Optical amplifiers

Signal Reshaping and Amplification

• In long distance communications, whether going

through wire, fibre or wave, the signal carrying the information
experience:
- Power loss
- Pulse broadening
which requires amplification and signal reshaping.

• In fibre optics communications, these can be done in two ways:

• Opto-electronic conversion
• All optical
2
• In order to transmit signals over long distances (>100 km) it is
necessary to compensate for attenuation losses within the fiber.
• To amplify an optical signal with a conventional repeater, one
performs photon-to-electron conversion, electrical
amplification,retiming,pulse shaping, and then electron-to –photon
conversion.
• Although this works well for moderate –speed single –wavelength
operation and not for high speed multiple wavelength system.
• To eliminate the transmission delay problem ,we develop optical
amplifiers.
• These devices completely operates in the optical domain to boost the
power levels of multiple light wave signals over spectral bands of 30
nm and more.
Applications and types of optical amplifiers
• Optical amplifiers have found widespread use in diverse applications
ranging from ultralong undersea links to short links in access
networks.
• 1.General applications
• Figure shows the general applications of the following three classes of
optical amplifiers.
a. Inline optical amplifiers
• In a single mode link, the effects of fiber dispersion may be small so
that main limitation to repeater spacing is fiber attenuation.
• Since such a link does not necessarily require a complete regeneration
of the signal, simple amplification of the optical signal is sufficient.
• Thus an inline optical amplifiers can be used to compensate for
transmission loss and increase the distance between regenerative
repeaters as fig. a illustrates.
b. Preamplifier
• Fig.11.b shows an optical amplifiers being used as a front –end
preamplifier for an optical receiver.

• Thereby ,a weak optical signal is amplified before photodetection so

that the SNR degradation caused by thermal noise in the receiver
electronics can be suppressed.
• Compared with other front end devices such as a avalanche
photodiodes or optical heterodyne detectors ,an optical preamplifiers
provides a larger gain factor and a broader bandwidth.
C. Power amplifier
• Power or booster amplifier applications includes placing the device
immediately after an optical transmitter to boost the transmitted
power, as in in fig c.
• This serves to increase the transmission distance by 10-100 Km
depending on the amplifier gain and fiber gain.
• One can also employ an optical amplifier in a local area network to
compensate for coupler-insertion loss and power splitting loss.
• Figure d shows an example for boosting the optical signal in front of a
passive star coupler so sufficient power arrives at each receiver.
2.Optical Amplifiers - Types

There are mainly three types:

• Semiconductor Laser (optical) Amplifier (SLA) (SOA)

• Active-Fibre or Doped-Fibre Amplifier
• Erbium Doped Fibre Amplifier (EDFA) –widely used in C-band
• Thulium Doped Fibre Amplifier (TDFA)- S band
• Fibre Raman Amplifier (FRA)

10
• All optical amplifiers increase the power level of incident light through a
stimulated emission or an optical power transfer process.
• In SOAs and DFAs the mechanism for creating the population inversion
that is needed for stimulated emission to occur is the same as is used in
laser diodes.
• Although the structure of such an optical amplifier is similar to that of a
laser, it does not have the optical feedback mechanism that is necessary
for lasing to take place.
• Thus an optical amplifier can boost incoming signal levels, but it cannot
generate a coherent optical output by itself.
• The basic operation is shown in fig below
 • Device absorbs energy supplied
from an external source called
Generic optical amplifier the pump.
• The pump supplies energy to
electrons in an active medium,
which raises them to higher
energy levels to produce a
population inversion.
• An incoming signal photon will
trigger these excited electrons
to drop to lower levels through
a stimulated –emission
process.
• Since one incoming trigger photon stimulates a cascade effect in which many excited
electrons emit photons of equal energy as they drop to the ground state, the result is an
amplified optical signal.
• In Raman amplification, there is a transfer of optical power from a high
power pump wavelength to light wave signals at longer wavelengths
• This Raman amplification mechanism is done without the need for a
population- inversion process.
• Alloys of semiconductor materials from group III and V
(eg:gallium,indium.arsenic) make up the active medium in SOAs.
• The attractiveness of SOAs is that devices can be made to work in the O-
band as well as in the C-band.
• They can be integrated easily on the same substrate as other optical
devices and circuits.
• Compared with DFAs they consume less electrical power, have fewer
components, and are more compact.
• SOAs have a more rapid gain response, this rapid response results in
both advantages and limitations.
• The advantage is that SOAs can be implemented when both switching
and signal processing functions are called for in optical networks.
• The limitation is that the rapid carrier response causes the gain at a
particular wavelength to fluctuate with the signal rate for speeds up
to several Gb/s.
• Since this fluctuations affects the overall gain, the signal gain at other
wavelength also fluctuates.
• Thus the rapid gain response gives rise to crosstalk effects when a broad
spectrum of wavelengths must be amplified.
• In DFAs ,the active medium for operation in the S,C and L is created by
lightly doping a silica (silicon dioxide)or tellurite(tellurium oxide)fiber core
with rare earth elements such as thulium, erbium(Er) or ,ytterbium(Yb).
• The DFAs for the O-band are achieved through doping fluoride –based
fibers (rather than silica fibers)with elements such as neodymium(Nd) and
praseodymium(Pr).
• The important features of DFAs include the ability to pump the devices at
several different wavelengths, low coupling loss to the compatible –sized
fiber transmission medium, and very low impedance of gain on light
polarization.
• DFAs are highly transparent to signal format and bit rate, since they exhibit
slow gain dynamics, with carrier lifetimes on the order of 0.1-10 ms.
• The result is that,incontrast to SOAs,the gain responses of DFAs are basically
constants for signal modulations greater than a few kilohertz.
• Consequently ,they are immune from interference effects (such as cross talk
and intermodulation distortion)between different optical channels when
wavelength channels in a broad spectrum.
• A Raman amplifier is based on a nonlinear effect called stimulated
Raman scattering(SRS), which occurs in fibres at high optical powers.
• Whereas a DFAs requires a specially constructed optical fiber for its
operation , Raman amplification take place within a standard
transmission fiber .
• The Raman gain mechanism can be achieved through either a
lumped (or discrete) amplifier or distributed amplifier.
• In the lumped Raman amplifier configuration ,a spool of about 80m
small –core fiber along with appropriate pump laser is inserted into
the transmission path as a distinct packaged unit.
• For the distributed Raman amplifier application, one or more Raman
pump lasers convert the final 20 to 40 km of the transmission fiber
into a preamplifier.
• Since the Raman gain in a particular spectral range is derived from the
SRS-induced transfer of optical power from shorter pump
wavelengths to longer signal wavelengths, these amplifiers can be
designed for use in any wavelength band.
Erbium –doped fiber amplifiers (EDFA)
• The active medium in an optical fiber amplifier consists of a nominally
10-to-30 m length of optical fiber that has been lightly doped with a
rare earth element, such as thulium, erbium(Er) ,ytterbium(Yb) or
praseodymium(Pr).
• The host fiber material can be standard silica, a fluoride based glass,
or a tellurite glass.
• The operating regions of these device depends on the host material
and the doping elements.
• A popular material for long haul telecommunication application is a
silica fiber doped with erbium, which is known as an EDFA .
• In some cases, Yb is added to increase the pumping efficiency and the
amplifier gain.
• The operation of a standard EDFA normally is limited to the 1530 to
1565 nm region.
• Actually the fact that an EDFA operates in the C-band(conventional
band)
1.Amplification mechanism
• Whereas semiconductor optical amplifiers use external current
injection to excite electrons to higher energy levels, optical fiber
amplifiers use optical pumping.
• In this process, one use photons to directly raise electrons into excited
states.
• The optical pumping process requires three or more energy levels.
• The top energy level to which the electron is elevated initially must
quickly lie energetically above the desired final emission level..
• After reaching its initial excited state, the electron must release some of
its energy and drop to a slightly lower energy level.
• A signal photon can then trigger the excited electron sitting in this
new low level into stimulated emission, whereby the electron releases
the remaining energy in the form of a new photon with a wavelength
identical to that of the signal photon.
• Since the pump photon must have a higher energy than the signal
photon, the pump wavelength is shorter than the signal wavelength.
• To analyse the working of an EDFA, We take the energy level structure
of erbium.
• The erbium atoms in silica are Er3+ ions ,which are erbium atoms that
has lost three of their outer electrons .
• In describing the transition of the outer electrons in these ions to
higher energy states ,it is common to refer to the process as “raising
the ions to higher energy levels”.
• Figure shows the simplified energy level diagram and various energy
level transition process of these Er3+ ions .
• The two principal levels for telecommunication application are a
metastable level(the so called level) and the pump level.
• The term metastable means that the lifetimes for transition from this
state to ground state are very long compared with the lifetimes of the
states that led to this level.
• The metastable, the pump, and the ground state levels are actually
bands of closely spaced energy levels that form a manifold due to the
effect known as stark splitting.

• To understand the various energy transition and photon emission

ranges ,consider the following conditions.
1. The pump band shown in the top left of fig .exists at a 1.27eV
separation from the bottom of the ground state. This energy
corresponds to the 980nm wavelength.
2 The top of the metastable band (level D) is separated from
the bottom of the ground state band(level A) by 0.841 eV .
This energy corresponds to the 1480 nm wavelength.
3. The bottom of the metastable band (level C) is separated
from the bottom of the ground state band(level A) by 0.841
eV .This energy corresponds to the 1530 nm wavelength.
4. The bottom of the metastable band (level C) is separated
from the top of the ground state band(level B) by 0.775
eV .This energy corresponds to the 1600 nm wavelength.
• This means that the possible wavelengths are 980nm to 1480nm.
• The photon emitted during transition of electrons between possible
energy levels in the metastable state and ground state bands can
range from 1530 to 1600 nm.
• In normal operation a pump laser emitting 980nm photons is used to
excite ions from the ground state to the pump level, as shown by
transition process 1 in fig .
• These excited ions decay(relax)very quickly from the pump band to
the metastable band, shown as transition process 2 in fig
Fig. 11-4: Erbium energy-level diagram
• During this decay, the excess energy is released as phonons or
,equivalently ,mechanical vibrations in the fiber.
• Within the metastable band, the electrons of the excited ions tend to
populate the lower end of the band.
• Another possible wavelength is 1480nm.
• The energy of these pump photons is very similar to the signal photon
energy, but slightly higher.
• The absorption of a 1480nm pump photon excites an electron from the
ground state directly to the lightly doped top of the metastable level as
indicated by transition process 3 in fig .
• These electron then tend to move down to the more populated lower
end of the metastable level(transition 4).
• Some of the ions sitting at the metastable level can decay back to the
ground state in the absence of an externally stimulating photon flux ,
transition process 5 in fig .
• This phenomenon is known as spontaneous emission and adds to the
amplifier noise.
• Two more types of transition occurs when a flux of signal photons
that have energies corresponding to the bandgap energy between the
ground state and the metastable level passes through the device.
• First, a small portion of the external photons will be absorbed by ions in the
ground state, which raises these ions to the metastable level, as shown by
transmission process 6.
• Second, in the stimulated emission process (transition process 7) a signal
photon triggers an excited ions to drop to the ground state, thereby emitting
a new photon of the same energy ,wavevector, and polarization as the
incoming signal photon.
• The widths of metastable and ground state levels allow high levels of
stimulated emissions to occur in the 1530-to-1560 nm range.
• The absorption and emission response of an EDFA depends on the
composition of the host glass and on the types of dopants, such as Ge,Al ,in
the glass.
2.EDFA architecture
• An optical amplifier consists of a doped fibre ,one or more pump lasers, a
passive wavelength coupler ,optical isolators ,and tap couplers as shown in fig .
• The dichoric (2-wavelength) coupler handles either 980/1550 nm or
1480/1550 nm wavelength combinations to couple both the pump and signal
optical powers efficiently into the fiber amplifier.
• The tap couplers are wavelength insensitive
• They are generally used on both sides of the amplifier to compare the
incoming signal with the amplified output.
• The optical isolators prevent the amplified signal from reflecting back into the
device, where it could increase the amplifier noise and decrease the amplifier
efficiently.
EDFA configurations
• The pump light is usually injected from the same direction as the
signal flow-codirectional pumping.
• It is also possible to inject the pump power in the opposite direction
to the signal flow-counterdirectional pumping.
• counter directional pumping allows higher gains, but codirectional
pumping gives better noise performance.
• In addition, pumping at 980nm is preferred ,since it produce less
noise and achieves larger population inversion than pumping at 1480
nm
3.EDFA POWER CONVERSION EFFICIENCY AND GAIN
• As is the case with any amplifier, as the magnitude of the output
signal from an EDFA increases, the amplifier gain eventually starts to
saturate.
• The reduction of gain in an EDFA occurs when the population
inversion is reduced significantly by a large signal, thereby yielding the
typical gain Vs power performance curve is shown in fig .11.3
Fig. 11-3: Amplifier gain versus power
• The
  input and output powers of an EDFA can be expressed in terms of
the principle of energy conservation.
• (1)
• -input pump power, and s -pump and signal wavelengths
• The fundamental principle is that the amount of signal energy that can
be extracted from an EDFA cannot exceed the pump energy that is
stored in the device.
• The inequality in eqn (1) reflects the possibility of effects such as pump
photons being lost due to various causes (such as interactions with
impurities) or pump energy lost due to spontaneous emission.
•  From eqn. 1,,maximum output signal power depends on the ratio .
• For the pumping scheme to work, < s ,and,to have an appropriate
gain,
• Thus, the power conversion efficiency (PCE),defined as
• PCE = 1 (2)
is less than unity.
The maximum theoretical value of PCE is .
•  For absolute reference purpose, we use quantum conversion
efficiency (QCE), which is wavelength independent and is defined by
• QCE = PCE (3)
• The maximum value of QCE is unity, in which case all pump photons
are converted to signal photons.
• We can also rewrite eqn .1 in terms of the amplifier gain G.assuming
there is no spontaneous emission ,then
• G= (4)
•  This is a important relationship between signal input power and gain.
• When the input signal power is very large, then the maximum
amplifier gain is unity.
• This means that the device is transparent to the signal.
• From eqn .4,to achieve a specific maximum gain, the input signal
power cannot exceed a value given by
• (5)
• In addition to pump power,the gain also depends on the fiber length.
• The maximum gain in a three level laser medium of length L ,such an
EDFA is given by
• Gmax= exp(ρσeL) (6)
• σe -signal emission cross section
ρ -rare earth element concentration
when determining maximum gain,eqn .4 and 6 must be considered
together.
•  The maximum possible EDFA gain is given by the lowest of the two
gain expressions.
• G (7)
• Since G= = the maximum possible EDFA output power is given by the
lowest of the output power expressions.
•} (8)
Fig. 11-6: Gain versus EDFA length• Figure shows the onsets of gain saturation
for various doped fiber lengths as the
pumping power increases.
• As the fiber length increases for low
pumping powers, the gain starts to decrease
after a certain length because the pump
does not have enough energy to create a
complete population inversion in the
downstream portion of the amplifier.
• In this case, the unpumped region of the
fiber absorbs the signal, thus resulting signal
loss rather than gain in that section.
 • Since an electron state in the metastable level
Fig. 11-7: Gain versus pump level in an EDFA has relatively long lifetime, it is
possible to obtain very high saturated output
powers.
• The saturated output power (the power at
which gain saturation occurs) is defined as
the 3-dB compression point of the small
signal gain.
• For large signal operation, the saturated gain
increases linearly with pump power, as can be
inferred from fig.11.7
• This figure shows that as the input power
increases for a given pump level ,the
amplifier gain remains constant until
saturation occurs.
• If we are considering more detailed power-gain Vs-amplifier length
calculations for 980nm and 1480nm pumping, several points can be
noted from the curves.
1. The amplifier length that yields a maximum gain becomes longer with
increasing signal wavelengths
2. if a specific amplifier length is chosen ,then the EDFA will amplify each
wavelength differently. This leads to gain skew among different
wavelengths.
3. The 980nm pumping yields a complete population inversion( maximum
gain)at shorter amplifier length than 1480nm pumping. This leads to a
lower amplifier noise figures when using 980nm pumping.
RAMAN AMPLIFIERS
• A Raman optical amplifier is based on a nonlinear effect called stimulated
Raman scattering (SRS)which occurs in fibres at high frequencies.
• The SRS effect is due to an interaction between an optical energy field
and the vibrational modes of the lattice structure in a material.
• An atom first absorbs a photon at a particular energy and then releases
another photon at a lower energy ,that is, at a longer wavelength than
that of absorbed photon.
• The energy difference between the absorbed and released photon is
transformed in to a phonon, which is a vibrational mode of the material.
• The power transfer to higher wavelength occur over a broad spectral
range of 80 to 100 nm.
• The shift to a particular longer wavelength is referred to as the stokes
shift for that wavelength.
• Figure shows the Raman gain spectrum for a pump laser operating at
1445 nm and illustrates the SRS-induced power transfer to a signal at
1535 nm ,which is 90 nm away from the pump wavelength .
• The gain coefficient is given in terms of the Raman gain coefficient gR
units of 10-14 m/W.
• Whereas an EDFA requires a specially constructed optical fiber for its
operation, a Raman amplifier make use of the standard transmission
fiber itself as the amplification medium.
• The Raman gain mechanism can be achieved through either a
lumped (or discrete) amplifier or distributed amplifier.
• In the lumped Raman amplifier configuration ,a spool of about 80m
small –core fiber along with appropriate pump laser is inserted into
the transmission path as a distinct packaged unit.
• For the distributed Raman amplifier application, optical power from
one or more Raman pump lasers is inserted in to the end of the
transmission fiber toward the transmitting end.
• This process converts the final 20 to 40 km of the transmission fiber
into a preamplifier.
• Hence the word distributed is used, since the gain is spread out over a
wide distance.
• Fig.11.19 shows this effect on a single wavelength for several different
pump levels.
• The term on-off Raman gain is defined as the increase in signal power at
the receiver amplifier output when the Raman pump lasers are turned
on.
• As the optical power from the pump travels upstream (from the receiver
towards the sender ),the SRS effect progressively transfers power from
shorter pump wavelengths to longer signal wavelengths.
• This occurs over the characteristic Raman gain length LG=gRP/Aeff ,where P
is the pump laser power and Aeff –effective area of the transmission fiber
which is approximately equal to the actual fiber cross sectional area.
• In general ,the noise factors limit the practical gain of a distributed
Raman amplifier to less than 20 dB.
• In practice, link design engineers use several pump lasers to generate
a flat wideband gain spectrum.
• Fig.11.20 shows a typical Raman gain spectrum for six pump lasers at
different wavelengths.
• As indicated in figure, when using several pump lasers it is important
to remember that there is a strong Raman interaction between the
pumps themselves.
• Since pumps at short wavelengths amplify the power
of longer wavelength pumps, the shorter
wavelengths pumps typically need to have higher
power levels.
• As a result of the asymmetry of the Raman gain
efficiency spectrum, the power level and wavelength
of the longest wavelength pump need to be
calculated carefully.
• For example, In the case of shown in fig.11.20,
although only 10 percent of the total pump power
comes from the laser operating at 1495 nm, since it is
amplified by the other pumps it contributes 80% of
the gain for the longest signal wavelength.
• The top right curve of figure illustrates that by using
an appropriate combination of pump wavelengths
and pump power ,it is possible to achieve a fairly flat
gain over a wide spectral range.
• Pump lasers with high output powers in the 1400 nm-to-1500 nm are
required for Raman amplification of C- and L–band signals.
• Lasers that provide fiber launch powers of up to 300mW are available
in standard 14-pin butterfly packages.
• Figure 11.21 shows the setup for a typical Raman amplification system.
• Here a pump combiner multiplexes the outputs from 4 pump lasers
operating at different wavelengths onto a single fiber .
• These pump-power couplers are referred to popularly as 14XX-nm
pump-pump combiners.
• This combined pump power then is coupled into the transmission
fiber in a counter propagating direction through a broadband WDM
coupler.
• The difference in the power levels measured between the two
monitoring photodiodes shown in fig. gives the amplification gain.
• The gain flattening fiber is used to equalize the gain at different
wavelengths.
Semiconductor optical amplifiers
• A SOAs is essentially an InGaAsP laser that is operating below its
threshold point.
• Analogous to LDs,the gain peak of an SOA can be selected in any
narrow wavelength band extending from 1280 nm in the O-band to
1650 nm in the U-band by varying the composition of the active
InGaAsP material.
• Most SOAs belong to the travelling wave amplifier category.
• This means that in contrast to the laser feedback mechanism, where
the optical signal makes many passes through the lasing cavity in the
SOA, the optical signal travels through the device only once.
• During this single passage the signal gains energy and emerges
intensified at the other end of the fiber.
• The SOA construction is similar to a resonator cavity structure of a laser
diodes.
• The SOA has an active region of length L, width w, and height d.
• The end facets have reflectivities R1 and R2 and are dramatically lower in
order for the optical signal to pass through the amplification cavity only
once.
• These low reflectivity values are achieved by depositing thin layers of
silicon oxide,silicon nitride or titanium oxide on the SOA end facets.
•1.External
  pumping
• External current injection is the pumping method used to create
population inversion needed for having a gain mechanism in SOAs.
• This is similar to the operation of laser diodes.
• Thus the sum of the injection stimulated emission and spontaneous
recombination rates gives the rate equations that governs the carrier
density n(t) in the excited state.
• (1)
•  =J(t)/qd (2)
• Is the external pumping rate from the injection current density J(t) in
to an active layer of thickness d , - the combined time constant
coming from spontaneous emission and carrier recombination
mechanisms.
• (3)
• is the net stimulation rate . -group velocity of the incident
light. -optical confinement factor, a-gain constant , -threshold carrier
density, -photon density and g –overall gain per unit length
•  Given that the active area of the optical amplifier has a width w, and a
thickness d, then for an optical signal of power Ps with photons of
energy and group velocity ,the photon density is
•= (4)
• In the steady state, =0,so that eqn .1 becomes
• (5)
• Now substitute eqn.2 for and ,the second equality in eqn.3 for ,and
the first equality in eqn.3 solved for n into eqn.(5).
•  Solving for g then yields the steady state gain per unit length.
• g= = (6)

•= (7a) saturation photon density

• = () (7b) medium gain per unit length in the absence of
signal input (when the photon density is zero),which is called zero-
signal or small signal gain per unit length
2.Amplifier gain
•  Signal gain or Amplifier gain G,is defined as the
• G= (8)
• and -output and input power respectively, of the optical signal being
amplified.
• The single pass gain in the active medium of the SOA is
G= exp[(gm- )L] (9)
-optical gain confinement factor, gm-material gain coefficient,
•
 -effective absorption coefficient of the material in the optical path.
• L-amplifier length, -overall gain per unit length
• Eqn.9 shows that gain increases with device length.
• However, the internal gain is limited by gain saturation.
• This occurs because the carrier density in the gain region of the amplifier depends
on the optical input intensity.
• As the input signal level is increased, excited carriers(electron- hole pairs) are
depleted from the active region.
• When there is a sufficiently large optical input power ,further increases in input
the signal level no longer yield an appreciable change in the output level because
there are not enough excited carriers to provide an appropriate level of stimulated
emission.
• The carrier density at any point z in the amplifying cavity depends on the signal
level Ps(z) at that point
•  An expression for the gain G as a function of the input power can be
derived by examining the gain parameter g(z) in eqn.9
• This parameter depends on the carrier density and signal wavelength.
• Using eqn. 4 and 6,at a distance z from the input end,g(z)is given by
• g(z) = (10)
• -unsaturated medium gain per unit length in the absence of signal
input. -internal signal power at point z, amplifier saturated power- -
defined as the internal power level at which the gain per unit length
has been halved.
•  The gain given by eqn.9 decreases with increasing signal power.
• In particular ,the gain coefficient in eqn.10 is reduced by a factor of 2
when the internal signal power is equal to the amplifier saturation
power.
• Given that g(z) is the gain per unit length, in an incremental length
dz,the light power increases by
• dP= g(z) dz (11)
• Substituting eqn.10 into eqn.11 and rearranging terms gives
• g0(z) dz=() dP (12a)
•  Integrating the eqn. from z=0 to z=L yields
(12b)
• Defining the single pass gain in the absence of light to be =exp(L),and
using eqn .8
• ,we then have
G=1+)= exp(- ) (13)
 • Here zero signal gain is 30 dB
Fig. 11-3: Amplifier gain versus input power • As the input power level is
increased, the gain first stays
near the small signal level and
then starts to decrease.
• After decreasing linearly in the
gain saturation region, it finally
approaches an asymptotic value
of 0 dB(unity gain)for high input
powers.
• Output saturation power -point
at which the gain is reduced by
3dB.
• The wavelength span over which the gain decreases by 3 dB w.r.t the
maximum gain is called the gain BW or the 3-dB optical BW.
Amplifier Comparison