transmission systems [6]. Among the WDM channels,
stimulated Raman scattering (SRS) causes a power transfer
from shorter wavelength channels to longer wavelength
channels [7]
When intense beam of light of high power level, which is
usually called pump, propagates in the optical fiber, then
interaction between vibration of molecules of SiO2 and light
cause stimulated Raman scattering.
This effect primarily affects the power distribution of the
input data channels and leads to channel to channel
crosstalk [8].

Figure 2. Eye Diagram

Output result for 11 input data channels, 3dBm input power & channel
spacing 100GHz

If the number of channels increases to 18 and power to

5dBm then we get following eye diagram, bit error rate and
max Quality Factor (Shown in figure 3)


Our proposed algorithm (shown in figure 1) consists of
WDM transmitter, WDM receiver, optical fiber & bit error rate
analyzer to analyze the output result.
In order to analyze the impact of XPM on optical fiber
communication system at various power levels, we vary the
input power & number of channel of WDM transmitter.
The optical transmission link in our algorithm is such
designed that the residual dispersion is almost 0 ps/nm/Km.
Transmission link consist 100Km single mode fiber (SMF)
with dispersion of 16 ps/nm/Km followed by 20Km long
dispersion compensating fiber (DCF) with dispersion of
-72ps/nm/km (shown in figure 1). One Optical amplifier (OA)
with gain of 35dB & zero noise figure is used after DCF. NRZ
nodulation type is used. Link is operating at 10Gbps. For the
visualization of BER, Q factor & eye diagram, BER analyzer is
used. Channels spacing is 100GHz.



Figure 3.
Eye Diagram
Output result for 18 input data channels & 5dBm input power , channel
spacing 100GHz

Now for 26 number of channels & 11dBm power, the

output eye diagram & bit error rate, Q factor etc are shown in
figure 4.



BER Analyzer


Figure 1

If transmitter consist 11 DWDM channels & 3dBm power

then eye & bit error rate, Q factor etc at optical receiver are
shown in figure 2.

Figure 4.
Eye Diagram
Output results for 26 input data channels , 11dBm input power, channel
spacing 100GH

By comparing the figure 2, figure 3 & figure 4, we can

analyze that as the power & number of channel trnsmitting
through fiber increase, the cross phase modulation effect also
incrrease. Figure 2, 3 & 4 shows the the bit error rate (BER) of

fiber communication system increase by increasing transmitter

power & number of data channels because of XPM.
A. Cross phase modulation compensation
In order to make DWDM fiber communication system
effective for long distance communication, it is necesssary to
reduce the non linear Cross phase modulation effect.
The penalty arising from XPM can be effectively decreased by
dispersion management technique & increasing channel
For the circuit shown in figure 1, increase the channel
spacing to 110GHz & reducing the length of DCF to 16 km to
leave some residual dispersion (almost 450 ps/nm/Km) in
fiber transmission link.
The Eye diagram, BER & Q factor for 11 WDM channel
& 3dBm power are shown below (Figure. 5)

Figure 6. Input spectrum for 8 data channels & -8dBm input WDM
transmitter power

Figure 5. Eye Diagram

Output result for 11 channels , 3dBm input power, channel spacing is 110GHz
& residual dispersion is not zero

Figure 2 & Figure 5 shows the output results of two

different fiber communication system having almost same
transmission length, input data channels & input power but the
residual dispersion & channel spacing for both are different.
So by comparing figure 2 & figure 5, we get that BER of
fiber system (whose output results are shown in figure 5)
decrease & Quality factor increase, by reducing XPM effect.

Figure 7. output spectrum for 8 data channels & -8dBm input WDM
transmitter power

Now if the number of channels increases to 16 & power to

2dBm, the output optical spectrum is shown in figure 8.


For analyzing the effect of FWM on DWDM fiber
communication system at different power level & number of
input data channel the proposed algorithm is almost same as
shown in figure 1. Transmission link consist 70Km SMF
followed by 16Km DCF. After DCF an optical amplifier with
25 dB gain is used. Dispersion of SMF is 17 ps/nm/Km & DCF
is -72ps/nm/Km. residual dispersion of link is almost zero.
Channel spacing is 100GHz. NRZ modulation type is used.
Input power and number of data channels are increases
gradually. Optical spectrum analyzer is used to analyze the
spectrum at input & output.
Spectrum of input signal is measured after the WDM
multiplexer, and after optical amplifier the output signal
spectrum is measured by Optical spectrum analyzer.
For 8 input channels & -8dBm input power, spectrum of
input & output signal is shown below in Figure 6 & Figure 7

Figure 8. Output spectrum for 16 data channels & 2dBm input WDM
transmitter power

For 30 data channels and 10 dBm input power, optical

spectrum of output signal is shown in figure 9.

Uneven channel spacing at the input side can also be used

to overcome the effect of FWM.
System which we selected for analysis of SRS consists of
CW lasers for input, ideal WDM multiplexer, optical fiber and
optical spectrum analyzer to observe the output results.
For SRS analysis in optical fiber communication system at
various power levels, we vary the number of channels and
power by keeping length of fiber unchanged.

Figure 9. Output spectrum for 30 data channels & 10 dBm input

WDM transmitter power

Following table (figure 10) gives the analysis that how

FWM effect increase with increasing the input power &
number of channel.

No of Obs

Input power
in dBm


-8 dBm
2 dBm
10 dBm

number of

FWM side
band power
-66 dBm
-47 dBm
-38 dBm

We design our system such that length of optical fiber

which we selected is 0.2 km. Channel spacing is not constant;
it varies from one channel to other. We keep the power of first
input channel greater than all of other remaining input
If we have 11 DWDM input data channels and power of
first channel is 50dbm and -99dbm of each of remaining 10
channels, then input signal spectrum after WDM Muxer and
output signal spectrum after fiber are shown in figure 12 &
figure13 respectively .

Figure 10.

A. Four wave mixing compensation

Rectangular optical filter can be very much effective to
compensate the four wave mixing effect.
For optical fiber communication system whose output is
shown in figure 7, in that system, if we use Rectangular optical
filter after optical amplifier. Bandwidth of Rectangular optical
filter is set 800 GHz & central frequency is 193.35GHz. The
output of this modified circuit is shown in figure 11.

Figure 12. Input signal spectrum for 11 channels

Figure 13. Ioutput signal spectrum for 11 input channels

Figure 11. Output spectrum for 8 I/P data channels, -8dBm input WDM
transmitter power & Rectangular optical filter is used at o/p

By comparing figure 7 & figure 11, we can conclude that

FWM can be completely removed by Rectangular optical filter.
Rectangular optical filter did not reduce the output power.

Now, we have 18 input channels, power of first channel is

54.7712 dbm and -99dbm of each remaining channels. Then
figure 14 shows the SRS effects.

Figure 14. output signal spectrum for 18 channels

Comparing the output spectrum shown in figure 13 &

figure 14, we conclude that SRS affect increase with increase
in power & number of input channels.
By increasing the effective area of fiber we can reduce SRS
in fiber communication system.
This paper briefly presents the analysis of XPM, FWM and
SRS effect in DWDM fiber optics communication system. The
analysis is done on the basis of result obtained from simulation
in OptiSystem. In this paper we show how the non linearitys
i.e. XPM, FWM and SRS increase in optical fiber
communication system by increasing the input power and
number of input channels. Further, some effective techniques
for compensating the XPM & FWM effects are discussed
briefly in this paper. Currently the work is underway, on
efficient compensation of SRS.






optical fiber telecommunication
N. Kikuchi, K. Sekine and S. Sasaki Analysis of cross-phase
modulation (XPM) effect on WDM transmission performance
ELECTRONICS LETTERS 10th April 1997 Vol. 33 No. 8
Milora cviteticd optical transmission system engineering
A.V.Ramprasad, M.Meenakshi, G.Geetha, R.Satheesh kumar
Suppression of Four Wave Mixing Crosstalk Components in DWDM
Optical Systems 1-4244-0340-5/06/$20.00 02006 IEEE
Bijayananda Patnaik , P.K.Sahu Optimization of Four Wave Mixing
Effect in Radio-over-Fiber for a 32-Channel 40-GBPS DWDM System

Jinsong Wang, Xiaohan Sun, and Mingde Zhang Effect of Group
Velocity Dispersionon Stimulated Raman Crosstalk inMultichannel
LETTERS, VOL. 10, NO. 4, APRIL 1998
Toshiaki Yamamoto, Student Member, IEEE, and Seiji Norimatsu
Statistical Analysis on Stimulated Raman Crosstalk in DispersionManaged Fiber Links JOURNAL OF LIGHTWAVE TECHNOLOGY,
VOL. 21, NO. 10, OCTOBER 2003
Sebastien Bigo, Member, IEEE, Stephane Gauchard, Alain Bertaina,
and Jean-Pierre Hamaide, Member, IEEE Experimental Investigation of
Stimulated RamanScattering Limitation on WDM TransmissionOver
Various Types of Fiber Infrastructures IEEE PHOTONICS