Fiber Optic Resource Center
ALL About Fiber Optics

Fiber AUTHORS Submit your "learning/education articles here"

'How This Works' .... SEE sponsor instructions right sidebar!
Fiberopticcleaners.net   HOME   Search Articles   All Articles   Submit an Article Random Article   Contact Us
  01/22/2018, 06:52:15 PM Welcome Guest, Please LOGIN to your account or SIGNUP now
  Search Articles
 
Keyword
Exact phrase
All words (AND)
Any word (OR)
 
 
  Categories
  Termination
  Polishing
  Inspection
  Cleaning
  Interferometry
  Cable Assembly
  Optical Test
  Theory of Fiber
  Applications
  Installers
  Mining
  Oil and Gas
  Epoxy, Adhesives,...
 
  Authors
  Wikipedia
  Fiber Optic...
  FOA
More Authors List 
 
 
 
 Editor's Pick
 DEFINING AND MEASURING PHYSICAL PARAMETERS OF PC POLISHED CONNECTORS
  DEFINING AND MEASURING PHYSICAL PARAMETERS OF PC POLISHED CONNECTORS
Author Eric Norland  Added On 06/21/2014
Rating (1)  Category Interferometry
 
 Article Categories
Termination(2) Polishing(2) Inspection(1)
Cleaning(1) Interferometry(3) Cable Assembly(7)
Optical Test(1) Theory of Fiber(9) Applications(4)
Installers(4) Mining(1) Oil and Gas(1)
Epoxy, Adhesives, Curing(12)
 
 Latest Articles
 Thesis on Plastic Optical Fiber
 

Mediterranea University of Reggio Calabria

Faculty of Engineering

Bachelor's Degree of Telecommunication Enginnering

Thesis

Plastic Optical Fibers Characteristics and applications

Supervisor Candidate

Prof. Eng. Francesco Giuseppe Della Corte Domenico Carlo Romeo domenicocarlo.romeo@gmail.com

Academic Year 2009 - 2010

To my family, to my relatives, to my colleagues and to my friends.

Contents

Introduction xv

1 Optical fibers 1

1.1 Features and structure..............................................1

1.2 Operating principle..................................................2

1.3 Characteristic parameters..........................................6

1.4 Classification........................................................7

1.5 Connections..........................................................9

1.6 Manufacturing methods............................................11

1.6.1 Double-crucible method....................................11

1.6.2 Preform method............................................12

1.7 Issues ................................................................13

1.7.1 Attenuation..................................................13

1.7.1.1 Attenuation inside optical fibers................13

1.7.1.2 Attenuation inside the connection between optical fibers ........................................16

1.7.2 Dispersion....................................................18

1.7.2.1 Modal dispersion..................................18

1.7.2.2 Chromatic dispersion............................20

1.7.2.3 Spatial dispersion................................21

1.8 Advantages and disadvantages......................................21

2 Optical communication systems 23

2.1 Architecture..........................................................23

2.1.1 The optical transmitter....................................24

2.1.2 The optical link....................... 27

2.1.3 The optical receiver .................... 28

2.2 Notes on devices for optical systems............... 30

2.2.1 LED............................. 31

2.2.2 The LASER diode..................... 34

2.2.3 The PIN photodiode........................................36

2.2.4 The APD....................................................37

2.2.5 The splitter..................................................38

2.2.6 The combiner................................................38

2.2.7 The directional coupler ....................................38

2.3 The WDM system..................................................39

2.4 Issues ................................................................41

2.4.1 Attenuation..................................................41

2.4.2 Four-Wave Mixing (FWM) ................................43

2.4.3 Optical retro-reflections....................................43

2.4.4 Synchronization ............................................43

2.4.5 Noise ........................................................44

2.4.6 Wavelength conversion......................................45

2.4.7 Cost..........................................................45

2.4.8 The reliability of the optical source........................45

2.5 Performance analysis................................................46

3 Plastic Optical Fibers (POF) 47

3.1 Structure............................................................47

3.2 Manufacturing ......................................................48

3.2.1 The continuous extrusion method..........................48

3.2.2 The coextrusion method....................................49

3.2.3 The preform method........................................50

3.3 Main characteristics ................................................50

3.3.1 Attenuation..................................................50

3.3.2 Dispersion....................................................53

3.3.3 Mechanical performance....................................54

3.4 Connections..........................................................56

3.5 Installation..........................................................57

4 Applications of POFs 59

4.1 Ethernet over POF (EoPOF)......................................59

4.1.1 Characteristics..............................................59

4.1.2 Devices......................................................61

4.1.3 Installation costs and time..................................65

4.2 Other applications..................................................67

4.3 Organizations and projects ........................................67

4.3.1 POFTO......................................................67

4.3.2 POF-ALL....................................................68

4.3.3 POF-PLUS..................................................69

4.3.4 POLYCOM..................................................71

5 Conclusions 73

A Notes on coherent light 75

B PLL (Phase Locked Loop) 77

C Eye pattern 79

Acknowledgements 81

Bibliography 83

List of Figures

0.1 John Tyndall experiment ..........................................xv

1.1 Structure of optical fiber............................................1

1.2 Size and forms of optical cables....................................2

1.3 Incident ray, refracted ray and reflected ray......................3

1.4 Total internal reflection............................................4

1.5 operating principle of optical fiber ................................4

1.6 Definition of numeric aperture......................................6

1.7 Optical fiber profiles................................................7

1.8 Mono mode optical fiber............................................8

1.9 Multi mode optical fiber............................................8

1.10 Types of optical fibers..............................................9

1.11 Connectors for optical fibers........................................10

1.12 Mechanical junction................................................10

1.13 Realization fusion junction..........................................10

1.14 Double-crucible method............................................11

1.15 Preform preparation................................................12

1.16 Spinning system ....................................................12

1.17 Attenuation..........................................................13

1.18 Dispersion............................................................13

1.19 Absorption ..........................................................14

1.20 Attenuation coefficient for GOF....................................14

1.21 Scattering............................................................15

1.22 Macrobend ..........................................................16

1.23 Microbend............................................................16

1.24 Difference between refraction indices..............................17

1.25 Difference between core diameters ................................17

1.26 Disallineamento assiale..............................................17

1.27 Angular disalignment................................................18

1.28 [Accostamento]......................................................18

1.29 Modal dispersion inside SI fibers..................................18

1.30 Modal dispersion inside SI fibers..................................19

1.31 Cancellation of modal dispersion in GI fibers ....................20

1.32 Chromatic dispersion................................................20

1.33 Spatial dispersion....................................................21

2.1 Architecture of an optical communication system................23

2.2 Optical transmitter..................................................24

2.3 Circuit scheme of intensity modulator............................25

2.4 Channel coding schemes............................................26

2.5 Channel coupler ....................................................27

2.6 Optical receiver scheme ............................................28

2.7 Responsivity of photodetector......................................29

2.8 Photodetector connected in cascade with amplification stages . 30

2.9 Homojunction structure............................................31

2.10 Heterojunction diode................................................32

2.11 Recombination of electron-hole....................................32

2.12 V-I characteristics of LED technologies............................33

2.13 I-P characteristics of LED varying the operating temperature . 33

2.14 SLED (top) e ELED (bottom) ....................................34

2.15 LASER diode........................................................34

2.16 Homojunction LASER diode (top) and heterojunction LASER diode (bottom)......................................................35

2.17 I-P characteristics of LASER diode at different operating temperatures ............................................................36

2.18 PIN photodiode......................................................36

2.19 PIN photodiode structure..........................................36

2.20 APD..................................................................37

2.21 APD structure......................................................37

2.22 Splitter ..............................................................38

2.23 Combiner............................................................38

2.24 Directional coupler..................................................39

2.25 Directional coupler in fusion fiber..................................39

2.26 Directional coupler in planar optical integration..................39

2.27 WDM system and dichroic filter ..................................40

2.28 Spatial distribution of SLED, ELED and LASER diodes .... 42

2.29 Optical system with repeater......................................42

2.30 Optical retro-reflections inside an optical channel................43

3.1 Molecular structure of PMMA ....................................47

3.2 Comparison between POF and GOF..............................47

3.3 Continuous extrusion method......................................49

3.4 Coextrusion method................................................49

3.5 Curve of attenuation coefficient for POF SI and POF GI . . . . 50

3.6 Curvature losses of SI POFs........................................52

3.7 Curvature losses of GI POFs ......................................52

3.8 Modal conversion....................................................53

3.9 Traction performance of POFs ....................................54

3.10 Flexion performance of POFs......................................55

3.11 Torsion performance of POFs......................................55

3.12 POF connectors......................................................56

3.13 POF connections with OptoLock..................................56

3.14 Ducts for POF......................................................57

4.1 Ethernet over POF PDU............................................61

4.2 POF-UTP media converter........................................62

4.3 POF-GOF media converter........................................62

4.4 Cabling between GOF and EoPOF networks through an FTTH access point and media converters ................................62

4.5 USB-POF adapter..................................................63

4.6 EoPOF switch ......................................................63

4.7 EoPOF router........................................................64

4.8 EoPOF network card................................................64

4.9 Example of EoPOF network wiring................................65

A.1 Light wave packets..................................................75

A.2 Coherent light........................................................75

A.3 Incoherent light......................................................75

B.1 PLL scheme ........................................................77

C.1 First example of eye pattern........................................79

C.2 Second example of eye pattern ....................................80

List of Tables

1.1 Optical windows for GOF..........................................15

2.1 Parameters of LED technologies....................................33

3.1 Transmission windows for POF SI ................................51

3.2 Transmission windows for POF GI................................51

4.1 Analysis wiring cost through UTP cables ........................66

4.2 Analysis wiring cost through POF cables..........................66

Introduction

The history of optical fibers goes back two centuries. This technology has been developed since the 1960's, while its operating principle dates back to the early 1800's.

The operating principle of optical fibers is based on total internal reflection1 , which was observed and demonstrated in the late 1800's by the physician John Tyndall (figure 0.1). Using a container filled with water and illuminating the latter from above, he proved that the light ray was propagated inside a water jet coming out of a hole made on the wall of the container. This discovery was the first experience of light fed inside an optical waveguide, but Tyndall did not find any practical applications of his discovery. We must therefore wait until the mid 1900's, the period when Lawrence Curtiss physically produced the first optical fibers thanks to the development of the gastroscope2. Despite this, these fibers had losses equal to about 100 dB/Km, so they were not acceptable for long-distance links (specifically, those extended to distances measurable in kilometers). [1, 2, 3]

Then in 1965, Charles Kao and George Hockham, researchers from STC Ltd.3published a very important article about optical fibers: after demonstrating that the strong attenuation from the fibers was only due to their material impurities and so would be overcome, the researchers fixed 20 dB/Km as the maximum limit of the attenuation factor associated to an optical fiber for longdistance links. Many competitors then entered a competition to manufacture a technology which matched these requirements. [2]

Unexpectedly the competition was won by Corning Glass Inc.4, a company wholly foreign to the world of telecommunications. In 1970 the company announced the production of a glass optical fiber with a loss factor equal to 17 dB/km using titanium-doped silicon. In the following years the American company contracted with many telecommunications companies, so starting off experimentation and production of optical fibers all over the world. Some of them also manufactured essential devices for optical communications such as LEDs, LASER diodes, photodetectors and optical amplifiers. [2]

Since the early 1980's optical fibers have been gradually replacing coaxial cables in backbones, radio links and transoceanic links. Currently the performance of this technology has reached extraordinary levels: a single piece of fiber can be extended to hundreds of kilometers and can transmit tens of thousands of telephone channels. Therefore the optical fiber is used for digital transmissions. [2]

Despite their progress, metallic cables (UTP, coaxial, etc.) have been surviving in the so-called last mile, the stretch of the network which connects the backbone networks to the end users. The installation of optical fibers in this stretch might be expensive and so it has been considered that substitution is not worth the cost where a zone is cabled with metallic cables. [2]

Plastic optical fibers, the subject of this thesis, can become ever used and replace transmission by metallic cables.

This thesis is structured as follows:

• Chapter 1 describes the features and structure of optical fibers, noting the operating principle, the characteristic parameters, classification, connections, manufacturing methods and main issues. Then we will list the advantages and the disadvantages of glass optical fibers compared to metallic cables;

• Chapter 2 discusses the architecture of telecommunication systems in optical fibers, mentioning the devices which compose them. Then we will explain the WDM system, the main issues and finally the performance of telecommunication systems in optical fibers;

• Chapter 3 illustrates the structure of plastic optical fibers, giving an overview of the methods of manufacture, the main features, the connections and the installation of this technology;

• Chapter 4 analyzes the main applications of plastic optical fibers, highlighting a very promising technology for the future: Ethernet over POF (in abbreviation EoPOF). Then we will discuss POFTO, an international organization which promotes the worldwide proliferation of plastic optical fibers in IT and some projects for the diffusion and development of communication systems based on this technology;

• Chapter 5 is the conclusion of the thesis.


Figure 0.1 John Tyndall experiment

Chapter 1 Optical fibers

1.1 Features and structure

Optical fibers are dielectric waveguides, or guiding structures which convey one or more light signals inside them. They are made of filaments of glass or plastic materials. In the first case they are called glass optical fiber or GOF5, while in the second case they are called plastic optical fibers or

POF6. [1]

Optical fibers are made of two concentric cylindrical layers: a central part, known as the core, and another part around this, known as the cladding (figure 1.1). The core is the internal part of the fibers in which light rays are propagated and always has a diameter less than that of the cladding (generally the beams are generated and modulated by optical transmitters, which will be discussed in chapter 2). Cladding is the external part of the fibers which reflects the beams inside the core and it always has a diameter larger than that of the core. [1]


Figure 1.1 Structure of optical fiber

To ensure a certain degree of protection from mechanical and thermal abrasion, optical fibers are coated with several layers of synthetic materials (PVC, Kevlar, Nylon, etc.). They are embedded in specific structures, called optical cables. The latter can contain up to thousands of individual small fibers and they are produced in different sizes and shapes (figure 1.2). [3]


Figure 1.2 Size and forms of optical cables

1.2 Operating principle

Before understanding the operating principle of optical fibers, it is necessary to recall some concepts of physics. A light beam propagates in vacuum at a speed of c0 = 3 ■ 108 m/sec. In a generic medium it propagates at a speed of:

V = [m/sec]

where

• v is the speed of light travelling in a generic medium

• co is the speed of light in vacuum

n is the refractive index of a generic medium

With reference to the scenario of figure 1.3, if a light beam strikes the surface of separation between two mediums such that n1 >n2 (n1 denotes the refractive index of the first medium, while n2 that of the second medium), then a part will be reflected from that surface and another part will be refracted in the second medium. [4]

In particular the behaviour of the beam is adjusted by Snell's laws7:

◊inc = ◊refl (1)

n1 sen 0inc = n2 sen 0refr (2)

where

0inc is the incident angle

Orefi is the reflection angle

0refr is the refraction angle

• ni is the refraction index of the first medium

n2 is the refraction index of the second medium

[4]

The relationship (1) denotes that the incident angle and that of reflection will be the same, while the relationship (2) describes a connection between the incident angle, the refraction angle and the refraction index of those mediums. From such relationships we can deduce that, increasing the incident angle 0inc, the refraction angle 0refr will increase faster until 9inc reaches a certain value called the limit angle8. As soon as 9inc reaches this angle, total internal reflection will consequently occur: the light ray will completely be reflected on the interface between the two mediums (figure 1.4). [4]


n1 >n2

n1 \


0inc^/\^

y/^erefl

n2


Figure 1.4 Total internal reflection

The operating principle of optical fibers is based on these theories. [4]

With reference to figure 1.5, let us hypothesise that:

• a light ray travelling in air impinges on the initial section of an optical fiber (the refraction index is n0);

• the refraction index of the core is larger than that of the cladding (ni > n2, where n1 denotes the refraction index of the core while n2 is the refraction index of the cladding).





A

Core

r>7




n, > n2

Air



Cladding

n2

Figure 1.5 operating principle of optical fiber

Since the air and the fiber have different refractive indices (no = n1 = n2), the initial ray will impinge on the initial section of the fiber with an angle of 90ext, called the acceptance angle of optical fiber. So it will be divided into a refracted ray and a reflected ray. The refracted ray will enter the core, while the reflected one will be rejected and it will follow its path into the air. It is possible to observe that the reflected ray forms, with the perpendicular to the initial section of the fiber, a same angle 90ext according to the relationship (1), while the refracted ray will propagate inside the core with an angle equal to 9c, called the critical angle. This angle can be calculated applying the relationship (2) to the scenario of figure 1.5:

no sen Ooext = ni sen 9c

where:

• n0 is the refraction index of air

90ext is the external angle

ni is the refraction index of the core

9c is the critical angle

When the refracted ray meets the core-cladding interface, it might partially reflect into the core and partially it might refract into the cladding. So the refracted ray will have some losses because a part of it wil be lost in the cladding. To avoid this happening, it is necessary that 90ext is such that 9c is equal to the limit angle. In this way total internal reflection in core will occur and so optical losses due to refraction inside the cladding will be null. So the only ray useful for transmission by optical fiber will be the travelling wave inside the core. [5]

1.3 Characteristic parameters

Optical fibers used for telecommunications are characterized by the following parameters (usually given in the datasheet):

• attenuation factor a (dB/Km)

This represents the power attenuated by a fiber after 1 Km. This parameter is specified graphically in the datasheet according to the wavelength

• acceptance angle 90ext (°)

This is the maximum angle at which the light ray can be injected inside the optical fiber. This angle corresponds geometrically to the half angle of the acceptance cone, or the conical region where the various rays can be injected inside optical fiber (see figure 1.6). So the incision of the rays inside the acceptance cone optimizes the coupling with the optical fiber.

• refraction index of the core n\

This is the refraction index of the core.

• numeric aperture NA9

This is defined as the sine of the acceptance angle.


Figure 1.6 Definition of numeric aperture

This relationship shows that the numerical aperture depends not only on the difference between the square of the refraction index of the core

and that of the cladding (in other terms on n2 — n|), but also on the acceptance angle of optical fiber.

bandwidth (MHz/Km)

This represents the maximum frequency of an optical signal which can be injected inside a stretch of a 1 km long fiber; it depends on the number of modes10.

[3, 8]

1.4 Classification

Optical fibers are classified according to their refractive index profile and the number of modes. [3]

According to the profile the fibers can be step index (in short, SI) or graded index (in short, GI). The former (figure 1.7(a)) are characterized by a constant refractive index inside the core which decreases sharply inside the cladding; whereas, the latter (figure 1.7(b)) have a graded refractive index from the center of the fiber towards the core-cladding interface. [3]

(a) Step index (b) Graded index

Figure 1.7 Optical fiber profiles

According to the number of modes, optical fibers can be single mode or multi mode.

Single mode fibers (figure 1.8 on the following page) can convey only one mode internally; if the fibers are made of glass, then the core will have a diameter of 8^10 mm, while the cladding a diameter of 125 mm. So they have got a very narrow SI profile because only one light ray propagates inside a very thin core. It has been shown that for V < 2.405 (where V = NA, where a denotes the ray of optical fiber) the cut-off condition is verified, or the condition for which the fiber is single mode. [3]






Cladding





Core

Figure 1.8 Mono mode optical fiber

Multimode fibers (figure 1.9) however can convey more modes internally; if the fibers are made of glass, the core will have a diameter of 50 or 62.5 mm while the cladding will have a diameter of 125 mm. So they have a large step index or graded index profile because more light rays can be propagated internally. It has been demonstrated that, using Maxwell's equations with the appropriate boundary conditions, the maximum number of possible modes of a multi mode fiber is given by the expression M = . Such approximation is valid only if V < 20. [3]


ow

</


Cladding


Core

Figure 1.9 Multi mode optical fiber

Therefore we can distinguish three types of optical fibers (figure 1.10):

• mono mode step index;

• multi mode step index;

• multi mode graded index.


step

Figure 1.10 Types of optical fibers

1.5 Connections

The following paragraph focuses on the connections of the GOF (the POF be

discussed in chapter 3).

GOFs may be connected with optical connectors, mechanical junctions and fusion junctions. [3]

Optical connectors are mechanical devices which link a fiber either with another one or with a communication device and there are various types of connectors: SC (in plastic, with a square shape), LC (in plastic, with a square shape), ST (in metal, with a circular shape and a bayonet plug), FC (in metal, with a circular shape and a screw plug), MT-RJ (in plastic, with the same shape as SC but slightly wider and flatter), SMA (in metal with a hexagonal shape and a crew plug) and TOSLINK (in plastic, metal or both, with the same shape as SC). [3]


Multi mode step index fiber


Multi mode graded index fiber

 

Refractive index


Figure 1.11 shows the different types of connectors cited before:

FC

LC


SMA TOSLINK

Figure 1.11 Connectors for optical fibers

SC

ST

MT-RJ

Mechanical junctions are sleeves that physically maintain the two ends in contact (figure 1.12). Inside them they need a gel interposed between the faces of the fibers in order to reduce loss of light signals. [3]

FI6CR END FACES

FIBER

CAPILLARV TUBE

INDEX-MATCHING GEL IN TUBE

Figure 1.12 Mechanical junction

Fusion junctions are obtained by fusion of the two sections of fiber using an instrument equipped with high-voltage electrodes (figure 1.13). The junction is mechanically fragile and therefore requires a mechanical protection. [3]


Figure 1.13 Realization fusion junction

In practice it is easier to connect the fibers with connectors rather than with mechanical or fusion joints. [3]

1.6 Manufacturing methods

The following paragraph is limited to manufacturing methods of GOF (POF manufacturing methods will be discussed in chapter 3 because they are produced in different ways).

The most common manufacturing methods of GOF are two:

• double-crucible method;

• preform method.

1.6.1 Double-crucible method


Figure 1.14 Double-crucible method

The double-crucible method consists in the insertion of glass in two concentric cylindrical crucibles (figure 1.14). In particular, the core compound will be deposited in the internal crucible, while the cladding compound will be put in the external crucible. Both crucibles are drilled from their base and they are heated in high temperature furnace to fuse the compounds. Then the optical fiber will be made taking the fused materials from that holes and pulling up to the required length. During this phase the glass will become solid because of the temperature difference between the crucibles, thus allowing the realization of the optical fiber. [5]

It is noted that the process described above is valid only for step-index fibers. If we want to create graded-index fibers, dopant should be directly inserted inside the liquid glass solution. [5]

1.6.2 Preform method

The preform method is the creation of the preform, a cylindrical bar made of two glass layers: the internal layer will become the core, while the external one will be the cladding during the creation of the preform, which will be discussed below. So it should have diameter larger than which will be obtained from the fiber. [5]

The preform is made by rotating a cylicndrical bar made of two glass layers as in figure 1.15. While this bar rotates, a flame is moved to create other layers in order to obtain the required size of core and cladding; the flame will be generated by a gas flow made of methane (CH4), oxygen (O2), silicon tetrachloride (SiCl4) and germanium tetrachloride (GeCl4). [3]

When the preform is ready, a spinning system is used to obtain the fiber as shown in figure 1.16. The preform is heated in a furnace and, acquiring a certain degree of viscosity, it collapses forming the core and the cladding. The fiber will have obtained the required diameter in this process and it will be covered by polymeric materials to protect the fiber from abrasion. [3, 5]


-•tjiik-K-t ithwnlrt

—(.ruii.uy Cualllit iii^lik-ilii

-nviiiny ciuing *y sletn —secondly coatug »pplifcal«

-tmUiijj curing »ystcui —iiKmtoi

Figure 1.16 Spinning system

SOOT PREFORM

E

Delivery Gases

Active oases ueiivery

y_|

K£S£3 FLAME DEPOSITION OF SOOT

Figure 1.15 Preform preparation

The process described above is valid only for step-index fibers. If we want to make graded-index fibers, we should make a set of layers each with the reffraction index less than the preceding layer. [3]


Figure 1.17 Attenuation Figure 1.18 Dispersion

The main issues of optical fibers are attenuation, dispersion and cost.

Attenuation reduces the amplitude of the signals propagating inside the fiber and it enhances increasing the diameter of the core. A graded-index fiber has an attenuation less than that of a step-index fiber, because the power is concentrated near the centre of the fiber (figure 1.17). [3]

Dispersion decreases the bandwidth and so increases the duration, rise and fall time of the signals propagating inside the fiber (figure 1.18).

Cost depends on the manufacturing times and the materials used for large-scale production of optical fibers.

So attenuation and dispersion degrade the optical link performance: if the optical link is longer, the degradation of the optical signals will be higher. Now let us examinate in detail the first two issues above.

1.7.1 Attenuation

Attenuation occurs:

1. inside optical fibers;

2. inside connections betweeen optical fibers.

Let us now examine the causes of these phenomenon.

1.7.1.1 Attenuation inside optical fibers

The main causes of attenuation inside optical fibers are:

• absorption;

• scattering;

• curvatures of fibers.

1.7 Issues

Absorption is the phenomenon by which light rays are removed by ions (figure 1.19). It is due to the presence of hydroxide ions (OH~), which penetrate into the fiber both as chemical sub-product during the manufacturing process of the fiber and as environmental humidity during the operational life of the fiber. Despite the complex treatment of the fibers, small quantities of substances will attenuate the light rays. However, absorption occurs more intensely only in some wavelength ranges. For this reason it has been demonstrated that optical fibers must operate on wavelength ranges in which absorption is minimum. These ranges are called windows, namely wavelength ranges in which absorption is minimum and so attenuation is limited. [3]


Figure 1.19 Absorption

With reference to the graph of figure 1.20 which shows the trend of the attenuation coefficient a for a GOF varying the wavelength, we can highlight 3 main windows in table 1.1:


Wavelength (nm)

Figure 1.20 Attenuation coefficient for GOF

 



Minimum

Attenuation of

Window

Wavelength range

attenuation coefficient

the strict 1 Km long

I

800 ^ 900 nm (near infrared)

2 dB/Km

2 dB

II

1250 ^ 1350 nm (infrared)

0, 4 dB/Km

0,4 dB

III

1500 ^ 1550 nm (farinfrared)

0, 2 dB/Km

0,2 dB

Table 1.1 Optical windows for GOF

Values about 0.2 dB/Km are not obtainable in copper cables (UTP, coaxial, etc.), because these types of link have the same values of least 2 dB/Km. So, with the same length of the link, attenuation in optical fibers is much lower than that of a metallic cable. This is one of the reasons for which the fibers will be useful for the future because the attenuation is very low in wide frequency bandwiths. Thanks to this, it is possibile to project long-distance links with a number of regenerators lower than that which would occur with a connection to the transmission metal link. [6]

Scattering is the phenomenon by which light rays are deviated from a straight trajectory on two or more paths due to localized non-uniformities in the medium. This situation occurs when the material of the core is not homogeneous (the refractive index of the core is not constant in the space). If this is true, the rays will leave the core and they will be dispersed inside the cladding (figure 1.21). In this way there will be some energy losses which are not removable, because they depend on the treatment of the material after the manufacturing process of the fiber. [7]


Figure 1.21 Scattering

Fiber curvatures are directional externally sourced deviations which cause dispersion of the incident ray towards the cladding. This situation occurs when the fiber is bent. Curvatures may be either macrobends or microbends. The former (figure 1.22) are visible to naked eye because they extend to very large arcs. This usually occurs during the installation of the optical cable. The latter (figure 1.23) are not visible to naked eye because they extend to very small arcs. This usually occurs when the fiber is mechanically stressed. [3]

Figure 1.22 Macrobend Figure 1.23 Microbend

1.7.1.2 Attenuation inside the connection between optical fibers

The main causes of attenuation inside the connection between optical fibers are:

• difference between refractive indices;

• difference between numeric apertures;

• difference between core diameters;

• axial alignment errors;

• axial angular errors;

• contact errors.

Passing from the core with index n1 to another core with index n2 where n1 = n2, the difference between the refractive indices of two fibers causes losses because the light ray will be transmitted only in part (figure 1.24). So there will not be any losses when n1 = n2.

Figure 1.24 Difference between refraction indices

The difference between numeric apertures causes losses when the first fiber has an NA larger than the second one, because some rays which come into the second fiber will be dispersed.

The difference between core diameters causes losses only if the diameter of the second fiber is less than the first one, because the losses are due to some rays which backwards (figure 1.25). In the opposite case, there will be not losses because this phenomenon will not occur.

T )

iw>j

Figure 1.25 Difference between core diameters

Axial alignment errors cause losses when the core axis does not coincide (figure 1.26).


Figure 1.26 Disallineamento assiale

Angular alignment errors occur when the axis of two connecting connecting fibers form a certain angle (figure 1.27).

Contact errors occur when the contact between the fibers is not perfect,

i.e. when two fibers are separated by an air gap (figure 1.28).

__






ni


ni



_*


Figure 1.28 [Accostamento]

1.7.2 Dispersion

Dispersion can be modal, chromatic (spectral) or spatial.

1.7.2.1 Modal dispersion

Figure 1.29 Modal dispersion inside SI fibers

Modal dispersion is the phenomenon by which the light rays do not arrive at the end of the fiber at the same time (figure 1.29). So among the various rays there will be a time difference, because a ray will propagate at a speed greater than that of the another rays which travels more slowly. Because of this, the form of the original signal will be time-spread and for this reason it will be distorted.

Let us discuss a way to avoid modal dispersion.

With reference to the scenario of figure 1.30 let us suppose that two rays propagate inside the core with different modes: the shortest ray is 1 and so it has an angle a = 0°, while the longest one is 2 and so it has an angle a = —max.

If we suppose that both paths travel inside the core at speed v = n0, we can

n1

obtain the expression of the delay between these rays as following:

t


Figure 1.30 Modal dispersion inside SI fibers

V = c0 with V

where

• l is the length of the single ray

• t is the time of the path of the single ray

n


At this point

ti = h n

1 1 c0

— with I2

S6n "max c0

sen"max

t2 = l2 "

z z co

At = t2 - ti = 2C0n2 NA

The final expression shows that the time difference between various rays depends on the propagation times of the single rays or on the characteristic parameters of the optical fiber. [8]

In multimode GI fibers rays will not propagate along straight lines, but along helical paths because of the gradual variation of the refraction index (figure 1.31). This means that the propagation time is constant in all rays and therefore the time difference At and so the modal dispersion are null. [1]

_n_

JL


Figure 1.31 Cancellation of modal dispersion in GI fibers

1.7.2.2 Chromatic dispersion

Chromatic (spectral) dispersion is the phenomenon by which the white

light transmitted into the fiber is decomposed into beams with different colours (red, orange, yellow, green, blue, indigo and violet). This situation is the same as when a beam strikes a prism as in figure 1.32). So the beams will assume different wavelengths and propagating velocities and for this reason the chromatic components will propagate inside the fiber with different times. However this causes the received signal will be different from the transmitted one. Therefore chromatic dispersion could be minimized injecting monochromatic light into the optical fiber. [6]


Figure 1.32 Chromatic dispersion

1.7.2.3 Spatial dispersion

Spatial dispersion is the phenomenon by which the rays disperse inside the cladding because of the impefections of the core-cladding interface (figure 1.33), which are usually due to manufacturing defects. [6]


Figure 1.33 Spatial dispersion

1.8 Advantages and disadvantages

With reference to what has been discussed in this chapter, we list the advantages and disadvantages of GOF according to metallic cables:

ADVANTAGES

• low attenuation

• high bandwidth

• low size and weight

• possibility to convey more signal with the same fiber

• absence of issues regarding electric security

• absence of crosstalk between near fibers

• absence of electromagnetic interference

• high resistance to temperature and environmental agents (rain, snow, smog, etc.)

DISADVANTAGES

• manufacturing cost

• installation cost

• difficulty in connecting

• difficulty of installation

Chapter 2

Optical communication systems

2.1 Architecture

Optical communication systems are sets of devices aimed at transmitting

and receiving information through propagation by optical fibers. An optical communication system is generally made of three essential components (figure 2.1):

• an optical transmitter;

• an optical link;

• an optical receiver.

[3]


Figure 2.1 Architecture of an optical communication system

The optical transmitter converts the input electric signal into a light signal to send it to the optical link. [3]

The optical link transports information flows generated by the transmitter towards the receiver. [3]

The optical receiver reconstructs the electrical signal from the signal sent from the optical link. [3]

To obtain the interoperability of the system, the components shown in figure 2.1 follow a common standard for electric, optical and timing requirements. In particular:

1. transmitter and receiver must use the same coding, modulation and multiplexing systems;

2. transmitter and receiver must use the same chip technology (TTL, ECL, etc.);

3. transmitter and receiver must follow the same optical requirements of the optical link (number of fibers, number of modes, etc.);

4. both transmitter and receiver must use the same type of connectors (SC, LC, ST, etc.).

[3]

Therefore optical systems must always follow the standard developed by standardization organizations such as ITU and ETSI. [9]

2.1.1 The optical transmitter

As can be seen in the figure 2.2, the transmitter has the following devices:

• a modulator;

• a polarization circuit;

• an optical source;

• a channel coupler.





POLARIZATION CIRCUIT





CARRIER












SUPPLY CURRENT






1

MODULATED


'

MODULATED


TRANSMISSION

INFORMATION

MODULATOR

SIGNAL (ELECTRIC)

OPTICAL

SIGNAL (OPTICAL)

CHANNEL

SIGNAL (OPTICAL)



SOURCE


COUPLER


Figure 2.2 Optical transmitter

A modulator is an electronic circuit which makes the information suitable for transmission by the optical channel. So the output signal of this device consists of a modulated signal according to a parameter of the information to be transmitted (amplitude, phase or frequency): the final result will be the frequency-shifting of the information spectrum to the highest frequencies. This can be obtained using an appropriate modulation technique.

Usually, in optical communication systems digital modulations (OOK, ASK, PSK, FSK, etc.) are more used than analogic ones because the former are more economical and efficient than the latter.

In particular intensity modulation consists of modulating the driving current of the optical source according to the information (see figure 2.3). [3]


MMftxfe ot Modulating LEDs or Laser Diodes

Figure 2.3 Circuit scheme of intensity modulator

4

There are two reasons for which modulation is preferred:

1. the information to be transmitted must be resistant to link noise and various disturbances (i.e. ISI);

2. the optical link will be shared through WDM systems, which will be discussed in paragraph 2.3.

If the modulator uses digital modulation technology, an encoder should be first connected and the decoder following. In this way information will be reconstructed by the decoder to detect and/or correct signal errors.

The most common coding techniques are: unipolar NRZ, polar NRZ, unipolar RZ, bipolar RZ and Manchester (figure 2.4).

BINARY DATA 0 10 0 1

volt

i

L

NRZ unipolar 0 encoding

Time

NRZ polar encoding

J

i

D

Ho

D

RZ polar encoding

i

ru

»

EL

n

tJ

RZ bipolar encoding

-A

l i i

i l

NRZ Manchester 0 encoding

: ir_ri .nr.

Figure 2.4 Channel coding schemes

The polarization circuit is a circuit which generates a stable polarization current for the optical source.

The optical source is an electronic component which generates light from the modulated signal. It can be a LED11 or LASER diode12, because they are easily coupled with the optical fiber and suitable for optical transmission (LED and LASER diode will be explained in paragraph 2.2). LED and LASER diode are encapsulated in a package which improves their coupling with the fiber. Thanks to this package the light intensity emitted inside the link will be maximized and thus the optical power will be maximized. [3]

The channel coupler (figure 2.5) is a device equipped with a microlens which focuses the signals generated by the source into the fiber. So it maximizes the power transmitted into the fiber and consequently the power which will arrive at the receiver. Thanks to this device optical receiving is improved.

fgfr»

2.1.2 The optical link

The optical link is made of one or more optical cables, which contains one or more fibers to convey different light rays. It generally produces a unidirectional or bidirectional communication between the transmitting and the receiving system.

2.1.3 The optical receiver

With reference to figure 2.6, the optical receiver contains the following devices:

• a channel coupler;

• a polarization circuit;

• a photodetector;

• one or more amplification systems;

• a demodulator.


Figure 2.6 Optical receiver scheme

The channel coupler focuses the signals from the optical link into the pho-todetector and so it maximizes the power injected by the photodetector and consequently optical power will arrive at the receiver. Thanks to this device optical receiving will be improved.

The polarization circuit sends a stable polarization current to the pho-todetector.

The photodetector is an electronic component which converts the incident radiation into an electric signal proportional to the former. It can be a PIN photodiode or APD13 because they are easily coupled with the optical fiber and so they are suitable for optical receiving (PIN photodiode and APD will be described in paragraph 2.2). [3]

Similarly to what happen to the LED and LASER diode, the PIN photodiode and APD are covered by a package to improve the coupling with the fiber. In this way photodetection will improve and thus the optical receiving will be accurate. [3]

The operating principle of the photodetector is based on the absorption of light by the material, which is the phenomenon by which each incident photon on the semiconductor material will be absorbed generating an electron-hole in the junction. This junction implies a certain output current of the device when it is under the action of an electric field from outside. This current is called photocurrent and it is equal to:

Ip = MRP

where:

• Ip is the photocurrent

• M is the gain of the photodetector (M = 1 for photodiode PIN, while M > 1 for APD)

• R is the responsivity14

• P is the incident optical power on the photodetector

It is noted that the photocurrent is graphically specified on datasheets according to the wavelength (figure 2.7). [3]

< 

<73

t ,0 >

!7i z o

Q.

V)

LU

DC


/



\








=t












°: I ;

WAVELENGTH [nm]

Figure 2.7 Responsivity of photodetector

The demodulator is an electronic circuit which performs the inverse task of the modulator. In other words, it reconstructs the information transmitted by the received signal, as well as that modulated by the transmitter. So information will be reconstructed using the same modulation technique as the transmitter.

If the demodulator uses digital technologies (ASK, PSK, FSK, etc.), then it is preferable that the decoder will be connected to it to improve the receiving. The decoder must follow the same coding scheme as the transmitter to ensure the operativity between the components of the optical system.

The amplifier stages generate an electric signal without noise from the output electric signal of the photodetector, which could be attenuated and distorted because of the non-ideal characteristics of the real channel. Figure 2.8 shows an example of a two-stage amplifier, where the first is a current-voltage converter while the second is a window comparator. It is noted that there are precautions for these devices to avoid saturation and so signal distortion. [3]

Current to voftage Converter


Figure 2.8 Photodetector connected in cascade with amplification stages

■^VW—

2.2 Notes on devices for optical systems

In the following subparagraphs we will discuss the main devices for optical systems.

2.2.1 LED

LED is a diode which emits incoherent light (see appendix A) when it is directly polarized. It exploits the optical properties of semiconductor materials to generate light radiation. This is due to spontaneous emission, namely the phenomenon by which photons, which are generated by the recombination of electrons and holes, are emitted from the diode junction in random propagation directions. [3, 9, 10]

LED is commonly based on a homojunction structure or on a hetero-junction. [3]

The first structure (figure 2.9) is made of two regions with the same material: the first is made up of a doped p-type semiconductor while the second of a doped n-type semiconductor; in this case the operating principle is practically the same as that of the p — n junction diode. [3]


Figure 2.9 Homojunction structure

So holes in a p-type region will move towards the n-type region and electrons will move towards a p-type region. At this point there will be many electron-hole couples in the junction, with consequent emission of photons in random directions and so of light towards the surrounding space. [3]

The second structure (figure 2.10) is similar to the above, except that the two regions are separated by an intrinsic material layer15; in this case the electric behaviour is not the same as that of the p — n junction diode. [3]

Figure 2.10 Heterojunction diode

As can be seen in the figure 2.11, p and n regions with maximum bandgap energy (Egi) give holes and electrons to the intrinsic region with another bandgap energy (Eg2). At this point in the intrinsic region holes and electrons will re-combine forming electron-hole couples and thus photons towards the surrounding space. So the radiation will be emitted in the intrinsic region of the LED. [3]

\ electrons electrons

^oooooo oooo_y r

Toooo

oooooo.

holes

Figure 2.11 Recombination of electron-hole

A LED is powered by DC direct polarization and its V-I characteristics recall silicon diodes (Si) but with various differences which depends on its materials (arsenicum, gallium, phosphorus, etc.). [3, 10]

Figure 2.12 compares the characteristics of LED technologies with silicon diodes, while table 2.1 shows the following parameters: maximum direct current (IFmax), direct voltage (VF) and wavelength (A); it is noted that the direct voltage of a LED is higher than that of a silicon diode and it is equal to 1,1 ^ 3,5 V according to it is made up of. [10]


Figure 2.12 V-I characteristics of LED technologies

 


Infrared

Red

Yellow

Green

Blue

Material

GaAs

GaAsP

GaP

GaP

SiC

1Fmax

150

100

60

60

50

Vf [V] with If = 20mA

1,1

1,6

2,1

2,1

3,5

/ [nm]

898

665

575

565

475

Table 2.1 Parameters of LED technologies

Also the I-P characteristics of an LED (current-power) is very similar to that of a silicon diode: the behaviour of the device is almost linear but it varies according to the operating temperature (figure 2.13). [11]



TG =273 K



-........

\jr


^298 K

343 K


\__Line

arzone



0 20 40 60 80 100

Drive current I (mA)

Figure 2.13 I-P characteristics of LED varying the operating temperature

According to the type of radiation emission, a LED can be an SLED16 type or an ELED17 type (figure 2.14). [12]


Figure 2.14 SLED (top) e ELED (bottom)

An SLED type is a LED which emits light radiations through a hole placed on its surface. It is protected by a plastic package which focuses the light as a lens. [12]

An ELED type is however a LED which emits light through a hole placed on its lateral surface. [12]

2.2.2 The LASER diode


Figure 2.15 LASER diode

The LASER diode (figure 2.15) is a diode which emits an intense coherent and monochromatic light (see appendix A). This radiation is, with some exceptions, concentrated in a straight beam when the device is directly polarized. It uses the optical properties of some semiconductor materials to generate light with the same properties. This is possible thanks to stimulated emission, which is generated inside the Fabry-Perot cavity for the electron-hole couples emitting. The cavity, which is placed between p-type and n-type regions, emits laterally like an ELED. So the structure of LASER diodes is very similar to that of ELEDs because the radiation will be generated laterally. [3, 13]

The LASER diode is based on a homojunction structure similar to that of the LED (figure 2.16), except that it contains a cavity capable of creating lateral photon emission. So the operating principle is similar to that of the LED, but the photon emission occurs inside the Fabry-Perot cavity. [3, 14]


Figure 2.16 Homojunction LASER diode (top) and heterojunction LASER diode (bot-

tom)

Like the LED, the LASER diode will be illuminated through DC direct polarization and its V-I characteristics are similar to those of the silicon diodes. However the LASER diode has very different I-P characteristics compared to that of the LED.

Figure 2.17 shows that optical power emitted by the LASER diode will only be generated above a threshold current. The behaviour of the device is linear and it varies by varying the operating temperature. [3, 11]


50 100

Drive current I (mA)

Figure 2.17 I-P characteristics of LASER diode at different operating temperatures

2.2.3 The PIN photodiode

\ v


Figure 2.18 PIN photodiode

Figure 2.19 PIN photodiode structure

The PIN photodiode (figure 2.18) is a photodiode with a heterojunction structure (figure 2.19), which is made up of p-type and n-type regions separated by an intrinsic material layer. So photons, impinging on a part of the semiconductor, will penetrate into the material of the photodiode and will then form electron-hole couples, producing an emission of photocurrent in the output device. [3]

2.2.4 The APD


Figure 2.20 APD

\ iii/

Figure 2.21 APD structure

The APD (figure 2.20) is essentially a photodiode with four layers of doped asymmetrically semiconductors (figure 2.21):

• doped p-type region;

• doped p+-type region;

• doped n+-type region;

• intrinsic region;

• doped n-type region.

[15]

p+ and n+ material regions represent the fundamental part of the APD because they cause the multiplicative effect of the charges. In other words, primary charges generated in the intrinsic zone for the photoconductive effect create an avalanche effect, which is able to generate a large number of secondary charges. These charges will generate the output current of the photodiode. [15] 2.2.5 The splitter

A splitter is a device which acquires the input optical flow and splits it into N outputs (N is equal to a power of 2). It is used to split the traffic in an optical network into different outputs. Figure 2.22 shows a 1:2 splitter, which takes an input optical signal and splits it into 2 outputs. [3]

r\


-T\

o —-


0


0



Figure 2.22 Splitter

2.2.6 The combiner

A combiner is a device which acquires N input optical signals and combines them into one single output (N is equal to a power of 2). It is used to merge traffic from different optical networks into a single output. Figure 2.23 shows a 2:1 combiner which combines 2 input optical signals into a single output. [3]

0 —



Ef-^

v

Figure 2.23 Combiner

2.2.7 The directional coupler

A directional coupler is a device used to bypass a station in case of malfunctions or to acquire optical power for the purposes of network monitoring (figure 2.24). It can be produced by fusion fiber (figure 2.25) or planar optical integration (figure 2.26). [3, 16]


Figure 2.24 Directional coupler

 


Figure 2.25 Directional coupler in fu- Figure 2.26 Directional coupler in pla-sion fiber nar optical integration

2.3 The WDM system

A WDM18 is the main technology for optical communication systems because it is specifically designed for these applications. It is also a multiplexing system which allows sharing the optical link dividing it into many sub-channels. Each of these sub-channels extends to a specific wavelength range divided from others by a guard band19. To modulate different sub-channels, we usually use different carriers with different wavelengths and the same modulation scheme. In this way it is possible to share the same channel among many users at the same time without causing mutual interference. [17]

Therefore a WDM is a sort of FDM20 but developed for optical multiplying. Currently it is the most used multiplexing technology in optical applications. This is true thanks to the vast capacity of traffic made possible: for instance, if an optical channel is made of 1024 fiber, with 1000 flow of 10 Gigabit/sec each one, then channel will have a capacity equal to:

C = 1024 • 1000 • 10 • 109 = 10, 24 • 1015 = 10, 24 Petabit/sec

So the optical channel is able to transport a vast amount of information. This implies that a WDM system with transport capacity and high performance components cannot generate bottlenecks.

As can be seen in figure 2.27, a WDM system uses a multiplexer on the transmission side to simultaneously send many signals, and a demultiplexer on the receiving side to split them. Filtering devices used on the receiving side are usually dichroic filters, namely optical filters with pass-band responses made of two plain glass plates placed in parallel. [17, 18]


Dichroic Filter

Figure 2.27 WDM system and dichroic filter

According to the spacing between the sub-channels in terms of wavelengths, WDM systems can be divided into three categories:

conventional WDM;

DWDM21;

CWDM22.

[19]

Conventional WDM provides up to 16 channels in the third window of GOFs (1500 ^ 1550 nm) with a channel spacing equal to 1 ^ 30 nm. [19]

DWDM provides up to 32 channels in the third window of GOFs with a channel spacing of 0.8 nm; a further DWDM variant exists which provides up to 40 channels with spacing of 0.4 nm: the UDWDM (Ultra Dense WDM). [19] CWDM provides up to 8 channels in the second and third window of GOFs (1250 ^ 1350 nm and 1500 ^ 1550 nm) but with a channel spacing of 20 nm. [20]

2.4 Issues

The main issues of optical communication systems are attenuation, Four-Wave Mixing (FWM), optical retro-reflection, synchronization, noise, wavelength conversion, cost and reliability of the optical source.

Let us examine in detail the issues concerning each one.

of the signals which arrive at the receiver

2.4.1 Attenuation

Attenuation reduces the amplitude and it may occur:

1. in the connections between the

optical source and the channel;

2. in the connections between the optical channel and the receiver;

3. in the channel;

4. in the splitter;

5. in the combiner;

6. in the directional coupler.

Connections between the optical source and the channel generally introduce strong losses and thus strong attenuation. These losses are due to the fact that the source emits light in various directions, in particular outside and/or inside the acceptance cone. The causes of attenuation in this point are mainly three:

1. only a part of the emitted light from the source will strike the core;

2. a part of the rays which strike inside the acceptance cone will be transmitted into the core regularly, the rest will disperse;

3. a part of the light which strikes the core will be reflected by it but will not penetrate into the fiber.

It is noted that attenuation in source-fiber coupling depends on the type of source (LED or LASER diode). Best results will be obtained using ELED and LASER diodes, because the emitted light will be conveyed in a narrower cone than that of SLED (figure 2.28). [3]

Ed9* «mlttlrvg LED

^1/ ^

I/: j.

Figure 2.28 Spatial distribution of SLED, ELED and LASER diodes

The connection point between the channel and the source is usually an area of light losses because most of the radiation which exists in the fiber will be picked up by the photodetector.

Losses inside the channel will be high when its distance and attenuation factor are rather high, because the attenuation of the channel depends on these parameters. These losses can be compensated by introducing a certain number of repeaters23 in the intermediate points of the optical link (see figure 2.29). [21]

Qj^QJ

TX



RX

Figure 2.29 Optical system with repeater

A splitter causes attenuation given by the following relationship:

A = 10 log10 N [dB]

where N denotes the number of splitter outputs and is equal to a power of 2: the attenuation of the splitter depends only on N.

The combiner causes attenuation too, because it is a reciprocal device of the splitter and so they behave symmetrically regardless of the technology. The attenuation of the combiner is given by:

A = 10 log10 N [dB]

where N denotes the number of combiner outputs and it is equal to a power of 2. So the attenuation of the splitter depends only on N.

The directional coupler introduces losses, in particular an excess loss of 0,1 = 1 dB and directivity loss of -40 = -50 dB.

2.4.2 Four-Wave Mixing (FWM)

FWM is an intermodulation which occurs because of the linear dependence of optical power on the refraction index of the fiber (Kerr effect). It introduces some spectral components which in part will overlap with WDM sub-channels, so generating receiving errors. To avoid this problem, it is necessary to increase the spacing between these sub-channels. [22]

2.4.3 Optical retro-reflections

Optical retro-reflections occur in bidirectional channels with intermediate devices such as splitters, combiners, connectors, optical junctions, etc. (figure 2.30). This problem can be solved by transmitting different wavelengths in various propagation directions. [16]


Figure 2.30 Optical retro-reflections inside an optical channel

2.4.4 Synchronization

In most cases digital data flows will be transmitted without clock for economic reasons. So the receiver must be able to reconstruct the clock from the receiving signal. To do this, the receiver will generate clock from an approximate reference frequency and then synchronizes it with the flow through PLL (see appendix B). This process is called clock recovering and its aim is to recover the clock from the receiving signal. [3]

To allow the correct working of this scheme, data flow must have frequent transitions in order to correct any receiving errors. To do this, it is necessary to use particular channel coding such as Manchester because it provides transitions for each bit interval. [3]

2.4.5 Noise

Noise develops principally in the receiver: the output signal of the transmitter almost always has a high signal-to-noise ratio. The noise margin of the optical link is negligible because the optical noise power density is -233 dBm/Hz (much less than the thermal noise power density, which is -174 dBm/Hz). [3, 6] The main causes of noise are:

1. the fluctuations of optical input, because photons pass into the input discretely and not continuously;

2. the photodiode, because it adds thermal noise during the spontaneous emission of photons;

3. the electronic schemes of amplifier stages, because there are four types of noise:

• thermal noise, namely the noise caused by the random motion of the electrodes inside a conductor;

• granular noise, namely the noise caused by the random variation of electron flow which moves in relation to the potential difference of the transistors;

• partition noise, namely the noise caused by fluctuations in the number of electrons which are collected by two or more electrodes;

• excess noise, namely the noise caused by fluctuations in the conductivity and the phenomenon of surface recombination.

[23]

2.4.6 Wavelength conversion

Wavelength conversion is obtained through semiconductor optical amplifiers. [24]

2.4.7 Cost

The cost depends on manufacturing time, materials and the components used for the production of optical devices.

2.4.8 The reliability of the optical source

The optical source is a component subject to optical and/or electrical failures. These anomalies can occur for instance because of current or temperature variations. For this reason we introduce the MTTF24, which is the parameter relevant for this issue. It denotes the average time after the optical source failure and it is given by:

MTTF = Ae^T [sec]

where

• A is the typical photodetector coefficient

Ea is the energy failure, which is equal to 0, 5 ^ 1, 2 eV

• k is the Stefan-Boltzmann constant, which is equal to 1,38 • 10_23 J/K

T is the temperature in K

[3]

It is noted that at environmental temperature (at T = 300 K ¥ 27 °C) MTTF is 25 years for a LED while it is 6 years for a LASER diode. [3]

2.5 Performance analysis

The performance of optical systems will be analyzed through the eye pattern (see appendix C) and by studying the system in terms of bit rate, signal-to-noise ratio, bit error ratio and optical power budget.

Bit rate denotes the number of bits transmitted by an optical link in a given time interval. Since information which travels in optical systems is digital, namely under form of bits, it is obvious that the rate is measured in bit/sec.

Signal-to-noise ratio is a numerical quantity that correlates the useful signal power with respect to that of the noise in a communication system (including optical noises). It is given by the following relationship:

SNR = PuPusignal [dB ]

Pnoise

where

SNR is the signal-to-noise ratio

• Pusefuisignai is the useful signal

Pnoise is the noise power in the optical system

This relationship shows that the signal-to-noise ratio is very important in the design of optical systems: the greater is the BER found, the greater is the useful signal power and thus the quality of the received signal is better.

The minimum threshold of SNR depends on receiving technologies; the objective for the design of optical systems is to guarantee the receiver a high SNR.

BER is the ratio between the number of bits received incorrectly by the receiver and the number of bits sent by the transmitter. It is a parameter qualitatively related to the SNR: the greater is the BER found, the smaller is the SNR and vice versa.

The optical power budget represents the difference between the output power of the transmitter and receiver sensitivity. It is a parameter used to establish if the received power is such as to ensure the desired SNR (or equiv-alently BER). The received power is the optical power of the transmitter subtracting the link loss and design margin.

Chapter 3

Plastic Optical Fibers (POF)

3.1 Structure

POFs are made of a core in PMMA25 (figure 3.1) and a cladding in fluorinated polymer. They work basically like the GOFs, but some differences between POFs and GOFs will be highlighted below. [25]

CHj CHj CH3 CH3 -C — CHj —C — CHj—C— CHJ — C— CHj —

c c c c

# \ ^ \ \

° och3 0 och,° och,° och3

Figure 3.1 Molecular structure of PMMA

The diameter of a POF is 980 mm, while that of the cladding is 1000 mm for requirements of mechanical and thermal resistance (figure 3.2): PMMA has a traction force equal to 70 = 80 N/mm26. [8, 25]


Figure 3.2 Comparison between POF and GOF

Thus POFs have a maximum number of modes and mechanical features better than GOFs in high size and mechanical resistance. As explained in chapter 1:

• monomode GOFs are made up of a core with a diameter of 8^10 mm and a cladding with a diameter of 125 mm;

• multimode GOFs are made up of a core with a diameter of 50 or 62.5 mm and a cladding of 125 mm; glass has a traction force of 33 N/mm2.

[3, 26]

3.2 Manufacturing

There are three common manufacturing methods of POFs:

• the continuous extrusion method;

• the coextrusion method;

• the preform method.

3.2.1 The continuous extrusion method

The continuous extrusion method (figure 3.3) consists in injecting MMA inside a high temperature reactor, which produces the core material by the polymerization of MMA (monomer). When the PMMA is ready, the solution produced is pushed through a gear pump in an extruder. The impurities in the compound will evaporate into the extruder and they will flow into the reactor to be reused. The fiber is obtained by a spinning block, in which the core will be coated by the cladding (obtained in the same way by an extrusion process). [27]

monomer feed


Figure 3.3 Continuous extrusion method

3.2.2 The coextrusion method

The coextrusion method (figure 3.4) is a process used to create graded-index POFs. It consists in injecting the core and cladding materials inside high temperature feeders. Then the two compounds will be placed in a coextruder, from which a filament of the two layers will be generated. During this process some monomers will be injected radially inside the fiber in order to obtain a graded-index fiber. Then the plastic material will be spun, producing the GI plastic fiber. [27]


Figure 3.4 Coextrusion method

3.2.3 The preform method

The preform method is similar to that seen in paragraph 1.6.2, but it employs the use of plastic materials rather than glass. [27]

3.3 Main characteristics

The main characteristics of POFs are numerous and for this reason they will be described in detail in the following subparagraph. Then we will list other properties, such as modal coupling, modal conversion, mechanical strength and thermal resistance.

3.3.1 Attenuation

The graph in figure 3.5 shows trends in the attenuation coefficient a varying the wavelength for POF SI and POF GI. From this graph it is important to highlight 4 transmission windows for each type of POF. [28]


Gl-POF

SI-POF

Figure 3.5 Curve of attenuation coefficient for POF SI and POF GI

Tables 3.1 and 3.2 list the characteristics of the 4 transmission windows shown in figure 3.5.

A comparison of tables 1.1, 3.1 and 3.2 shows that POFs have a minimum attenuation on wavelengths different from those of GOFs. In particular, graded-index POFs have attenuation higher than step-index POFs because of the doping materials are used to obtain a gradual profile of the refractive index. [28]


Wavelength

Minimum

Attenuation of

Attenuation of

Window

attenuation

the strict 1

the strict 100


range

coefficient

Km long

m long

I

e

o ^ ^ s •I- cS

o 10

90 dB/Km (0, 09 dB/m)

90 dB

9 dB

II

490 ^ 530 nm (green)

75 dB/Km (0, 075 dB/m)

75 dB

7, 5 dB

III

565 ^ 590 nm (yellow)

75 dB/Km (0, 075 dB/m)

75 dB

7, 5 dB

IV

630 ^ 760 nm

(red)

150 dB/Km (0,15 dB/m)

150 dB

15 dB

Table 3.1 Transmission windows for POF SI


Wavelength

Minimum

Attenuation of

Attenuation of

Window

attenuation

the strict 1

the strict 100


range

coefficient

Km long

m long

I

nm

0) 9e ^ s

•I- cS

0 5 4

400 dB/Km (0, 4 dB/m)

400 dB

40 dB

II

490 ^ 530 nm (green)

348 dB/Km (0, 348 dB/m)

348 dB

34,8 dB

III

565 ^ 590 nm (yellow)

225 dB/Km (0, 225 dB/m)

225 dB

22, 5 dB

IV

630 ^ 760 nm

(red)

198 dB/Km (0,198 dB/m)

198 dB

19,8 dB

Table 3.2 Transmission windows for POF GI

This means that:

• similar values in GOFs are not obtainable in POFs: POFs and GOFs are not compatible because PMMA is made of a C == O or C — H molecular bond, which increases the optical loss in the infrared region. As we will see in chapter 4, it is possible to connect a GOF cable to a POF through special devices;

• SI and GI POFs are not suitable for kilometer-long distances (to be clear, those extended over long distances for the provision of network broadband access);

• only SI POFs are suitable for hundreds meter long distances (to be clear, those installed in residential home and business applications).

[8, 29, 30]

As regards losses due to curvature, POFs are subject to curvature losses up to a certain value of the curvature radius according to the profile. Figures 3.6 and 3.7 respectively show the loss in function of the curvature ray for SI and GI POFs. In the former the attenuation are 0,5^0,01 dB if the radius is 5^29 mm but null if the radius is greater than 30 mm; in the latter, the attenuation is 0,5 dB only if the radius is greater than 50 mm. [8]


Figure 3.7 Curvature losses of GI POFs

ben

d los:

;es [e

B]







































J=f

>—f

be

id rac

ius [n

nm]

0 10 20 30 40 50 60 70 80

Figure 3.6 Curvature losses of SI POFs

As regards the other types of losses (attenuation inside connections, between the transmitter and the fiber, between the fiber and the receiver), POFs are subject to these types of losses but they will be less than those of GOFs, because the core diameter is greater than that of GOFs.

Therefore the transmission of visible light by SI POF is the best solution for domestic and business applications. For this reason it is best to use LED or LASER diodes which generate blue, green, yellow or red light.

3.3.2 Dispersion

As we all know, modal dispersion is caused by rays which arrive at the fiber end at different times. [1]

As regards SI POFs, it has been shown that their dispersion is negligible for the following reasons:

1. experiments have demonstrated that spatial dispersion increases the bandwidth. This occurs because the energy redistribution between different modes allows slow rays to travel fast or vice-versa;

2. when the POFs are curved, at the start and end of the curve there are two straight tracts and so we can adopt a suitable system according to the curvature. This implies a modal conversion (figure 3.8), namely the phenomenon by which the propagation direction of the light changes when the POF is curved.

[8]

3.3.3 Mechanical performance

With reference to paragraph 3.1, the mechanical characteristics of POFs derive directly from their plastic materials. For this reason POFs have a very strong mechanical performance. [8]

Figure 3.9 shows that:

• a force of 250 N at -40°C causes an elongation of about 20 mm, with a consequent fall in the percentage of transmission27 below 50%;

• a force of about 150 N at +23°C causes an elongation of about 190 mm, with consequent fall in the percentage of transmission below 50%;

• a force of 50 N at +85°C causes an elongation of about 220 mm, with consequent fall in the percentage of transmission of about 85%.


0 50 100 150 200 250 0 50 100 150 200 250

Figure 3.9 Traction performance of POFs

[8]

Figure 3.10 shows that the transmission of flexed POFs will degrade in the following cases:

• after 4.000 flexions with a curvature ray of 20 mm and a sheath cable in PE28;

• after 400.000 flexions with a curvature ray of 40 mm and a sheath cable in PE;

• after 1.000.000 flexions with a curvature ray of 40 mm and a sheath cable in PA29.

110 100 90 80 70 60 50

, , mn - transmiss

I I III

ion [%]
























































-R -R -R


20 in

nm (P nm (P nm (P

E)

F)








= 40

A)








.....1 ........1

1 10 100 1,000 10,000 100,000 1,000,000

number of flexing cycles

Figure 3.10 Flexion performance of POFs

[8]

Finally, figure 3.11 shows that the transmission of POF under torsion will degrade in the following cases:

• after 600 torsions at 40°C;

• after 2.500 torsions at 23°C;

• after 10.000 torsions at 85°C.

transmission

:%]
















\



4

s







1

































-T -T -T

~

-40°

c

r>










+85°C/85%

I

H.





number of torsion cycles

10 100 1,000 10,000 100,000

Figure 3.11 Torsion performance of POFs

[8] 3.4 Connections

POFs can be connected by optical connectors, mechanical joints or fusion joints.

As regards optical connectors, POFs can be connected to the same connectors seen in paragraph 1.5 and also by: F07 (in plastic, in a rectangular shape), OptoLock (in plastic, with a block-plug), RCC45 (in plastic, in a square shape) and SMI (in plastic, in a pseudo-rectangular shape). [31]

Figure 3.12 shows the types of connectors noted above:

n

OptoLock SMI

Figure 3.12 POF connectors

OptoLock is the most commonly used connector because it is economical and practical to use. The design of this connector allows the cutting of POFs at the exact length required by the application. So it helps inexperienced users to terminate the cable in naked fiber (see figure 3.13). Thus it offers the advantages of optical fibers with DIY simplicity and low costs to domestic users. [32] 3.5 Installation


RCC45


€> Insert

© Link

Figure 3.13 POF connections with OptoLock

o

Slice

The installation of POFs is carried out taking advantage of existing ceilings and ducts inside or outside the walls of a building, in particular in ducts used for electrical telephone wiring (figure 3.14). This is possible thanks to the size, the capacity of electromagnetic immunity and the optimum mechanical characteristics of plastic fibers. [32]

Therefore POFs are non-invasive and they do not impact on performance.

[32]

POF CABLE

i


Figure 3.14 Ducts for POF

Chapter 4

Applications of POFs

4.1 Ethernet over POF (EoPOF)

When discussing GOF for the wiring of LAN, the first two things that come to mind are the need for high performance and high cost. These aspects, which are due to the fact that we are talking about a very powerful technology but expensive and difficult to install, often forces us to pose UTP cables in large quantities to enable telematics infrastructure to evolve depending on the size of the infrastructure. The low cost and simplicity of installation seem to be the main points in favor of UTP cables, but we shall not neglect the rapid saturation of the capacity to accommodate these cables in ducts and false ceilings. This is due to the bulkiness of the cable, which is normally made up of four pairs of twisted wires. In this case it is possible that additional time and expense related to the enhancement of carrier infrastructure (ducts and false ceilings), will lead to a substantial increase in the time and cost of installation.

4.1.1 Characteristics

Ethernet over POF is the alternative of Fast Ethernet30 over UTP. It also a standard for LAN31 which provides:

• the use of a pair of POFs of length equal to or greater than 100 m, of which one is for transmission and the other for receiving (bidirectional link);

• the use of network devices equipped with LEDs or class 1 LASER diodes32to avoid damaging the eyes;

• the use of bidirectional connectors for POF such as MT-RJ, F07, Op-toLock, RCC45 and SMI (most EoPOF devices are equipped with Op-toLock for their low cost and practicality);

• the base band transmission with 4B/5B33 and MLT-334 coding: the former makes the transmitted signal immune to noise, while the latter allows it to compensate for the total band required by the system;

• a transmission speed of 100 Mbps;

• WDM in visible spectrum [32, 33, 34, 35]

It is noted that EoPOF does not use CSMA/CD35 because it exploits bidirectional communication thanks to the pair of plastic optical fibers. So there will be neither collisions nor retroflections. [36]

As we will see in subparagraphs 4.3.2 and 4.3.3, experiments have succeeded in bringing EoPOF to a speed greater than 1 Gbps on 100 m long fibers. [37, 38] The PDU36 of Ethernet over POF has the format shown in figure 4.1, which is very similar to a conventional Ethernet frame.

It is made of the following elements:

• preamble: string 10101010 which allows clock recovering by the receiver;

SF (start frame): the initial part of the frame;

destination address: the MAC address of receiver;

source address: the MAC address of transmitter;

frame length: length of the field data;

data: the information to be transmitted;

PAD (padding): ensures that the minimum length of the frames is not less than 64 bytes;

FCS (frame check sequence): allows the receiver to correct and detect frame errors.

f fFCsf

V+ 7 16 6 1 O-lSOO 0-4* A

^ Length in byte

Figure 4.1 Ethernet over POF PDU

[39]

4.1.2 Devices

The main devices for EoPOF networks are:

• the media converter;

• the USB-POF adapters;

• the switch;

• the router;

• the EoPOF network card.

The media converter is a device which makes an interface between the POF and the UTP cable (figure 4.2) or between the POF and the GOF (figure 4.3). Thus they can be used:

• to implement a point-to-point connection between terminals which do not use the same cable;

• to link a FTTH37 access point to a EoPOF network (figure 4.4).

Figure 4.2 POF-UTP media converter Figure 4.3 POF-GOF media converter

FTTH Access Pa ml > Termination


POF - Home Netvwr* Terminaton / Interface

AcMtre Media -Convener Patch Pane* POF - RJ45 100 Mt>ps — REG Mounted Type OMS 100 REG

CPE.

Access Router Etc

Figure 4.4 Cabling between GOF and EoPOF networks through an FTTH access point and media converters

[30, 40]

The USB-POF adapter (figure 4.5) is a device which provides an interface between a USB port of a personal computer and a POF cable. It may support WakeOnLAN38. [40]

Figure 4.5 USB-POF adapter

The switch (figure 4.6) is a device which selectively forwards the received frames from the RJ4539 or POF input towards a POF output port, which is towards a specific destination. The switch makes point-multipoint connections and it usually has a RJ45 port and a number of POF interfaces. [40]


Figure 4.6 EoPOF switch

The router (figure 4.7) is a network device which forwards data packets between EoPOF nodes. It provides many types of scenarios such as point-to-point links, point-to-multipoint links, bus topology, tree topology, etc. The router usually has a RJ45 port and POF interfaces, but it may also have a number of antennas for wireless network access. [40, 41]


Figure 4.7 EoPOF router

The EoPOF network card (figure 4.8) is a device which is usually installed on the PCI slot of a personal computer to create an interface between the PCI bus and the POF. Generally it is plug and play40 and provides a compatible driver with GNU/Linux, Apple Mac OS X and Microsoft Windows. [40]

Figure 4.8 EoPOF network card

Thus the devices cited above are able to construct different topologies in perfect continuity with an existing LAN in an absolutely transparent way as regards the services and protocols to be transmitted. They can also be configured in the same way as conventional Ethernet over UTP and wireless Ethernet devices. [32]

4.1.3 Installation costs and time

To establish the installation costs and time of an EoPOF network, let us suppose that we want to wire a LAN with 8 terminals, a switch and 120 m POF cables as in figure 4.9.

In the installation cost, in addition to considering the unit prices of the components of the LAN (switch, wall outlets, etc.), it is also necessary to consider:

• external ducts to facilitate the drafting of UTP cables (not required for POF because they can be passed in existing ceilings and ducts inside or outside the walls of a building);

• the time taken for the preparation of 120 m cable (2h for UTP, 0,5 h for POF);

• the cost of cable laying, which is equal to about 60 $/h;

• the time taken for a single cable termination (5 $/minute for UTP, 1 $/minute for POF);

• the cost of a single cable termination ($1 both for UTP and POF);

• the time and cost needed for the maintenance of the network, which are lower for the POF thanks to the features highlighted in the chapter 3.

[30]

So the termination cost is $1x5 $/minute = $ 5 for UTP and $1x1 $/minute = $ 1 for POF. [30]

The wiring cost using UTP and POF are summarized respectively in tables 4.1 and 4.2:

1

8 port switch for UTP ca

bles




$ 180,00

120 m

UTP cable


(120,00 m

x 0,88 $

/m = $ 106,60)

$ 106,60

120 m

External duct


(120,00 m

x 5 $ /r

n = $ 600,00)

$ 600,00

2 h

Cable installation


(2 h x 60 $

/h = $

120,00)

$ 120,00

8

RJ45 wall outlets


(8 x $ 4,91

= $ 39,28)

$ 39,28

16

RJ45 UTP terminations


6

x $

5

= $ 80)


$ 80,00

TOTAL

$ 1.125,88

Table 4.1 Analysis wiring cost through UTP cables

 

1

8 port POF switch

$ 320,00

120 m

POF cable

(120,00 m x

1 $ /m = $ 120,00)

$ 120,00

0,5 h

Cable installation

(0,5 h x 60

/h = $ 30,00)

$ 30,00

8

POF wall outlets

(8 x $ 75 =

$ 600,00)

$ 675,00

16

OptoLock POF connectors

6

x $

=

$ 16)

$ 16,00

TOTAL

$ 1.086,00

Table 4.2 Analysis wiring cost through POF cables

[30]

These results shows that POF wiring is more advantageous than that with UTP cable, because it provides lower cost and time installation than that of UTP wiring. It is noted that the prices shown above are indicative and they will be reduced as the technology is adopted and spreads more widely in various users. The example above is applicable to any other POF application.

4.2 Other applications

There are many specific applications for which the use of POF is suitable for their characteristics highlighted in chapter 3. This means that POF applications do not regard only LAN, but also sectors: from video surveillance to home automation, from aeronautics to naval, from industrial automation to robotics, and so on. [42]

Moreover, it is likely that in the future there will be further applications of POF involving other sectors (e.g. in wireless networks). [42]

4.3 Organizations and projects

For a number of years POFs have raised the interest of many researchers and companies around the world. For this reason some organizations such as POFTO41, an international organization which promotes the development and diffusion of POF all over the world, has been formed. In Europe some projects have been created and funded by the European Union to further this dynamic:

• POF-ALL, started in 2006 and was followed two years later by POF-PLUS (ended in 2011);

• POLYCOM, started in 2006 and ended in 2009.

[37, 38, 43, 44]

In the following subparagraphs we will describe POFTO and various other projects.

4.3.1 POFTO

POFTO was responsible for the following activities:

• the diffusion of technical and practical knowledge of POFs amongst companies and individuals;

• training about POF aimed at companies and/or individuals;

• the active promotion of POF in IT;

• the development of standards for the implementation and installation of POF systems.

[43]

Different companies and universities take part in this organization, among then:

• Electronic BRICKS42;

• Firecomms Ltd.43;

• Luceat S.p.A.44;

• Mitsubishi International Corp.;

• Fraunhofer Institute;

• California University (Irvine headquarters).

[43]

4.3.2 POF-ALL

POF-ALL aimed at developing low-cost technologies based on POF that allow access to the Internet at speeds much higher than those of a normal ADSL connection. [37]

The partners of POF-ALL were:

• Istituto Superiore Mario Boella45;

• Luceat S.p.A.;

• DieMount GmbH46;

• POF Application Center47;

• Fraunhofer Institute;

• Duisburg-Essen University;

• Eindhoven University;

• Fastweb S.p.A.;

• Siemens;

• Teleconnect GmbH48;

• ST Microelectronics.

[37]

The main activities of POF-ALL were:

• experimentation on a system of transmission of 100 Mbps across connections of 200-300 m in length;

• experimentation on a system of transmission of 1 Gbps connections across a length exceeding 100 m;

• research and development of transmitters and receivers for POF.

[37]

Further developments on similar topics are the subject of the POF-PLUS project, coordinated by the same partners as POF-ALL. [38]

4.3.3 POF-PLUS

POF-PLUS focalises on developments similar to those of POF-ALL, in particular on POF communication system with Gbps speed. It promised the same impact expected by POF-ALL. [38]

The main partners of this project were:

• Istituto Superiore Mario Boella;

• Luceat S.p.A.;

• POF Application Center;

• Fraunhofer Institute;

• Firecomms Ltd.;

• Rosetta IP Ltd.49;

• Telecom Italia Lab50;

• France Telecom;

• Eindhoven University.

[38]

The activities of the project have produced numerous concrete results. These include an experiment on transmission systems with speeds of 6^10 Gbps across 50-60 m long links. [38]

The success of the experiments by POF-PLUS will impact significantly in the following areas of application:

• LAN home and business at speeds in the order of Gbps (POF-PLUS will allow the network to match the performance corresponding to that of FTTH outer delivered to the end customer);

• cluster51;

• SAN52.

[38]

4.3.4 POLYCOM

POLYCOM had as its goal the development of POF and low-cost devices capable of amplification and optical switching at very high speed. Funded by the European Union and interested in following the trend of developments and studies on POF, POLYCOM thinks the impact of POFs were similar to that of POF-ALL and POF-PLUS. [44]

The partners in this project were:

• Polytechnic University of Milan;

• Luceat S.p.A.;

• Sheffield University;

• Imperial College;

• Wildau University;

• Lisbona University.

[44]

The activities of POLYCOM produced the following results:

• a demonstration of the ability of high-gain and high-speed optical switching of the PFO 53;

• the creation of a POF doped with PFO, with very low losses and high speed optical switching;

• the creation of PFO LASER diodes, with high speed optical amplification and switching.

[44, 45]

Chapter 5

Conclusions

To improve the performance and the cost of certain technologies, substitutes often have been sought. Sometimes these will not give the expected result but can open the way for unpredictable innovations and they may also be complementary to existing technologies.

Plastic optical fibers enter this scenario because they can substitute metal cables thanks to those properties illustrated in chapter 3 and those characteristics not present in glass optical fibers. Thus the size, low loss curvature, low modal dispersion and excellent mechanical properties of POFs have favoured their use in those applications illustrated in chapter 4.

Nevertheless, it has not been possible yet for POFs to rival GOFs in some fields, such as that of attenuation. POFs perform well over short distances whereas GOFs are suitable for the quality of their large-range transmission. Thus POFs are not yet worthy substitutes for GOFs but both can substitute metal cables.

The interest in POF is anyway recent and for this reason the current studies and research are just the beginning of a new era for this technology. The fact that POF is innovative in several areas will undoubtedly lead to their continuous development and an improvement in their characteristics.

Appendix A

Notes on coherent light


Figura A.2 Coherent light

A light beam is made up of many wave packets, each with its own phase. Therefore the peaks and troughs of the various wave packets do not match (figure A.1). [6]

frequency


frequency

Figura A.1 Light wave packets

From a spatial point of view, a beam light is coherent when the phase difference between the various wave packets is null and constant in time (figure A.2). [6]

It follows that a beam light is incoherent when the phase difference between various wave packets is neither null nor constant in time (figure A.3).


Figura A.3 Incoherent light

Appendix B

PLL (Phase Locked Loop)

PLL54 is a circuit used to recover the clock signal in telecommunication applications. It consists of a negative retroaction system made of three devices (figure B.1):

• a phase comparator;

• a low-pass filter;

• a VCO55.

[46]


Figure B.1 PLL scheme

The phase comparator compares the phase of the input signal (^) with that of the output signal generated by (^0). It is made of a XOR port and it gives an output signal ve proportional to the input-output phase difference (A^ = y>i - ^0), where ve denotes the error signal. [46]

The low-pass filter removes the frequencies from the error signal extracting its average vem. It is based on a first order RC circuit. [46]

The VCO generates a signal with frequency fo proportional to the value of

vem. [46]

A PLL works as follows:

The input frequency fi will be compared with that of the VCO output signal and the output of the comparator is a function of A^>. If fi = f0 the output of the comparator is the sum between the two signals with frequencies fi — f0 and fi + fo. The low-pass filter will attenuate strongly fi + f0, so the error signal, which will drive the VCO, will have a frequency of fi — f0. At this point the frequency can be made proportional to the control voltage. The frequency of the oscillator will be varied around fo by the error signal. Finally, if the input signal is in the locked phase range, the oscillator will generate a frequency of fi. [23]

Appendix C Eye pattern

Eye pattern is a diagram used generally to evaluate the performance of optical systems. The term derives from the shape of the diagram, which is similar to that of an eye. Diagrams are shown using an oscilloscope applying the data sequence from a system to the vertical input of the instrument, while the time reference is synchronized with the transmission data rate: the screen of the oscilloscope will show overlapped individual pulses with the deformations produced due to the noise of the system.

Figure C.1 shows an example of an eye pattern for a binary sequence, obtained under the following assumptions:

• the system has altered the rising and falling edges but the adjacent bits are not influenced (no intersymbol interference);

• the time reference of the oscilloscope is synchronized to the signal;

• the trace starts from time instances t0,t3,...

Figure C.1 First example of eye pattern


fo '? 'j '< '5

Figure C.2 shows another example of an eye pattern, from which we can highlight some characteristics of the transmission system:

• the opening of the pattern denotes a region without impulses: the length of the segment AA' allows the evaluation of the noise margin of the system: the fact that AA is shorter than the ideal situation implies that DD indicates a reduction in impulse width and so a deterioration in the SNR;

• the segment DD', which represents the maximum opening of the eye pattern, denotes the optimum sample conditions;

• the rapidity with which the eye is closed, namely whether shape of the segment BAB' is more or less accentuated, indicates the sensitivity of the system to an error at the instant of sampling variation;

• the length of the segment BC gives the maximum variation of the zero crossing and so allows the detection of an error with which the signal transitions between 1 and 0 can be obtained in time at the exit.

D


Figure C.2 Second example of eye pattern

[23]

Acknowledgements

I think acknownledgements are not a formal conclusion of a thesis, but the knowledge that in this time of hard work we are not alone.

For this reason, I would first like to thank Prof. Eng. Francesco Giuseppe Della Corte for the constant, active and kind professionalism shown to me. I also thank Eng. Valter Foresto, president and founder of ElectronicBricks, and Prof. Eng. Roberto Gaudino, professor at the Polytechnic University of Turin and scientific coordinator of POF-ALL and POF-PLUS projects, for providing useful information for the realization of the thesis.

Bibliography

[1] http://it.wikipedia.org/wiki/Fibra_ottica

[2] http://luigi.bonavoglia.eu/fibra_ottica.phtml

[3] Green L. D., Fiber Optic Communications, CRC Press

[4] http://www.ilmondodelletelecomunicazioni.it/fibre_ottiche/ richiami_di_ottica.htm

[5] http://it.emcelettronica.com/le-fibre-ottiche-principio-di-funzionamento-e-metodi-di-fabbricazione

[6] http://users.libero.it/sandry/fibre.pdf

[7] http://www.xraptor.it/Elettronica.htm

[8] Ziemann O., Krauser J., Zamzow P. E. & Daum W., POF Handbook: Optical Short Range Transmission Systems, Springer

[9] http://it.wikipedia.org/wiki/Comunicazioni_in_fibra_ottica

[10] Mirandola S., Corso di Elettronica - Elettronica analogica con laboratorio, Calderini Edagricole

[11] http://www.iet.unipi.it/rn.luise/com_ottiche/cap2Frm.html

[12] http://www.iet.unipi.it/p.bagnoli/led.doc

[13] http://it.wikipedia.org/wiki/Laser

[14] Sze M. S., Dispositivi a semiconduttore, Hoepli

[15] http://it.wikipedia.org/wiki/Fotodiodo_a_valanga

[16] http://home.dei.polimi.it/maier/ro_PON_0910_v01.pdf

[17] http://it.wikipedia.org/wiki/Wavelength_Division_Multiplexing [ 18] http://it.wikipedia.org/wiki/Filtro_dicroico

[19] DeCusantis C., Fiber Optic Data Communication: Technological Trends and Advances, Academic Press

[20] http://searchnetworking.techtarget.com/definition/coarse-wavelength-division-multiplexing

[21] http://web.tiscali.it/federicomoretti/amplificatori_ottici.htm

[22] http://net.infocom.uniroma1.it/corsi/ro/materiale/dispWDM.pdf

[23] Biondo G. & Sacchi E., Manuale di Elettronica e Telecomunicazioni, Hoepli

[24] http://it.wikipedia.org/wiki/Amplificatore_ottico

[25] http://litewire.it/tecnologia/

[26] http://en.wikipedia.org/wiki/Ultimate_strength

[27] http://www.engineers.or.kr/2005pof/2005pof-4.pdf

[28] http://www.telecommunicationsengineering.it/uploads/Seminars/

Visani_2.pdf

[29] http://it.wikipedia.org/wiki/Fibre_ottiche_polimeriche

[30] http://t-k.it/obj/Presentazione%20POF.pdf

[31] http://www.pofac.de/downloads/itgfg/fgt24/ FGT24_Berlin_Mulligan_POF-Connectors.pdf

[32] http://www.consoft.it/news/pdfStampa/POF_pag.44.pdf

[33] http://www.cartesio-luxemburg.net/listing/didattica/cisco/CCNA1/ CCNA1-Cap07-Ethernet%20Technologies_2pg.pdf

[34] http://www.electronicbricks.it/contenuti/pdf/

etsi_ts_10517501v010101p.pdf

[35] http://www.civinini.net/index.php/varie-kbmenu-34/123-ethernet-uno-sguardo-alla-base-delle-reti

[36] http://www.ge.infn.it/ brunengo/corsi/alet/Lezione%2009.pdf

[37] http://www.ist-pof-all.org/

[38] http://www.ict-pof-plus.eu/

[39] http://www.dii.unisi.it/ benelli/scienze_della_comunicazione/dispense/ 2004_05/FastEthernet.pdf

[40] http://www.ebricks.it/p/prodotti.html

[41] http://hwaying.en.alibaba.com/product/329421232-210253677/POF_Wireless_Broadband_Router.html

[42] http: //litewire.it/applicazioni/

[43] http://www.pofto.org/home/nodeZ9

[44] http://www.fisi.polimi.it/polycom

[45] http://cordis.europa.eu/documents/documentlibrary/115484541EN6.pdf

[46] Mirandola S., Corso di Elettronica - Le conversioni elettroniche con laboratorio, Calderini Edagricole

85

1

Total internal reflection is a phenomenon by which a light ray which strikes a separation surface between two mediums with refraction indexes m and n2, a part of it penetrates inside the medium n2 and the another part is partially reflected by the same surface (see paragraph 1.2).

2

A gastroscope is a medical instrument used to visualize the inside of the upper digestive

tract (esophagus, stomach and duodenum).

3

STC Ltd. (acronym which stands for Standard Telephone and Cable Ltd.) was a British

telecommunication company acquired in 1991 by Nortel Networks Co.

4

Coring Glass Inc., American company specialized in glass manufacturing.

5

GOF, acronym which stands for Glass Optical Fiber.

6

POF, acronym which stands for Plastic Optical Fiber.

7

Snell's laws are theories that describe the behavior of rays travelling within mediums with different refractive indices.

8

Limit angle represents the incident angle corresponding to a refraction angle Orefr = 90°.

9

NA, acronym which stands for Numeric Aperture.

10

The number of modes denotes the number of light rays propagated inside a waveguide like an optical fiber.

11

i i i

12

LASER, acronym which stands for Light Amplification by Stimulated Emission of Radiation.

13

APD, acronym which stands for Avalanche PhotoDiode.

14

Responsivity is the ratio between the photocurrent generated by the photodetector and the incident optical power. It is a function of the wavelength.

15

Intrinsic material is a semiconductor material which contains a small quantity of impurities compared to the number of electrons and holes generated thermally. [14]

16

SLED, acronym which stands for Surface LED.

17

ELED, acronym which stands for Edge LED.

18

WDM, acronym which stands for Wavelength Division Multiplexing.

19

Band guard avoids the mutual interference between the sub-channels of a multipling technology like WDM.

20

FDM, acronym which stands for Frequency Division Multiplexing.

21

DWDM, acronym which stands for Dense WDM.

22

CWDM, acronym which stands for Coarse WDM.

23

A repeater is a device which regenerates an optical signal by converting it into an electrical signal, processing that electrical signal and then retransmitting an optical signal.

24

MTTF, acronym which stands for Mean-Time-To-Failure.

25

Polymethylmethacrylate, plastic polymers formed by the methyl methacrylate (MMA)

26

(CbH8O2). It is a compound used to produce Plexiglass.

27

Transmission percentage is quantity of information in percent (%) which will be effectively

transmitted inside an optical fiber.

28

PE, acronym which stands for polyethylene.

29

PA, acronym which stands for polyamide (nylon).

30

Fast Ethernet, standard which provides Ethernet LAN with speed of 100 Mbps.

31

LAN (acronym which stands for Local Area Network), type of computer network with a territorial extension of about a few kilometers.

32

Class 1, classification of LASER diodes which include those with emitted optical power less than 0.04 mW.

33

4B/5B, channel coding scheme which allows a 4 bit sequence to be converted into a 5 bit code by a special conversion table.

34

MLT-3 (MultiLevel Tone-3), channel coding scheme based on the using of 3 logical levels: -1, 0, +1.

35

CSMA/CD (acronym which stands for Carrier Sense Multiple Access with Collision Detection), protocol used by Ethernet and Fast Ethernet to detect and manage collisions. It implements the following procedure:

• CSMA/CD listens to the channel before transmitting;

• if a collision is detected during the transmission, CSMA/CD stops and signalizes this collision to all another stations and tries to retransmit again later.

36

PDU (acronym which stands for Protocol Data Unit), information unit for a network protocol.

37

^FTTH (acronym which stands for Fiber To The Home), broadband network architecture with cabling up to the the boundary of the living space.

38

WakeOnLAN, standard which allows a computer to be turned on by a message from a remote node.

39

RJ45, connector used for LANs with UTP cabling.

40

Plug and play is a technology which facilitates the discovery of a hardware component in a system without the need for physical device configuration or user intervention in resolving resource conflicts.

41

POFTO, acronym which stands for POF Trade Organization.

42

ElectronicBricks, Italian company which offers low-cost IT solutions based on POF.

43

Firecomms Ltd., Irish company specialized in optoelectronic components for POF.

44

Luceat S.p.A., Italian company which offers low-cost IT solutions based on POF.

45

Istituto Superiore Mario Boella, IT research institute in collaboration with Polytechnic

University of Turin and companies from Turin.

46

DieMount GmbH, German company which manufactures devices and components for POF.

47

POF Application Center, German research institute specializing in POF applications.

48

Teleconnect GmbH, German IT company.

49

Rosetta IP Ltd., Israeli company which produces VLSI components.

50

Telecom Italia Lab, research division of Telecom Italia.

51

Cluster is a set of computers typically connected by a LAN. Its purpose is to distribute

the complex computing on many computers.

52

SAN (acronym which stands for Storage Area Network) is a high-speed Gbps made up of storage devices. Its purpose is to make these devices available to any computer connected to it.

53

PFO (acronym which stands for polifluorene), plastic material made of fluorene polymers (C13H10).

54

PLL, acronym which stands for Phase Locked Loop.

55

VCO, acronym which stands for Voltage Controlled Oscillator.


 

Author Domenico Carlo Romeo  Added On 11/02/2014
Rating (0)  Category Theory of Fiber
 How to Remove Epoxy
 
  • 129,271 views
  • 3 Editors

  • Edited 46 weeks ago

Three Methods:Use Heat to Remove EpoxyFreeze the EpoxyUse Chemicals to Remove Epoxy

Epoxy is a permanent adhesive used on many types of surfaces, from plastic to metal. Once epoxy has hardened, removing it can be tricky. Epoxy starts in a liquid state. As it is mixed, the substance temperature heats up until it starts to cool off and harden. You can remove epoxy by getting it back to a liquid, or at least gel-like, state so that you can scrape it from the surface.

Method 1 of 3: Use Heat to Remove Epoxy

  1. 1
    Aim a heat gun at the epoxy for several minutes. If the epoxy is on a plastic or wooden surface, keep an eye on the surface so you don't warm it up too much and burn it.
    • If the epoxy is attached to a wooden surface, soak the area in acetone for an hour or more before using heat to soften the epoxy.
      Remove Epoxy Step 1Bullet1.jpg
    • Acetone will only soak into a wooden surface. When dealing with epoxy on plastic, marble, cement, vinyl or metal, any chemical will interact with the top of the surface, but does not penetrate down into the layers like it can on wood.
      Remove Epoxy Step 1Bullet2.jpg
    Ad
  2. Remove Epoxy Step 2.jpg
    2
    Use a putty knife, razor or any sharp objects to scrape the epoxy off the surface. Keep reheating and scraping until all the epoxy is gone.

Method 2 of 3: Freeze the Epoxy

  1. Remove Epoxy Step 3.jpg
    1
    Spray a refrigerant spray on the epoxy. The spray will quickly drop the temperature of anything it touches. The epoxy will freeze and turn brittle.
  2. Remove Epoxy Step 4.jpg
    2
    Chip away the brittle epoxy. Use a putty knife or hit it with a rubber mallet or hammer. The epoxy should be cold enough that it turns to crystals and breaks off easily.
    • Be careful not to damage your surface by putting too much pressure on the epoxy. If it doesn't break apart easily, try adding more spray cooler to chill the resin even more.
      Remove Epoxy Step 4Bullet1.jpg

Method 3 of 3: Use Chemicals to Remove Epoxy

  1. Remove Epoxy Step 5.jpg
    1
    Choose a chemical that will soften epoxy, but not damage the surface to which it is stuck. Chemicals may damage certain surfaces such as cloth, plastic or vinyl. The strong chemicals may actually eat away at surfaces before they soften the epoxy glue.
    • Try a paint thinner. The acetone in most paint thinners can soften up hardened epoxy, but be prepared to let the epoxy and object to which it is stuck to soak for at least an hour.
    • Use the chemical methylene chloride. This highly toxic chemical can damage many surfaces, including plastics and wood. Methylene chloride has been classified as a carcinogen so it is not widely available.
  2. Remove Epoxy Step 6.jpg
    2
    Scrape epoxy resin from the surface once it is softened by the chemical. If some epoxy is still stuck to the surface, soak the remaining epoxy in the chemical for a while longer before trying to scrape it off.
    Ad

Author WikiHow  Added On 09/18/2014
Rating (0)  Category Epoxy, Adhesives, Curing
 Clean Up and Removing Epoxy
 
Clean Up & Removing Epoxy

Clean Up

Contain large spills with sand, clay or other inert absorbent material. Use a scraper to contain small spills and collect as much material as possible. Follow up with absorbent towels. Uncontaminated resin or hardener may be reclaimed for use.


DO NOT use sawdust or other fine cellulose materials to absorb hardeners.

DO NOT dispose of hardener in trash containing sawdust or other fine cellulose materials-spontaneous combustion can occur.

Clean resin or mixed epoxy residue with lacquer thinner, acetone or alcohol. Follow all safety warnings on solvent containers. Clean hardener residue with warm soapy water.

Dispose of resin, hardener and empty containers safely. Puncture a corner of the can and drain residue into the appropriate new container of resin or hardener.

DO NOT dispose of resin or hardener in a liquid state. Waste resin and hardener can be mixed and cured (in small quantities) to a non-hazardous inert solid.

CAUTION! Large pots of curing epoxy can get hot enough to ignite surrounding combustible materials and give off hazardous fumes. Place pots of mixed epoxy in a safe and ventilated area, away from workers and combustible materials. Dispose of the solid mass only if cure is complete and the mass has cooled. Follow federal, state or local disposal regulations.

Removing Epoxy

Removing uncured or non-curing epoxy. Uncured epoxy is removed as you would spilled resin. Scrape as much material as you can from the surface using a stiff metal or plastic scraper-warm the epoxy to lower its viscosity. Clean the residue with lacquer thinner, acetone, or alcohol. Follow safety warnings on solvents, and provide adequate ventilation. After recoating wood surfaces with epoxy, it's a good idea to brush the wet epoxy (in the direction of the grain) with a wire brush to improve adhesion. Allow solvents to dry before recoating.

Removing fiberglass cloth applied with epoxy.

Use a heat gun to heat and soften the epoxy. Start in a small area a near a corner or edge. Apply heat until you can slip a putty knife or chisel under the cloth (about 200°F). Grab the edge with a pair of pliers and pull up on the cloth while heating just ahead of the separation. On large areas, use a utility knife to score the glass and remove in narrower strips. Resulting surface texture may be coated or remaining epoxy may be removed as follows.

Removing cured epoxy coating.

Use a heat gun to soften the epoxy (200°F). Heat a small area and use a paint or cabinet scraper to remove the bulk of the coating. Sand the surface to remove the remaining material. Provide ventilation when heating epoxy.

Author unknown  Added On 09/18/2014
Rating (0)  Category Epoxy, Adhesives, Curing
All Articles 
 
 
  Login Here
 
Username
Password
Signup Now
Forgot password
 
 
  Top Rated
 
NEW TECHNIQUES FOR...
Fiber termination's...
DEFINING AND...
 
 
  Most Popular
 
Fiber Teminations...
Thesis on Plastic...
Fiber Testing
 
 
  Editor's Pick
 
DEFINING AND...
 
 
  Sponsor
 

Contribute your White Papers and Articles here. Patent papers ere welcomed. This is a FREE Learning Center with Application oriented solutions. The Basics of Fiber and Advanced Topics are shared here. Hopefully, by sharing your Fiber Optic knowledge others will benefit! This is a great place to say thanks to your industry by helping someone else learn about Fiber Optic disciplines and Application Theory. Please respect the content of these Authors and the Copyrights expressed. Many of the Authors expressed herein have listed alternate web locations for viewing additional Authoring's.
 
   
 
   
 HOME | LOGIN | SIGNUP
SUBMIT AN ARTICLE | SEARCH ARTICLES | ALL ARTICLES
TERMS OF USE | PRIVACY POLICY | LEGAL POLICY | CONTACT US

Real Time Web Analytics
Powered By Article directory software