FIBRE OPTIC

COMMUNICATION
Syllabus
• Unit 1: Introduction to vector nature of light, propagation of light,
propagation of light in a cylindrical dielectric rod, Ray model, wave model.
• Unit 2: Different types of optical fibers, Modal analysis of a step index fiber,
Signal degradation on optical fiber due to dispersion and attenuation,
Fabrication of fibers and measurement techniques like OTDR.
• Unit 3: Optical sources- LEDs and Lasers, Photo-detectors, pin-diodes,
APDs, detector responsivity, noise, optical receivers, Optical link design,
BER calculation, quantum limit, power penalties.
• Unit 4: Optical switches-coupled mode analysis of directional couplers,
electro-optic switches.
• Unit 5: Optical amplifiers-EDFA, Raman amplifier.
• Unit 6: WDM and DWDM systems, Principles of WDM networks.
• Unit 7: Nonlinear effects in fiber optic links, Concept of self-phase
modulation, group velocity dispersion and solition based communication.
Text/Reference Books
• 1. J. Keiser, Fibre Optic communication, McGraw-Hill, 5th Ed. 2013
(Indian Edition).
• 2. T. Tamir, Integrated optics, (Topics in Applied Physics Vol.7),
Springer-Verlag, 1975.
• 3. J. Gowar, Optical communication systems, Prentice Hall India, 1987.
4. S.E. Miller and A.G. Chynoweth, eds., Optical fibres
telecommunications, Academic Press, 1979.
• 5. G. Agrawal, Nonlinear fibre optics, Academic Press, 2nd Ed. 1994.
6. G. Agrawal, Fiber optic Communication Systems, John Wiley and
sons, New York, 1997.
• 7. F.C. Allard, Fiber Optics Handbook for engineers and scientists,
McGraw Hill, New York (1990).
Course Objectives

• Introduce the concept of optical fiber communication, different

models of optics.
• Introduce the concept of optical fiber characteristics (dispersion and
attenuation) and understand fabrication of fibers and measurement
techniques.
• Introduce the concept of optical sources (LED and LASER), photo
detector, Optical switches and Optical amplifiers.
• Introduce the basic concept of WDM, DWDM system and nonlinear
effect on optical fibers.
Course Outcomes

• On successful completion of this course, the students should be able

to:
• Understand the principles fiber-optic communication, the
components and the bandwidth advantages.
• Understand the properties of the optical fibers and optical
components.
• Understand operation of lasers, LEDs, and detectors.
• Analyze system performance of optical communication systems.
• Design optical networks and understand non-linear effects in optical
fibers
INTRODUCTION
Fiber-Optic Communication is the most modern and advanced
mode of data communication which has very recent roots
dating back to not more than 40 years ago. Communication
Scientists all over the world were in an incessant search of a
wideband and low-loss medium of data communication which
could be used at high data rates with the least amount of lost
possible. This constant search, for such a medium, led to
development of optical fiber communication. Let us have a
quick glimpse into history of communication.
HISTORY
• The first revolution in the field of communication came when
Sir Alexander Graham Bell successfully converted voice
signals into electrical signals which were transmitted on
electrical wires and then converted back to voice signals.
This was the major break-through in the field of
communication. Right from this time there has been a
continuously increasing need of bandwidth for
communication due to continuously increasing number of
users. More people wanted to communicate and thus large
bandwidths were required thereby forcing communication
scientists to look for new possibilities. This increasing trend,
of need of large bandwidths, even continues today.
ELECTROMAGNETIC SPECTRUM
Initial communications started at lower operating frequencies of about
30MHz. The bandwidths then required were also low. Since then the
operating frequencies have drastically increased due to large
requirements in bandwidths. Let us take a look into the electromagnetic
spectrum to get an idea of our discussion.

Fig.1: Electromagnetic Spectrum

• The medium of transmission that were used for operating frequencies upto about
1GHz were coaxial cables in which there was a centre conductor surrounded by a
layer of dielectric material and the dielectric material was surrounded by outer
metallic layer. The electromagnetic energy travelled, along the lengths of these
cables and was confined in between the two metallic layers. These cables had a
loss figure of about 20db/km.
• When operating frequencies increased further the coaxial cables proved to be
inadequate and lossy, thereby giving rise to the need of another medium called
waveguides.
• Waveguides are basically hollow structures which guide the electromagnetic
energy from one point to another through them. But as the operating frequency
further increased to few hundreds of gigahertz these waveguides too proved to
be inadequate as there were no supporting electronic circuitry available that
could operate at such high frequencies.
• The reason behind this was that at such high frequencies, even the size of the
electronic component started to show some variations in the circuit behaviour
and the electronic components could no longer be treated as lumped elements.
Hence this led to a strong need of a search for other alternatives because though
there seemed to have appeared a halt in the available technology.
• Scientist all over the world started to explore new possibilities and
looked in the optical domain which was already being used in
laboratory experiments. The idea was that, if the already well-
established relationship between bandwidth (BW) and operating
frequency (f0) held good at optical frequencies then we would
emerge with a new option for communication that would increase
the existing bandwidth by 1000 to 10,000 times. Thorough
investigations showed that optical domain had the potentiality to be
used for communication. Two very obvious questions then come to
the mind that whether or not there are transmitters and receivers
available for this new communication technology and the second
question as to whether or not there exists such a wideband and loss-
less medium for carrying optical signals.
Communication

Fig.2 Basic block diagram of communication system

Fig.3 Basic Block diagram of Optical communication link

OPTICAL
FIBRE

Fig.4 Basic block diagram Digital optical of communication link

Fig.5 Basic model of an long haul Optical Communication Link
ADVANTAGES of OFC
• Long distance transmission due to low attenuation
• Small size and low weight
• Larger information capacity due to wider bandwidth

DISADVANTAGES of OFC
• OFC is delicate so has to be handlled carefully
• Communication is not totally in optical domain, so repeated
electrical-optical-electrical conversion is needed
APPLICATIONS
• Bandwidth: Optical Communication provides an ultra-high bandwidth
for communication of the order of Terahertz (THz)(). This advantage
meets the first requirement of a high quality reliable communication
system.
• Low Losses: The loss figure of optical communication is very low,
about 0.2 dB/Km. So this system has high SNR values. This advantage
provides a reliable communication system. No other medium today,
can provide such low loss figures as optical medium.
• Less Interference: Optical communication systems have very low or
even negligible electromagnetic interference (EMI).
• Security: Optical Communication provides high security data
transmission. This is because optical signal travels through optical
fibers which do not allow leakage of light energy. So tapping of
transmitted information is very difficult in optical communication.
• Low Cost: Optical communication systems have very low
manufacturing cost. Whatever cost is incurred is only due to the
technology. This is because optical medium glass is made from silica,
which is freely abundant in nature. So, the only cost is in moulding it
to a form of optical medium like optical fibre. The cost per voice
channel of an optical fibre is also very much smaller than cost per
voice channel of any other medium like coaxial cable though the two
may have comparable costs per kilometre. This is because the
bandwidth of optical fibre is almost 1000 times larger than a coaxial
cable. So, cost per channel of an optical fiber would be one
thousandth of that in a coaxial cable.
• Light Weight: Applications where space and weight are constraints,
optical fiber serves adequately because optical fibers have low weight
and low volume compared to other media.
CHARACTERISTICS OF LIGHT
• We know that light is a form of electromagnetic radiation. We are also
familiar with the propagation of light in the form of rays which help us
to prove different laws like laws of reflection, Snell’s law, interference,
diffraction, etc. We are well acquainted with phenomena like
photonic emission, photo-electricity which define a particle nature for
light in the form of photons. So basically, the nature of light depends
on the context of which we talk about. So, in this course we first treat
light in the context of the Ray Model and study different phenomena
based on the ray model of light. Proceeding further, when we find the
ray model is inadequate in explaining some phenomena, we depart
from the ray model and then adopt a higher model for light which is
the wave-model where light is treated as an electromagnetic
wave. And in those situations where we find even the wave-model
inadequate in explain certain phenomena like interaction light with
matter, we adopt the quantum model of light where light will be
treated as a photons. So, in this course, we will treat light in the
following three models:
• Ray Model
• Wave Model
• Quantum Model
CHARACTERIZATION OF A LIGHT SOURCE
A source of light can be characterized by the following factors:
• Intensity of the light: Intensity of light is defined as the power per unit
solid angle. So for a given power of the source, if the emitted light is
scattered into a very wide solid angle then the source has low
intensity. If the emitted light is confined to very narrow cone, the
source appears to be very bright because its intensity increases. This
happens in case of a LASER whose light appears to be much brighter
and travels long distance than a normal 60W bulb though the power
of the LASER is much smaller than 60W. The intensity of the source is
indicative of how focussed is the emitted light.
• Wavelength of Light (λ):
The second characteristic on which a source is characterized is the
wavelength of the emitted light. The wavelength of light is indicative of
the colour of the light and so many a times it is also called as the colour
of the source. The visible light lies within a wavelength range of 400nm
to 700nm. Glass which appears so transparent to us in daily life is not
actually that transparent to wavelengths of 400nm to 700nm. In fact, it
is much more transparent to lights of wavelengths 1300nm and
1550nm, which lie in the infrared region. Since these regions are not in
the visible range, colour does not have any meaning, yet we may retain
the colour as one of the characteristic to categorize light. Depending on
the desired loss performance of the optical communication system λ
can be chosen either 1300nm or 1550nm. So, the choice of wavelength
of transmission has a direct relation to the SNR of the transmission.
• Spectral Width of Source (∆λ): It is basically the wavelength range over
which the emission takes place. In other words, it is the range of
wavelengths emitted by the source. Thus the spectral width may be
considered to be indicative of the purity of the colour of the light source.
That is, if we have a source with a wide spectral width, say for example if it
emits all the wavelengths ranging from blue to red, we get a light from the
source which will look like white light. If we reduce the spectral width to
near red, we would get a sharp red colour light. If we reduce it to near
blue, we would get a blue coloured light and so on. Thus reducing the
spectral width increases the purity of the colour. Spectral width is a very
important parameter of a source because we would later discover that
spectral width of a source is related to the data rate upto which a source
can be used as a transmitter of optical signal. Smaller the value of ∆λ more
will be the purity of the source and also higher would be the data rate of
the source. In other words, higher will be the bandwidth of the
communication system. So, the choice of ∆λ has a direct relation to the
bandwidth of the transmission.
• Light energy can be modelled in three different forms which relate
the particular model of light to the context in which it is talked about.
Light can be characterized in any one of the following models
• Ray Model
• Wave Model
• Quantum Model
• In the simplest possible context, light is treated as a ray and the
different phenomena exhibited by light are explained in terms of the
ray-model of light. Some phenomena exhibited by light are not
adequately explained by the Ray-Model of light. In that case, we
resort to the more advanced nature of light such as the wave and the
quantum models. In this section we shall mainly deal around the ray
model of light and attempt to explain the propagation of light in an
optical fiber treating light as a ray.
 • Constructionally, an optical fiber is a solid cylindrical glass rod called
the core, through which light in the form of optical signals
propagates. This rod is surrounded by another coaxial cylindrical shell
made of glass of lower refractive index called the cladding.

Fig. 6: Constructional Details of an Optical Fiber

The diameter of the cladding is of the order of 125 µm and the diameter of the core is even smaller than that. Thus
it is a very fine and brittle glass rod that we are dealing with. In order to provide mechanical strength to this core-
cladding arrangement, other coaxial surrounding called the buffer coating and jacketing layers are provided. They
do not play any role in the propagation of light through the optical fiber, but are present solely for providing
mechanical strength and support to the fiber.
• The light energy in the form of optical signals propagates inside the
corecladding arrangement and throughout the length of the fiber by a
phenomenon called the Total Internal Reflection (TIR) of light. This
phenomenon occurs only when the refractive index of core is greater
than the refractive index of cladding and so the cladding is made from
glass of lower refractive index. By multiple total internal reflections at
the core-cladding interface the light propagates throughout the fiber
over very long distances with low attenuation. We shall now discuss
the essential requirements of the propagation of light through an
optical fiber, over long distances with minimum loss, in detail.
• Figure 7 shows a section of the core of an optical fibre. If a ray of light
is incident on the core of an optical fibre from the side, the ray of light
simply refracts out from the fibre on the other side. The ray shown in
figure 7(in green) demonstrates the situation.
 Figure 7: Launching of light into an optical fiber
No matter what the angle of incidence of the light is, any light that enters the fiber from
the side does not propagate along the fiber. The only option thus available with us is to
launch the light through the tip of the fiber. That is, in order to guide light along the fiber,
the light must be incident from the tip of the optical fiber. The red ray of light in figure .7
explains this situation. In other words, if the tip of the optical fiber is not exposed to light,
no light will enter the fiber. Although there may be ambient light, as long as the tip is
protected, no light from the sides propagates along the fiber. Equivalently, if there was
propagation of light through the fiber, no light would emerge from the sides of the fiber.
This characteristic of the optical fiber imparts the advantage of information security to
the Optical Fiber Communication Technology.
• At this juncture, one basic question that may come to the reader’s
mind is that whether a partial reflection at the core-cladding interface
suffices the propagation of light along the fiber over long distances?
The answer to this question is very clearly a no. The reason is that, at
each reflection a part of the optical energy launched into the optical
fiber would be lost and after a certain distance along the length of the
fiber the optical power would be negligibly low to be of any use. Thus
total internal reflection is an absolute necessity at each reflection for
a sustained propagation of optical energy over long distance along
the optical fiber. This precisely is the sole reason of launching light
into the fiber at particular angles so that light energy propagates
along the fiber by multiple total internal reflections at the core-
cladding interface.
• We have already stated that for explaining propagation of light in an
optical fiber, the Ray-Model of light shall be used. The Ray-Model of
light obeys the Snell’s laws. Following figure depicts a situation of a
typical refraction phenomenon taking place at the interface of two
optically different media having refractive indices n1 and n2:

Figure 8: Refraction of light at a media interface

• The angles measured in the expression for Snell’s law are measured with respect to the
normal to the media interface at the point of incidence. If n2 > n1 , then the angle of
refraction is greater than the angle of incidence and the refracted ray is said to have
moved away from the normal. If the angle of incidence (θ1) is increased further, the angle
of refraction (θ2) also increases in accordance with the Snell’s law and at a particular
angle of incidence the angle of refraction becomes 90o and the refracted ray grazes along
the media interface. This angle of incidence is called the critical angle of incidence (θc) of
medium 2 with respect to medium 1. One should note here that critical angle is media-
relative. That means, the same optically denser medium may have different critical angles
with respect to different optically rarer media. If θ1 is increased beyond the critical angle,
there exists no refracted ray and the incident light ray is then reflected back into the
same medium. This phenomenon is called the total internal reflection of light. The word
‘total’ signifies that the entire light energy that was incident on the media interface is
reflected back into the same medium. Total Internal Reflection (TIR) obeys the laws of
reflection of light. This phenomenon shows that light energy can be made to remain
confined in the same medium when the angle of incidence is greater than the angle of
reflection. Thus we can see that there are two basic requirements for a TIR to occur:
• 1. The medium from which light is incident, must be optically denser than the medium to
which it is incident. In figure.8 n2 > n1.
• 2. The angle of incidence in the denser medium must be greater than the critical angle of
the denser medium with respect to the rarer medium.
• The above three characteristic treat light as sources of energy. When
we move on to discuss the propagation of light in an optical fibre, this
notion of light as an energy source is inadequate and we have to treat
light as an electromagnetic wave. Under this adoption, if the
dimensions of the medium of propagation are very large compared to
the wavelength of the light, light can be considered as a transverse
electromagnetic (TEM) wave. This means that the direction of electric
field, direction of magnetic field and the direction of propagation of
light are mutually perpendicular to one another according to the right
hand thumb rule as shown in figure 9 (a) & (b). Fig. 9(a): TEM nature
of Light Fig. 9(b): TEM nature of Light
Fig. 9(a): TEM nature of Light

Fig. 9 (b): TEM nature of Light

• The electric and magnetic fields of light are hence related to each other
through the medium parameter η which is called the intrinsic
impedance of the medium. That is,

permeability of medium is the measure of the resistance of a material against the

formation of a magnetic field.
permittivity of medium is the measure of resistance of a medium in an electric field.

So, if the electric field is known completely, the magnitude of magnetic field can be
determined using the above relation and its direction would be perpendicular to the
direction of electric field. Thus with the knowledge of electric field the magnetic
field can also be determined. The behaviour of the electric field as a function of
time is called the polarization of light.
RAY-MODEL OF LIGHT
• With the above background of information, let us now move on to
discuss the simplest possible model of light, the ray-model.
Fundamentally, we have two main aspects of the ray nature of light.

Fig. 10: Spherical and Plane Wave-Fronts

• Light rays actually are fictitious lines which in reality represent the
direction of propagation of what are called as phase-fronts of light as
shown in figure 10. Phasefronts are nothing but constant phase
surfaces in which the phase difference between any two points is
zero. In reality they represent the spatial nature of propagation of a
wave and hence are also called as wave-fronts. These phase-fronts
may be either spherical or planar in nature and accordingly, we get
two aspects of ray nature of light. A line drawn perpendicular to a
phase front at every point gives the direction of propagation of light
energy at that point and is conventionally called as a light-ray. So,
light-rays are actually imaginary lines that determine the direction of
propagation of light energy. In other words, what we actually have is
not light ray but the direction of propagation of phase-fronts of light
energy that are represented by directed line segments called light
rays.
• If we have an isotropic source of light situated at finite distance, as in
figure 10(a), we get spherical phase-fronts because the light source
would emit light in all directions and light waves would travel
spherically outwards from the source of light. If we now draw lines
perpendicular to these phase-fronts at any point, these lines would
give the direction of flow of light energy at that point and the line
drawn would then be called a light ray. Due to their spherical nature,
the light waves can be called as spherical waves.
• If the same source is now placed at a very large distance from the
point of observation (ideally at infinity) the wave-fronts would appear
almost planar and parallel to each other as shown in figure 10(b). So
the light rays too would now not look divergent, but appear parallel
to each other because they are perpendicular to the phase fronts at
every point. The light waves now can be called plane waves.
WAVE-MODEL OF LIGHT
• The “Wave-Model” of light introduces light as an electromagnetic wave. If light is treated as
an electromagnetic wave then light must be expressible in terms of the generalized wave
function, which is given as:

The wave function is a generalized function of space (x) and time (t). The term
(ωt-βx) is the phase function of ψ(x,t). Thus the phase of the wave is a function of
space and time. If we now freeze space, i.e. take x=constant or in other words
observe at a particular point, we see that there is a sinusoidal variation of the
wave amplitude as a function of time having an angular frequency of ω rad/s. If
we freeze time, i.e. take t=constant or in other words observe the whole wave
simultaneously, we see that the amplitude of the wave has a sinusoidal variation
with a phase constant β rad/m. So these two phenomena together constitute a
wave phenomenon represented by the generalized wave equation.
• The phase constant β is defined as the phase change per unit
distance. The wavelength (λ) of a wave is the distance between two
consecutive points on the wave which are in the same phase. The
phase difference between two points in the same phase is either zero
or an integral multiple of 2π. Thus the wavelength of a wave is
measured between two points that have a phase difference of 2π.
Hence the phase constant β can be calculated as: Β= 2π/ λ
• So, if we know the wavelength (λ) of a wave in a medium, we can
calculate the phase constant of the wave in that medium and vice–
versa.
• Once the frequency of light (f) is known, the wavelength (λ) can be
readily calculated from the relation: λ=c/f
• Where, c = velocity of light, then what is so special about β?
• At this point the velocity of light is a medium dependent parameter. It
is c (3x108m/s) in vacuum, but is different in different media. In the
same medium too, the velocity of light may vary if the medium is
made bound. Thus the velocity of light is a medium dependent
parameter. So, the wavelength of light is not always given by equation
λ=c/f but can be found from:
• λ medium = velocity of light in the medium/frequency of light
• And β medium= 2π/ λmedium
• Once the frequency of light is known, we may calculate the value of
β(medium) and then , we can calculate the value of λ(medium) and
then find the velocity of light from equation .
• Treating light as an electromagnetic wave, let us now define an
important optical parameter of a medium called the refractive index
of the medium.
• Refractive index of a medium is defined as the ratio of the velocity of light
in vacuum to the velocity of light in that medium. It is denoted by n. Since
refractive index is a ratio of two velocities, it is a pure number and has no
unit.
• For most media, n(medium) < 1, i.e. the velocity of light reduces from its
value in vacuum. In fact, light travels fastest in vacuum and in any other
medium it slows down. For example, the refractive index of material glass
is about 1.5, i.e. light travels 1.5 times faster in vacuum than in glass.
Similarly the refractive index of water is 1.33. In other words, refractive
index of a medium indicates the factor by which the speed of light reduces
in the medium. As we study further, we would encounter another type of
refractive index, called the effective refractive index which is denoted by
neff and is given by,
neff= c/v’
• Where v’ is not the necessarily the velocity of light in an unbound medium
but is a generalized velocity of light in a bound medium like optical fibres,
optical waveguides, etc. and it is structure or size dependent velocity.
 For example, the actual refractive index of glass in unbound form is
1.5. But if glass is made in the form of a bound medium (say an optical
fiber) its refractive index changes because the velocity of light in it
changes. The refractive index of this particular bound form or structure
of glass is then the effective refractive index of this structure of glass
and would change again if we alter this form.
We have already seen that in general, whenever we discuss issues
related to light, conventionally, we discuss them in terms of the
wavelength of light rather than frequency. This was precisely the
reasons for defining the three windows of optical communication in
terms of wavelengths and not in terms of frequencies. We have also
come across the quantity called the spectral width of the source of light
which is the range of wavelengths that are emitted by a source of light.
In general, wavelength of light is given by the relation. That is,
• λ =v/f
Where λ = wavelength of light in a medium
v = velocity of light in the medium.
f = frequency of the light under study
The spectral width can be calculated from as

The quantity (v/f) is the wavelength and so, the above relation can be rewritten as

Let us now take a small example. Let us assume an operating wavelength (λ) of about 1000
nm and an optical window of operation (∆λ) of about 1000 nm, then from the above
relation , we find the value of bandwidth (∆f) of our assumed optical communication
system to be about 3x1013 Hz. Though this is a very crude and small example, but yet it
gives an idea about the enormous bandwidth that optical communication can provide. Thus
an optical communication system is really a very wideband system compared to a
conventional microwave communication system.
• LAUNCHING OF LIGHT INTO AN OPTICAL FIBER
• Light propagates inside an optical fiber by virtue of multiple TIRs at the
corecladding interface. The refractive index of the core glass is greater than
that of the cladding. This meets the first condition for a TIR. All the light
energy that is launched into the optical fiber through its tip does not get
guided along the fiber. Only those light rays propagate through the fiber
which are launched into the fiber at such an angle that the refracted ray
inside the core of the optical fiber is incident on the corecladding interface
at an angle greater than the critical angle of the core with respect to the
cladding. But before delving into rigorous mathematical calculations, let us
first visualise how light energy can be launched into a fiber. Figure 11
shows one of the possibilities of launching light into an optical fiber where
the light ray lies in a plane containing the axis of the optical fiber. Such
planes which contain the fiber axis are called meridional-planes and
consequently the rays lying in a meridional-plane are called meridional-
rays. Meridional rays always remain in the respective meridional plane.
 Figure 11: Launching of Meridional Rays

There may be infinite number of planes that pass through the axis of the fiber and consequently there are an
infinite number of meridional planes. This indirectly indicates that there are an infinite number of meridional rays
too, which are incident on the tip of the fiber making an angle with the fiber-axis as shown in the above figure.
These meridional rays which get totally internally reflected at the corecladding boundary meet again at the axis of
the optical fiber as shown in the figure 12 below. In the figure the meridional plane is the plane of the paper which
passes through the axis of the fiber and the incident rays, refracted rays and the reflected rays lie on the plane of
the paper. Though only two rays are shown in the figure for the sake of clarity, in practice there would be a bunch
of rays that would be convergent at the same point.
• Meridional rays are classified into bound and unbound rays. The rays
that undergo TIR inside the fiber core remain inside the core at all
times along the propagation and are called as bound rays. The rays
that fail to undergo TIR inside the core are lost into the cladding and
are called unbound rays. The dotted ray shown in figure 11 is an
unbound meridional ray.

Figure 12: Meridional Rays meeting at the axis.

• Since all the reflected rays meet at the same point a region of high
optical intensity is generated at that point (point A in figure 12). Since
these rays undergo multiple TIR at the core-cladding boundary, they
meet repeatedly at the axis at regular intervals along the fiber. This
causes multiple regions of maximum intensity along the axis of the
fiber. Also, different sets incident rays would have different angles of
reflection at the core-cladding boundary and consequently have
different points of maximum intensities along the fiber axis. Thus it
can be visualized easily that at the output end of the optical fiber,
maximum intensity will be in the axial region of the fiber core and the
intensity would gradually decrease as we move away from the axis
towards the periphery of the core.
• Another way of launching a light ray into an optical fiber is to launch
it in such a way that it does not lie in any meridional plane. These rays
are called skew rays. A pictorial representation of launching a skew
ray is shown in the figure 13 below.
 Figure 13: Launching of Skew Rays
Skew rays propagate without passing through the central axis of the fiber. In fact the skew
rays go on spiralling around the axis of the optical fiber. The light energy carried by them
is effectively confined to an annular region around the axis as shown in figure 13.
Consequently, at the output, skew rays will have minimum energy at the axis of the optical
fiber and it will gradually increase towards the periphery of the core.
Thus when light energy is launched into an optical fiber, there arises two possible energy
distributions; one, which has maximum intensity at the axis due to meridional rays and
the other, which has minimum intensity at the axis due to the skew rays. Thus, on the
whole, there are two ways of launching light into an optical fiber; light can be launched
either as meridional or as skew rays.
• Assuming that light is launched as meridional rays into the optical fiber, let
us now carry out a simple analysis.
• For that let us concentrate on figure 14 below. The figure shows a cross-
section of an optical fiber with a core of refractive index n1 and a cladding
of refractive index n2. The incident ray AO (shown by dotted line) is
incident at an angle ϕ with the axis of the fibre.
• The refracted ray for AO in the core (dotted line ON1) fails to be incident on
the core-cladding interface at angle greater or equal to the critical angle of
the core w.r.t. cladding and hence refracts out of the core and is lost to the
cladding. In other words, the angle of incidence of a refracted ray at the
core-cladding interface in turn depends on the initial angle at which the
incoming ray was launched into the fiber.
• If this launching angle (with the fiber axis) is decreased, the angle of
incidence which the refracted ray makes at the corecladding interface
increases. If this increase is such, as to exceed the critical angle of the core-
cladding interface, then total internal reflection of the refracted ray takes
place and the light remains in the core and is guided along the fiber.
If this increase is such, as to exceed the critical angle of the core-cladding
interface, then total internal reflection of the refracted ray takes place
and the light remains in the core and is guided along the fiber. The ray CO
is launched into the fiber at such an angle ‘α’ that its refracted ray is
incident at the core-cladding boundary at its critical angle ‘θc’. If any light
ray is launched at anangle more than α then the refracted ray just
refracts out to the cladding because the angle of incidence of its
refracted ray at the core-cladding interface is less than the critical angle.
Thus the angle α is indicative of the maximum possible angle of
launching of a light ray that is accepted by the fiber. Consequently, the
angle α is called the angle of acceptance of the fiber core. Since the
optical fiber is symmetrical about its axis, it is very clear that all the
launched rays, which make an angle α with the axis
 Figure 14: Launching of Light into an Optical Fiber

considered together, form a sort of a cone. This cone is called the

acceptance cone of the fiber as shown in the above figure. Any launched ray
that lies within this cone is accepted by the fiber and the light of this ray is
guided along the fiber by virtue of multiple TIRs as shown by the red ray BO
in the figure 14.
NUMERICAL APERTURE OF OPTICAL FIBER
• With the same initial assumption of meridional launching of light into an
optical fiber, let us consider the figure 15 below.
• The figure 15 shows a cross-section of a core of refractive index n1 and a
cladding of refractive index n2 that surrounds the core glass. An incident
ray AO is incident from medium1 at the tip of the fiber making an angle α
with the axis of the fiber, which is the acceptance angle of the fibre.
• The refracted ray for this incident ray in the core then is incident at the
core-cladding interface at the critical angle θc of the core with respect to
the cladding.
• The angle of refraction for critical angle of incidence is 90° and the
refracted ray thus grazes along the core-cladding boundary along BC as
shown in the figure 14.
• According to Snell’s laws, the incident and the refracted rays lie in the
same meridional plane, which is the plane of the paper in this case.
• Applying Snell’s law at the medium1-core interface we get

………..(i)
From the figure it is clear that,

and so substituting this in above

equation(i), we get: ….(ii)

From the basic trigonometric ratios,

Figure 15
……(iii)

Applying Snell’s law at the core-cladding interface we get: (α=𝜃𝑐 ; 𝜃= 90° )

……………….(iv)

Substituting equation (iv) in equation (ii) we get:

• Since the initial medium 1 from which the light is launched is air most
of the times, n = 1. The angle α is indicative of light accepting
capability of the optical fiber.
• Greater the value of α, more is the light accepted by the optical fiber.
In other words, the optical fiber acts as some kind of aperture that
accepts only some amount of the total light energy incident on it. The
light accepting efficiency of this aperture is thus indicated by sin α
and hence this quantity is called as the numerical aperture (N.A.) of
the optical fiber. Thus for an optical fiber in air, with core refractive
index n1 and cladding refractive index n2 and having an acceptance
angle of α is given by
………..(v)
• Numerical Aperture is one of the most fundamental quantities of an optical fiber.
• It indicates the light collecting efficiency of an optical fiber.
• More the value of N.A. better is the fiber.
• For greater values of N.A. the difference on the right hand side of equation (v)
has to be maximized.
• For maximizing the difference, either the refractive index of the core (n1) has to
be increased or the refractive index of the cladding (n2) has to be reduced.
• Since the core used is always glass, the value of its refractive index n1 is thus fixed
(approximately 1.5). The only option thus available with us is to reduce the value
of n2. But it too has a limitation of the lowest value of 1 for air because till date
no material is known which has a refractive index lower than that. If we make n2
=1, we would then get the maximum possible N.A. for an optical fiber.
• But then we are basically talking about removing the cladding because, if there is
a cladding, the value of n2 will always be greater than 1.
• Thus one can clearly say that from the point of view of light accepting efficiency,
the presence of a cladding is undesirable.
• The above discussion suggests that although the optical fiber is made of
core and cladding, the presence of cladding is undesirable because it
reduces the light accepting efficiency of the optical fiber.
• However, the prime concern behind prolonged research on optical fibers
was not just to put light inside an optical fiber with the best efficiency but
also to propagate the light over long distances with the least attenuation.
• That means if we have a source of optical signal and an optical fiber with
the highest light accepting efficiency but is incapable of propagating the
accepted light; the optical fiber is of no use in spite of its high N.A.
• Thus judging the need of a cladding just on the basis of light launching
efficiency would be highly inappropriate. In other words, light launching
efficiency is just one of the key characteristic aspects of an optical fiber.
• There are other attributes too which have to be given importance while
determining the quality of an optical fiber.
• One of such attributes of an optical fiber is its bandwidth. Large
bandwidths are desirable for high data rates of transmission.
• When optical fiber is used for transmission of information, light signal launched
into it cannot be of continuous nature.
• For a carrier signal to carry information, one or more of its characteristics has to
be altered in accordance to the data signal.
• In an optical fiber light is launched in the form of optical pulses to transmit the
required information.
• Light energy launched into the fiber may be considered to travel in the form of
numerous rays in accordance to the Ray-Model.
• These rays travel different paths inside the core of an optical fiber because
different light rays are incident on the tip of the optical fiber at different angles
within the acceptance cone itself.
• This causes different light rays in the acceptance cone to travel along different
paths in the core of the optical fiber and accordingly take different time intervals
to travel a given distance too, which leads to a phenomenon of pulse broadening
inside the core of the optical fiber.
• Thus the pulse of light which might originally be of width T seconds now might be
of T+∆T seconds inside the fiber core. The figure 16 below depicts a pictorial
description of how light pulse broadens inside the core of the fiber.
 Figure 16: Pulse-Broadening inside optical fiber core

Any incident ray that lies within the acceptance cone gets guided inside the optical fiber by virtue
of multiple total internal reflections. Since the angle of refraction of different incident rays are
different, they travel along different paths in the optical fiber as shown in the above figure 16.
This causes the initially launched narrow light pulse to broaden as shown.
• The amount of broadening is measured in terms of the increase in the pulse time
width and is denoted by ∆T. the value of ∆T is given by:

……….(vi)

• Where, ∆T= Pulse Broadening; c = velocity of light in free space;

n1 = refractive index of core and n2 = refractive index of the cladding.
• The quantity L is the horizontal distance travelled before suffering the first total
internal reflection by the refracted ray OB which corresponds to the incident ray
AO, incident at the acceptance angle as shown in the figure 16.
• The amount of pulse broadening is effectively the difference in time of travel
between the ray travelling along the axis and the incident ray AO. This pulse
broadening effect signifies that if a second pulse is now launched into the fiber
within the time interval T+∆T, the two pulses will overlap and no identifiable data
would be obtained on the output.
• Thus for a given length L, there would be a corresponding value of ∆T (from
equation vi) which would limit the rate at which light pulses can be launched into
the optical fiber.
• In other words, it limits the rate at which data can be transmitted
along the fiber. This indirectly limits the bandwidth available on the
fiber. Thus we can say that more the pulse broadening lower the
bandwidth. That is:
…….(vii)
• Equation (vii) suggests that for higher bandwidth of transmission the
pulse broadening, ∆T should be as low as possible.
• In equation (vi), we see that the value of ∆T is dependent on the
value of L, the difference (n1 – n2) as well as the value of n1/n2. But
reducing the value of L would signify the reduction in the length of
the optical fiber, which is not desirable. As 1<n2<n1the ratio, n1 / n2
is very close to 1. Thus for low ∆T values, the only option available
with us is to decrease the value (n1–n2) or in other words, to increase
the refractive index of the cladding n2.
• One can now notice that a contradictory situation has been generated as to whether
the cladding should be removed for high NA or to use a cladding of large refractive
index value for higher bandwidth?
• The answer to this query is purely application specific. That means if an optical fiber is
used as a sensor (say), where lowest possible light has to be accepted, we use fiber with
low n2 values.
• When the optical fiber is used for data communication, fibers with high values of n2 are
used.
• For practical communication purposes the value of (n1 – n2) is made of the order of
about 10-3 to 10-4 .
• If the cladding is removed, the value of n2 becomes 1 and the value of the above
difference becomes about 0.5.
• The bandwidth corresponding to this value of n1-n2 is of the order of few Kilohertz,
which is far worse than that of a normal twisted pair of wires.
• Thus cladding is an extremely important requirement for optical fiber when the
bandwidth is the prime concern of the application and its refractive index is made as
close to that of the core as the available technology permits, but not made equal.
• This is brought about by varying the amount of doping in a single glass rod. The
differently doped regions have different refractive indices and serve as core and
cladding of the optical fiber.
• If we refer to the electromagnetic wave theory of light, it shows that at total
internal reflection, the light intensity inside the cladding is not completely zero.
• Instead, there exist some decaying fields in the cladding, which do not carry any
power but support the total internal reflection phenomenon by satisfying the
boundary conditions at the core-cladding interface. These fields are called as
evanescent fields.
• The Ray-model of light does not offer any explanation about the evanescent
fields, which indeed are as equally important as the fields in the core for total
internal reflection to occur.
• The importance of these evanescent fields in the TIR can be clearly ascertained
from the fact that even the slightest disturbance to these fields in the cladding
could lead to the failure of the TIR at the core-cladding boundary accompanied by
leakage of optical power to the cladding.
• This is one of the instances when the ray-model of light becomes inadequate in
explaining the phenomena exhibited by light.
• Though the evanescent fields are decaying fields, they never become zero, atleast
theoretically.
• In other words, they remain present upto infinite distance from the core-
cladding boundary.
• But in practice, these fields decay down to a negligibly small value as we
move away from the core-cladding boundary deeper into the cladding.
• Larger the value of the angle of incidence of the incident ray at the core-
cladding boundary, sharper is the decay of the evanescent fields.
• Thus there must me a sufficient thickness of cladding provided for these
evanescent fields to be accommodated so that they decay to a negligibly
small value in the cladding and cannot be disturbed by external sources.
• With this backdrop of knowledge about the propagation of light in an
optical fiber, let us now look into certain finer aspects of such propagation
and investigate the condition required for successful propagation of light in
the fiber.
• Figure 17 below shows two parallel rays that are launched into an optical
fiber and they propagate as shown. The dotted lines represent the wave-
fronts of the rays. The refractive indices of core and cladding are n1 and n2
respectively.
 Figure 17: Propagation of Light rays in an Optical fiber

The diameter of the core is ‘d’. The phase-front AE is common to both Ray 1 and Ray 2. The
phase-front DB is common to Ray 2 and BF. The Ray 2 is thus common to both the phase-
fronts. Hence for a sustained constructive interference, the distance between these two
phase-fronts must be multiples of 2 π .In other words, it can be said that the phase
difference between the phase change undergone by Ray 1 in travelling distance s1 and the
Ray 2 in travelling s2 must be 0 or integral multiples of 2π
• Mathematically, ……..(viii)

…….(ix)

• If δis the phase change undergone in each TIR of Ray 1, then the total
phase change undergone by Ray 1 in travelling s1 is given by
……..(x)
• Where n1 = refractive index of core; =Wavelength of the light in the
core. The phase change undergone by Ray 2 in travelling s2 is given by
…….(xi)
• For a sustained constructive interference, both φ1 and φ2 must have
a phase difference of either 0 or integral multiples of 2π. That is, for
an integer m (=0,1,2,3,…) the following condition must be satisfied:
 ……(xii)

The significance of the equation (xii) is that only those rays, which are incident
on the tip of the fiber at angles such that their angle of refraction in the core
satisfies equation (xii), can successfully travel along the fiber.
If we concentrate on equation (xii), we find that since ‘m’ can take only
discrete integral values, the value of angle θ is also discrete.
This suggests that there are only some discrete launching angles within the
acceptance cone (N.A. cone) for which the rays can propagate inside the fiber
core.
• A 3D visualisation reveals the significance of this observation, i.e. the
acceptance cone can no longer be assumed as a solid cone of rays,
launched at all possible angles (smaller than acceptance angle), but has to
be viewed as composed of discrete annular conical rings of rays which are
launched at the tip of the fiber core at angles which satisfy equation (xii).
• Thus the condition that the launching angle of the incident ray should be
within the acceptance cone is necessary but not sufficient.
• This angle has to be such that the equation (xii) is satisfied. Thus light can
only be launched at certain discrete angles within the N.A. cone leading to
a further decrease in the light gathering efficiency of the optical fiber.
• Any ray that is not launched at these discrete angles will not propagate
inside the optical fiber.
• This discretization in the values of launching angles lead to formation of
what are called as modes in an optical fiber, which are nothing but
different patterns of light intensity distribution around the axis of the core.