Characterization of an Optical Communication System

Utilizing Dispersion Compensating Fiber and

Nonlinear Optical Effects

Miziya.K , S.K.Sudheer and Alina C. Kuriakose

Department of Optoelectronics

University of Kerala

Thiruvananthapuram -695 581

India

miziyak@gmail.com

Abstract— Nonlinear effects and dispersion in optical fibers have

become an area of academic research and of great importance in

optical fiber based systems. Presence of these nonlinear effects

and dispersion in the optical fiber communication systems

adversely affect the communication between two receiving ends.

In the present investigation, the authors are reporting the design

and analysis of a dispersion compensated link of single mode

fiber (SMF) for total span of 100Km at 1550nm.

The effect due to nonlinear dependence of the refractive

index on pulse intensity and power (self phase modulation) for

different modulation schemes NRZ Rectangular, NRZ Raised

Cosine, RZ Rectangular, RZ Soliton is analyzed.The degradation

of the performance of fiber optic link due to the nonlinear effects

arise in presence of booster amplifier is simulated and results are

discussed. The above investigations help to reduce the length of

dispersion compensating fiber (DCF) and thereby the cost of

implementation will be low without compromising the quality of

transmission and reception.

The variation of BER was studied with respect to bit rate,

fiber length and booster gain. The simulation was repeated for

different values of booster gain and various modulation schemes.

The simulation studies are carried out using OPTSIM software

from RSoft Inc.

Keywords― Nonlinear effect, DCF, chromatic dispersion,

anomalous dispersion, SPM, SMF, Optical amplifier, NRZ, RZ, RZ

soliton.

I. INTRODUCTION

Nonlinear effects and dispersion in optical fibers have become an area of academic research and of great importance in the optical fiber based communication systems. Presence of these nonlinear effects and dispersion in the optical fiber communication systems adversely affect the performance of

communication between two receiving ends. When the effects of nonlinearity and dispersion are considered together, the situation changes. In some circumstances, the nonlinearity could counteract the dispersion [1].

Out of the several techniques for dispersion compensation, DCF is the most widely deployed dispersion compensator, as they are cascadable, commercially available and compatible with all optical network concepts [2].Inside each DCF span, the pulses of a longer wavelength channel travel faster than pulses of a shorter wavelength channel , whereas the opposite is the case inside each SMF span [3] .The condition for perfect dispersion compensation for a pulse to regain its original shape, is given by D1L1 + D2L2 = 0, where D1 and D2 are dispersion coefficients of SMF and DCF, L1 and L2 are the corresponding length respectively. For practical reasons, L2

should be as small as possible. [1, 4].

The most dominant fiber nonlinear effect in a standard SMF is the self-phase modulation (SPM), which is caused by the nonlinear dependence of the refractive index on pulse intensity [5].Deepak Gupta etal reported that appropriately choosing the pulse profile and the pulse power, SPM and GVD can be made to exactly compensate for each other [6].

It is known that fiber has lowest loss at 1550nm and also that efficient amplifiers operate around this wavelength. So it is desirable to operate around 1550nm [4].With increased bitrates it has been shown that Return-to-Zero (RZ) modulation formats offer certain advantages over NRZ, as they tend to be more robust against distortions [7].

II. MODELING AND SIMULATION OF OPTIC AND NON-

OPTIC EFFECT.

OPTSIM is advanced optical communication system simulation packages which can be used to design optical

IEEE - 31661

4th ICCCNT 2013 July 4-6, 2013, Tiruchengode, India

communication systems and simulate them to determine their performance given various component parameters. To guarantee the highest possible accuracy and real-world result, OPTSIM simulation software is used for the work.

A. System Description

The simulation setup is shown in the Fig.1. This system consists of three major sections i.e., transmitter, fiber section and receiver section.

Figure 1 . Simulation set up with NRZ Rectangular driver

1) Transmitter Section: The transmitter consists of

pseudorandom bit sequence generators, which generate bit

sequence, which is fed to coder to produce an electrical coded

signal. The pulses, after passing through a low pass filter are

then modulated with continuous wave (CW) laser radiation of

wavelength 1550nm to obtain optical pulses .Mach Zender

Modulator is used as external modulator to obtain optical

pulses from electrically coded signal. The PRBS has a starting bit rate of 10Gbps. NRZ and RZ

coders are used here to analyze different modulation scheme effects. The optical source used here is continuous wave Lorentzian laser which has a Center emission frequency, center emission wavelength and CW power of 193.4THz, 1547-1560 nm and 1dB respectively. Electrical low pass filter used is Bessel filter.

2) Fiber Section: Optical fiber section consists of single

mode fibers and dispersion compensated fibers in the channel.

The optical signal after boosting is fed into the single mode

fiber. The 100 km fiber span consists of two segments, SMF

of length L1 km and DCF of length L2km. L1 and L2are kept

variables. Fixed gain amplifiers are used as boosters and pre-

amplifiers. Preamplifier gain is fixed as 30dB with figure of

merit 4.Gain of booster is varied from 10dB to 24 dB. For the design of communication channels, the parameters

of SMF and DCF are as shown in the Table1.

3) Receiver Section: Optical receiver is composed of

PIN diode as photoelectric detector, Lorenzian band pass filter

as optic filter, Bessel low pass filter as electrical filter.PIN

photodiode converts the optical signal into an electrical signal.

The Bessel filter has a cut-off frequency determined by the

type of the waveform used for modulation. . Finally a

visualization tool called Scope, provided by OPTSIM is used

to display results at the output. It is an optical or electrical

oscilloscope with numerous data processing options, eye

display and BER estimation features [12]. The parameters

used in receiver section used are given table 2:

Serial

number Parameter SMF DCF

1 Fiber length (km) 100-75(L1) 0-25(L2)

2 Fiber dispersion D

(ps/nm/km) 17 -95

3 Effective come Area Aeff

(nm2) 80 20

4 Dispersion slope

( ps/nm2/km) 0.08 -0.1

5 Non linear refractive index

n2(x 10-20) 2.5 2.5

6 Attenuation α (dB/km) 0.25 0.5

Table 1 . Parameter Of SMF and DCF

Serial

number Center Frequency 194.4THz

1 No: of Poles 5

2 Quantum Efficiency 0.7

3 Responsivity 0.875A/W

4 Dark Current 0.1nA

Table 2 . Parameter Of Receiver

III. RESULTS AND DISCUSSIONS

The effect due to nonlinear dependence of the refractive index on pulse intensity and power (self phase modulation) for different modulation schemes is analyzed.

IEEE - 31661

4th ICCCNT 2013 July 4-6, 2013, Tiruchengode, India

A dispersion compensated transmission link of total 100km span is designed and realized using OPTSIM. Within the span of 100km, the SMF length L1 and DCF length L2 are kept variables.SMF length is varied from 75km to 100km and DCF length from 25km to nil. Minimum bit error rate of 1x 10

-40 (BERmin) and wide eye opening were obtained for a

DCF length of 15km.The variation of BER and quality of eye pattern were studied for different values of bit rate and booster gain (G). A degradation of the performance of fiber optic link was observed due to the nonlinear effects arose in the presence of booster amplifier. Better performance was observed for smaller DCF length, because in anomalous dispersion region, SPM counteracts with dispersion.

1) Variation Of BER with Bit rate: The bit rate of the

signal is varied in PRBS generator and corresponding values

of BER are noted. The plot of BER with respect to bit rate is

shown in fig 2.

Figure 2 . Variation of BER with Bit rate

Figure 2 shows the variation of BER with bit rate, where L1and L2are 88km and 12 km respectively. NRZ rectangular driver is used for modulation. For our simulation, the maximum bit rate with permissible BERis observed to be 10Gbps.The experiment when repeated for booster gains of 10dB, 16dB and 22dB, the starting point of flattening region move downwards. The explanation can be given that SPM effect increases with increase in booster gain and thus BER is decreased. When the selected bit rate was 10.83Gbps, 4.3x 10

-

4, 3x 10

-4, and 2.4x10

-4 are the obtained BER for booster gain

values 10db, 16dB, and 22dB respectively.

2) Variation Of BER and Q with DCF length: Fig. 3

shows the variation of BER with DCF length for booster gain

10dB. A minimum BER of 1x10-40

was observed for a range

of DCF length varying from 11km to 19km.Within this range

better eye pattern was obtained for DCF length 14km which

has the maximum Q value. The variation of Q value with DCF

length is shown in fig. 4.

Figure 3 . Variation Of BER with DCF length

Figure 4 . Variation Of Q factor with DCF length

3) Variation Of BERmin and Qmax with booster

gain:Fig.5 shows the range of DCF length for which a BER of

1x10-40

is maintained. The distance between the green line

(DCF max) and blue line (DCF min) gives the range. The

DCF length at which maximum Q value obtained is also

shown in fig.5.For each gain value, better eye diagram is

observed at the Qmax points within the BER minimum range

of DCF. From the analysis of fig.5, two results could be observed

with the increase in booster gain. 1) The range of DCF length for which BERmin (1 x10

-40) obtained has reduced and shifted

to lower values of DCF. 2) DCF length for which the maximum Q value obtained is shifted downwards to the lower values of DCF.As a result the minimum length of DCF with

IEEE - 31661

4th ICCCNT 2013 July 4-6, 2013, Tiruchengode, India

lowest BER and good eye diagram has reduced with increase in booster gain. It can be explained with the nonlinear dependence of the refractive index on pulse intensity. With the increase in booster gain, SPM effect increases. As we work in the anomalous region, it reduces GVD.As a result reduced length of DCF is enough to compensate the acquired dispersion through SMF. The best eye diagrams obtained for different booster gain are shown in fig.6.

Figure 5 . Variation of BERmin and Qmax DCF lengths with booster gain

(using NRZ rectangular driver)

From fig.6, fig.6 (d) is the best eye pattern with less noise which is at DCF length13km.Fig.6 (f) is also a good eye pattern for DCF length 12km.Further the DCF length reaches to 10km whose performance is not stable with lower booster gain value. It can be understood from fig.5. Hence we could say that L2 = 12km is the minimum DCF length that can be used with a stable performance.

Figure. 6 (a) Eye diagram observed for L1 = 86km,L2=14km and G = 10dB

Figure.6 (b) Eye diagram obtained for L1=86km , L2=14km and G=12dB

Figure.6 (c) Eye diagram observed for L1 = 87km, L2=13km and G = 14dB

Figure.6(d) Eye diagram obtained for L1=87km ,L2=13km and G=16dB

Figure.6 (e) Eye diagram observed for L1 = 88km, L2=12km and G =18dB

IEEE - 31661

4th ICCCNT 2013 July 4-6, 2013, Tiruchengode, India

Figure.6 (f) Eye diagram obtained forL1=88km, L2=12km and G=20dB

Figure.6 (g) Eye diagram observed for L1 = 88km,L2=12km and G = 22dB

Figure.6 (h) Eye diagram obtained for L1=90km, L2=10km and G=24dB

4) Observations with different modulation schemes:In

fig.7, the performances of the dispersion compensated fiber

optic link with different modulation schemes are shown. Fig.5

shows 1) The range of DCF length for which BER of 1x10-40

is maintained. 2) The DCF length at which maximum Q value

obtained. Fig.7 (a), Fig.7 (b) and Fig.7(c) shows the same

when NRZ raised cosine driver, RZ rectangular driver and RZ

soliton driver are used. Comparing Fig.5 and Fig.7 (a) the following observations

are noted. When NRZ raised cosine driver is used, DCF minimum value with BER minimum is constant within the range of different gain values in our study. It was observed that for a specific gain value and DCF length, the BER is found to be less for NRZ rectangular driver. Comparing Fig.7(b) and Fig.7(c), the observation made is that RZ soliton driver is less affected with booster gain.

Figure.7 (a) Variation of BERmin and Qmax with booster gain using NRZ

cosine driver

Figure.7 (b) Variation of BERmin and Qmax with booster gain using RZ

rectangular driver.

Figure.7(c) Variation of BERmin and Qmax with booster gain using RZ

soliton driver.

IEEE - 31661

4th ICCCNT 2013 July 4-6, 2013, Tiruchengode, India

In general the following observations made are with NRZ and RZ modulation schemes. (a) RZ modulation is less affected with variation in booster gain. (b) The range of DCF length for which the BER is minimum is more for RZ modulation. The power launched into the fiber using RZ modulation is less when compared to NRZ modulation is the explanation that can be given to the observations noted.

IV. CONCLUSION

In the present investigation, the variation of BER was studied with respect to bit rate, fiber length and booster gain. The simulation was repeated for different values of booster gain and various modulation schemes.

The maximum bit rate with permissible BER is observed to be 10Gbps.When a particular bitrate is selected, the obtained BER decreases with increase in booster gain. The degradation of the performance of fiber optic link due to the nonlinear effects arise in presence of booster amplifier is simulated and discussed. The above investigations help to reduce the dispersion compensating fiber length (DCF).RZ modulation is found to be less affected with variation in booster gain. Also the range of DCF length for which the BER is minimum is more for RZ modulation.

The BER and eye diagram technique have been employed as a good means for evaluating the system performance in the present paper.

REFERENCES

[1] S Sujith and K G Gopchandran,”A Simulation study on DCF compensated SMF using OptSim”, Modern Telecommunications and Control Systems and Workshops (ICUMT), 2010

[2] M. Basu, S. Roy, “Design considerations of depressed clad W-shaped single mode dispersion compensating fibers”, Optik117, 377–387, (2006).

[3] Christos Kouloumentas, “Theoretical analysis of the all-fiberized, dispersion managed regenerator for simultaneous processing of WDM channels” ,Optics Communications 284, 4340–4349, (2011).

[4] I.C. Goyal*, A.K. Ghatak and R.K. Varshney,” Dispersion Compensating Fibers ”,lCTON, 2002

[5] Muhammad Anisuzzaman Talukder, Mohammed NazrulIslam ”A long haul wavelength division multiplexed system using standard single-mode fiber in presence of self-phase modulation” , Optik 120 , 356–363, (2009).

[6] Deepak Gupta, Gautam Kumar, K. Thyagarajan ,”Nonlinear pulse propagation in dispersion decreasing fibers”, Optics Communications 237, 309–317, (2004) .

[7] Anu Sheetala, AjayK .Sharmab, R.S.Kalerc, Impact of optical modulation formats on SPM-limited fiber transmission in 10and40Gb/s optimum dispersion-managed light wave systems , Optik 121 ,246–252, (2010).

[8] A E Willner in B.D.Guenther, A.Miller ,L.Bayvel and J.E Midwinter, ”Encyclopedia of Modern Optics” Academic press,2004.

[9] DjaferK. Mynbev and Lowell L.Scheiner, “Fiber-Optics Communications Technology”,Pearson publication, 2001.

[10] John M Senior,“Optical Fiber Communications”, Pearson publication,2010.

[11] Agrawal.G.P., “Application of Nonlinear FiberOptics”, Academic press, USA, 2007.

[12] OPTSIM Application Notes and Examples, Rsoft Design Group, Inc

IEEE - 31661

4th ICCCNT 2013 July 4-6, 2013, Tiruchengode, India