Dr Ali Y.

Fattah
Communication Engineering Department
University of Technology

Optical Fiber Communication

Systems II
2. Multichannel systems

1
Multiplexing
A process where multiple analog message signals or digital
data streams are combined into one signal over a shared
medium.

Types
1. Optical Time Division Multiplexing ( OTDM ).
2. Optical Subcarrier Multiplexing ( OSCM ).
3. Orthogonal Frequency Division Multiplexing ( OFDM ).
4. Wavelength Division Multiplexing ( WDM ).
5. Optical Code Division Multiplexing ( OCDM ) , or
Optical Code Division Multiple Access ( OCDMA ).
6. Hybrid Multiplexing .

2
1. Optical Time Division Multiplexing (OTDM)

A block schematic of an OTDM system which has demonstrated

160 Gb / s transmission over 100 km is shown below:

3
2. Optical Subcarrier Multiplexing ( OSCM )
The microwave frequency or RF electrical subcarriers are modulated
with an optical carrier and then are transmitted using a single wavelength
signal .
Microwave subcarrier multiplexing enables multiple broadband signals
to be transmitted over single-mode fiber .

4
3. Orthogonal Frequency Division Multiplexing ( OFDM )

Orthogonal frequency division multiplexing (OFDM) is a multicarrier

transmission technique which is based on frequency division
multiplexing (FDM) . In OFDM several subcarrier frequencies
orthogonal to each other (i.e. perpendicular) and therefore they do not
overlap.
Hence this technique can squeeze multiple modulated carriers
tightly together at a reduced bandwidth without the requirement for
guard bands while at the same time keeping the modulated signals
orthogonal so that they do not interfere with each other .

5
6
4. Wavelength Division Multiplexing (WDM )
A powerful aspect of an optical communication link is that
many different wavelengths can be sent along a single fiber
simultaneously in the 1300 to-1600-nm spectral band .
The technology of combining a number of wavelengths on to
the same fiber is known as Wavelength Division Multiplexing
WDM .

7
8
 Multiple channels of information carried over the same fiber, each
using an individual wavelength . The

 The fact that one can use different wavelengths over the same channel
increases bandwidth capacity enormously.

 Most WDM systems work in the C-band around 1550 nm.

 ITU channel spacing are 0.4 nm, 0.8 nm and 1.6 nm (50, 100 and
200 GHz). Proposed spacing of 0.2 nm (25 GHz) and even 0.1 nm
(12.5 GHz)

9
 Older systems which were spaced at 200 GHz are known as WDM
whereas systems with denser packing are called Dense WDM
(DWDM) .

 Still older ITU specification, referred to now a days as CWDM

(Coarse- WDM) specified a 20 nm spacing in the wavelength range
1270 to 1610 nm.

The number of channels could be reaching up to 128 so that a single

fiber can give a bandwidth of over 300 M bit/sec.

Modern systems can easily pack channels to give a bandwidth of 400

G bits/ sec. up to 1.6 Terra bits/sec.

 WDM can support multiple protocols. Each signal can be carried at

different bit rates.

10
 Why WDM?
Capacity upgrade: The application of WDM has been to
upgrade the capacity of existing fiber networks (without
adding fibers).
Transparency: Each optical channel can carry any
transmission format. Thus, using different wavelength ( fast
or slow asynchronous or synchronous, analog or digital data)
can be sent simultaneously, and independently over the same
fiber without a common signal structure.
Wavelength routing : Wavelength is used as another
dimension to time and space. Wavelength routed networks
use the actual wavelength of a signal as the intermediate or
final address.
Wavelength switching: Allow reconfigurations of the
optical layer. Components for implementing these networks
include optical add/drop multiplexers, optical cross connects,
and wavelength converters.
11
 Principles of WDM
Since an optical source has a narrow linewidth , this type of
transmission makes use of only a very narrow portion of the
transmission bandwidth capability of a fiber.
BW of a modulated DFB laser: 10-50 MHz  0.001 nm
Typical channel spacing: 0.4 – 1.6 nm.
80 nm or 14 THz @1300 nm.
120 nm or 15 THz @ 1550 nm
With such spectral bandwidths, simplex systems make use of
only a small portion of the transmission bandwidth capability of
a standard fiber. This can be seen from the curve, which depicts
the attenuation of light in a silica fiber as a function of
wavelength.

12
ITU-T Standard Transmission WDM
windows
 c 
   2  
 

13
 A key reason for selecting a fixed frequency spacing,
rather than a constant wavelength spacing, is that when
locking a laser to a particular operating mode it is the
frequency of the laser that is fixed.
The range of standardized channel grids includes 50 GHz
(0.4 nm), 100 GHz(0.8 nm), 200 GHz(1.6 nm) spacing.
Wavelength spacing practically depends on:
 laser line width
 optical filter bandwidth
Discrete wavelengths form individual channels that can be
separated, routed, and switched individually.
The optical intensity is kept sufficiently low to prevent
nonlinear effects, such as stimulated Brilouin scattering
and four-wave mixing processes, from degrading the link
performance.
14
 WDM , CWDM and DWDM
First WDM networks used just two wavelengths, 1310 nm and 1550
nm.
In the late 1980s,with the advent of tunable lasers that have extremely
narrow linewidths, one then could have very closely spaced signal
bands .This is the basis of dense WDM.
DWDM is a historically derived designation, it refers to the spacing
denoted by ITU-T G.692.
Dense WDM (DWDM) has dense channel spacing (0.8 nm) which
allows simultaneous transmission of 16 wavelengths – high capacity.
Today's DWDM systems utilize 16, 32,64,128 or more wavelengths in
the 1550 nm window .
Each of these wavelength provide an independent channel (Ex: each
may transmit 10 Gb/s).
Coarse WDM (CWDM) has wider channel spacing (20 nm)-low cost .
15
Typical WDM network
At transmitting end :
 Different signals from different channels are used to
modulate laser sources separately.
 The laser sources have different frequency or wavelength
(each emitting signals) at unique wavelength.
 The output signals from the different sources are then
combined in to a serial spectrum of closely spaced wavelength
signal and couple them onto a single fiber by WDM
multiplexer .
Multiplexer provides a low-loss path from each optical
source to the multiplexer output .

16
At receiving end :
After the combined optical signal is passed through the fiber :
Optical signals are separated by WDM de-multiplexer to
separate the different wavelengths.
 The separated signals are then detected by separated photo
detectors and received by the receivers for signal processing.
 Photodetectors include all the WDM channels since they
are sensitive over a broad rang of wavelengths.
The implementation of WDM networks requires a variety of
passive and active devices to combine, distribute, isolate, and
amplify optical power at different wavelengths.

17
Multiplexers & Demultiplexers
Optical Multiplexers are components specifically designed for
wavelength division multiplexing (WDM) systems. The main function
of an optical multiplexer is to couple two or more wavelengths into the
same fiber.
The Demultiplexer undoes what the multiplexer has done ; it separates
a multiplicity of wavelengths in a fiber and directs them to many
fibers .

18
 Optical Add-Drop Multiplexers (OADM)
The OADM selectively removes (drops) a wavelength from a
multiplicity of wavelengths in a fiber , and thus from traffic on the
particular channel. It then adds in the same direction of data flow the
same wavelength, but with different data content.

F1 signifies a filter selecting wavelength λ1 while passing through all

other wavelengths, and M1 signifies a multiplexer that multiplexes all
wavelengths.
19
A better view of OADM function is shown below. This function is
especially used in WDM ring systems as well as in long-haul with drop-
add features .

The main function of an optical drop-add multiplexer is to selectively

remove a wavelength and add the same wavelength in the fiber (OAs are
optional).

20
A possible structure of the optical add-drop function using an optical
2 X 2 switch is illustrated (for one wavelength) below :

More optical switches may be added between demultiplexer and

multiplexer to support more drop-add wavelengths.

21
 Optical Couplers
Optic couplers either split optical signals into
multiple paths or combine multiple signals on
one path.
The number of input (N)/ output (M) ports,
( N x M size) characterizes a coupler.
Fused couplers can be made in any
configuration, but they commonly use multiples
of two (2 x 2, 4 x 4, 8 x 8, etc.).

22
The amount of power transmitted into fibres depend on the coupling
length .
The coupling length changes with the wavelength .
The splitting ratio can be tuned choosing the coupling length .
By choosing carefully the coupler length, it is possible to combine or
separate two different wavelengths .
23
Coupler - Performance Parameters

Coupling ratio or splitting ratio

Power from any single output P
CR   t
Total power out to all ports PT out
Input power Pi
Excess Loss Le 
Total outputpower PT out

Power input Pi
Insertion Loss Li  
Power output Pj

Isolation Loss or Crosstalk

Reflected power back into other input port
Liso 
Input power at one port
24
Star Couplers
The principle role of all star couplers is to combine the power
from N inputs and divide them equally among M output ports.
The splitting is done uniformly for all wavelengths. so each of M
output receives 1/M of the power entering the device.
Techniques for creating star couplers: fused fibers, gratings,
micro-optic technologies, and integrated-optics schemes.
The fibre-fusion technique has been a popular method for N X M
star couplers.
An ideal star coupler, the optical power from any input is evenly
divided among the output ports.
The total loss consists of its splitting loss plus the excess loss in
each path through the star.

25
 Basic Star Coupler
May have N inputs and M outputs

 Up to N =M = 64, typically N, M < 10

 Types of star couplers:
♠ transmission star coupler
the light at any of the input port is split equally through all output port.
♠ reflection star coupler
26
Fused- Fiber Star Coupler

Combines power from N inputs and divided them between M outputs

1 1
P1

PN
N M

1
Splitting loss  10 log10  N   10 log10 N
 
 P 
 10 log  in 
Excess loss  NP 
 i out,i 
10

1
Power at any one output Po,i  ( P1  P2  .......PN )
N
27
The 2x2 Fiber Coupler
A device with two inputs and two outputs.
Fused fibre coupler
Fabricated by twisting together, melting, and pulling two
single- mode fibres so they get fused together over a uniform
section of length W.
Each input and output fibre has a long tapered section of
length L, the transverse dimensions are gradually reduced
down to that coupling.
The total length = 2L+W.
This device is known as a fused Biconical Tapered Coupler.

28
 Fused- Biconical Coupler OR
Directional Coupler

P3, P4 extremely low (-50 to -70 dB below P0)

P0 input power, P1 throughout power, P2 power
coupled in the seconed fiber
Coupling / Splitting Ratio = P 2/(P1+P2)
If P1=P2  It is called 3-dB coupler
29
 2X2 Coupler - Performance Parameters

 P2 
Coupling ratio CR    X 100%
(splitting loss)  P1  P2 
 P0 
Excess loss Le  10 log10  
 P1  P2 

Pi
Insertion Loss Li 
Pj

Isolation Loss  P3 
Liso  10 log10  
or Crosstalk  P0 

30
Example : A 2x2 biconical tapered fiber coupler has an input
optical power level of Po= 200 mW. The output powers at the
other three ports are P1=90 mW, P2=85 mW, P3=6.3 nW.
Determine
 85 
1- coupling ratio CR    X 100% = 48.6%
 90  85 
 200 
2- Excess loss Le  10 log10   = 0.58 dB
 90  85 
3- Insertion losses are:
 P0 
Insertion loss (Po to P1) Liso  10 log10   = 3.47 dB
 P1 
 P0 
Insertion loss (Po to P2) Liso  10 log10   = 3.72 dB
 P2 
 P3  = -45 dB
4- Crosstalk ( Isolation loss) Liso  10 log10  
 P0 
31
Star Coupler - 8 X 8

Star couplers are optical couplers with more than four ports
1 1, 2, ... 8
2

3
4

N/2 5
6

7
8 1, 2, ... 8
log2 N

N
No. of 3dB coupler N c 3dB  log 2 N
2 32
 This device formed by using twelve 2x2 couplers.
It made from either fused-fiber or integrated- optic components.
A fraction 1/N of the launched power from each input port
appears at all output ports.
A limitation to the flexibility or modularity of this technique is
that N is a multiple of 2; that is , N= 2n with the integer 1  n.
If an extra node needs to be added to a fully connected NxN
network, the NxN star needs to be replaced by a 2Nx2N star,
thereby leaving 2(N-1) new ports being unused.
One extra 2x2 coupler can be used at one port to get N+1 outputs
These two new ports have an additional 3-dB loss.

33
Star Coupler - 8 X 8 – Performance

 If a fraction of power traversing each 3 dB coupler = FT

where 0< FT < 1.
Then, power lost within the coupler = 1- FT.

Excess loss Le  10 log10 ( FTlog2 N )

Coupling ratio 1

CR  10 log10    10 log10 N
(splitting loss) N

Total loss = splitting loss + excess loss

LT  10 (1  3.322 log10 F ) log10 N

Try Ex: 10-5
Keiser 2000 34
5. Optical Code Division Multiplexing ( OCDM ) , or
Optical Code Division Multiple Access ( OCDMA )
OCDMA is a digital technique where, instead of each channel occupying
a given wavelength, frequency or time slot, the information is transmitted
using a coded sequence of pulses. Each channel employs a specific code
to transmit and recover the original signal.
It utilizes the basic principle of spread spectrum transmission where all
users share the fiber channel bandwidth simultaneously.
The basic OCDM technique is illustrated in the following figure where
multiplexing of three channels is accomplished by transmitting a unique
time-dependent series of short pulses.

35
Each bit to be transmitted is subdivided into a number of small intervals
n known as chips (e.g. n = 64 or 128). Each user is then assigned a
unique chip sequence of the n bit code. Many coding schemes exist for
the generation of the chip sequences to encode/decode OCDM channels.
The greater the number of unique sequences needed (i.e. number of
users), the larger the code sequence required .
A decoder is then used at the receiving end to recover the particular
channel employing autocorrelation with the original chip sequence.
However, OCDM is not a bandwidth-efficient multiplexing technique.
In order to enable an increasing number of OCDM channels to be
transmitted on a single fiber , ultra short pulses (i.e. 10-15 second pulse
duration) are required to be used in OCDM systems.
Furthermore, the larger the number of channels, the longer the code
sequences needed to provide a unique channel code.

36
6. Hybrid Multiplexing
When two (or more) different multiplexing techniques are combined to
allow optical signal multiplexing for several optical signals, the resultant
is referred to as hybrid multiplexing.
It should be noted that different multiplexing strategies exhibit their own
advantages and drawbacks and therefore the combination of different
multiplexing techniques can be used to overcome the problems
associated with a specific technique. A hybrid multiplexing system can
comprise either optical or electrical domain multiplexing, or combination
of both signal types.
Common examples of optical hybrid multiplexing are WDM being
combined with OTDM, OCDM or OSCM.

37