Dispersion Effects Of Ocdma Performance

MEE09:51 

Dispersion Effects on  OCDMA System Performance  Abdul Gafur  This thesis is submitted in partial fulfillment of the  Degree of Master of Science in Electrical Engineering     

Blekinge Institute of Technology    September 2009  Blekinge Institute of Technology School of Computing Supervisor: Dr. Doru Constantinescu Examiner: Dr. Doru Constantinescu

MEE09:51 

ii

MEE09:51 

Abstract  This thesis presents the effect of dispersion and Multi Access Interference (MAI) of optical fiber on the Bit Error Rate (BER) performance of a Direct Sequence Optical Code Division Multiple Access (DS-OCDMA) network by means of intensity modulation and optical receiver correlators. By using Matlab simulations, Signal-to-Noise Ratio (SNR) versus Received Optical Power (ROP) of an OCDMA transmission system can be evaluated with a so-called 7chip m-sequence for different numbers of system users. This can be done for the ROP versus BER for various lengths of single mode optical fiber by taking into consideration the dispersion effect in the optical fiber. Matlab simulations can be performed in order to illustrate the reduction of the dispersion index gamma, or to visualize different scenarios, e.g., what amount of transmitted power is required in order to obtain a BER of 10-9 when the length of the optical fiber is increased.

iii

MEE09:51 

 

 

iv

MEE09:51 

Acknowledgements All my gratitude and thanks to ALLAH, who is the heavenly power of the Earth and the store of every information and knowledge. It is my pleasure to state the delight earnest admiration and philosophical esteem to my respectable supervisor and examiner, Dr. Doru Constantinescu at School of Computing, at Blekinge Institute of Technology, Sweden, for his help, pedagogic supervision, steady encouragement, precious advices and collaboration for doing well with my thesis work. I expand my earnest thanks to my family members and particularly to my beloved parents. Furthermore, I want to thank all of my associates who assisted me during this period of the thesis research work.

v

MEE09:51 

vi

MEE09:51 

Table of Contents  1.  Introduction ........................................................................................................................................... 1  1.1 Introduction .................................................................................................................................... 1  1.2 Related Work .................................................................................................................................. 2  1.3 Thesis Motivation .......................................................................................................................... 3  1.4 Objectives of the Thesis .............................................................................................................. 3  1.5 Outline of the Thesis .................................................................................................................... 4  2. Optical Fibers ......................................................................................................................................... 5  2.1. Advantages of Optical Fiber Transmission ....................................................................... 5  2.1.1 Distance .................................................................................................................................... 5  2.1.2 Bandwidth ............................................................................................................................... 5  2.1.3 Electrical Isolation ............................................................................................................... 6  2.1.4 Reliability ................................................................................................................................. 6  2.2. Shortcomings of Optical Fiber Transmission ................................................................... 6  2.3 Characteristics of Optical Fibers ............................................................................................. 7  2.3.1 Linear Characteristics ......................................................................................................... 7  2.3.2 Non Linear Characteristics ............................................................................................. 12  2.4 Coupling of Light ......................................................................................................................... 14  2.5 Intensity Modulation and Optical Amplification ........................................................... 16  2.5.1 Erbium Doped Fiber Amplifier ..................................................................................... 17  3. System Description ............................................................................................................................ 19  3.1. Future of OCDMA Systems ..................................................................................................... 19  3.2. System Description.................................................................................................................... 19  3.3 Noise in OCDMA ........................................................................................................................... 21  3.3.1 Dark Current Noise ............................................................................................................ 22  3.3.2 Thermal Noise ...................................................................................................................... 22  3.3.3 Quantum Shot Noise .......................................................................................................... 22  3.4. System Analysis .......................................................................................................................... 23  3.5. Advantages of OCDMA ............................................................................................................. 25  3.5.1 Bandwidth ............................................................................................................................. 25  3.5.2 Network Control ................................................................................................................. 26  3.5.3 Value‐added Services ........................................................................................................ 26  3.5.4 Security ................................................................................................................................... 26  vii

MEE09:51 

3.6. Drawbacks of OCDMA .............................................................................................................. 27  3.6.1 Noise ......................................................................................................................................... 27  3.6.2 Error Correction .................................................................................................................. 27  3.6.3 Encoding/Decoding ........................................................................................................... 28  3.6.4 Security Integration ........................................................................................................... 28  4. System Performance ......................................................................................................................... 29  5. Conclusion and Future Work ......................................................................................................... 41  Appendix A: Acronyms .......................................................................................................................... 43  References .................................................................................................................................................. 45 

viii

MEE09:51 

List of Figures    Figure 1: Light path in optical fiber .................................................................................................................. 7  Figure 2: Attenuation of light in single mode ............................................................................................... 8  Figure 3: Chromatic dispersion in optical fiber ........................................................................................ 10  Figure 4: Polarization dispersion ................................................................................................................... 11  Figure 5: Cross‐phase modulation ................................................................................................................. 13  Figure 6: Four wave mixing .............................................................................................................................. 14  Figure 7: Acceptance cone ................................................................................................................................. 15  Figure 8: Stimulated emission ......................................................................................................................... 17  Figure 9: OCDMA transmitter .......................................................................................................................... 20  Figure 12: BER vs. ROP performance for 12/17 users .......................................................................... 31  Figure 13: Performance comparison ‐ BER vs. ROP ............................................................................... 32  Figure 14: SNR vs. ROP performance for 3/6 users ............................................................................... 33  Figure 15: SNR vs. ROP performance for 12/17 ...................................................................................... 34  Figure 16: Performance comparison ‐ SNR vs. ROP ............................................................................... 35  Figure 17: Eye‐diagram ‐ gamma = 0.1 ........................................................................................................ 36  Figure 18: Eye‐diagram ‐ gamma = 0.2 ........................................................................................................ 37  Figure 19: Eye‐diagram for 70 km fiber length ........................................................................................ 38  Figure 21: Eye‐diagram for 100 km fiber length ..................................................................................... 40 

ix

MEE09:51 

x

MEE09:51 

1. Introduction  1.1 Introduction  In the decade 1985-1995, four significant events heralded the possibility of optical networking namely that both transmission and switching could be based on fiber optic communication. This was realized due to four main factors: 1. Realization of optical amplifiers. 2. Economic deployment of Wavelength Division Multiplexing (WDM). 3. Introduction of an Optical Cross Connect (OXC) enabling rapid reconfiguration of

light paths based on the wavelength channels. 4. Convergence of services and transport transmission rates [6].

Most up to date WDM transmission technologies realize optical link capacities exceeding 10 Tbit/s per fiber based on 40 Gbit/s per wavelength channel [11]. Through optical fiber, light propagates using a so-called total internal reflection technology, i.e., the carrier signal is the light. However, the transmission signal may become mixed due to several reasons. Among these, we mention optical amplifier noise, which creates amplified spontaneous emissions and the effect of nonlinear and chromatic dispersion to optical fibers. Our main objective in this thesis is to analyze Optical Code Division Multiple Access (OCDMA) networks. We analyze the main causes for Multiple Access Interference (MAI) which may reduce the performance of OCDMA networks. We also analyze the limitations that occur due to dispersion in OCDMA networks. Dispersion can also reduce the performance of the passive optical network. Furthermore, we present in this report our simulations results for BER versus received power for a multiple users system. We present the simulation results of the effects of

1

MEE09:51 

dispersion for various scenarios, e.g., different received optical power for multiple users and various optical fiber lengths in OCDMA systems. Development of the optical fiber communications technology has evolved rapidly in order to achieve larger transmission capacity and longer transmission distances [12]. Nowadays, OCDMA systems are highly interesting as they offer several sought-after features such as asynchronous access, privacy, secure transmissions, and ability to support variable bit rates and busy traffic and provide high scalability of the optical network [3]. In OCDMA, one great feature is that all subscribers can access the network asynchronously. In this case, a great advantage is that no conversion of optical to electrical signals is needed nor use of timing devices is required.

1.2 Related Work  Current research on OCDMA focuses on direct time spread OCDMA, spectral encodingdecoding, pulse-position modulation OCDMA, asynchronous phase encoding OCDMA and frequency hopping OCDMA [5]. However, in [5] chromatic dispersion of fiber is not considered. Chromatic dispersion can reduce system performance and occurs when increasing the inter-chip interference and decreasing the receiver optical power. At present, the performance of an asynchronous phase encoded OCDMA system considering fiber chromatic dispersion has been reported in [13] in the case of standard single mode optical fiber, while systems with dispersion shifted optical fiber are presented in [5]. Intensity modulation with direct detection On-Off Keying (OOK) OCDMA and Pulse Position Modulation (PPM) OCDMA systems are analyzed in [1]. The capacity of these networks is limited because the number of signature sequences available with good correlation properties for a given sequence length is small [2]. 2

MEE09:51 

1.3 Thesis Motivation  There are three windows of wavelength used in optical fiber communications. These windows of wavelength are as follows: 850 nm, 1300 nm and 1550 nm. Among these, the 1550 nm wavelength is considered in our thesis work because of its low attenuation, around 0.25 dB/km. Furthermore, recent developments in coherent OCDMA encoders/decoders allow for the efficient separation of large number of simultaneously users providing thus a feasible solution for low-cost applications in multi-user Local Area Networks (LAN) environments [6]. In this thesis, the analysis is carried out for direct sequence OCDMA system with intensity modulation and direct detection sequence inversion keyed receiver [14] considering both MAI and chromatic dispersion. OCDMA is one promising candidate for the next generation broadband multiple access technique due to full asynchronous transmission, low latency access as well as soft capacity on demand [15]. MAI may be seen as a kind of noise. The MAI noise is minimized in this thesis work by using the so-called m-sequence signature code. In case of practical OCDMA network applications, the capacity of asynchronous multiuser access is essential [15]. In addition, an aspect of dispersion, namely the limitation of the OCDMA system is also presented here. The main focus of the thesis is to do research on what can be done in order to reduce the dispersion of the OCDMA network such as to obtain a given BER.

1.4 Objectives of the Thesis  The main objectives of this thesis work are as follows: •

Carry out the analysis of an OCDMA system based on a star coupler.

•

Carry out a MAI analysis of the OCDMA network. 3

MEE09:51 

•

Determine optimum system parameters in the design of an OCDMA system.

•

Find out the BER versus received optical power for various numbers of users with the help of m-sequences.

•

Find out the SNR versus received power of an OCDMA transmission system with m-sequence and different numbers of system users.

•

Find out the penalty of the eye diagram of power for various values of dispersion fiber indexes and various lengths of fiber.

1.5 Outline of the Thesis  This thesis report consists of five chapters. Chapter 1 gives an introduction of the OCDMA system and presents the motivation and the objectives of this thesis. Chapter 2 describes the light propagation mechanisms through optical fiber. This chapter also presents the characteristics of optical fibers and describes linear and nonlinear phenomenon in optical fibers. We also present here the optical modulation and the optical amplification. In Chapter 3 we introduce the OCDMA system analysis and derive the mathematical formulas that we will simulate later on by using the Matlab software. Chapter 4 is dedicated to the simulation results. In this chapter, we present the results for different parameters that improved the performance of the OCDMA network. Finally, Chapter 5 concludes this thesis work. This chapter presents several recommendations that should be taken into consideration by companies providing OCDMA system access in order to improve their systems.

4

MEE09:51 

2. Optical Fibers  The aim of this chapter is to describe the system of optical fiber communications and to also describe how light propagates through optical fibers. This chapter presents the characteristics of optical fiber as well as various concepts and techniques which may be of help in the design of optical transmission lines.

2.1. Advantages of Optical Fiber Transmission  2.1.1 Distance    Without a passive or active optical repeater, we can transmit the light signal in optical fibers for a distance of about 100 km. Such distances are easily achieved in optical fiber communications. This is to compare with copper and metallic based cables, where we can transmit the electrical signal for only a few kilometers without the need of repeaters. 2.1.2 Bandwidth  In optical fiber communications, infrared light is used. The frequency of infrared light is expressed in Hertz (Hz). Also, it is a known fact that the frequency depends on the available bandwidth. In addition, there is no Electromagnetic Interference (EMI) in optical fibers. The main reason for this is that the optical fiber has three layers. The layers in optical fibers are the core, the cladding and the jacket. The core and the cladding consist of the same type of materials. A mechanism called total internal reflection is employed for propagating the light through optical fibers. Further, metallic components are not used at all in optical fibers. For these reasons, optical fibers have no electrical conductivity.

5

MEE09:51 

2.1.3 Electrical Isolation  The manufacturing process of optical fiber use polymers of either plastic or glass. This is also the reason why optical fibers do not maintain an earth loop. Here earth loop or ground loop means an unwanted electrical current. Two terminals in an electrical conductor are adjusted to the same potential while the ground is connected to the opposite potential. Consequently, when current flows through the conductor an electromagnetic field is produced. As a result, EMI occurs in the electrical conductor. On the other hand, as stated before, the optical fiber has three layers. The refractive density of the core and the cladding is different. As a result, crosstalk or EMI is minimal in optical fiber. For this reason, optical fiber communication is perfect for use in hazardous electrically conditions since optical fibers cannot produce electrical short circuits. The manufacturing medium for optical fiber is primarily glass. The glass, in turn, is produced from sand, which is a widely available natural resource. Consequently, the manufacturing process for optical fiber allows for low potential cost when compared to the manufacturing process of copper conductors. 2.1.4 Reliability  Reliability is very high in optical fibers and losses are very low in optical fiber transmissions. For boosting transmitted signal, line amplifiers and repeaters are often used. That is why the reliability of optical transmission systems is higher than in the case of electrical conductors.

2.2. Shortcomings of Optical Fiber Transmission  Optical fiber transmissions have two fundamental shortcomings: transmission angles and bandwidth limitations. Furthermore, impurities and the phenomenon of glass absorption also create losses in optical fibers. 6

MEE09:51 

Losses occur in optical fibers mainly when the fiber is bended at an angle. As a result, the light can leak out from the cladding. There are three windows of wavelengths used in optical communications. In addition, losses are also related to the wavelength. For instance, the 850 nm wavelength has a loss of 4-5 dB/km. For the 1310 nm wavelength, the loss is 3 dB/km and for 1550 nm wavelength the loss is 1 dB/km.

Figure 1: Light path in optical fiber

The loss depends on the light paths because light may have different paths within an optical fiber. Figure 1 illustrates the loss for different light paths.

2.3 Characteristics of Optical Fibers  The characteristics of optical fiber communications are divided into two main categories, namely linear and nonlinear. 2.3.1 Linear Characteristics  The main linear characteristics of optical fibers are: attenuation, chromatic dispersion, polarization mode dispersion and optical SNR.

7

MEE09:51 

Attenuation: Nowadays, optical fiber technology is quite popular because it is optically transparent and is able to transmit the light signals over existing systems of modern communications. There are two types of attenuation: intrinsic and extrinsic. The intrinsic attenuation occurs when there is absorption of the ultra-violet and infrared light. In the 0.2-2 µm range, the resonance of ultra violet and infrared light cannot be absorbed in optical fibers. As a result, the impurities of the optical fiber must be also taken into consideration. From 1.23-1.4 µm range, the OH ions in the optical fiber are responsible for the attenuation (illustrated in Figure 2). The wavelength range of 1.26-1.62 nm is more suitable for optical transmission because, in this case, the attenuation of signal is 0.5 dB/km. Furthermore, intrinsic attenuation consists of Rayleigh scatterings due to variations of n. The attenuation due to Rayleigh scattering is sketched in the dashed line in Figure 2. Extrinsic attenuation appears when the light path is bending or due to micro impurities from the manufacturing process. Attenuation

3.5 dB

0.4 dB 0.2 dB 850 nm

1300 nm

1500 nm

Figure 2: Attenuation of light in single mode

8

Wavelength

MEE09:51 

Figure 2 illustrates the attenuation of three wavelength windows. These windows are the following: 850 nm, 1300 nm and 1500 nm. The attenuation (measured in decibels (dB)) for the 850 nm, 1300 nm and 1500 nm window is 3.5 dB, 0.4 dB and 0.2 dB respectively. In our thesis we use the 1500 nm wavelength. In the above figure there are two lines. One solid line and one dashed line. The solid line illustrates the typical shape which is followed in the 1990’s. The dashed line shows the actual shape of the attenuation of the single mode fiber. Chromatic Dispersion: There are several reasons for the reduced performance of optical fiber communications. Chromatic dispersion is such an effect which can reduce the performance of passive optical networks. Chromatic dispersion is the combinations of mainly two factors: dispersion of material and dispersion of the waveguide. Mathematically we can write

DT = DM + Dwg

(1)

where DT denotes the total dispersion, while DM and Dwg defines the material and the waveguide dispersion respectively [7]. Chromatic dispersion is the effect of pulse spreading (or broadening) and can reduce the integrity of a received signal unless appropriate dispersion modules are included in the optical communication system [7]. Simulated results of the power at the input and output of the optical fiber are shown in Figure 3 [8].

9

MEE09:51 

Figure 3: Chromatic dispersion in optical fiber

Here Tc is the power of the optical fiber per chip. From the figure we can see that data is not uniformly changed. The reason for this is that the power is increased when coding of two data streams are superimposed for the same duration in the chip. As a result, more superimposed coded data spreads the available data per chip. This phenomenon interferes with adjacent chips. For this reason, errors are also increased which in turn, reduces optical system performance. Mainly, this effect occurs when multiple users are using the system. Consequently, we can conclude that data spreading adversely affects the OCDMA system and shortens the light pulses. In such cases, it is necessary to take into account the fiber dispersion effects and compare them to MAI limitations [8]. Polarization Mode Dispersion: Polarization mode dispersion occurs when various light planes propagates through the optical fiber at slightly various speeds. Consequently, polarization mode dispersion reduces the performance of the optical network at higher data rates.

10

MEE09:51 

When light propagates through the fiber it follows the so-called polarization modes via two axes of the light path. The paths travel towards the receiver at right angles with each other. For this reason, the light pulses are overlapped and this changes the shape of the original light pulses, an unwanted phenomenon at the receiver. A light pulse will propagate through these two axes at different velocities. The differential group delay between the two polarization modes is illustrated in Figure 4 which results in a time broadening in their correct detection [16].

Figure 4: Polarization dispersion

Optical SNR: The Optical SNR (OSNR) belongs to the receiver section. It determines the total signal power and the total noise power. This is the reason why OSNR is a fundamental and important design part in optical fiber systems. At the receiver, an important parameter is the so-called Q-factor. This evaluates the qualitative performance of the receiver section. Consequently, the Q-factor follows the minimum SNR in order to get the desired BER within the system. As a result, when BER is high then OSNR will be also high.

11

MEE09:51 

2.3.2 Non Linear Characteristics    The scope of the non linear effects is to describe various characteristics of the optical fiber that occur when the power of the optical fiber is defined between 50-65 µm. For this reason, the intensity of the electromagnetic field behaves nonlinear.   In this section we will try to differentiate between several nonlinearities, both elastic and inelastic and as a result, the energy is exchanged in the medium. The major non linear characteristics in optical fibers are as follows: •

Cross Phase Modulation

•

Self Phase Modulation

•

Four Wave Mixing

Cross Phase and Self Phase Modulation: Are the most significant parts of nonlinear characteristics and are derived from the refractive index of the optical fiber. Self phase modulation is defined as the phase shift of light self induction when propagating through the optical fiber. Cross phase shift modulation is defined as the nonlinear phase shift originating from the optical field for various wavelengths. When two optical fields, with frequencies W1 and W2 are polarized along the x axis, they propagate simultaneously inside the fiber as

E=

1 x[E1 exp(− iw1t ) + E 2 exp(− iw2 t ) + c.c ] 2

(2)

This formula, expressed through self phase and cross phase modulation becomes [16]:

φ NL = n 2 K 0 L( E + 2 E ) 2

2

(3)

In (2), E is defined as the envelope of the frequency, x is the polarization direction, E1 is the envelope of frequency of cross phase modulation, E2 is the envelope of frequency of self phase modulation, and c.c is the counter clockwise direction. Further,

exp(− iwt ) is the exponential function where w is the angular frequency, t is the time period and i is the imaginary part of the complex function. 12

MEE09:51 

In (3), φ NL defines the nonlinear phase shift, n2 is the coefficient of nonlinear index, K o = 2π

λ , λ is the wavelength and L is the fiber length. Equation (3) represents the

combination of self phase modulation and cross phase modulation induced by the nonlinear phase shift. In (3) the contribution of cross phase modulation is twice the size of the self phase modulation of the nonlinear phase shift. An illustration of the cross phase modulation is given in Figure 5.

Figure 5: Cross-phase modulation

Four Wave Mixing: The nonlinear characteristic of the so-called four waves mixing is defined as the product of inter-modulations which, in turn, are the effect of various frequencies at different levels of interactions. For instance, when three wavelengths, λ1 , λ 2 and λ3 propagate through the optical fiber, they create a fourth wavelength λ 4 , produced by the incident scattering of the three photons. This mechanism is referred to as a four wave mixing and affects the optical transmission on OCDMA systems. A photon is defined as the quantum of the electromagnetic field with the unit photon as the light. The photons are regarded as a bundle of discrete packets. When two pump 13

MEE09:51 

(incident) photons (lights) are annihilated, two new photons are created: the first one is created at the signal frequency, while the other one is created at a complementary frequency called idler [10]. In a WDM context, this resulting power transfer impairs the transmission since it produces crosstalk between the transmission channels [9]. The four wave mixing phenomenon is illustrated in Figure 6.

Figure 6: Four wave mixing

2.4 Coupling of Light  The optical fiber consists of core, cladding and jacket where the core and the cladding are made from the same materials. By using total internal reflection technology, light propagates through the optical fiber. As a result, there is a so-called acceptance cone. The acceptance cone provides all light rays access to the fiber such that they are able to transmit optical signals. The acceptance cone has a so-called acceptance angle which is defined as half the angle of the cone. Mathematically, we can write this as NA = n0 * sin θ a where n0 is the refractive index of the air, which is defined as n0 = 1. NA is the so-called numerical aperture. Numerical aperture is the measurement of the maximum acceptance angle which

14

MEE09:51 

adjusts the light and transmits the light through a fiber. Consequently, NA is equivalent to the sine of the acceptance angle. The acceptance cone is illustrated in Figure 7.

Figure 7: Acceptance cone

In Figure 7, (1) indicates the accepted light ray and (2) indicates the light ray that was not accepted. Consequently, we can define the range for the propagating light through the optical fiber as n0 ∠ n 2 ∠n1 where n0 is the refractive index of the air, n1 is the refractive index of the core and n2 is the refractive index of the cladding. Furthermore, there are two light propagation modes in optical fibers. These propagation modes depend on the size of the core. These are: •

Single Mode Fiber

•

Multimode Fiber

The diameter size of the core in single mode fiber is usually small. That is why light coupling with single mode fiber is difficult when using as a transmitter source a Light Emitting Diode (LED). For this reason, in order to compensate the coupling losses, LASER is used as transmitting source in single mode optical fibers. This is also the reason why, single mode fiber is used for transmissions over long distances. 15

MEE09:51 

2.5 Intensity Modulation and Optical Amplification  There are three physical characteristics of the optical fibers that allow transmitting information. These are: 1. Polarization 2. Phase 3. Intensity

However, several characteristics of the optical fibers such as linearity and nonlinearity directly affect the performance of OCDMA. These are Polarization Mode Dispersion (PMD), Group Velocity Dispersion (GVD), Four Wave Mixing (FWM) and Cross Phase Modulation (XPM). We know light travels towards the receiver through the optical fiber with each other at right angles. As a result, for the PMD, the pulses of light are overlapped and change the original shape of the light pulses. GVD appears due to various velocities of the various frequency parts of the pulse propagating through the optical fiber. GVD spreads the short pulse of light in time. FWM is the result of different frequencies at various levels of interactions. XPM appears for different wavelengths in the optical field. To overcome these impairments, cost effectiveness and efficient transport of the spectral optical system must be realized. When OSNR is decreased, the optical signal power is increased due to the nonlinear effects of the optical fiber. This reduces the performance of a passive optical network. To limit this problem, we need to compensate the effects of dispersion with intensity modulations in order to achieve higher transmissions per channel. Consequently, in order to overcome the physical limitations of the optical fiber we need various technical solutions such as optical signal amplifiers.

16

MEE09:51 

2.5.1 Erbium Doped Fiber Amplifier  One such technical solution to the problem of optical amplifiers is the Erbium Doped Fiber Amplifier (EDFA). This particular amplifier is able to amplify the signal without first conversion of the signal from the electrical domain to the optical domain. By doing so, there is less noise produced in the signal than when an amplifier needs to do conversion of the electrical signal. Among the three wavelength windows, EDFA uses only the 1550 nm window. EDFA is made of solid state material and the Erbium ion ( E r 3 + ) is doped in the optical fiber´s core. When transition occurs, there are two types of emissions that take place here. These are the spontaneous emission and the stimulated emission. An illustration of the stimulated emission is depicted in Figure 8.

Figure 8: Stimulated emission

17

MEE09:51 

 

18

MEE09:51 

3. System Description     3.1. Future of OCDMA Systems  Although in the Code Division Multiple Access (CDMA) system soft capacity is obtained, the system faces interference in case of two users simultaneously access the communication channel which, in turn, degrades the performance of the CDMA system. Consequently, the main shortcoming of the CDMA system is multiple users’ access of the communication channel. For this reason, scientists and researchers are looking at systems that enable transmission without interference when considering multiple users. That’s why scientists turned their attention on the Optical CDMA, a system that tries to improve the shortcoming of the classic CDMA system. Nevertheless, there are several differences between the electrical and the optical CDMA. The optical CDMA is very important and becoming increasingly popular due to its high available bandwidth and elimination of cross talks. In the OCDMA system, multiple users can access the same channel with help of various coding techniques. These codes help maintaining low correlation between users and also help maintain low interference for each user.

3.2. System Description  The OCDMA technology is a relatively new, that emerged and gained focus for the research community during the last twenty years. The OCDMA implementation depends on several factors such as for instance the desired number of users. In OCDMA, the transmission signal may be subjected to conversion from electrical-tooptical, optical-to-optical or optical-to-electrical signal domain.

19

MEE09:51 

In this thesis, the OCDMA system consists of five main sections: 1. Data source (i.e., transmitting computer). 2. Optical CDMA encoder. 3. Optical star coupler: Device that accepts one input signal and is able to output to

several. At last, using the PN sequence receiver can receive his desired signal. However star coupler has a loss. But this is very poor. th

4. The 4 section is the optical CDMA decoder. 5. Data sink (i.e., receiving computer).

The schematic block diagram of an OCDMA communication system is depicted in Figure 9 and 10, for an OCDMA transmitter and for an Optical Correlator Receiver (OCR) with switched sequence inversion keying, respectively [1].

Figure 9: OCDMA transmitter

Figure 10: OCR with switched SIK

In the OCDMA transmitter, every user preserves different signature codes modulated as binary. Data are actually electrical signals sent to the optical drive which converts the electrical signals into optical signals.

20

MEE09:51 

The encoded signal is further sent to the star coupler. The star coupler used depends on the topology of the network which can be either a LAN or an access network. In case of a LAN, the star coupler is N:N, while in an access network, the star coupler is 1:N. Further, in OCDMA every user shares the same channel. For this reason, crosstalk which is interference due to multiple accesses is introduced here. In order to reduce this unwanted interference, every user uses various signature sequences. On the other hand, in the OCR with switched sequence inversion keying, an optical switched correlator is used. Consequently, a bipolar reference sequence is correlated directly with the channel’s unipolar signature sequence in order to recover the original data [1]. The unipolar-bipolar correlation is practically realized in an optical correlator, by spreading the bipolar reference sequence into two complementary unipolar reference sequences. In addition, the optical correlator provides unipolar switching functions for despreading the optical channel signal [5]. The PIN photodiode is also known as the p-i-n photo-receiver. Here, i is the intrinsic region which is un-doped between the doped regions of n and p. Finally, the PIN photodiode cancels the de-spreaded signal integrated with the periodic data. This occurs before the detection of the zero threshold voltage.

3.3 Noise in OCDMA  In OCDMA systems there are various noises creating spontaneous fluctuation, namely: •

Dark current noise.

•

Thermal noise.

•

Quantum shot noise.

21

MEE09:51 

3.3.1 Dark Current Noise  The dark current noise is defined as the noise occurring in an OCDMA system when there is no optical power incident on the photo detector. In this case, a small reverse leakage current still flows from the device terminals and can be mathematically expressed as [18]

i 2 d = 2eBe I d

(1)

Here, e represents the electron charge and the dark current is denoted I d . This type of noise has an effect on the whole system.

3.3.2 Thermal Noise  The thermal noise is defined as the spontaneous fluctuation due to thermal interaction between, say, the free electrons and the vibrating ions in a conducting medium. Thermal noise is especially prevalent in resistors at room temperature and can be mathematically expressed as [18]

i 2t =

4 KTBe R

( 2)

Here, K is the Boltzmann constant, Be is the bandwidth of the system and T is the absolute temperature.

3.3.3 Quantum Shot Noise  Detection of light by a photodiode is a discrete process and the signal emerging from the detector depends upon the statistics of photon arrivals [18]. The quantum shot noise is defined as [19] N sh = (2qRKPR ) / 4T

(3)

Here, q is the electron charge, R is the responsivity of the photo receiver, and PR is the optical received power.

22

MEE09:51 

3.4. System Analysis  At the OCDMA transmitter, the SIK (Sequence Inversion Keying) modulated signal is sent to the optical drives through a laser diode. Consequently, the output for K th users can be mathematically expressed as [3] N −1

S k (t ) = ∑ PT Bk (t ) ⊗ Ak (t − lTc )

( 4)

l =0

In (4), S k (t ) represents the transmitted output pulse shape for different users in single mode fiber while l represents the period of the chip and PT is the optical power of the chip. Furthermore, Bk and Ak represent the user´s binary signal and signature codes, respectively. The operator ⊗ represents the sequence inversion key modulation so that when Ak is transmitted for a ”1”, Ak is transmitted for a ”0”, respectively. Furthermore, Tc is the pulse interval. In the OCR with switched sequence inversion keying, due to chromatic dispersion of the optical fiber, the output can be expressed mathematically as [3] n −1

S output (t ) = ∑ l =0

1

πγ

e

⎡ 1 ⎡ ( t −lT ) ⎤ 2 π ⎤ c − j ⎢( ) ⎢ ⎥ − ( ) signγ ⎥ ⎢ γ ⎣ Tc ⎦ ⎥ 4 ⎣ ⎦

* sin c

(t − lTc ) πγTc

(5)

Here, γ represents the index of chromatic dispersion of the optical fiber which, in turn, can be expressed mathematically as [5]

γ =

(λ2 ) 2 Dbc L (π )(c)

(6)

In (6), λ represents the wavelength of the optical carrier, c is the light velocity and L is the fiber length. Further, D describes the coefficient of chromatic dispersion of the optical fiber while the rate of the chip is bc .

23

MEE09:51 

Consequently, we can state that when the fiber length is increased, the index of chromatic dispersion in the fiber is also increased. The receiver section handles the de-spreaded signal. The signal is sent to the photo detector and is integrated in the output of the correlator for the ith user. This can be mathematically expressed as [3]

Z i (t ) =

RPR 2

T K N −1

T

0 K =1 l =0

0

∫ ∑∑ BK (t )S out (t ) ⊗ AK (t − lTc ) * { Ai (t − lTc ) − Ai (t − lTc )} * dt + ∫ n0 (t )dt

(7)

Here, the photodiode´s responsivity is given by R, K represent the concurrent number of users, n 0 is the noise of the channel which as seen in the output of the optical correlator. PR represents the optical received power given by [5] PR = PT − Pf

(8)

In (8), PT is the transmitted optical power while the loss in the optical fiber is Pf .

The first part of the output signal described by (4) is compensated by considering the mean of the signature code, denoted as U while the remaining part represents the noise occurring in the channel due to multiple accesses of the channel, chromatic dispersion and various noises for the spontaneous signal fluctuations in the receiver. This is described by the variance of the system, denoted as σ 2 . The mean of Z i (t ) is given by [3]

U=

RPR 4T

T N −1

∫∑S 0 l =0

out

(t − lTc )dt

(9)

The interference variance due to multiple accesses is given by [20]

σ 2 =U2

2( K − 1) 3N

(10)

The noise of the variance n0 (t ) , is a combination of thermal noise and noise of the shot power and is given by N o = N SH + N TH 24

(11)

MEE09:51 

The thermal noise NTH and the shot noise of the photo detector are given by [5]

N TH =

(4 K B Tr ) * Br RL

(12)

N SH =

2qRKPR 4T

(13)

In (9) and (10), K B represents the Boltzmann constant while the bandwidth of the receiver is given by Br . Further, the temperature of the receiver is Tr and the charge of the electron is q. The resistance of the receiver load is denoted RL . The SNR and BER of the OCDMA system are given by [1] SNR =

U2 σ 2 + N0

(14)

1 SNR BER = ( )erfc( ) 2 2

(15)

3.5. Advantages of OCDMA  Some of the main advantages of using OCDMA systems can be summarized as follows: •

Equal distribution of available bandwidth.

•

Control and organization of the network.

•

Provisioning of value-added services.

•

Security.

3.5.1 Bandwidth  In OCDMA, users can access equal portions of the available channel bandwidth. Here, the bandwidth is equally shared by all active users and can be divided into virtual chan25

MEE09:51 

nels. Consequently, the OCDMA system users can equally share access of the available resources of the network. Further, this is also the reason why, no single user one can block another from accessing the optical channel, one major advantage of OCDMA networks.

3.5.2 Network Control  In OCDMA technology, the optical codes of optical are distributed in such a way that, the peak autocorrelation for the shifted and non shifted optical signals can be alternatively small or large. Consequently, the optical receivers can manipulate asynchronously these signals without the need of global clock synchronization between them. In this way, the OCDMA technology can properly manipulate and control the signal transmitted within the whole system.

3.5.3 Value­added Services  OCDMA uses different types of optical codes. In this way, various services for different types of traffic can be easily introduced. For example, the code rate for high and low importance traffic can be set different such that transfer of e.g., real time audio/video signals and electronic mail can be give different priorities. Consequently, OCDMA plays an important role in providing customized or value-added services to its users.

3.5.4 Security  If we consider an OCDMA system with 41 wavelengths and 961 time chips, it will require 1350 years trying all possible combinations before the code could be broken. Meanwhile, in one second, the OCDMA system can have more than 107 such codes are used. For this reason, the security of the OCDMA system is inherent within the OCDMA technology, which is a major advantage of OCDMA-enabled networks. 26

MEE09:51 

3.6. Drawbacks of OCDMA  In spite of their many advantages, OCDMA systems suffer from several drawbacks which may be summarized as follows: •

Noise. 

•

Error correction. 

•

Encoding/Decoding. 

•

Security integration. 

3.6.1 Noise    Beat and shot noise are both technological barriers of the physical channel which degrade the performance of the OCDMA network. Beat and shot noise are not appearing on the same wavelengths in case of multiple accesses. This is why, for a fixed receiver, the energy is used for a single channel wavelength. On the other hand in OCDMA systems, the total bandwidth is distributed. That’s why beat and shot noise may be introduced in the wavelengths of the same transmission channel. In OCDMA with the same wavelengths the channel bandwidth is allocated which is the optical power from other user which guide o the shot noise. Shot noise is defined as the optical root square of the received power and is direct proportional to the number of users. This type of noise reduces the scalability of the OCDMA network.

3.6.2 Error Correction  Forward error correction is costly and unusable in OCDMA because the speed for carrying the information in electrical cables and optical fiber is not same. For this reason, we have to design specialized encoding and/or decoding devices in order to correct the errors in case of optical signal transmissions. 27

MEE09:51 

It is possible to design codes for forward error correction devices which exclusively depend on optical signal processing, such as optical multiplexing and wavelength shifting. As a result of such codes, we can manage error free transmissions in case of optical signal processing.

3.6.3 Encoding/Decoding  The optical signal follows two dimensional codes. Fiber Bragg Grating (FBG) is a periodic perturbation of the refractive index along the fiber length which is formed by exposure of the core to an intense optical interference pattern [21]. The optical encoder supports FBG and has a predefined center frequency and temperature. For this reason a wavelength control loops or robust encoding device is required in order to ease this effect.

3.6.4 Security Integration  Integration of hybrid laser technology represents a monetary cost barrier in optical communication technologies. As a waveguide based encoder and as a waveguide modulator, an array of tunable lasers integrated on the same substrate. Substrate refers to the manufacturing materials. Semiconductor devices (i.e., glass) are manufactured from this material. Consequently, a waveguide modulator and demodulator are cheaper to manufacture than a monolithic LASER integration.

 

28

MEE09:51 

4. System Performance  The OCDMA system performance is validated with a rate of 10*109 chips per second. We evaluate the OCDMA system performance by looking at the BER for various users and at the eye diagram penalty for 7 chip m-sequence signature (m-signature chip used in our simulations was 1110010). In our simulations we have considered single mode optical fiber at 1550 nm wavelength, with coefficient of chromatic dispersion of 17 ps/km-nm and a receiver load resistance of 50 Ω. Table 4.1 presents the parameters of the evaluated OCDMA system.

Symbol

Significance

Value

λ

Operating wavelength

1550 nm

bc

Chip rate

10 G chip/s

Q

Electron charge

1.6e-19 c

K

Boltzmann constant

1.38e-23 W/K. Hz

Tr

Receiver temperature

3000 k

RL

Load resistance of receiver

50 Ω

R

Responsivity of each p-i-n photodiode 0.85 Ω

L

Length of fiber

245.05 km

PSdBm

Received optical power gain

-20

29

MEE09:51 

Idk

Dark current

10 nA

Nth

Thermal current

1 pA2Hz-1

D

Coefficient of chromatic dispersion

17 ps/km-nm

Table 1: Simulation parameters

Figure 11 presents the system performance for BER versus received optical power (ROP) for 3 and 6 users, respectively. We observe that the BER decreases when ROP is increased. For instance, in case of 10-15 BER, the ROP equals -13 dBm for 3 users and -12 dBm for 6 users. 0

10

3 users 6 users

-5

10

-10

Bit Error Rate(BER)

10

-15

10

-20

10

-25

10

-30

10

-35

10

-18

-17

-16

-15 -14 -13 -12 -11 Optical Received Power(dBm)

-10

Figure 11: BER vs. ROP performance for 3/6 users

30

-9

-8

MEE09:51 

Figure 12 illustrates the BER performance versus optical received power for 12 and 17 users, respectively. For instance, in case of 10-10 BER, the optical received power is -12 dBm and -10 dBm for 12 and 17 users, respectively.

0

10

12 users 17 users

-5

10

-10

Bit Error Rate(BER)

10

-15

10

-20

10

-25

10

-30

10

-35

10

-18

-17

-16

-15 -14 -13 -12 -11 Optical Received Power(dBm)

-10

-9

-8

Figure 12: BER vs. ROP performance for 12/17 users

Figure 13 illustrates the BER performance versus ROP for up to 23 users. We observe again that BER decreases when the ORP and the number of users is increased. For instance, when we consider a 10-5 BER, the ROP is -14.8 dBm for 19 users while for 23 users, this becomes -14.2 dBm.

31

MEE09:51 

0

10

users users users users users users

-5

10

of 3 of 6 of 12 of 17 of 19 of 23

-10

Bit Error Rate(BER)

10

-15

10

-20

10

-25

10

-30

10

-35

10

-18

-17

-16

-15

-14 -13 -12 -11 Optical Received Power(dBm)

-10

-9

-8

Figure 13: Performance comparison - BER vs. ROP

The SNR influences the receiver section of the OCDMA system. If SNR is increased it will increase the performance of the system. Figure 14 illustrates the SNR performance versus ROP of the OCDMA transmission with 3 and 6 users respectively, by considering an m-signature of the chip sequence of 1110010. For instance, when the SNR is 50 dB, the ROP is -13.5 dBm for 3 users and it is equal to -13 dBm in case of 6 users.

32

MEE09:51 

100 k=3 k=6

90 80 70

SNR

60 50 40 30 20 10 0 -20

-18

-16 -14 -12 Received optical power(dBm)

-10

-8

Figure 14: SNR vs. ROP performance for 3/6 users

Figure 15 shows the SNR performance as simulated in Matlab of the SNR versus ROP for 12 and 17 users, respectively. For instance, when the SNR is 60 dB, the ROP is -9 and -8 for 12 and 17 users, respectively.

33

MEE09:51 

100 k=12 k=17

90 80 70

SNR

60 50 40 30 20 10 0 -20

-18

-16 -14 -12 Received optical power(dBm)

-10

-8

Figure 15: SNR vs. ROP performance for 12/17

Figure 16 presents the SNR versus ROP of the OCDMA system for up to 23 users. For instance, it is observed, that when the SNR is 30 dB, the ROP is -12 and -10 dBm for 19 and 23 users, respectively.

34

MEE09:51 

100 k=3 k=6 K=12 k=17 k=19 k=23

90 80 70

SNR

60 50 40 30 20 10 0 -20

-18

-16 -14 -12 Received optical power(dBm)

-10

-8

Figure 16: Performance comparison - SNR vs. ROP

The eye diagram for the evaluated OCDMA system is also simulated in Matlab. We used two signal levels in our simulations: signal level 1 and signal level 0. The top line shows the output level of the signal level 1. The bottom line represents the output level of signal level 0. When the eye is “opened” and the line is spiky, it means a better performance of the OCDMA system. On the other hand the eye is “distorted” when dispersion occurs in the system. We evaluated the OCDMA performance by looking at the eye diagram for a chip rate of 10 G chip/s and a coefficient of fiber dispersion of 17 ps/km-nm and different indices of chromatic dispersion γ . 35

MEE09:51 

Figure 17 depicts the eye-diagram of the OCDMA system when γ = 0.1.

1.6 1.4

Eye diagram of index of chromatic dispersion=0.1 and rate of chip=10Gc/s coefficient of dispersion=17ps/km-nm

The output of current

1.2 1 0.8 0.6 0.4 0.2 0

50

100

150 200 250 number of samples in the chip

300

350

Figure 17: Eye-diagram - gamma = 0.1

Figure 18 represents the eye-diagram of OCDMA when γ = 0.2.

We observe in figures 17 and 18, that the eye is more closed when the index of chromatic dispersion of fiber, γ = 0.2 than when this is 0.1.

We conclude that in order to obtain a good performance of the OCDMA system we need to reduce the value of the index of chromatic dispersion in the optical fiber.

36

MEE09:51 

1.6 Eye diagram of index of chromatic dispersion =0.2 & rate of chip=10Gchip/s &coefficient of dispersion=17ps/km-nm 1.4

The output of current

1.2

1

0.8

0.6

0.4

0.2

0

50

100

150 200 250 number of samples in the chip

300

350

Figure 18: Eye-diagram - gamma = 0.2

Next, we looked at the performance of the OCDMA system when considering the length of the optical fiber. Consequently, we considered an OCDMA system with a chip rate of 10 Gchip/s, the coefficient of dispersion of the fiber = 17 ps/km-nm and the index of chromatic dispersion of fiber ( γ ) = 0.05.

Figure 19 illustrates the eye-diagram of the OCDMA system for a fiber length of 70 km while in Figure 20 the length was set to 85 km. We observed from our simulations that the eye-diagram is more opened for a length of 70 km than in the case of 85 km. 37

MEE09:51 

1.6 1.4

The output of current

1.2

Eye diagram of of index of chromatic dispersion=0.05 &Chip rate=10Gchip/s Coefficient of dispersion=17ps/km-nm Length of fiber is 70km

1 0.8 0.6 0.4 0.2 0

50

100

150 200 250 number of samples in the chip

300

Figure 19: Eye-diagram for 70 km fiber length

38

350

MEE09:51 

1.6 1.4

Eye diagram of index of chromatic dispersion=0.05 Chip rate is 10Gch/s & coefficient of dispersion is 17ps/km-nm Lengths of fiber is 85 km

The output of current

1.2 1 0.8 0.6 0.4 0.2 0

50

100

150 200 250 number of samples in the chip

300

350

Figure 20: Eye-diagram for 85 km fiber length

Furthermore, we also looked at the eye-diagram of the OCDMA system for a fiber length of 100 km (Figure 21). We observed from all our simulations that the eye is more closed for longer fiber lengths with the same index of chromatic dispersion. We also observed that, in order to maintain a better performance of the OCDMA system we need also to reduce the index of the chromatic dispersion of the optical fiber.

39

MEE09:51 

1.6

1.4

Eye diagram of index of chromatic dispersion=0.05 chip rate is 10Gch/s & coefficient of dispersion=17ps/km-nm Length of fiber is 100km

The output of current

1.2

1

0.8

0.6

0.4

0.2

0

50

100

150 200 250 number of samples in the chip

300

Figure 21: Eye-diagram for 100 km fiber length

40

350

MEE09:51 

5. Conclusion and Future Work  

According to our OCDMA system analysis, the performance of BER is evaluated at an operating wavelength of 1550 nm, a chip rate of 10*109 chip/s and for different numbers of users. We also considered the shot and the thermal noise with MAI. We observed that a higher power of the optical transmitter is required in order to maintain a 10-9 BER for increasing number of users. We also observed the behavior of the OCDMA system by looking at the eye diagram of the OCDMA network with a coefficient of fiber dispersion of 17 ps/km-nm and an index of chromatic dispersion of the optical fiber γ = 0.05 and when considering different lengths of the optical fiber. The more closed the eye-diagram is, the worse performance the OCDMA system has. For instance, we found that for an index of chromatic dispersion of 0.05 and a coefficient of fiber dispersion equal to 17 ps/km, the eye-diagram is more opened at 70 km than at 100 km. We noticed also that when the fiber length is decreased, the index of chromatic dispersion of the optical fiber increases. In addition, BER performance degrades due to dispersion effects in the OCDMA system. The BER be reduced by adding the chips while the effect of the chromatic dispersion is reduced by sinking the power of the optical transmitter. It is further observed that in order to advance development of OCDMA networks, optimum transmitting power is needed, due to the sensitivity of the optical receiver.

41

MEE09:51 

As future work, we suggest a performance analysis of the effect of chromatic dispersion of OCDMA networks for different number of system users while varying the signature sequence code of the Pseudo Random Number Generator (PRNG), for the gold and mchip sequence.

42

MEE09:51 

Appendix A: Acronyms  BER

Bit Error Rate

CDMA

Code Division Multiple Access

EDFA

Erbium Doped Fiber Amplifiers

EMI

Electromagnetic Interference

FBG

Fiber Bragg Grating

FWM

Four Wave Mixing

GVD

Group Velocity Dispersion

LASER

Light Amplification by Stimulated Emission of Radiation

LED

Light Emitting Diode

MAI

Multiple Access Interference

OCDMA

Optical CDMA

OOK

On-Off Keying

OSNR

Optical SNR

OXC

Optical Cross Connect

PMD

Polarization Mode Dispersion

PPM

Pulse Position Modulation

PRNG

Pseudo Random Number Generator

SIK

Sequence Inversion Keying

SNR

Signal to Noise Ratio

WDM

Wavelength Division Multiplexing

WDM

Wavelength Division Multiplexing

XPM

Cross Phase Modulation

43

MEE09:51 

 

44

MEE09:51 

References  [1] S.P. Majumder and Md. Forkan Uddin, “The effect of four wave mixing on bit error rate performance of a direct sequence optical code division multiple access system”, 2005 Asia-Pacific Conference on Communications, Perth, Western Australia, 3-5 October 2005. [2] T.O. Farrell and S. Lochmann, “Performance analysis of an optical correlator receiver for SIK DS-CDMA communications systems”, Electronics Lett., Vol.30, No.1, pp. 63-64, 6th January 1994. [3] S.P.Majumder, Member, IEEE, Afreen Azhari, and F.M. Abbou, “Impact of fiber chromatic dispersion on the BER performance of an Optical CDMA IM/DD transmission system”, IEEE Photonics Technology Letters, Vol.17, No. 6, June 2005. [4] David D. Sampson, Mark Calleja, and Robert A. Griffin, “Crosstalk Performance of Coherent Time-Addressed Photonic CDMA Networks”, IEEE transactions on communications, Vol. 46, No. 3, March 1998. [5] S. P. Majumder, Afreen Azhari, “Performance Limitations of an Optical CDMA System Impaired by Fiber Chromatic Dispersion”, 0-7803-8783-X/O4/$20.00 0 2004 IEEE. [6] Michael J. O’Mahony, Christina Politi, Dimitrios Klonidis, Reza Nejabati and Dimitra Simeonidou, “Future Optical Networks” Journal of Light wave Technology, Vol.24, No.12, December 2006. [7] L.N. Binh and Y.L. Cheung, “DWDM Optically Amplified Transmission Systems – SIMULINK Models and Test-Bed: Part III –DPSK”, ECSE Monash university 2005. http://www.ds.eng.monash.edu.au/techrep/reports/2005/MECSE-4-2005.pdf [8] “Fiber Optic Chromatic Dispersion Effects on High Speed Optical Code Division Multiplexing Access (OCDMA) ” http://www.ee.kent.ac.uk/research/theme_project.aspx?pid=94 [9] S. Song, “High-order four-wave mixing and its effect in WDM systems”, Optics Express, vol. 7, pp. 166–171, August 2000. [10] David Boivin, “Optical phase-modulated systems: numerical estimation and experimental measurement of phase jitter”, Georgia Institute of Technology, 9-Nov-2006. 45

MEE09:51 

http://smartech.gatech.edu/handle/1853/14130 [11] Kohsuke NISHIMURA, Ryo INOHARA, Masashi USAMI, and Shigeyuki AKIBA, “All-Optical Regeneration by Electro-Absorption Modulator”, IEICE TRANS. ELECTRON., VOL.E88–C, NO.3 MARCH 2005. [12] Ng Wai Ling, F.M. Abbou, A. Abid, and H. T. Chuah, “Performance evaluation of FH-OCDMA in the presence of GVD and SPM”, IEICE Electronics Express, Vol.2, No 23, pp. 583-588. [13] C. H. Chua, F.M. Abbou, H.T. Chuah and S.P. Majumder, "Performance analysis on phase-encoded OCDMA communication system in dispersive fiber medium", IEEE Photon. Techno/. Letts, vol. 16, pp. 668-670, Feb. 2002. [14] T. O. Farrell and S. I. Lochmann, "Switched Correlator receiver architecture for optical CDMA networks with bipolar capacity”, Electron. Lett. vol. 31, pp. 905-906, May 1995. [15] Xu Wang and Naoya Wada, Taro Hamanaka and Ken-ichi Kitayama, Akihiko Nishiki, “10-user, truly-asynchronous OCDMA experiment with 511-chip SSFBG en/decoder and SC-based optical thresholder” http://www.ofcnfoec.org/materials/PDP33.pdf [16] Muhammad Haris, “Advanced modulation formats for high bit rate optical networks”, School of Electrical and computer engineering, Georgia Institute of Technology, August, 2008. [17] Abdul Gafur, Md. Humayun Kabir, and Waqas Ahmad, “Optical CDMA Network Based on Star Coupler Aspects of BER, Received Optical Power and Number of Users”, BTH transaction on Engineering Research Methodology , Karlskrona, Sweden, November 2008. [18] Dr.Yuliya Semenova, “Optical Communications Systems”, Dublin Institute of Tech-

nology, School of Electronics and Communication engineering http://www.electronics.dit.ie/staff/ysemenova/OCS/Optical%20Receivers.pdf [19] S. P. Majumder, Afreen Azhari, “Performance Analysis of an Optical CDMA in the Presence of Fiber Chromatic Dispersion”, Bangladesh University of Engineering and Technology.

46

MEE09:51 

[20] T. O'Farrell and S. I. Lochmann, “Switched correlator receiver architecture for optical CDMA networks with Bipolar capacity”, Electron. Lett, vol. 31, pp. 905-906, May. 1995. [21] Kenneth O. Hill and Gerald Meltz, “Fiber Bragg Grating Technology Fundamentals and Overview”, Journal of Lightwave Technology, Vol.15, No.8, August 1997.

47