EC405 Optical Communication Lecture Note

MODULE IV
DIGITAL TRANSMISSION SYSTEMS

4.1 DESIGN OF INTENSITY MODULATED DIRECT DETECTION( IMDD) LINKS

A modulation scheme in which an information signal changes the driving current and the latter
modulates the amplitude of the output light is called intensity modulation (IM). At the receiver
end, modulated light is directly converted into electrical current by a photodiode; that is, the
receiver provides direct detection (DD). In short, the method is called intensity
modulation/direct detection (IM/DD)

4.1.1 LINK DESIGN

The user generally specifies the distance over which the information is to be sent and the data
rate to be transmitted. The Designer then has to find the specification of the system
components.
The designer generally has to define some additional criteria either as per the standards or as
per the user specifications.
The Design criteria are given in the following.

1. Primary Design Criteria

Data Rate
Link length
2. Additional Design Parameters

 Modulation format eg. Analog/digital

Depends upon the type of signals user want to transmit. For example if it is a TV signal, then
may be analog transmission is more suited as it requires less bandwidth and better linearity.
On the other hand if data or sampled voice is to be transmitted, digital format may be more
appropriate. The digital signals have to be further coded to suite the transmission medium and
also for error correction.
 System fidelity: BER, SNR
The system fidelity defines the correctness of the data received at the receiver.
For digital transmission it is measured by the Bit Error Ratio (BER) . The BER is defined as

In optical system the BER has to be less than 10-9 . For analog system the quality parameter is
the Signal-to-noise (SNR) ratio. Also there is a parameter called the inter-modulation distortion
which describes thelinearity of the system.
 Cost : Components, installation, maintenance
Cost is one of the important issues of the link design. The cost has three components,
components, installation and maintenance. The component and the installations cost are the

Sanish V S ,Asst.Professor , ECE,JCET |1

EC405 Optical Communication Lecture Note

initial costs. Generally the installation cost is much higher than the component cost for long
links. This is especially true for laying the optical cable. It is therefore appropriate to lay the
cables keeping in view the future needs. The optical link is suppose is supposed to work for at
least 25years. The maintenance costs are as important as the initial cost. An initial cheaper
system might end up into higher expenses in maintenance and therefore turn out to be more
expensive as a whole.
 Upgradeability
The optical fiber technology is changing very rapidly and the data rates are increasing steadly.
The system should be able to adopt new technology as well as should be able to
accommodate higher data rates with least possible changes.
 Commercial availability
Depending upon which part of the world one is, the availability of the components and the
systems may be an issue.

DESIGN OF A SIMPLE POINT-TO-POINT OPTICAL LINK.

A simple point to point link is shown in the following Fig.

The link has primarily 3 components to design.

1. Optical Transmitter.
2. Optical Fiber
3. Optical receiver

The Fig. shows the typical optical power which LEDs and and Lasers can deliver and the
photo-detector needs for a BER of 10-9

Sanish V S ,Asst.Professor , ECE,JCET |2

EC405 Optical Communication Lecture Note

As the data rate increases the power delivering capacity of the source reduces and at the
same time the power requirement of the detector increases.
The following table gives the combination of the sources and fibers for different link
capacity and distance.

Considering the cost, speed etc, first choose the laser and the detector. Also the type of
fiber is chosen from the above table. Generally a multi-core fiber is laid even if the
immediate requirement is only one or two fibers.
The link design then reduces to finding locations of the repeater on a long link.
Two calculations are carried out in the link design
The Fig. show the power loss model of an optical fiber link. The power is lost in various
components like, fiber, connectors, splicing.

The fiber loss depends upon the wavelength and also the physical conditions of the fiber.
The fiber loss is generally higher than that the specified by the manufacturers. This is
primarily due to micro-bending of the fiber. Also the micro-bending loss is higher for
1550nm compared to 1310nm. Therefore the over all loss could be higher at 1550nm than at
1310nm, although intrinsically silica glass has minimum loss at 1550nm. Typical loss at
1550nm may lie in the range 0.4-0.5 dB/km.
Sanish V S ,Asst.Professor , ECE,JCET |3
EC405 Optical Communication Lecture Note

The splice loss could be between 0.05-0.1 dB per splice.

The connector loss is higher and could be 0.2-0.3 dB per connector.
POWER BUDGET CALCULATIONS
Ps = Power from the Transmitter in dBm
Pr = Sensitivity of receiver in dBm for given BER

 Maximum permissible loss

 Power limited link length

Beyond this distance the SNR is below the acceptable limit

Sanish V S ,Asst.Professor , ECE,JCET |4

EC405 Optical Communication Lecture Note

System margin is generally taken to be 6 dB to accommodate deterioration of components over

time.

RISE TIME BUDGET CALCULATION

 Rise time analysis gives effective bandwidth of the link

 For satisfactory operation of the link ( for NRZ format)

√
Beyond this distance the signal distortion is unacceptable

 Rise time of a system or component = 1/bandwidth

Here,

 = Total system rise time.

 ttx = Transmitter rise time
 trx= Receiver rise time. Generally trx˃˃ ttx.
 D = Dispersion of the fiber
 = Spectral width of the transmitter
 L= Length
 Tb= Data bit duration =

Note: For RZ data the system rise time should be .

 In the link design two lengths, the power budget length and the rise time budget
length are calculated.
 The repeater has to be installed at a distance min( .

Generally, the links are power limited and the repeaters are installed at . Typical
 repeater length is about 50-60 km in practice.

Example : 1

Let us take typical parameters for a link.

Data rate = 1 GHz. ; DFB Laser spectral width = 0.1nm ;
SM fiber dispersion at 1550nm = -20 ps/km/nm = -0.02 ns/km/nm
Rise time of the receiver = 0.1 nsec ; Rise time of the transmitter = 0.1nsec
Fiber loss = 0.4dB/km ; Transmitter power -3 dBm
Min Detectable power -40 dBm ; Neglect splice and connector losses.

Sanish V S ,Asst.Professor , ECE,JCET |5

EC405 Optical Communication Lecture Note

{ }

{ }

( )

Since , the link is power limited and the repeater has to be installed at a
distance of less than 77.5Km

Questions for practice

Refer Text Book : Fiber optic communication system –Systems & Components by Mishra
& Ugale page nos. 283-288

Sanish V S ,Asst.Professor , ECE,JCET |6

EC405 Optical Communication Lecture Note

4.2 BASIC COHERENT SYSTEM

The dashed lines enclose the main elements which distinguish the coherent system from its
IM/DD equivalent. At the transmitter a CW narrow-line width semiconductor laser is shown
which acts as an optical frequency oscillator. An external optical modulator usually provides
amplitude, frequency or phase shift keying of the optical carrier by the information signal.

At the receiver the incoming signal is combined (or mixed) with the optical output from a
semiconductor laser local oscillator. The combined signal is then fed to a photo detector for
direct detection in the conventional square law device. When the optical frequencies (or
wavelengths) of the incoming signal and the local oscillator laser output are identical, then the
receiver operates in a homodyne mode and the electrical signal is recovered directly in the
baseband. For heterodyne detection, however, the local oscillator frequency is offset from the
incoming signal frequency and therefore the electrical spectrum from the output of the
detector is centred on an intermediate frequency (IF) which is dependent on the offset and is
chosen according to the information transmission rate and the modulation characteristics. This
IF, which is a difference signal (or difference frequency), contains the information signal and
can be demodulated using standard electrical techniques.

The electrical demodulator is required in particular for an optical heterodyne detection system
which can utilize synchronous or asynchronous/ nonsynchronous electrical detection.
Synchronous or coherent demodulation implies an estimation of phase of the IF signal in
transferring it to the baseband. Such an approach requires the use of phase-locking techniques
in order to follow phase fluctuations in the incoming and local oscillator signals. Alternatively,
asynchronous or noncoherent (envelope) IF demodulation schemes may be employed which
are less demanding but generally produce a lower, performance than synchronous detection
techniques .

Sanish V S ,Asst.Professor , ECE,JCET |7

EC405 Optical Communication Lecture Note

Optical homodyne detection is by definition, however, a synchronous demodulation scheme

and as the detected signal is brought directly into the baseband, then optical phase estimation
is required.

4.2.1 COHERENT DETECTION PRINCIPLES

Figure: Basic coherent receiver model

A simple coherent receiver model for ASK is displayed in Figure. The low-level incoming signal
field eS is combined with a second much larger signal field eL derived from the local oscillator
laser. It is assumed that the electromagnetic fields obtained from the two lasers (i.e. the
incoming signal and local oscillator devices) can be represented by cosine functions and that
the angle Φ represents the phase relationship between the incoming signal phase
and the local oscillator signal phase defined at some arbitrary point in time. The two fields
may be written as

eS = ES cos(ωSt + ) and eL = EL cos(ωLt )

where ES is the peak incoming signal field and ωS is its angular frequency, and EL is the peak
local oscillator field and ωL is its angular frequency. The angle Φ(t) representing the phase
relationship between the two fields.
For heterodyne detection, the local oscillator frequency ωL is offset from the incoming signal
frequency ωS by an intermediate frequency such that:
ωS = ωL + ωIF

where ωIF is the angular frequency of the IF, the IF is usually in the radio-frequency region and
may be a few tens or hundreds of megahertz.
By contrast, within homodyne detection there is no offset between ωS and ωL and hence
ωIF = 0. In this case the combined signal is therefore recovered in the baseband. The two wave
fronts from the incoming signal and the local oscillator laser must be perfectly matched at the

Sanish V S ,Asst.Professor , ECE,JCET |8

EC405 Optical Communication Lecture Note

surface of the photo detector for ideal coherent detection. This factor creates the normal
requirement for polarization control of the incoming optical signal.

In the case of both heterodyne and homodyne detection the optical detector produces a signal
photocurrent Ip which is proportional to the optical intensity (i.e. the square of the total field
for the square-law photodetection process) so that:

Ip ∝ (eS + eL)2

Ip ∝ [ES (cos ωSt + φ) + EL cos ωLt]2

Removing the higher frequency terms oscillating near the frequencies of 2ωS and 2ωL which
are beyond the response of the detector and therefore do not appear in its output, we have:

Then recalling that the optical power contained within a signal is proportional to the square of
its electrical field strength, expression may be written as:

where PS and PL are the optical powers in the incoming signal and local oscillator signal
respectively.

where η is the quantum efficiency of the photo detector, e is the charge on an electron, h is
Planck’s constant and f is optical frequency. When the local oscillator signal is much larger than
the incoming signal, then the third a.c. term in Eq. may be distinguished from the first two d.c.
terms and Ip can be replaced by the approximation IS where

Equation allows the two coherent detection strategies to be considered. For heterodyne
detection ωS ≠ ωL

For the special case of homodyne detection, however, ωS = ωL and therefore Eq. reduces to:

Or
Sanish V S ,Asst.Professor , ECE,JCET |9
EC405 Optical Communication Lecture Note

where R is the responsivity of the optical detector. In this case the output from the photodiode
is in the baseband and the local oscillator laser needs to be phase locked to the incoming
optical signal.
When the local oscillator signal power is much greater than the incoming signal power then the
dominant noise source in coherent detection schemes becomes the local oscillator quantum
noise. In this limit the quantum noise may be expressed as shot noise.

Substituting for IpL ,where the photocurrent generated by the local oscillator signal is assumed
to be by far the major contribution to the photocurrent, gives:

The detected signal power S, being the square of the average signal photocurrent

Hence the SNR for the ideal heterodyne detection receiver when the local oscillator power is
large (ignoring the electronic preamplifier thermal noise and photo detector dark current noise
terms) may be obtained

IF amplifier bandwidth BIF is assumed to be equal to 2B (i.e. BIF = 2B)

Hence the SNR limit for optical homodyne detection is:

Sanish V S ,Asst.Professor , ECE,JCET | 10

EC405 Optical Communication Lecture Note

4.2.2 SENSITIVITY OF COHERENT RECEIVER

BER and Sensitivity of coherent systems depends on modulation format and demodulation
scheme used by coherent receiver.

Table 1 : Sensitivity of synchronous receiver

Table 2 : Sensitivity of asynchronous receiver

4.2.3 SENSITIVITY DEGRADATION

Sensitivity of practical systems may degrade due to phase noise, intensity noise, polarization
mismatch and fiber dispersion.

1. Phase Noise
An approach to solve the phase-noise problem is designing special receivers known as phase-
diversity receivers . Such receivers use two or more photo detectors whose outputs are
combined to produce a signal that is independent of the phase difference ;.
This technique works quite well for ASK, FSK, and DPSK formats. Figure shows schematically a
multiport phase-diversity receiver. An Optical component known as an Optical hybrid
Sanish V S ,Asst.Professor , ECE,JCET | 11
EC405 Optical Communication Lecture Note

combines the signal and local-oscillator inputs and provides its output through several ports
with appropriate phase shifts introduced into different branches. The output from each port is
processed electronically and combined to provide a current that is independent of . In the
case of a two-port homodyne receiver, the two output branches have a relative phase shift of
90°, so that the currents in the two branches vary as , and . When the two
currents are squared and added, the signal becomes independent of . In the ease of three-
port receivers, the three branches have relative phase shifts of 0, 120 °, and 240°. Again, when
the currents are added and squared, the signal becomes independent of .

Figure : Phase diversity receiver

2. Intensity Noise

A solution to the intensity-noise problem is offered by the balanced coherent receiver made
with two photo detectors. Figure shows the receiver design schematically. A 3-dB fiber coupler
mixes the optical signal with the local oscillator and splits the combined optical signal into two
equal parts with a 90° relative phase shift. The operation of a balanced receiver can be
understood by considering the photocurrents I+ and I-, generated in each branch. Using the
transfer matrix of a 3-dB coupler, the currents I+ and I- are given by

Where ⁄

The subtraction of the two currents provides the heterodyne signal. The dc term is eliminated
completely during the subtraction process when the two branches are balanced in such a way
that each branch receives equal signal and local-oscillator powers. More importantly, the
intensity noise associated with the dc term is also eliminated during the subtraction process.
The reason is related to the fact that the same local oscillator provides power to each branch.
As a result, intensity fluctuations in the two branches are perfectly correlated and cancel out

Sanish V S ,Asst.Professor , ECE,JCET | 12

EC405 Optical Communication Lecture Note

during subtraction of the photo currents I+ and I-. It should be noted that intensity fluctuations
associated with the ac term are not cancelled even in a balanced receiver. However, their
impact is less severe on the system performance because of the square-root dependence of the
ac term on the local-oscillator power.

Balanced receivers are commonly used while designing a coherent lightwave system because
of the two advantages offered by them. First, the intensity-noise problem is nearly eliminated.
Second, all of the signal and local-oscillator power is used effectively. A single-port receiver
rejects half of the signal power PS (and half of PLO) during the mixing process. This power loss is
equivalent to a 3-dB power penalty. Balanced receivers use all of the signal power and avoid
this power penalty. At the same time, all of the local-oscillator power is used by the balanced
receiver, making it easier to operate in the shot-noise limit.

3. Polarization mismatch

The most commonly used approach solves the polarization problem by using a two-port
receiver, with two branches process orthogonal polarization components. Such receivers are
called polarization-diversity receivers as their operation is independent of the polarization
state of the signal received.

Figure shows the block diagram of a polarization-diversity receiver. A polarization beam

splitter is used to separate the orthogonally polarized components which are processed by
separate branches of the two-port receiver. When the photocurrents generated in the two
branches are squared and added, the electrical signal becomes polarization independent. The
power penalty incurred in following this technique depends on the modulation and
demodulation techniques used by the receiver. In the case of synchronous demodulation, the
power penalty can be as large as 3 dB. However, the penalty is only 0.4-0.6 dB for optimized
asynchronous receivers.

Sanish V S ,Asst.Professor , ECE,JCET | 13

EC405 Optical Communication Lecture Note

Figure: Polarization –diversity coherent receiver

4. Fiber dispersion
Dispersion becomes a major limiting factor for systems designed with standard fibers when
transmission distance is increased using optical amplifiers. Dispersion management would
solve this problem. Electronic equalization can be used for compensating dispersion in
coherent systems . The basic idea is to pass the intermediate-frequency signal through a filter
whose transfer function is the inverse of the transfer function associated with the fiber.

4.3 COMPARISON BETWEEN IMDD & IMCD SYSTEMS

SL.NO PARAMETERS IMDD IMCD

Intensity modulated direct Intensity modulated

1 Name
detection system coherent detection system

More open eye. (The more

open the eye the better is
2 Eye diagram Less open eye
the quality of the received
signal)

3 sensitivity Low sensitivity High sensitivity

4 Gain Low gain High gain

5 Power penalty low high

Unamplified
6 low high
transmission distance

7 Splitting ratio low high

The higher receiver sensitivity, higher splitting ratio and a long unamplified transmission reach
makes coherent detection scheme a technology of choice for application in FTTH technology.

Sanish V S ,Asst.Professor , ECE,JCET | 14

EC405 Optical Communication Lecture Note

4.4 SOLITON SYSTEMS

We can use SPM to compensate for dispersion-caused pulse Spread to keep the pulse width
constant over the entire transmission distance. Such a pulse, having a constant width, is
called a soliton. A soliton either keeps its width constant or changes it periodically, but its
width never exceeds a given value. A soliton, then, is essentially a nonlinear effect because it is
based on SPM.

Role of positive GVD and SPM:

SPM leads to the spectrally enriched positively chirped pulse by lowering of the leading edge
frequencies i.e. red shifted, and raised for the trailing edge frequencies i.e. blue shifted.
Consequently, propagation through a medium with positive GVD, the high frequency
components in the trailing edge of the pulse will increasingly lag behind the low frequency
components in the leading edge upon propagation as if these two end are being pull apart. This
results in the enhanced temporal broadening of the pulse.

Role of -ve GVD and SPM

The positive chirp of spectrally enriched pulse produced by SPM can be removed with
subsequent pulse compression. In this case, the high frequency components from the back end
try to catch with the low frequency component from the front end due to negative GVD.
Consequently, the pulse is compressed as shown in figure

Sanish V S ,Asst.Professor , ECE,JCET | 15

EC405 Optical Communication Lecture Note

If the two effects, balance each other, the pulse will retain a constant shape which moves like a
particle and result in a pulse which retains its shape on propagation which is dynamically and
structurally stable like a particle and is called a fundamental soliton. Such a particle like state of
the pulse will have strong implications for the transport of signal or information. It is also
possible that it rebuilds its shape periodically in space. This is the case of higher order solitons.

4.4.1 SOLITON TRANSMISSION

The soliton pulse able to keep its shape and width steady as a result of mutual compensation of
dispersion-broadening and self-phase-modulated narrowing processes.

Figure: soliton propagation in one soliton period.

Solitons can be transmitted over a very long fiber link without amplification and dispersion
compensation, they look like the most promising transmission technology.

Sanish V S ,Asst.Professor , ECE,JCET | 16

EC405 Optical Communication Lecture Note

The major challenge in the development of commercial soliton transmission systems is

accomplishing soliton transmission over already installed multimillion-mile networks of a
standard fiber.

LONG DISTANCE SOLITON TRANSMISSION SYSTEM

Desirable attributes of a communication system are

 Large data bit rates.

 Low transmission losses.
 Easy scalability.

Since transmission bandwidth or the data bit rate of a communication system is proportional
to the carrier wave frequency, light waves being the highest frequency waves are therefore, the
best suited candidates as information carrier. Further, the availability of low loss fibers, fiber
amplifiers and high speed optical modulators has made the optical communication a reality.

Long distance communication systems like submerged trans-Atlantic communication link

require that the integrity definition of the data bits or optical pulses is maintained. However,
pulse broadening by group velocity dispersion effect in optical fibers will spoil this definition.
This can be managed by using repeater stations at regular periods to recover the data bits and
retransmit these. Too frequent use of repeater stations add to the complexity and cost of the
system. Optical nonlinearity can be exploited to compensate the broadening effects of the
group velocity dispersion and generate soliton pulses in optical fibers which maintain their
shape with propagation. The long distance soliton communication can, therefore, have great
prospects. Nakazawa et-al of NTT(Japan) demonstrated 10Gbits soliton transmission over 1
million kilometer in a land mark paper. Many refinements have been reported since then. The
obvious wave length choice for soliton system is 1500nm fiber window as it offers lowest
1.55μm and –ve GVD.

For a typical fiber having Aeff =50μm operating λo= 1500nm the soliton power, energy and
period are given by

Sanish V S ,Asst.Professor , ECE,JCET | 17

EC405 Optical Communication Lecture Note

To achieve high bandwidths, soliton pulse duration should be small. However smaller
duration soliton requires large peak powers and energy. The choice of input pulse width then
will depend upon the semiconductor laser source.

As an example let us consider the duration of soliton pulses

Achievable using semiconductor laser sources

need amplification for

To avoid soliton- soliton interaction of adjacent pulse (e.g. 11000 type pattern) separation
between solitons should be several times the soliton width(say times) i.e.

There are several factors that affect the performance of the long distance soliton system. In the
following we address some of these limitations and remedies to overcome these.

4.4.2 LIMITATIONS OF SOLITONS

1. FIBER LOSS:
The power requirement of the fundamental soliton of given pulse width is

If fiber loss is small over soliton period, Zo, then pulse will adiabatically adjust its width to
conform to power requirement for N=1 Soliton. However, this limits the communication
distance. To overcome this limitation, one can use the distributed Raman gain
amplification .Taking fiber loss to be 0.2dB/km, 3 dB loss will result after propagation
distance L=15 km. Also when input power drops by a factor of 2, the soliton will become a
normal dispersive pulse i.e. non-soliton propagation. Hence for long distance it is necessary to
amplify the signal periodically.

Solitons arise as a balance of the phase shifts arising from nonlinearity & dispersion

 The balance occurs for

Sanish V S ,Asst.Professor , ECE,JCET | 18

EC405 Optical Communication Lecture Note

By making length of fiber amplifier so that the phase change in each amplifier
segment is small, average dispersion can be cancelled by the average nonlinear phase
shift. Hence Soliton System can even be realized with lumped amplification which can
be easily realized using Erbium doped fiber amplifiers arranged periodically in a chain.

The significance of Er doped fiber amplifier is that

 It can provide amplification for signal in the low loss region of 1520-1580 nm range.
 Pump power are modest ~tens of mw.
 Pumping possible with 1480 nm In Ga As/ in gap MQW Lasers.
 Gain of up to 25dB.

The gain factor of the amplifier has to be such that the gain exactly compensates the
loss.
For a fiber having loss of

0.2 dB/km

Let

To compensate for the loss we need to launch power greater than necessary for exciting N=1
soliton.
Thus to have “average N=1 ” soliton it can be shown that we require

For the example of 25ps fundamental soliton N=1 in a lossless system, one requires

Therefore power required for N =1.3 soliton is

Sanish V S ,Asst.Professor , ECE,JCET | 19

EC405 Optical Communication Lecture Note

2.SOLITON- SOLITON INTERACTIONS

Maximum data rate is determined by how close two adjacent pulse/bits can be packed. To
determine how close two solitons can propagate without interacting one needs to solve
nonlinear Schrödinger equation numerically with initial condition

where is their initial separation and θ is their relative phase. Note that
bound soliton while will correspond to two independent single solitons. Evolution of two
co-propagating solitons with is shown in figure . It can be seen that these collapse and
separate with an oscillation period
This process is referred as “Brother Soliton”.

If the relative phase “θ” between adjacent solitons is such that then they experience
repulsive force.

Propagation over Length results in collapse of two in-phase soliton and distance
apart at input. By choosing the soliton travel length soliton collapse can be avoided.
In practice soliton separation has to be 5-10 X soliton width. Hence using 1 ps soliton data
transmission rates ~100-200 G bits /s can be realized. 10 ps soliton will provide data rates of
10-20 Gbits/s and 25ps soliton will enable only 4-8Gbits/s data rates.

Sanish V S ,Asst.Professor , ECE,JCET | 20

EC405 Optical Communication Lecture Note

1. AMPLIFIER INDUCED PULSE JITTER & THE GORDON-HAUS (GH) EFFECT

The soliton central frequency can also affected by ASE induced by optical amplifiers. ASE
induced noise produce random frequency variations ( ,which can be translated into time
variations at the pulse output. This phenomenon is called Gordon–Haus effect. Soliton
transmission is susceptible to Gordon –Haus effect, which results in a reduction of overall
system transmission capability.

The soliton pulses are affected by two sources.

 Fluctuation of pulse energy of soliton

 Pulse arrival time at the end of fiber.

The Gordon-Haus effect is due to the combination of fluctuations in the center frequency of
optical pulses with the timing. Therefore, a change in the center frequency causes a change in
group velocity, which in turn affects the pulse timing.

Gordon and Haus considered the noise in a fiber-optic link in periodically spaced fiber
amplifiers. In these amplifiers, it is assumed that they have a wavelength-independent gain that
introduces a quantum noise. This noise randomly causes a shift in the optical center frequency.
With more and more shift in the center, the timing deviations of pulses accumulate. The
cumulative effect of these deviations causes growing timing errors.

The Gordon-Haus analysis was based on soliton pulse propagation for long-haul data
transmission. This jitter can be suppressed considerably by using regularly spaced optical
filters, or amplifiers having a limited gain bandwidth. In case of mode locked lasers there is an
effect of timing drift, due to the Gordon-Haus jitter.

The perturbing Gordon-Haus effect can be reduced considerably by adopting the following
measures.Use narrow-band frequency-guiding filters periodically distributed along the
transmission line. Key is that the filters are opaque to noise but transparent to solitons.

Use post- transmission dispersion compensation. This can reduce jitter up to a factor of 2.

Sanish V S ,Asst.Professor , ECE,JCET | 21

EC405 Optical Communication Lecture Note

4.4.3 ADVANTAGES OF SOLITON BASED COMMUNICATION SYSTEMS

Solitons are unaffected by an effect called PMD due to the imperfection in the circular
symmetry fiber which leads to a small and variable difference between the propagation
constants of orthogonal polarized modes. This dispersion becomes a major problem over long
distances and at high data rates.
 Solitons are well matched with all optical processing techniques. Our long term goal is to
create networks in which all of the key high-speed functions, including routing,
demultiplexing and switching are performed in the optical domain. So the signals need
not be converted into an electrical form on the way. Most of the devices and techniques
designed for these tasks work only with well-separated optical pulses, which are
particularly effective with solitons.
 If the solitons are controlled properly they can be more robust than NRZ pulses. Schemes
have been devised that can not only provide control over the temporal positions of the
solitons, but also remove noise added by amplifiers. Such schemes would allow the
separations between amplifiers to be many times greater than in the schemes that are
used with NRZ pulses. The particle nature of solitons can be employed for sliding-
frequency guiding optical filters along the link. With these cantered at slightly reducing
wavelengths along the path, the soliton is capable of following this change without any
degradation.
 The use of in-line saturable absorbers, which work in the time domain to suppress noise.
The particle feature of solitons is also very useful to perform various all optical functions
such as switching.
 Yet another and very important particle feature is the fact that solitons tend to stay
together in presence of a walk-off between different polarization components - so called
PMD. The soliton PMD robustness may be a key to success when upgrading existing fiber
links to high speed.
 Solitons would replace the traditional NRZ with RZ modulations, which are used in
almost all commercial terrestrial WDM systems. Typically the design of a conventional
WDM system involves an effort to increase the power as much as possible to counteract
attenuation and noise without introducing too much nonlinearity. Thus NRZ and RZ
systems are often called linear system. Recent advancements in soliton communication
with 3.2T b/s have been demonstrated.

4.4.4 APPLICATIONS OF SOLITONS

Sanish V S ,Asst.Professor , ECE,JCET | 22

EC405 Optical Communication Lecture Note

The effects due to nonlinearity and dispersion are destructive in OFC but useful in Optical
Soliton Fiber Communication (OSFC) systems.
The soliton type pulses are highly stable. Their transmission rate is more than 100 times better
than that in the best linear system. They are not affected by the imperfections in the fiber
geometry or structure. Soliton can be propagated without any distortion if the nonlinear
characteristics like amplitude, intensity of the pulse-depending on velocity and the dispersion
characteristics like frequency-depending on velocity of the media, are balanced. Soliton can
also be multiplexed at several wavelengths without interaction between the channels, though
they usually suffered in Non Return to Zero (NRZ) systems. Nowadays, most of the
communication systems use RZ format, for example Transoceanic Transmission (TOT) where
the transmission rate is 10 G b/s per channel, transmits the information transfer in dispersion
managed fibers. This format is the only stable form for pulse propagation through the fiber in
the presence of fiber nonlinearity and dispersion in all optical transmission lines with
minimum loss.

In dispersion managed fibers, a large pulse width is allowed, pulse height is reduced and
nonlinear interactions between adjacent pulses as well as among different wavelength
channels are reduced . Not only in he field of communication, solitons also find application in
the construction of optical switches , soliton laser, pulse compression and the like.

Sanish V S ,Asst.Professor , ECE,JCET | 23