Download Analysis and Compensation of Four Wave Mixing Products in

ISSN : 2230-7109 (Online) | ISSN : 2230-9543 (Print)
IJECT Vol. 5, Issue 4, Oct - Dec 2014
Analysis and Compensation of Four Wave Mixing Products
in Wavelength Division Multiplexing System
1
Himakshi Sharma, 2Er. Gian Thakur
P.G. Scholar, Dept. of ECE, L.R.I.E.T Solan (H.P), India
Asst. Professor, Dept. of ECE, L.R.I.E.T Solan (H.P.), India
1
2
Abstract
In this paper we had investigated the non-linearity i.e. Four Wave
Mixing in Wavelength Division Multiplexing system. Also the
effect of phase mismatching (dispersion) and Optical signal
to noise ratio is analyzed. The effect of channel spacing and
increasing number of channels is taken into account. This works
aims at finding the new possibilities to mitigate the effect of four
wave mixing in Wavelength Division Multiplexing systems. The
outputs are analyzed in optical spectrum analyzers.
where N is the number of wavelengths to transmit such a high
capacity over long distance requires operation in the 1550-nm
window of dispersion shifted fiber. In order to preserve an adequate
signal to noise ratio, a 10 Gbps system operating over a long
distance and having minimal optical repeater spacing of 100 km
needs optical power launch of around 1mW per channel. For such
WDM system, the simulation requirements of high launch power
and low dispersion give rise to the generation of new frequency
due to four wave mixing [8].
Keywords
Four Wave Mixing (FWM), Stimulated Raman Scattering (SRS),
Stimulated Brillouin Scattering (SBS), Bit Error Rate (BER),
Optical Spectrum Analyzer (OSA), Optisystem, Wavelength
Division Multiplexing (WDM).
III. Experimental Setup
The simulation arrangement for FWM configuration is shown in
fig. 1. Block diagram of the system is divided into three sections:
transmitter, channel and receiver. In transmitter section first block
is data source where data rate and bit is customized. The data
source is the pseudo number sequence. Electrical driver is used to
convert binary sequence into electrical pulses; continuous wave
lorentzian laser is used with various values of power. Booster can
also be used which is an EDFA with fixed output power and further
fixed gain optical amplifiers are used. Channel section basically
consists of optical fiber and band pass Bessel electrical filter. In
receiver section an optical spectrum analyzer is used to access
the non-linear output of optical spectrum along with that a power
meter is used to view the power ratings in receiver side.
I. Introduction
This article provides concepts of Four Wave Mixing which is a
type of Nonlinearity. Due to the widespread growth in wireless
communication, network operators are facing difficulty in
accommodating the growing traffic. WDM appears to be a viable
solution to such problems posed by the micro cellular system
[1]. WDM system however suffers from nonlinear effect in
fiber [2]. Communications in WDM generate new waves under
appropriate condition through a variety of Non-linear phenomena
such as Stimulated Raman Scattering (SRS), Stimulated Brillouin
Scattering (SBS), Self-Phase Modulation (SPM), and Four Wave
Mixing (FWM). Four wave mixing or four photon mixing is the
process whereby optical power in one channel in the WDM system
is spilled over into an adjacent channel [3]. Such powers transfers
not only results in the power loss for the channel but also induces
inter channel crosstalk that degrades the system performance
severely. FWM is also useful in designing light wave systems.
It is often used for demultiplexing channels when time-division
multiplexing is used in the optical domain [4]. It can also be used
for wavelength conversion. Wavelength conversion is a significant
function in the future broadband multichannel light wave system
because it makes many other possible useful functions such as
wavelength reuse and dynamic wavelength routing and switching
[5]. FWM in optical fibers is occasionally used for generating a
spectrally inverted signal through the process of optical phase
conjugation.
II. Four Wave Mixing
Four wave mixing also termed as four photon mixing is one of
the most governing signal degradation effect in the WDM system.
Four wave mixing is the third order non-linearity in the in silica
fiber that is analogous to intermodulation distortion [6].
It mostly occurs with the dense channel spacing and low chromatic
dispersion. Channels in the WDM system are equally spaced which
tends to the generation of new waves which will fall at channel
frequency by FWM, and thus give rise to crosstalk [7]. Dense
WDM transmission in which individual wavelength channel are
modulated at the rates of 10Gbps offer volumes of N*10 Gbps,
134
International Journal of Electronics & Communication Technology
Fig. 1: Experimental setup of FWM
IV. Results and Discussion
Two WDM channels were launched over a single optical fiber
of span 10Km. Dispersion was completely compensated at each
channel. The fiber dispersion value was varied from -6 (ps/nm/
km) to 14 (ps/nm/km) through individual runs. The optical power
spectrum of the received signal shows that the FWM product
decreased with increasing dispersion. Also negative dispersions
were used which could be easily compensated locally. The OSNR
seemed to show high rate at using negative dispersion.
Fig. 2(a) and fig. 2(b) had shown the optical signal spectrum at
the transmitter end and receiver end at zero dispersion values. By
matching them we had found that at receiver end two extra peaks
had appeared which nothing but the noise was in the system due
to FWM phenomena.
There are certain parameters which need to be preset:
1. Attenuation =0
2. Effective area = 53µm2
3. Input power =1mW
4. Degeneracy =6
5. Fiber length =10km
w w w. i j e c t. o r g
IJECT Vol. 5, Issue 4, Oct - Dec 2014
ISSN : 2230-7109 (Online) | ISSN : 2230-9543 (Print)
Table 1: Dispersion vs. OSNR
Dispersion
(ps/nm/km)
-5
-4
0
4
8
14
Received signal
power(dbm)
-45.950
-46.088
-45.956
-45.951
-45.949
-46.983
OSNR
47.86
47.84
47.58
47.02
46.97
46.07
or decreasing value beyond “0” actually increases phase mismatch
which help in vanishing extra peaks. Fig. 2(d and e) showed
outputs at “4” and “-5” values. Outputs were also calculated for
each of the values showed in Table 1.
Fig. 2(c): Output of Optical Spectrum Analyzer at “0”
dispersion.
Fig. 2(a): Output of Optical Spectrum Analyzer at transmitter
end.
Fig. 2(d): Output of Optical Spectrum Analyzer at “4”
dispersion
Fig. 2(b): Output of Optical Spectrum Analyzer at receiver end
at “-5”dispersion.
Fig. 2(b) showed the Dual OSA output at dispersion value -5
(ps/nm/km). Clearly it could be seen from table 1.That we got
best result at “-5” value with highest OSNR which was required.
Similarly fig. 2(c) showed result at “0” dispersion value that has
maximum noise and that was due to phase matching. By increasing
w w w. i j e c t. o r g
Fig. 2(e): Output of Optical Spectrum Analyzer at “-5”
dispersion
International Journal of Electronics & Communication Technology 135
IJECT Vol. 5, Issue 4, Oct - Dec 2014
Fig. 2(f): Plot between OSNR vs. Dispersion.
ISSN : 2230-7109 (Online) | ISSN : 2230-9543 (Print)
Himakshi Sharma received her B.
Tech. degree in Electronics and Comm.
from Himachal Pradesh University
(HPU), Shimla, India, in 2012,
the M. Tech. degree in Electronics
and Communication Engineering
from Himachal Pradesh Technical
University (HPTU), Hamirpur, India,
in 2014. Her research interests include
digital signal processing, electronic
measurement techniques,antenna and
optical technique and networks. At present, she is engaged in
research in optical fiber nonlinearities.
Fig. 2(f) showed the comparison graph between proposed and
existing scheme. And the OSNR came out to be high which was
required.
V. Conclusion
For the given experimental set up various values of dispersion had
been changed accordingly. The results given in fig 2(a, b, c, d and
e) had been showing the plot between optical power and dispersion
values. Here the spectrum of the received signal had shown that
FWM product decreased with increasing dispersion or more
exactly the best value of OSNR came out t “-5” dispersion. Thus
in conclusion it had been illustrated that FWM products in WDM
system was stronger at lower fiber dispersion values. However at
zero dispersion value due to phase-matching condition the FWM
effect came out to be extreme. By increasing the fiber dispersion
the phase mismatch increased and FWM effect decreased.
References
[1] Naresh Kumar, Ajay K. Sharma, Vinod Kapoor,“XPMInduced Crosstalk with Variety of Fiber in SCM-WDM
Optical Transmission Link”, pp. 2125-2127.
[2] Naresh Kumar, Ajay K. Sharma, Vinod Kapoor, “Improved
XPM-Induced Crosstalk with Higher Order Dispersion in
SCM-WDM Optical Transmission Link”, pp. 941-944.
[3] Govind P. Agrawal,“Fiber-Optic Communication System”,
3rd ed., Wiley, New York, 2002.
[4] Govind P. Agrawal,“Nonlinear Fiber Optics”, Third
Edition.
[5] Gerd Keiser,“Optical Fiber Communications”, McGraw-Hill,
fourth edition.
[6] C. Siva Ram Murthy, Mohan Gurusamy,“WDM Optical
Networks Concepts, Design, and Algorithms”, Prentice-Hall
India.
[7] S. Pachnicke, E De Man, S. Spalter, E. Voges,“Impact of the
Inline Dispersion-Compensation Map On FWM - Impaired
Optical Networks”, IEEE Photonic Technology Letters, Vol.
17, No. 1 January 2005.
[8] Jian Hui Zhou, Namkyoo,“Four Wave Mixing Wavelength
Conversion Efficiency in semiconductor Travelling wave
Amplifiers Measured to 65nm of Wavelength shift”, IEEE
Photonics Technology Letter, Vol. 6, No. 8, August 1994.
[9] T. Sabapathi, S. Sundaravadivelu,"Analysis of Bottlenecks
in DWDM Fiber Optic Communication System”, Optik 122,
pp. 1453– 1457, 2011.
136
International Journal of Electronics & Communication Technology
w w w. i j e c t. o r g