UNIT 4

Source to fiber Power

Launching

Coupling Efficiency
PF
power coupled into the fiber

power emitted from the sourse Ps

Ps
Source

PF
Optical Fiber

[5-1]

Output Patterns
Optical output of a luminescent source is usually
measured by its radiance B at a given diode
current.
Radiance: It is the optical power radiated into a
unit solid angle per unit emitting surface area and
is generally specified in terms of watts per square
centimeter per steradian.
The angle that, seen
from the center of a
sphere, includes a
given area on the
surface of that sphere.

Output Patterns
The angle that, seen from the center of a sphere,
includes a given area on the surface of that
sphere.
The value of the solid angle is numerically equal
to the size of that area divided by the square of
the radius of the sphere
Radiance = Power / per unit solid angle x per
unit emitting surface area

Radiance (Brightness) of the source

B= Optical power radiated from a unit area of the source into a

unit solid angle [watts/(square centimeter per stradian)]

Surface emitting LEDs output pattern:

B ( , ) B0 cos

This is lambertian pattern,

which
means
that
the
source is equally bright
when viewed from any
direction.

Power Coupled from source to the fiber

As and s : area and solid emission angle of the source

A f and f : area and solid acceptance angle of fiber

PF

B( A ,

Af

)d s dAs

2 0 max

rm

B( , ) sin dd

d s rdr

Total
coupled
power
is
summing
up
the
contribution
from
each
individual
emitting
pint
source
of

Power coupled from LED to the Fiber

2

rs

2B0
0

P
0

rs

B0

0 max

d s rdr
cos

sin

2
sin
0 max d s rdr

rs

B0

NA

d s rdr
2

PLED,step rs B0 ( NA) 2 rs B0 n1
2

Power coupling from LED to step-index fiber

Total optical power from LED: If we consider that the Ps is emitted from
the source of area As

Ps As

2 / 2

B( , ) sin dd

Ps rs 2B0
2

/2

2
cos

sin

rs B0

Ps ( NA) 2

PLED,step a

Ps ( NA) 2
rs

if rs a

if rs a

Power coupling from LED to graded-index fiber

Fiber separation

Power launching
During transmission, optical power
launched into a fiber is independent of
the wavelength of the source but
depends only on its brightness.
The number of modes that can
propagate in graded-index fiber of core
radius a and for parabolic profile is
Number of modes operating in 900nm
will be two times than no. of modes at
1300 nm.
The radiated power per mode

Equilibrium Numerical
Aperture
Manufactures supplied
light source with a short fiber

pigtail which is then attached to the source to a

system.
This pigtail should be connected to a system fiber with
identical NA and core diameter.
At this junction around 0.1 to 1 dB optical power is
lost.
An excess power loss will occurs in the system fiber in
addition to the coupling loss, which is due to the
modes scattering out of fiber.
Optical power is measured when the launched modes
have come to equilibrium and fiber is of larger length.
The launched modes attain equilibrium at approx. 50
mP50
from
the power
fiber measured
starting at
point.
: optical
50 m.
(distance
at which modes become equilibrium)

Optical Detector

Requirement for the optical detector

The detector must satisfy following requirements
for better performance.
High sensitivity at the operating wavelengths.
High fidelity.
Large electrical response to the received optical signal.
A minimum noise introduced by the detector.
Stability of performance characteristics.
Small size.
High reliability.
Low cost.

Optical detection principles

The basic detection process in an intrinsic absorber is illustrated

in Figure 8.1 which shows a pn photodiode.
This device is reverse biased and the electric field developed
across the pn junction sweeps mobile carriers (holes and
electrons) to their respective majority sides (p- and n-type
material).
A depletion region or layer is therefore created on either side of
the junction.
This barrier has the effect of stopping the majority carriers
crossing the junction in the opposite direction to the field.
However, the field accelerates minority carriers from both sides
to the opposite side of the junction, forming the reverse leakage
current of the diode.
Thus intrinsic conditions are created in the depletion region.

Optical detection principles

A photon incident in or near the depletion region of this

device which has an energy greater than or equal to the
band-gap energy Eg of the fabricating material (i.e. hf Eg)
will excite an electron from the valence band into the
conduction band.
This process leaves an empty hole in the valence band and is
known as the photo-generation of an electronhole (carrier)
pair, as shown in Figure 8.1(a).
Carrier pairs so generated near the junction are separated and
swept (drift) under the influence of the electric field to
produce a displacement by current in the external circuit in
excess of any reverse leakage current (Figure 8.1(b)).
Photo-generation and the separation of a carrier pair in the
depletion region of this reverse-biased pn junction is
illustrated in Figure 8.1 (c).

Optical detection principles

The depletion region must be sufficiently

thick to allow a large fraction of the incident
light to be absorbed in order to achieve
maximum carrier pair generation.
However, since long carrier drift times in the
depletion region restrict the speed of
operation of the photodiode it is necessary to
limit its width.
Thus there is a trade-off between the number
of photons absorbed (sensitivity) and the
speed of response.

Optical detection principles

Absorption
The absorption of photons in a photodiode to
produce carrier pairs and thus a photocurrent is
dependent on the absorption coefficient 0 of the
light in the semiconductor used to fabricate the
device.
At a specific wavelength and assuming only
band-gap transitions the photocurrent I p
produced by incident light of optical power P o is
given by

Optical detection principles

Quantum efficiency
The quantum efficiency is defined as the
fraction of incident photons which are absorbed
by the photo-detector and generate electrons
which are collected at the detector terminals:

where rp is the incident photon rate (photons per

second) and re is the corresponding electron rate
(electrons per second).

Optical detection principles

Responsivity
The expression for quantum efficiency does not involve
photon energy and therefore the responsivity R is
often of more use when characterizing the
performance of a photo-detector.
It is defined as:
where Ip is the output photocurrent in amperes and P o is
the incident optical power in watts (i.e. output optical
power from the fiber).
The responsivity is a useful parameter as it gives the
transfer characteristic of the detector (i.e.
photocurrent per unit incident optical power).

Physical Principles of Photodiodes

The most common semiconductor photo-detector
is the pin photodiode.

The device structure consists of p and n regions

separated by a very lightly n-doped intrinsic (i)
region.

Contd
In normal operation a sufficiently large reverse
bias voltage is applied, so that the i region is fully
depleted of carriers.
When an incident photon has an energy E g of
semiconductor material, the photon give up its
energy and excite an electron from valance band
to the conduction band.
This process generates the free electron-hole
pairs, known as photo-carriers.
As shown in figure on next slide.

Contd
The detectors are designed so that these carriers
are generated in the depletion region where most
of the light absorbed.
The high electric field present in the depletion
region causes photo-generated carriers to
Separate and be collected across the reverse
biased junction.
This give rise to a current Flow in an external
circuit, known as photocurrent.

Photocurrent
Optical power absorbed in a distance x, in the depletion region can
be written in terms of incident optical power P0 , P(x) :

P( x) P0 (1 e

s ( ) x

Absorption coefficient s ( ) strongly depends on wavelength.

The upper wavelength cutoff for any semiconductor can be
For longer wavelength, the
determined by its energy gap as follows:
photon energy is not
1.24
sufficient to excite an
c ( m )
electron from the valance
E g (eV)
band to conduction band.

Taking entrance face reflectivity into consideration, the absorbed

power in the width of depletion region, w, becomes:

P( w) P0 (1 e

s ( ) w

q
I p P0 (1 e s ( ) w )(1 R f )
h

Optical Absorption Coefficient

Responsivity

The primary photocurrent resulting from absorption is:

q
Ip
P0 (1 e s ( ) w )(1 R f )
h

Quantum Efficiency:

# of electron - hole photogenerated pairs

# of incident photons
IP / q

P0 / h

Responsivity:

IP
q

P0
h

[A/W]

Responsivity vs. wavelength

Avalanche Photodiode (APD)

APDs internally multiply the
primary photocurrent before it enters
to external circuitry.
In order to carrier multiplication
take place, the photo-generated
carriers must traverse along a high
field region.
In this region, photo-generated
electrons and holes gain enough
energy to ionize bound electrons in
VB upon colliding with them.
This multiplication is known as
impact ionization.
The newly created carriers in the
presence of high electric field result
in more ionization called avalanche
effect.

Optical radiation
Reach-Through APD structure (RAPD)
showing the electric fields in depletion
region and multiplication region.
Below
the
diode
breakdown
voltage a finite carriers are
created,
whereas
above
breakdown the number of carrier
used structure for
sCommonly
are infinite.
achieving the multiplication is
RAPD

 This configuration is known as

p+pn+ reach-through structure.
The layer is basically an
intrinsic material.
The term reach-through arises
from photodiode operation.
When a low RB voltage is
applied, most of the potential
drop is across pn+ junction.
The depletion layer is widen
with increasing the bias until a
certain voltage is reached at
which the peak electric field at
the pn+ junction is about the 510 % below that needed to
cause
the
avalanche
breakdown.
At this point, the depletion
layer just reach through to the
nearly intrinsic region.

It is in this high-field
region that carrier
multiplication takes
place.

In general, the RAPD is operated in the fully

depleted mode.
Light enters the device through the p + and is
absorbed in the material, which acts as a
collection region for photo-generated carriers.
Upon being absorbed the photon give up its
energy, and created the electron-hole pairs.
These carriers drift through the region in the
pn+ junction, where a high electric filed exists.

Responsivity of APD

The multiplication factor (current gain) M for all carriers generated in the
photodiode is defined as:

IM
M
Ip

Where I M is the average value of the total multiplied output current & I P
is the primary photocurrent.

The responsivity of APD can be calculated by considering the current gain

as: where R0 is the unity gain responsivity.

APD

M 0 M
h

Photo-detector Response Time

The response time of a photo-detector with its output circuit depends mainly
on the following three factors:
1.
2.
3.

The transit time of the photo-carriers in the depletion region.

Diffusion time of photo-carriers outside depletion region.
RC time constant of the circuit. :

The photodiode parameters responsible for above factors are the absorption
coefficient s, the depletion region width w, the photo diode junction and package
capacitances, the amplifier capacitance, the detector load resistance, the amplifier
input resistance, and the photodiode series resistance.

Photodetector Response Time

1. The transit time of the photocarriers in the depletion region. The response
speed of a photodiode limits by the time it takes photo-generated carriers to
travel across the depletion region.
The transit time t ddepends on the carrier drift velocity
and v
thed
w
depletion layer width w, and is given by:

td

2.

vd

Diffusion time of photo-carriers outside depletion region. The diffusion

processes are slow compared with the drift of carriers in the high-field
region.
Therefore, to have a high speed photodiode, the photo-carriers must be
generated in the depletion region or so close to it that diffusion times
are less or equal to the drift time.
This response time is described by the rise time and fall time of the
detector output when the detector is illuminated by a step input of optical
radiation. As shown in the figures on next two slides.
For fully depleted region the rise time and fall time are the same.

Photodiode response to optical pulse

Fully depleted region

Photodiode response to optical pulse

Typical response time of the photodiode that is not fully

depleted

Photodetector Response Time

3.

RC time constant of the circuit. The circuit after the photo-detector

acts like RC low pass filter with a pass-band given by:

1
B
2RT CT

RT Rs || RL and CT C a C d

Various optical responses of photo-detectors:

Trade-off between quantum efficiency & response time
To achieve a high quantum
efficiency, the depletion
layer width must be larger
1/ s
than
(the inverse of the absorption
coefficient),
so that most of the light
will be absorbed.
At the same time with large
width, the capacitance is
small
and
RC
time
constant getting smaller,
leading
to
faster
response, but wide width
results in larger transit
time in the depletion
region. Therefore there is a
trade-off between width
and QE. It is shown that the
best is:

1/ s w 2 / s

Structures for InGaAs APDs

Separate-absorption-and multiplication (SAM) APD

light
InP substrate
InP buffer layer
INGaAs Absorption layer

InP multiplication layer

Metal contact

InGaAs APD superlattice structure (The multiplication region is composed

of several layers of InAlGaAs quantum wells separated by InAlAs barrier
layers.

Temperature effect on avalanche gain

The gain mechanism of an
avalanche photodiode is very
temperature-sensitive because
of the temperature dependence
of the electron and hole
ionization rates.
This temperature dependence is
critical at high bias voltage
where small changes in
temperature
cause
large
changes in the gain.
As shown in the figure for si
avalanche photodiode.

Comparison of photodetectors

Pin
diode

Example
Bandgapandphotodetection
(a)Determinethemaximumvalueoftheenergygapwhichasemiconductor,usedasa
photoconductor,canhaveifitistobesensitivetoyellowlight(600nm).
(b)Aphotodetectorwhoseareais5102cm2isirradiatedwithyellowlightwhose
intensityis20mWcm2.Assumingthateachphotongeneratesoneelectronhole
pair,calculatethenumberofpairsgeneratedpersecond.
Solution

(a) Given,=600nm,weneedEph=h=Egsothat,
Eg=hc/ =(6.6261034Js)(3108ms1)/(600109m)=2.07eV
(b) Area=5102cm2andIlight=20103W/cm2.
Thereceivedpoweris
P=Area Ilight=(5102cm2)(20103W/cm2)=103W
Nph=numberofphotonsarrivingpersecond=P/Eph
=(103W)/(2.059602181019J/eV)
=2.9787photonss1 =2.9787EHPs1.

Example
BandgapandPhotodetection
(c)FromtheknownenergygapofthesemiconductorGaAs(Eg=1.42eV),calculatethe
primarywavelengthofphotonsemittedfromthiscrystalasaresultofelectronhole
recombination.Isthiswavelengthinthevisible?
(d)WillasiliconphotodetectorbesensitivetotheradiationfromaGaAslaser?Why?
Solution

(c)ForGaAs,Eg=1.42eVandthecorrespondingwavelengthis
=hc/Eg=(6.6261034Js)(3108ms1)(1.42eV61019J/eV)
=873nm(invisibleIR)
ThewavelengthofemittedradiationduetoEHPrecombinationis873nm.

(d) ForSi,Eg=1.1eVandthecorrespondingcutoffwavelengthis,
g=hc/Eg=(6.6261034Js)(3108ms1)(1.1eV61019J/eV)
=1120nm
Sincethe873nmwavelengthisshorterthanthecutoffwavelengthof1120nm,
theSiphotodetectorcandetectthe873nmradiation(Putdifferently,thephoton
energycorrespondingto873nm,1.42eV,islargerthantheEg,1.1eV,ofSiwhich
meanthattheSiphotodetectorcanindeeddetectthe873nmradiation)

Example
Absorption coefficient
(a)

If d is the thickness of a photodetector material, Io is the

intensity of the incoming radiation, the number of photons

absorbed per unit volume
sample
I 1 of
exp(
d ) is

n ph

d h

Solution

(a)IfI0istheintensityofincomingradiation(energyflowingperunitareaper
second),I0exp( d)isthetransmittedintensitythroughthespecimenwith
thicknessdandthusI0exp( d)istheabsorbedintensity

Example

InGaAspinPhotodiodes
ConsideracommercialInGaAspinphotodiodewhoseresponsivityisshowninfig.
Itsdarkcurrentis5nA.
(a) Whatopticalpoweratawavelengthof1.55mwouldgiveaphotocurrent
thatistwicethedarkcurrent?WhatistheQEofthephotodetectorat1.55
m?
(b) Whatwouldbethephotocurrentiftheincidentpowerinawasat1.3m?
WhatistheQEat1.3moperation?
Responsivity(A/W)
1
0.8
0.6
0.4

The responsivity of an InGaAs

pin photodiode

0.2
0

800 1000 1200 1400 1600 1800

Wavelength(nm)

Solution

(a) At =1.55106m,fromtheresponsivityvs.wavelengthcurvewe
haveR0.87A/W.Fromthedefinitionofresponsivity,

I ph
Photocurrent ( A)
R

Incident Optical Power (W ) P0

wehave

2 I dark 2 5 10 9 ( A)
P0

11.5 nW
R
R
0.87 A / W )
I ph

Fromthedefinitionsofquantumefficiency
andresponsivity,

e
e

h
hc

hcR (6.62 10 34 J sec)(3 108 m / s )(0.87 A / W )

0.70 (70 %)
19
6
e
(1.6 10 coul )(1.55 10 m )
Notethefollowingdimensionalidentities:A=Cs1andW=Js1sothatAW1=CJ1.
Thus,responsivityintermsofphotocurrentperunitincidentopticalpowerisalsocharge
collectedperunitincidentenergy.

Opticalreceiveroperation

Fundamental receiver operation

An optical Rx consists of photo-detector, an amplifier,
and signal processing circuitry.
It has the task of first converting the optical energy
into an electrical signal, and then amplifying this
signal to a certain level for further processing.
In these processes, various noises and distortions will
unavoidably be introduces, which can lead to errors in
the interpretation of the received signal.
The design of an optical Rx is much more complex
than that of an optical Tx because the Rx must first
detect weak, distorted signals and then make decisions
on what type of data was sent based on an amplified
version of this distorted signal.

Digital Signal Transmission

Fundamental receiver operation

The transmitted signal is two-level binary data
stream consisting of either a 0 or 1 in a time
slot of duration Tb.
This slot is known as bit period.
To transmit these bits we assume ASK and
Coding technique is NRZ.

Error Sources
Errors in the detection mechanism can arise from various
noises and disturbances associated with the signal
detection system.
The term noise can be defined as any unwanted
components of electrical signal that tend to disturb the
transmission and processing of the signal in a physical
system, and over which we have no control.
There are various sources of noise like internal, external
Here, our focus is only noise due to internal source like
shot noise and thermal noise.
Which is due the spontaneous fluctuations of current and
voltage in electric circuit.

Error Sources

Error Sources
The random arrival rate of signal photons
produces a shot noise at the photo-detector.
This noise depends on the signal level, when
using the avalanche photodiode in Rx, an
additional shot noise arises due to multiplication
process.
This noise is increases with increasing the
avalanche gain M.
Additional photo-detector noise come form the
dark and leakage current, which are independent
with photodiode illumination.

Error Sources
Thermal noises arising from the detector load
resistor and from the amplifier .
The analysis of the noises and resulting error
probabilities associated with the primary
photocurrent generation and the avalanche
multiplication are complicated, since neither of
these process is Gaussian.
The primary photocurrent generated by the
photodiode is a time-varying process resulting
from the random arrival of photons at the
detector.

Error Sources
If the detector is illuminated by an optical signal
p(t), then average number of electron-hole pairs N
generated in a time is

E
N

P (t )dt

h
h
0

The actual number of electron-hole pairs n that are

generated fluctuates from the average value
according to the Poisson distribution. P (n) N n e N
r

n!

It is not possible to predict exactly how many

electron-hole pairs are generated by a known optical
power incident on the detector is the origin of the
type of shot noise called quantum noise.

Error Sources
A further noise source is ISI, which results from
the pulse spreading in the fiber.

Receiver Configuration
A schematic diagram of Rx is shown below. The
3 basic stage of Rx are a photo detector, an
amplifier, and an equalizer.

Receiver Configuration
Photo-detector may be avalanche with a gain M or pin for
which M=1.
The photodiode has a quantum efficiency and a capacitance
cd.
The detector bias resistor R b which generates the thermal
noise current ib(t).
The amplifier has an i/p impedance which is a parallel
combination of Ra and Ca.
Voltage appearing across this impedance causes current to
flow in the amplifier output.
The amplifier is basically voltage-controlled current source.
The equalizer that follows the amplifier is a linear frequencyshaping filter to rectify the ISI effect.

Receiver Configuration
The binary digital pulse train incident on the
photo-detector

P(t )

b h

(t nTb )

The mean output current from the photodiode at

time t resulting from the pulse train Eq.
mentioned above.

q
i (t )
MP(t ) 0 M bn h p (t Tb )
h
n

Where, p(t) is o/p optical power; bn is the

amplitude, hp is the received pulse shape, and R 0
is the responsitivity.

Digital Receiver Performance

In a digital Rx the amplified and filtered signal
o/p of the equalizer is compared with the
threshold level once per time slot to determine
whether or not a pulse is present at the photodetector in the time slot.
Ideally, the o/p signal vout(t)would always exceed
the threshold voltage when 1 is present and would
be less than the threshold when no pulse was sent.
But in actual system, deviation from the average
value of
vout(t) caused by various noises,
interference from adjacent pulses, etc.

Probability of Error
One common way to calculate the error rate or bit
error rate (BER), divide the number Ne of errors
occurring over a certain time interval t by the
number Nt of pulses transmitted during this interval.
Ne Ne
BER

Where B=1/Tb
Nt

Bt

Typical
error
rates
for
optical
fiber
telecommunications system range from 10-9 to 10-12.
Meaning the one error for every billion pulses sent.
To compute the BER at Rx, we have to know the
prob. Distribution of signal at the equalizer o/p.

P1(v) is the prob. That the equalizer o/p voltage is less than v when a logical 1 pulse is sent.

P0(v) is the prob. That the equalizer o/p voltage is exceed v when a logical 0 pulse is sent.

Probability of Error
If the threshold voltage is vth then the error prob. Pe
is defined as
P aP (v ) bP (v )
e

th

th

Where a & b are determined by the a priori

distribution of the data. Where f0(y) is Gaussian

distribution
P0 (vth ) P ( y / 0)dy f 0 ( y )dy
P0 (vth )

vth

vth

vth

vth

P( y / 1)dy f ( z )dz
1

f (s)
P0 (vth )
P1 (vth )

Mean is b
and 2 is
variance

( s m ) 2 / 2 2

1
2 off
1
2 on

exp

vth

2 off

dv

(bon v) 2
exp 2 on 2 dv
vth

(v boff ) 2

1
BER Pe (Q)

x2

dx

Q/ 2

Q 2 / 2

1
Q
1 e
1 erf ( )
2
2
2 Q
vth boff bon vth
Q

off
on
2
erf ( x)

e
0

y2

dy

The Quantum Limit

When all system parameters are ideal and the
performance is limited only by the photo-detection
statistics.
In other words, suppose we have an ideal photodiode
having =1 and which produces no dark current; i.e. no
electron-hole pairs are generated in the absence an
optical pulse.
Given this condition, it is possible to find the minimum
received optical power required for a specific BER
performance in a digital system.
This minimum received optical power level is known
as the Quantum limit.

Analog Rx
Although, digital transmission through optical link
having wide usage, there are many applications for
analog links as well.
These range from individual 4 KHz voice channel
to microwave links operating in GHz region.
For analog Rx, the performance is measured in
terms of a signal-to-noise ratio.
The simplest analog link use AM for signal
transmission.
Shown in the next slide.

Analog Rx
Transmitted optical power p(t) and modulation
p(t ) Pt 1 ms(t )
index are in the form
m

I
IB

Pt is the average transmitted optical power, s(t) is

analog modulation signal, I is the variation current
about the bias point, and IB is the bias current.
In order to minimize the distortion, it is desire to
confine the modulation process in the linear region.
At the Rx end, the photocurrent generated by
analog optical signal is is (t ) 0 MPr 1 ms (t )
I p M 1 ms (t )

Analog Rx
Where Ip=R0Pr is the primary photocurrent.
The mean square signal current at the photo detector
o/p is
1
1
2
2
i 0 MmPr MmI p
2
2
2
s

The mean square noise current is

iN2 2q ( I p I D ) M 2 F ( M ) B 2qI L B

4k BTB
Ft
Req

ID is the primary dark current, IL surface leakage

current, F(M) excess photodiode noise factor, B noise
BW, Req equivalent resistance of photo-detector load
and amplifier, and Ft noise figure of baseband amplifier.

Analog Rx

With the assumption, the negligible leakage

current, the S/N
1
2

0 MmPr
2
S i
For large
optical signal
2

S m Ip

N 4qB
m 2 0 Pr

4qB

When the
optical incident
power on the
photodiode is
small, then
noise term
dominant

s
2
N

4k BTB
2q ( 0 Pr I D ) M F ( M ) B
Ft
Req
2

1
2

I p Mm
2

4k BTB
2q ( I p I D ) M F ( M ) B
Ft
Req
2

1
1 2 2 2
2

I p m
m 0 Pr
2
2
; forM 1( pin)
4k BTB
4k BTB
Ft
Ft
Req
Req