University of Southern California Dissertations and Theses

Optical signal processing for enabling high-speed, highly spectrally efficient and high capacity optical systems

(USC Thesis Other)

Optical signal processing for enabling high-speed, highly spectrally efficient and high capacity optical systems

PDF

Download a page range

Download transcript

Copy asset link

Request this asset

Transcript (if available)

Content
OPTICAL SIGNAL PROCESSING
FOR ENABLING HIGH-SPEED,
HIGHLY SPECTRALLY EFFICIENT
AND HIGH CAPACITY OPTICAL SYSTEMS.

by

Muhammad Irfan Fazal

__________________________________________________________________

A Dissertation Presented to the
FACULTY OF THE USC GRADUATE SCHOOL
UNIVERSITY OF SOUTHERN CALIFORNIA
In Partial Fulfillment of the
Requirements for the Degree
DOCTOR OF PHILOSOPHY
(ELECTRICAL ENGINEERING)

May 2012

Copyright 2012 Muhammad Irfan Fazal
ii

Dedication

To my parents
Saleema Begum (late) and Fazal Maulla
my wife Safina , my kids Annan, Hasaan, Yumna & Haider
my siblings Sarwat, Adnan, Imran, Deeba, Zili and Affan
for their everlasting love, support and understanding.
iii

Acknowledgements

I would like to thank my academic advisor and dissertation committee chairman,
Dr. Alan E. Willner, for his guidance and mentorship throughout my graduate
studies. I would also like to extend my great appreciation to Professor Michelle
Povinelli and Professor Andrea Armani for serving on my dissertation and qualifying
exam committees. I would also like to thank Professor John O’Brien, Professor
Moshe Tu and Professor Steier for their support and guidance during my qualifying
examination.
I would like to extend my warmest thanks to the many members of the Optical
Communications Lab (OCLab) for many years of insightful discussions and
collaboration. OCLAB became a part of my life and I regard each one of you as my
family. In particular, Dr. Jian Wang, Dr. Jeng-Yuan Yang, Dr. Scott Nuccio, Dr.
Omer Faruk Yilmaz, Dr. Xiaoxia Wu, Salman Khaleghi and Nisar Ahmed were all
instrumental to the completion of the experimental results presented in this
dissertation. I wish you all great success in your careers.
Finally, I would like to extend my greatest thanks to Diane Demetras, Gerrielyn
Ramos and Anita Fung (Staff of Electrical Engineering) who have been a
phenomenal help during my Ph.D. years.
iv

Table of Contents
Dedication ....................................................................................................................ii
Acknowledgements.................................................................................................... iii
List of Figures .............................................................................................................vi
Abstract ................................................................................................................ xiii
Chapter 1: Introduction ................................................................................................1
Chapter 2: Demonstration of High Capacity Data Link using Orthogonal
Orbital-Angular-Momentum Modes and WDM ........................................5
2.1 Motivation for Orbital Angular Momentum modes.....................................5
2.2 Concept and experimental results ................................................................6
Chapter 3: Highly Spectrally Efficient Data Link using Polarization-
multiplexed OAM-based Spatial Modes..................................................15
3.1 Introduction................................................................................................15
3.2 Concept ......................................................................................................16
3.3 Experimental results and discussions.........................................................18
Chapter 4: System Optimization using Chirp-inducing Wavelength
Converters based on SOA-MZI in a Dispersion-compensated
Optical Link .............................................................................................25
4.1 Introduction................................................................................................25
4.2 Concept ......................................................................................................27
Chapter 5: System Optimization using Black-box Model of a Chirp-
inducing Network Module in a Dispersion-compensated
Optical Link .............................................................................................32
5.1 Introduction................................................................................................32
5.2 Concept ......................................................................................................34
5.3 Modeling and Fiber Simulation .................................................................35
5.4 Results and Discussion...............................................................................40
v

Chapter 6: System performance of DPSK signals transmitted through
broadband SBS-based slow light element and reduction of
slow-light-induced data-pattern dependence............................................41
6.1 Introduction................................................................................................41
6.2 Concept of slow light on phase-encoded optical signals ...........................44
6.3 Experimental results of slow light on 10-Gb/s NRZ-DPSK signals..........47
6.4 DPSK data-pattern dependence .................................................................49
6.5 Reduction of DPSK data-pattern dependence............................................53
6.6 System performance comparison between 2.5-Gb/s NRZ-DPSK
and RZ-DPSK .............................................................................................54
Chapter 7: SOA-Assisted Data-Polarization-Insensitive Wavelength
Conversion in a PPLN Waveguide ..........................................................58
7.1 Introduction................................................................................................58
7.2 Concept ......................................................................................................60
7.3 SHG:DFG in a PPLN waveguide...............................................................61
7.4 Data-inverting and non-inverting modes ...................................................63
7.5 Characterization of the polarization sensitivity of PPLN
waveguide ...................................................................................................65
7.6 Reduction of the PPLN waveguide’s polarization sensitivity using
an SOA module...........................................................................................66
7.7 Polarization modulation of PPLN waveguide pump by cross
polarization modulation in HNLF...............................................................72
Chapter 8: Optical Data Packet Synchronization and Multiplexing using a
Tunable Optical Delay based on Wavelength Conversion and
Inter-channel Chromatic Dispersion ........................................................75
8.1 Introduction................................................................................................75
8.2 Experimental setup and results ..................................................................81
8.3 Reconfiguration time of optical delay........................................................85
Chapter 9: All-optical time domain 160 Gb/s ADD/DROP based on pump
depletion and nonlinearities in a single PPLN waveguide.......................88
9.1 Introduction................................................................................................88
9.2 Experimental setup and results ..................................................................90
References ..................................................................................................................96

vi

List of Figures

Fig. 2-1 Experimental Setup for generating 25 WDM channels on
two OAM spatial modes at 40-Gb/s per channel.
7
Fig. 2-2 CCD images of the light beam at various positions through
the experimental setup.
7
Fig. 2-3 Spectra of back-to-back and two reconverted Gaussian
beams.
9
Fig. 2-4 BER curves for back-to-back and reconverted Gaussian
beams for OAM (0,16) for DPSK signals.
10
Fig. 2-5 BER curves for back-to-back and reconverted Gaussian
beams for OAM (0,8) for DPSK signals.
10
Fig. 2-6 Selected eye diagrams and power penalty for reconverted
Gaussian beams for OAM (0,16) and OAM (0,8) for DPSK
signals.
11
Fig. 2-7 BER curves for back-to-back and reconverted Gaussian
beams for OAM (0,16) and OAM (0,8) for directly
detected OOK signals at 40Gb/s.
12
Fig. 2-8 BER curves for back-to-back and reconverted Gaussian
beams for OAM (0,16) and OAM (0,8) for coherently
detected QPSK signals at 10Gbaud/s.
13
Fig. 2-9 Detected constellation diagrams and eye diagrams for
OAM modes (OAM0,+4, OAM0,+8, OAM0,-8 and
OAM0,+16) for QPSK at 10Gbaud/s.
13

vii

Fig. 3-1 Experimental setup for generation of pol-muxed four OAM
modes. PC: polarization controller; BPF: bandpass filter;
Pol.: polarizer; OC: optical coupler; Col.: collimator;
HWP: half-wave plate; SLM: spatial light modulator; BS:
beam splitter; PM: power meter; Att: attenuator; LO: local
oscillator; DSP: digital signal processing.
17
Fig. 3-2 (a1)-(a4) Phase patterns with charges of (a1) -4, (a2) +8,
(a3) -8, (a4) -16 applied to SLMs 4 thru 1. (b1)-(b5)
Intensity profiles of generated pol-muxed OAM modes
(b1) OAM0,+4, (b2) OAM0,+8, (b3) OAM0,-8, (b4)
OAM0,+16.
18
Fig. 3-3 Intensity profiles of super-imposed generated pol-muxed
OAMs.
19
Fig. 3-4 BER performance of demultiplexed (b) Y-pol. of different
pol-muxed OAM modes (OAM0,+4, OAM0,+8, OAM0,-8
and OAM0,+16) without and with crosstalk.
19
Fig. 3-5 Constellations of 16-QAM for (a) back to back. 20
Fig. 3-6 Constellations for demultiplexed (b)-(e) X-polarization and
(f)-(i) Y-polarization of different pol-muxed OAM modes.
(b)(f) OAM0,+4. (c)(g) OAM0,+8. (d)(h) OAM0,-8. (e)(i)
OAM0,+16.OAM0,-8, (b4) OAM0,+16.
21
Fig. 3-7 Spectra of back to back and reconverted Gaussian beams
from different polarizations of pol-muxed OAM modes
(resolution: 0.01 nm).
22
Fig. 3-8 BER performance of demultiplexed (a) X-polarization of
different pol-muxed OAM modes without and with
crosstalk.
22
Fig. 3-9 BER performance of demultiplexed (b) Y-polarization of
different pol-muxed OAM modes without and with
crosstalk.
23

viii

Fig. 4-1 Concept: Chirp produced by wavelength conversion in
SOA-MZI necessitates positive residual dispersion in fiber
link for optimal performance. Exp. Setup: LD: Laser
Diode, Mod: Modulator, PC: Pulse carver, SOA-MZI:
Semiconductor optical amplifier Mach-Zehnder
interferometer, BPF: Band-pass filter, TDCM: Tunable
dispersion compensation module, Rx: Receiver.
27
Fig. 4-2 Simulation Results: Extinction ratio and Output pulse-
width variation with pump inter-pulse delay.
28
Fig. 4-3 Variation of chirp-profile at points A, B and C. 28
Fig. 4-4 Robustness (in terms of eye height) to residual dispersion
of the various converted signals.
29
Fig. 4-5 Experimental Results: Measured chirp of converted signal
(-18Ghz).
30
Fig. 4-6 BER of wavelength-converted signal with varying residual
dispersion.
31
Fig. 4-7 Power penalty normalized with respect to the optimum
performance point of each signal with varying residual
dispersion.
31
Fig. 5-1 Illustrative block diagram of utilizing an arbitrary chirp
profile generator to compare chirp regimes for optimizing
the overall system performance. One (25) span(s) of 80km
compensated SMF is used with -13dBm (-3 dBm) launch
power, to explore the interaction of various chirp profiles
with fiber dispersion and nonlinearities.
34
Fig. 5-2 Normalized chirp profile. Normalized chirp indicates the
significance of chirp impairments in fiber transmission.
34
Fig. 5-3 Various basic chirp profiles inside a pulse. Power penalties
almost the same for CPx/CNx pairs.
36
Fig. 5-4 Power penalty at 10-9 BER versus average normalized
chirp on the edges for 8ps/nm dispersion for (a) Pin= -
13dBm, 80km SMF (b) Pin= -3dBm, 25×80km SMF
37
ix

Fig. 5-5 (a) Different chirp peak locations. (b) Low NLs: Pin= -
13dBm, 80km SMF. (c) High NLs: Pin= -3dBm, 25×80km
SMF. (BER>10-9 for CRef. @10-9 BER).
38
Fig. 5-6 (a) Different chirp regimes. (b) Low NLs: Pin= -13dBm,
80km SMF. (c) High NLs: Pin= -3dBm, 25×80km SMF.
(BER>10-9 for CRef. Power penalties at 10-9.)
39
Fig. 6-1 Concept of slow light on phase-modulated optical signals. 44
Fig. 6-2 Slow-light induced data-pattern dependence on
demodulated two output ports. Simulation result of phase
patterns of a 10-Gb/s DPSK signal before and after 8GHz
BW slow light element. Phase is delayed by 46 ps.
46
Fig. 6-3 Experimental Setup for DPSK slow-light based on
broadband SBS.
48
Fig. 6-4 Observation of DPSK slow-light: continuous delay of up
to 42 ps for a 10.7Gb/s DPSK signal.
48
Fig. 6-5 Slow-light-induced data-pattern dependence: 10.7-Gb/s
NRZ-DPSK through an 8-GHz slow light element. Bit
patterns before and after demodulation are shown.
50
Fig. 6-6 BER of DB port from 10.7-Gb/s DPSK signals after SBS
slow light element. Data-pattern dependence and Rayleigh
crosstalk are the two main reasons for DPSK signal
degradation.
52
Fig. 6-7 Power penalty comparison between 2.5-Gb/s and 10-Gb/s
NRZ-DPSK shows that data-pattern dependence is bit-rate
specific.
53
Fig. 6-8 Reduction of DPSK data-pattern dependence by detuning
the SBS gain peak: 3-dB Q factor improvement on the
AMI port is achieved.
54

x

Fig. 6-9 Delay for 2.5-Gb/s NRZ and RZ-DPSK with the same 5-
GHz SBS BW. The fractional delays for both NRZ and
RZ-DPSK are comparable.
55
Fig. 6-10 RZ-DPSK outperforms NRZ-DPSK by as much as 2dB,
which shows its robustness to data-pattern dependence.
56
Fig. 7-1 The polarization of the SOA probe (observed on a
polarimeter) changes as the SOA pump alternates between
ON and OFF.
60
Fig. 7-2 Second harmonic generation followed by difference
frequency generation in a periodically poled lithium
niobate waveguide
62
Fig. 7-3 (a) Data-inverting mode, where XGM supports XPolM. (b)
Data-non-inverting mode, where XGM opposed XPolM.
63
Fig. 7-4 Polarization sensitivity of DFG in a PPLN waveguide. The
wavelength converted output varies by >30 dB with
changes in the pump polarization.
66
Fig. 7-5 Experimental setup. Mod (intensity modulator), EDFA
(erbium doped fiber amplifier), SOA (semiconductor
optical amplifier), BPF (bandpass filter), PPLN
(periodically poled lithium niobate waveguide), and RX
(optical receiver).
67
Fig. 7-6 PPLN waveguide output spectrum. Non-inverted mode of
“multicast” wavelength conversion using our proposed
technique.
68
Fig. 7-7 Polarization-insensitive operation of our technique. The
input signal’s polarization affects the output by <1 dB.
70
Fig. 7-8 BER curve for inverted mode of operation, where XGM
and XpolM assist each other (2.5 Gb/s). The output eye
shows the slow recovery time of the SOA.
71

xi

Fig. 7-9 Experimental setup for using HNLF for polarization
modulation of PPLN waveguide pump.
72
Fig. 7-10 Conversion from polarization modulation of PPLN input
pump to amplitude modulation of the converted
wavelength at output of PPLN waveguide. (a) Converted
wavelength at PPLN output. (b) Error-free eye at optical
receiver.
73
Fig. 8-1 Packet 3 (P3) passes through the delay module which
consists of periodically-poled lithium-niobate (PPLN) λ-
converters, a dispersion compensating fiber (DCF) and a
chirped fiber Bragg grating (FBG).
77
Fig. 8-2 In the two scenarios, λ1 is converted to different λc’s,
resulting in different group velocities due to inter-channel
dispersion. The undesired intra-channel dispersion is
compensated by a FBG. (PPLN : periodically poled lithium
niobate waveguide).
78
Fig. 8-3 λ1 (input signal) and λpump-1 constitute the pumps for
PPLN-1. By tuning λdummy-1, the output λc can be tuned.
79
Fig. 8-4 Spectral arrangement of PPLN-2. λ1 is the output. 80
Fig. 8-5 Experimental setup: LD (laser diode), Mod (modulator),
PPG (pulse pattern generator), PPLN (periodically-poled
lithium-niobate), DCF (dispersion compensating fiber),
FBG (fiber Bragg grating), PC (polarization controller),
Circ (circulator), EDFA (erbium doped fiber amplifier) and
Rx (receiver). Note that ovals are simple passive couplers.
81
Fig. 8-6 Packet delay shown at 10 and 26.4 ns. Final output signal
is 1546.4 nm.
82
Fig. 8-7 Delay as a function of converted wavelength. 83
Fig. 8-8 Packet streams λ2 (non-delayed) and λ1 (delayed by 26.4
ns) synchronized and multiplexed. MUX = multiplexer.
Our multiplexer is a simple 3-dB passive coupler.
84

xii

Fig. 8-9 Measured bit error rate (BER) for back-to-back, delayed
single packet stream and multiplexed data stream. Power
penalty of 2.5 dB is observed.
85
Fig. 8-10 (a) Output spectra of the two PPLNs when the switch is in
the OFF position. (b) Output spectra when the 2x2 switch
is turned ON.
86
Fig. 8-11 Experimental setup for measuring the reconfiguration time
of the delay scheme. The inset shows that the
reconfiguration time is 276 ps.
86
Fig. 9-1 Add/drop working principle. 90
Fig. 9-2 Add/drop experimental set up. 91
Fig. 9-3 Eye-diagrams of the input, survived and new 40Gb/s
frame, and dropped channel. Clear eye opening can be
observed on all channels.
93
Fig. 9-4 Optical spectrum at the PPLN output (top), max hold for
different pump positions to tune the drop channel
wavelength (center), optical spectrum at the HNLF out in
the optical demultiplexer.
94
Fig. 9-5 BER measurements. The results about a sample channel
are reported. The penalty difference is lower than 1.5dB.
95

xiii

Abstract

The unabated demand for more capacity due to the ever-increasing internet
traffic dictates that the boundaries of the state of the art maybe pushed to send more
data through the network. Traditionally, this need has been satisfied by multiple
wavelengths (wavelength division multiplexing), higher order modulation formats
and coherent communication (either individually or combined together). WDM has
the ability to reduce cost by using multiple channels within the same physical fiber,
and with EDFA amplifiers, the need for O-E-O regenerators is eliminated. Moreover
the availability of multiple colors allows for wavelength-based routing and network
planning. Higher order modulation formats increases the capacity of the link by their
ability to encode data in both the phase and amplitude of light, thereby increasing the
bits/sec/Hz as compared to simple on-off keyed format. Coherent communications
has also emerged as a primary means of transmitting and receiving optical data due
to its support of formats that utilize both phase and amplitude to further increase the
spectral efficiency of the optical channel, including quadrature amplitude modulation
(QAM) and quadrature phase shift keying (QPSK). Polarization multiplexing of
channels can double capacity by allowing two channels to share the same
wavelength by propagating on orthogonal polarization axis and is easily supported in
coherent systems where the polarization tracking can be performed in the digital
xiv

domain. Furthermore, the forthcoming IEEE 100 Gbit/s Ethernet Standard, 802.3ba,
provides greater bandwidth, higher data rates, and supports a mixture of modulation
formats. In particular, Pol-MUX QPSK is increasingly becoming the industry’s
format of choice as the high spectral efficiency allows for 100 Gbit/s transmission
while still occupying the current 50 GHz/channel allocation of current 10 Gbit/s
OOK fiber systems. In this manner, 100 Gbit/s transfer speeds using current fiber
links, amplifiers, and filters may be possible.
Recently, interest has increased in exploring the spatial dimension of light to
increase capacity, both in fiber as well as free-space communication channels. The
orbital angular momentum (OAM) of light, carried by Laguerre-Gaussian (LG)
beams have the interesting property that, in theory, an infinite number of OAMs can
be transmitted; which due to its inherent orthogonality will not affect each other.
Thus, in theory, one can increase the channel capacity arbitrarily. However, in
practice, the device dimensions will reduce the number of OAMs used.
In addition to advanced modulation formats, it is expected that optical signal
processing may play a role in the future development of more efficient optical
transmission systems. The hope is that performing signal processing in the optical
domain may reduce optical-to-electronic conversion inefficiencies, eliminate
bottlenecks and take advantage of the ultrahigh bandwidth inherent in optics. While
40 to 50 Gbit/s electronic components are the peak of commercial technology and
100 Gbit/s capable RF components are still in their infancy, optical signal processing
xv

of these high-speed data signals may provide a potential solution. Furthermore, any
optical processing system or sub-system must be capable of handling the wide array
of data formats and data rates that networks may employ.
The work presented in this Ph.D. dissertation attempts at addressing the issue of
optical processing for advanced optical modulation formats, and particularly
explores the state of the art in increasing the capacity of an optical link by a
combination of wavelength/phase/polarization/OAM dimensions of light. Spatial
multiplexing and demultiplexing of both coherently and directly detected signals at
the 100 Gbit/s Ethernet standard is addressed. The application of a continuously
tunable all-optical delay for all-optical functionality like time-slot interchange at
high data-rates is presented. Moreover the interplay of chirp generated by differently
cross-phase modulation wavelength-convertors based on SOA-MZI with the residual
dispersion of a fiber link is studied and the optimal operating conditions are
explored.

1

Chapter 1: Introduction

Lightwave telecommunications was inaugurated with the introduction of low-
loss optical fiber in the period 1966-1970. By the mid-80’s commercial deployment
of fiber optic links was started which could operate upto to a speed of tens of Mbit/s.
In the last three decades since then, the field of fiber optic telecommunication has
advanced tremendously. With the initial tentative steps into the world of
communications which was dominated at that time by copper-based transmission
lines, fiber optic has literally taken over long-haul transmission. A recent study by
Cisco, predicts that the internet traffic is growing at an annual rate of about 43% [1].
Other studies by Duetche Telekom puts this growth even higher at around 60%. In
order to meet this unabated demand for more capacity efforts are being made to pack
more transmit information into the optical link at increasingly high rates. However,
to be commercially deployable, this increase in capacity has to be done in a cost-
effective and efficient manner at ultra high speeds. Deployed links currently operate
up to 10 Gbit/s (OC-192) with 40 Gbit/s links beginning to be deployed. However, it
seems like commercial 100 Gbit/s is soon going to take over the 40Gb/s era with
demonstrations of single-channel rates at 1-Tb/s and beyond already underway in the
research community.
2

Intensity modulation followed by direct detection, known as IMDD was
commonly used at the start of the optical communication era[2]. In this technique
light is simply turned ON or OFF to denote a ONE and ZERO. At the receiver a
square-law detector converts the optical signal into the electrical domain. Coherent
techniques, well known in the RF communication world, soon followed. However
the advent of EDFA amplification, delayed the commercial deployment of coherent
communication for about a decade. In the research world, in contrast, continual R&D
was done in the coherent communication domain. Over the last couple of decades
techniques for transmission and detection of optical information have gone through
many phases[3]. Coherent detection incorporates a local oscillator (LO) laser to mix
with the incoming signal, thereby generating an electrical beat signal carrying the
modulating signal. This gives a receiver sensitivity improvement of about 20dB
compared to IMDD. Moreover, since complete information about the phase and
intensity of the light field can be recovered at the receiver side, not only high order
complex modulation formats can be used to increase spectral efficiency and fiber
link capacity, but also transmission impairments can be potentially removed with
proper filter design. These gains also come with the price of increased complexity
including a phase-locked-loop and a narrow linewidth LO.
Photon is the fundamental carrier of information in optical communication. It has
various dimensions that can be explored to pack more information onto it, including
phase, intensity, arrival time, polarization frequency, orbital angular momentum
3

(OAM), linear momentum, superposition states, correlation, entanglement etc.
Photon represents a critical resource in the communication link (both free-space as
well as fiber link). While spin angular momentum relates to the polarization of light,
orbital angular momentum is associated with its phase structure[4]. The earlier can
have a maximum of two orthogonal states the later can have an infinite number of
orthogonal states. This provides a tremendous opportunity from the perspective of a
communication system since tremendous amount of information can be encoded into
the photon. Laguerre Gauss (LG) beams of light render themselves particularly
suited to carry OAM[5]. The phase front of a LG beam spirals as it propagates
forward such that within a distance of one wavelength it spirals ‘n’ times, such that n
is the degree of OAM it is carrying. Recently, there has been a tremendous interest in
the research community to utilize the OAM or spatial dimension of light to increase
the link capacity and spectral efficiency.
Additionally, as link capacity and speed is increased, the network elements
within the high-speed communication setup, can increasingly become the bottleneck.
Specially, if the routers and (de)multiplexers are not capable of scaling to the ultra-
high speed that the rest of the photonic link is operating at, then it defies the purpose
of operating at the ultra-high speed in the first place. One solution to reduce such O-
E-O bottleneck at routers and (de)multiplexers is to accomplish these tasks all-
optically[6]. Specially with the WDM capability, routing/multicastin/contention-
resolution can be achieved by all-optical wavelength-converters such as cross-phase
4

modulation in semi-conductor optical amplifier Mach-Zehnder interferometers
(SOA-MZI), sum-frequency generation/difference –frequency generation (SFG-
DFG) in periodically poled Lithium Niobate (PPLN) waveguides or four-wave
mixing in highly-nonlinear fibers etc[7]. While SOA-MZI, due to its smaller
footprint is more suitable for highly compact integrated solution, the later can be
used in more high-speed links due to its ultra-high speed of response. However, one
critical issue that needs to be addressed when using such all-optical wavelength
convertors is to quantify the degrading effects of such wavelength coverts and its
interaction with the fiber plant. For example, it is well-known that SOA-MZI based
wavelength converters can induce chirp on the converted signal due to the finite
recovery time of the SOA carriers. The effect of this chirp can be significant if the
fiber link is not 100% compensated. In an all-optical network, where the signal is
likely to travel various distances and links, it is highly likely that the signal will
encounter some residual dispersion. There is, therefore, a need to optimize the link
performance by considering both the chirp (and other degrading effects) induced by
the wavelength convertor and the residual dispersion of the optical link.
In this dissertation, novel methods for increasing the spectral efficiency and
capacity of optical link and techniques for all-optical signal processing capable of
scaling to ultra-high bit-rates are presented.

5

Chapter 2: Demonstration of High Capacity Data Link
using Orthogonal Orbital-Angular-Momentum Modes and
WDM

This chapter introduces the concept of orbital angular momentum (OAM) which
is associated with the phase structure of a propagating light beam. Laguerre-Gauss
(LG) beams can carry OAM. Spatial light modulators (SLM) are utilized to convert
a Gaussian beam into LG beam by imposing the phase structure uploaded to the
SLM with a computer. A desired phase structure is developed by the interference of
the phase structure with a plane wave and generating a JPEG image of the desired
LG profile. This image is then uploaded to the SLM such that the pixels of the SLM
can then superimpose the same phase structure on the Gaussian beam by inducing a
phase delay corresponding to the image. Using two such independent OAM-carrying
LG beams, the capacity of an optical link is shown to be doubled and a total capacity
of 2-Tb/s is demonstrated.
2.1 Motivation for Orbital Angular Momentum modes
A critical issue in optical communications research is the attempt to meet the
inevitable growth in data transmission capacity, even over short distances. Much
work has focused on increasing capacity by utilizing higher-order modulation
formats[3]. Another potentially complementary approach that has gained more
6

attention recently is to transmit independent data streams, each on a different spatial
mode [8]. These spatial modes can be in free-space or in fiber, such that adding
orthogonal modes can significantly increase capacity. Orbital angular momentum
(OAM) of photons, carried by Laguerre Gaussian (LG) beams, has recently been
keenly investigated as such spatial dimension to carry information [9][10]. In
principle, due to its inherent orthogonality, an arbitrary number of bits/photon can be
transmitted with OAM [4]. Recent research, showing OAM excitation in fiber, can
potentially open the possibility of taking advantage of these attributes in fiber also
[11][12]. Previously reported data transmission using OAM includes 2x10-Gbit/s on-
off-keying (OOK) using one wavelength that carried one OAM mode and one
conventional Gaussian mode [13]. A laudable goal would be to show the potential
use of OAMs for transmission by demonstrating a data link that can accommodate
different modulation formats, different detection schemes, and be compatible with
current wavelength-division-multiplexing (WDM) techniques in the
telecommunications spectral region.

2.2 Concept and experimental results
In this chapter, we experimentally demonstrate a 2-Tbit/s data link using two
orthogonal OAM modes and 25 WDM channels on each OAM mode[14]. We
quantify the link performance using bit error rate (BER) for 40-Gbit/s NRZ-DPSK,
40-Gbit/s NRZ-OOK and coherently-detected 10-Gbaud/s QPSK modulation
7

formats. Less than 4.5 dB power penalty is obtained for all channels for 40-Gbit/s
DPSK. For 40-Gbit/s OOK, the maximum power penalty is ~2.7 dB. Coherently-
detected QPSK performs with a power penalty of ~ 0.5 dB.

Fig. 2-1. Experimental Setup for generating 25 WDM channels on two OAM
spatial modes at 40-Gb/s per channel.

(a) OAM
0,16
(after SLM-1)

(b) OAM
0,8
(after SLM-2)

(c) OAM
0,16
demuxed (after SLM-3) (d) OAM
0,8
demuxed (after SLM-3)

(e) Separated with Pinhole (f) Separated with Pinhole
Fig. 2-2. CCD images of the light beam at various positions through the
experimental setup.

8

Laguerre-Gaussian (LGp,l) modes have an azimuthal angular dependence of
exp(-il ) where l is the azimuthal mode index. The angular momentum of an OAM
mode is proportional to l which describes the number of 0-2 phase spirals the light
beam undergoes as it travels a distance of one wavelength. “p” determines the radial
singularities of the modes. “p” is kept zero in our case, such that our LG modes are
essentially hollow single donut beams. As illustrated in Fig. 2-1, we used nematic
liquid crystal based reflective spatial light modulators (SLM)[15], designed for
telecom wavelengths, with the appropriate phase hologram to convert a data-carrying
Gaussian light beam into a OAM mode. We used OAM(0,8) and OAM(0,16) modes
for our two OAM branches, which are spatially multiplexed together using a non-
polarizing beam splitter. The two multiplexed OAM modes are incident upon a third
SLM which acts as a spatial de-multiplexer using a superimposed phase hologram.
As a result, the two OAM beams are converted back into Gaussian beams and
reflected spatially in different directions, to be collected by separate collimators and
fed into fiber-based detection system.
100-GHz ITU-grid-compatible 25 wavelength channels (from Ch.1: 1537.40
nm to Ch.25: 1556.55 nm) are combined together through an arrayed waveguide
grating (AWG). The transmitter (Tx) provides 40-Gbit/s NRZ-DPSK, 40-Gbit/s
NRZ-OOK and 10-Gbaud/s (20-Gbit/s) coherent QPSK signals, respectively, which
are launched into two separate collimators with a beam-waist of 3 mm. These
Gaussian beams are reflected off two separate SLMs (512 x 512 pixels and 7.68 mm
9

x 7.68mm size), to convert them into OAM(0,16) and OAM(0,8) modes,
respectively. Spiral phase patterns (shown in Fig. 2-1 insets) with topological
charges of 16 and 8 are used for these two SLMs. The two OAM modes are
multiplexed using a beam splitter (BS) and subsequently spatially de-multiplexed
using a third SLM. The resulting reconverted Gaussian beams are coupled into fiber
for detection. The state of light beam measured at different points of the setup is
shown in Fig. 2-2 (a)-(f).

Fig. 2-3. Spectra of back-to-back and two reconverted Gaussian beams.

10

Fig. 2-4. BER curves for back-to-back and reconverted Gaussian beams for
OAM (0,16) for DPSK signals.

Fig. 2-5. BER curves for back-to-back and reconverted Gaussian beams for
OAM (0,8) for DPSK signals..

11

Fig. 2-6. Selected eye diagrams and power penalty for reconverted Gaussian
beams for OAM (0,16) and OAM (0,8) for DPSK signals.

Fig. 2-2 shows the optical spectra of the 25-channel 40-Gbit/s NRZ-DPSK
signals. There is a power fluctuation of ~3 dB among the 25 channels on both the
OAM modes. BER data was taken for all 25 channels on both the OAM modes for
40-Gbit/s NRZ-DPSK as shown in Fig. 2-4 and 2-5.

12

Fig. 2-7. BER curves for back-to-back and reconverted Gaussian beams for
OAM (0,16) and OAM (0,8) for directly detected OOK signals at 40Gb/s.

There was a maximum of ~3.7 dB power penalty observed for the channels on
OAM(0,16) and ~4.5 dB for OAM0,8 at a BER of 10-9 shown in Fig. 2-6, while the
average power penalty over the 25 channels was ~2.2 dB for OAM(0,16) and ~2.9
dB for OAM(0,8). Shown in the insets of Fig. 2-6(c) are the balanced eyes for typical
DPSK channels.

13

Fig. 2-8. BER curves for back-to-back and reconverted Gaussian beams for
OAM (0,16) and OAM (0,8) for coherently detected QPSK signals at
10Gbaud/s..

Fig. 2-9. Detected constellation diagrams and eye diagrams for OAM modes
(OAM0,+4, OAM0,+8, OAM0,-8 and OAM0,+16) for QPSK at 10Gbaud/s.
14

BER performance for 40-Gbit/s NRZ-OOK (Ch1: 1537.40 nm, Ch8: 1549.32
nm, Ch16: 1550.12 nm, and Ch25: 1556.55 nm) signals is shown in Fig 2-7. The
observed maximum power penalty was ~2.7 dB for OAM(0,16) and ~2.1 dB for
OAM(0,8) at a BER of 10-9. For coherently-detected QPSK signals at 10-Gbaud/s,
the BER curves are shown in Fig 2-8 for Ch.1, Ch.16 and Ch.25. The observed
maximum power penalty was ~0.5 dB at a BER of 2x10-3 (enhanced FEC
threshold). Constellation diagrams and in-phase and quadrature eyes for Ch1
(1537.40 nm) are shown in Fig. 2-9.
In summary, we demonstrated a 2-Tbit/s data link using WDM over OAM
modes. It was shown that the link was transparent to modulation formats, i.e. both
intensity and phase modulated signals were quantified in terms of BER and power
penalty. In addition, both direct detection and coherent detection schemes were
demonstrated.

15

Chapter 3: Highly Spectrally Efficient Data Link using
Polarization-multiplexed OAM-based Spatial Modes

This chapter presents the second of two investigations on the spatial multiplexing
to increase data link capacity and spectral efficiency. 16 QAM signal (at base rate of
10.7Gb/s) is independently superimposed on four OAM-carrying LG beams with
different charges. Polarization multiplexing is used to double the number of OAM
modes to a total of 8 spatial modes. With this scheme a spectral efficiency of 25.6
bits/s/Hz is demonstrated[16].
3.1 Introduction
In the recent years, various approaches have been reported to increase the
spectral efficiency in optical communication systems, including higher-order
modulation formats, polarization-division multiplexing (PDM), space-division
multiplexing (SDM) and mode-division multiplexing (MDM) [17,18]. Similar to
MDM in fiber optical communication links using linearly polarized (LP) modes in
specially-designed few-mode fibers[8] [19], in free-space optical communication
links, orthogonal spatial modes carrying orbital angular momentum (OAM) are
laudable candidates for MDM. An OAM mode can be described as a light beam
containing an azimuthal phase term exp(il ) [20]. Since OAM has an infinite
number of orthogonal eigenstates corresponding to different values of charge l, it is
16

possible to modulate independent data onto different OAM modes to improve the
spectral efficiency.
In this chapter, we demonstrate a high spectral efficiency of 25.6 bit/s/Hz by
modulating 16-quadrature-amplitude-modulation (16-QAM) signals onto
polarization-multiplexed (pol-muxed) four OAM modes. Less than 3.5-dB optical
signal-to-noise ratio (OSNR) penalty is observed at a bit-error rate (BER) of 2e-3
(enhanced forward error correction (EFEC) threshold).
3.2 Concept
Fig. 3-1 illustrates the experimental setup for
generation/multiplexing/demultiplexing of pol-muxed four OAM modes. We prepare
a 10.7-Gbaud 16-QAM signal at 1550.12 nm from the vector addition of two QPSK
signals via polarization controllers (PCs), a tunable differential group delay (DGD)
element, and a polarizer (Pol.). Pre-filtering in the optical domain is followed to
reduce the signal spectral width. The 16-QAM signal is split into four paths,
relatively delayed with fibers, and delivered over four OAM modes which are
converted from four collimated Gaussian beams (3-mm beam size) by four reflective
spatial light modulators (SLM1-4) loaded with different phase patterns. Four OAM
modes are multiplexed through three non-polarizing beam splitters (BS1-3). Two
more polarizing beam splitters (BS4, BS5) are then used to enable the polarization
multiplexing. As a consequence, 16-QAM signals over pol-muxed four OAM modes
are obtained. For demultiplexing, a half-wave plate (HWP) and a polarizer are used
17

to select the polarization of pol-muxed signals. Another SLM (SLM5) loaded with a
specified phase pattern is employed to demultiplex one of the multiplexed OAM
modes back to the Gaussian beam which is coupled into the fiber for coherent
detection. The adopted SLM1-5 have 512 x 512 pixels, 7.68 mm x 7.68 mm size, and
a fast response time less than 20 ms.

Fig. 3-1. Experimental setup for generation of pol-muxed four OAM modes. PC:
polarization controller; BPF: bandpass filter; Pol.: polarizer; OC: optical coupler;
Col.: collimator; HWP: half-wave plate; SLM: spatial light modulator; BS: beam
splitter; PM: power meter; Att: attenuator; LO: local oscillator; DSP: digital
signal processing.

18

3.3 Experimental results and discussions

(a1) (a2)

(a3) (a4)

(b1) (b2)

(b3) (b4)
Fig. 3-2. (a1)-(a4) Phase patterns with charges of (a1) -4, (a2) +8, (a3) -8, (a4) -
16 applied to SLMs 4 thru 1. (b1)-(b5) Intensity profiles of generated pol-
muxed OAM modes (b1) OAM0,+4, (b2) OAM0,+8, (b3) OAM0,-8, (b4)
OAM0,+16.

19

Fig. 3-3. Intensity profiles of super-imposed generated pol-muxed OAMs.

Fig. 3-4. BER performance of demultiplexed (b) Y-pol. of different pol-muxed
OAM modes (OAM0,+4, OAM0,+8, OAM0,-8 and OAM0,+16) without and
with crosstalk.

20

Fig. 3-5. Constellations of 16-QAM for (a) back to back

21

(b) (c)

(d) (e)

(f) (g)

(h) (i)
Fig. 3-6. Constellations for demultiplexed (b)-(e) X-polarization and (f)-(i) Y-
polarization of different pol-muxed OAM modes. (b)(f) OAM0,+4. (c)(g)
OAM0,+8. (d)(h) OAM0,-8. (e)(i) OAM0,+16.OAM0,-8, (b4) OAM0,+16.

22

Fig. 3-7. Spectra of back to back and reconverted Gaussian beams from
different polarizations of pol-muxed OAM modes (resolution: 0.01 nm).

Fig. 3-8. BER performance of demultiplexed (a) X-polarization of different pol-
muxed OAM modes without and with crosstalk.

23

Shown in Fig. 3-2(a1)-(a4) are the computer-generated phase patterns with
charges of -4, +8, -8 and -16 which are applied to SLM4, SLM3, SLM2 and SLM1,
respectively. After reflecting off the four SLMs, azimuthal phases exp(il ) (l=-4,
+8, -8, -16) are added to the Gaussian beams which are converted to OAM(0,+4),
OAM(0,-8), OAM(0,+8) and OAM(0,+16). Note that the charge sign of the OAM
mode reflected off the SLM is opposite to that of phase pattern applied to the SLM,
which is due to the “mirror” image relationship of reflection. Following the
multiplexing of four OAM modes via BS1-BS3, the superimposed multiple OAM
modes after BS3 are OAM0,+4, OAM0,+8, OAM0,-8 and OAM0,+16.

Fig. 3-9. BER performance of demultiplexed (b) Y-polarization of different
pol-muxed OAM modes without and with crosstalk.

24

Pol-muxed four OAM modes are achieved by BS4 and BS5 with their intensity
profiles shown in Fig. 3-2(b1)-(b4). “Donut” shapes are observed with the ring
radius increasing with | l |. Fig. 3-2(b5) depicts the superimposed pol-muxed four
OAM modes. To demultiplex one of the pol-muxed OAM modes, a specified phase
pattern with the opposite charge is loaded onto SLM5 to remove the azimuthal phase
of OAM mode and converted it back to the Gaussian beam, which can be picked
from other OAM modes by pin hole and fiber collimator.
Fig. 3-3 shows spectra of reconverted Gaussian beams from different
polarizations of pol-muxed OAM modes. Benefiting from pre-filtering in the optical
domain, power suppression of ~30 dB is achieved at an offset frequency of 12.5 GHz
from the center, which could be used in the 12.5-GHz grid. Considering 10.7-Gbaud
16-QAM signals over pol-muxed four OAM modes, a spectral efficiency of 25.6
bit/s/Hz is obtained, including the 7% FEC overhead. Fig. 3-4 depicts the BER
curves without (only X-/Y-polarization of a pol-muxed OAM mode is on) and with
(all pol-muxed four OAM modes are on) crosstalk. Less than 1.4-dB OSNR penalty
at a BER of 2e-3 (EFEC threshold) is measured without crosstalk. A total OSNR
penalty of less than 3.5 dB at a BER of 2e-3 is observed with crosstalk. Shown in
Fig. 3-5 are constellations of 16-QAM for demultiplexed X- and Y-polarization of
different pl-muxed OAM modes with measured error vector magnitude (EVM)[16].
25

Chapter 4: System Optimization using Chirp-inducing
Wavelength Converters based on SOA-MZI in a
Dispersion-compensated Optical Link

This chapter presents the study on the system-level optimization of a chirp-
inducing wavelength converter operated in a dispersion-compensated optical link.
The degrading effects induced by a differential cross-phase modulation (DXPM)
based wavelength converter using semi-conductor optical amplifier Mach-Zehnder
interferometer (SOA-MZI) are explored with numerical simulation as well as
experimental demonstration at 40Gb/s. This is the first of two investigations into the
effect of chirp with residual dispersion[21]. A >2dB system improvement can be
achieved by operating the wavelength converter in optimal conditions w.r.t. the
residual dispersion in the optical plant. In this chapter the device considered is based
on a SOA-MZI, while the subsequent chapter considers a general black-box model,
not limited by the physics of any particular device
4.1 Introduction
Optical wavelength conversion is considered a key enabler for efficient routing in
WDM networks. Several approaches have been reported, including using highly-
nonlinear fiber, periodically-poled Lithium Niobate, and semiconductor optical
amplifiers (SOAs)[22,1,23]. Some benefits of SOAs are that integration might make
26

any SOA-based structure more manufacturable and include larger number of
elements. Specifically, a type of SOA-based wavelength converter that improves the
speed and performance is the differential-cross-phase-modulation (DXPM) device,
for which SOAs are embedded within the arms of a Mach-Zehnder interferometer
(MZI)[24]. Most of the research reported on DXPM wavelength converters deals
with optimizing the output signal quality of the wavelength converter, especially in
terms of extinction ratio and signal-to-noise ratio (SNR)[25]. However, the gain and
phase recovery issues can produce chirp in the output data signal. A key remaining
question are the deleterious issues and design guidelines for optimizing the
performance of the entire transmission system, which might be different than
optimizing the performance of the wavelength converter module itself.
In this chapter, we characterize the chirp profile of a DXPM-based wavelength
converter with simulations and experiment[26]; and explore the optimal conditions
of operation when the converted signal is transmitted through dispersion-
compensated fiber links. We show that due to the negative chirp produced by SOA-
MZI, a system improvement of >2-dB can be achieved by inducing +30-ps/nm
residual dispersion in a dispersion compensated SMF link.
27

4.2 Concept

The conceptual and experimental block diagram of our technique is shown in
Fig. 4-1. A 40 Gb/s RZ-OOK (33%) signal ( pump ~1551.13 nm) is generated
using a Mach-Zehnder modulator driven by a 215-1 pseudo-random binary sequence
(PRBS) followed by a pulse-carver[27]. This signal is amplified with an EDFA, split
with a 50% coupler and fed into the two arms of SOA-MZI as control (pump) signal,
with an optical delay and attenuator on one arm. The power at one arm is 10dBm and
2.7dBm at the other arm; the latter is delayed by 10-ps with the optical delay. The
SOA s are biased at 350-mW and 346-mW respectively. A CW probe signal
( probe ~1542.3 nm) at power level of 0dBm is fed into SOA-MZI. With DXPM,
whereby each pair of control pulses produces a -phase shift in the probe signal,
wavelength conversion is achieved. The converted signal is filtered with a 1-nm filter

Fig. 4-1. Concept: Chirp produced by wavelength conversion in SOA-MZI
necessitates positive residual dispersion in fiber link for optimal performance.
Exp. Setup: LD: Laser Diode, Mod: Modulator, PC: Pulse carver, SOA-MZI:
Semiconductor optical amplifier Mach-Zehnder interferometer, BPF: Band-pass
filter, TDCM: Tunable dispersion compensation module, Rx: Receiver
28

at the output of SOA-MZI, and transmitted through a 80-km SMF and 18-km of DCF
with each fiber span followed by an EDFA. To vary the residual dispersion in the
link a tunable dispersion compensator module (TDCM) based on temperature
controlled fiber Bragg gratings is used. The signal after passing through the TDCM
is received with a pre-amplified receiver and analyzed by taking bit error rate (BER)
at an electrically de-multiplexed rate of 10-Gb/s.

Fig. 4-2. Simulation Results: Extinction ratio and Output pulse-width variation
with pump inter-pulse delay

Fig. 4-3. Variation of chirp-profile at points A, B and C .

29

Fig. 4-4. Robustness (in terms of eye height) to residual dispersion of the
various converted signals.

Fig.4-2 shows that by tuning the pump inter-pulse delay the extinction ratio,
pulse-width and chirp of the converted signal is varied. Fig.4-3 shows that optimum
E.R. (point-A) does not correspond to optimum performance (point-B) when
dispersion in a fiber link is not 100% compensated. EH (average eye height) defined
as log10{(µ
1
- µ
0
-3( 1
+ 0
)/µ
1
- µ
0
)} is improved by 0.2-dB when the fiber link has
positive residual dispersion. Moreover, the converted signal has significantly reduced
tolerance to residual negative dispersion in the fiber link.
To experimentally verify this behavior, BER of a wavelength converted signal
was taken while varying the residual dispersion in the fiber link and compared to that
of a non-converted reference signal. The measured chirp on the converted signal was
-18-Ghz. While the non-converted signal has the lowest power penalty at zero
residual dispersion, the converted signal, however has a minima at a net residual
dispersion of positive 30 ps/nm corresponding to a 2.4-dB system improvement. This
can be attributed to the interaction of the negative chirp of the converted signal and
30

the net positive dispersion of the link. Moreover at a reference power penalty of 2-
dB, the non-converted signal can survive about 55 ps/nm of residual dispersion,
while for the converted signal this window of operation reduces to 38 ps/nm. Hence,
the robustness to residual dispersion for the converted signal is reduced by about 17-
ps/nm which corresponds to about a km of SMF.

Fig. 4-5. Experimental Results: Measured chirp of converted signal (-18Ghz)

31

Fig. 4-6. BER of wavelength-converted signal with varying residual dispersion

Fig. 4-7. Power penalty normalized with respect to the optimum performance
point of each signal with varying residual dispersion; .

32

Chapter 5: System Optimization using Black-box Model of
a Chirp-inducing Network Module in a Dispersion-
compensated Optical Link

In this chapter, the second of the two investigations into the interaction of chirp
with residual dispersion of an optical link is explained. While the study in the
previous chapter was based on DXPM in a SOA-MZI wavelength converter, this
chapter considers a black-box chirp-inducing model, where any arbitrary shaped
chirp profile can be attached to a signal[21,26]. The conclusions from this study can
be applied to a broad range of chirp-inducing network elements at 40Gb/s.
5.1 Introduction
Future WDM networks will likely require wavelength converters in order to
enable efficient and high-throughput routing[28]. Optical wavelength converters
have been researched as to their suitability for performing this function transparently
and at high speed. Several approaches have been reported, including using highly-
nonlinear fiber, periodically poled Lithium Niobate, semiconductor optical
amplifiers (SOAs), and electro-absorption modulators. Most of the research reported
on wavelength converters deals with optimizing the output signal quality of the
wavelength converter, especially in terms of extinction ratio (ER) and signal-to-noise
ratio (SNR). These parameters can be optimized at the output of the converter
module. However, due to many different effects that might occur in the different
33

techniques, there might be a chirp induced on the output data signal. This chirp will
interact with the chromatic dispersion of the transmission fiber and degrade the
system performance. Such effects might be beyond the ability for simple dispersion
compensators to fully correct. A key remaining question are the deleterious issues
and design guidelines for optimizing the performance of the entire transmission
system, which might be different than optimizing the performance of the wavelength
converter module itself. Moreover, since each type of module will induce a different
type of chirp for different operating regimes, it becomes quite valuable to explore
system optimization for a wide variety of chirp regimes for given a certain data
signaling.
In this chapter, we explore many different chirp regimes and simulate system
operation. The simulations show with ±10 ps/nm residual dispersion, a symmetric
chirp distribution around the center of the pulse has higher robustness to dispersion
variation by giving ~2dB lower power penalty. When nonlinearities and chirp
interact in fiber, for a chirp peak located on leading or trailing edge power penalty at
BER=10-9 is improved by ~2 dB by the choice of optimum residual dispersion, but
±5 ps/nm deviation from the optimum dispersion cancels this performance
enhancement.
34

Fig. 5-1. Illustrative block diagram of utilizing an arbitrary chirp profile
generator to compare chirp regimes for optimizing the overall system
performance. One (25) span(s) of 80km compensated SMF is used with -13dBm
(-3 dBm) launch power, to explore the interaction of various chirp profiles with
fiber dispersion and nonlinearities.

Fig. 5-2. Normalized chirp profile. Normalized chirp indicates the significance
of chirp impairments in fiber transmission.
5.2 Concept
SOAs and modulators can induce frequency chirping on signal. Although alpha
factor is a useful metric to quantify the device chirp, when a signal passes through a
subsystem such as an SOA-MZI wavelength converter, alpha factor definition gives
a time dependent parameter which may not be as useful for quantifying fiber
35

transmission behavior of the new chirp waveforms [29]. As depicted in Fig. 5-1, the
goal is to build a model to generate a variety of chirp profiles and explore their
transmission performance, to isolate the effect of chirp from extinction ratio (ER)
and signal-to-noise ratio (SNR). The study gives some guidelines for tailoring the
subsystem induced chirp to an optimum fiber transmission. The parameters in the
subsystem, such as an XPM based SOA-MZI wavelength converter[30,31,32], could
then be optimized to achieve the desired chirp profile, at the cost of degrading ER
and/or SNR.
5.3 Modeling and Fiber Simulation
A variety of arbitrary chirp profiles, shown in Fig. 5-2 and 5-3, are generated in
Matlab and are fed into VPI Systems software for fiber transmission simulation. A
40 Gb/s RZ-OOK signal ( c ~1553 nm) with first order Gaussian pulse shape and
FWHM=10 ps is generated and is driven by a 211-1 PRBS data. The extinction ratio
is set to 30 dB. Each bit is sampled at 128 points to capture the impacts of sub-
pulsewidth chirp inside the pulse. Average chirp peak is varied between ±50 GHz.
For the fiber transmission, a low nonlinearities (NLs) regime (launch power of -13
dBm through one amplified and fully compensated span of 80 km SMF/DCF) and a
high nonlinearities regime (launch power of -3 dBm into 25 fully compensated
spans, resulting in 2000 km of SMF) is simulated to study the interaction of chirp
and nonlinearities. The SMF (DCF) has a first order dispersion of 16 ps/nm/km (-90
ps.nm/km), nonlinear index of 2.6 10-20 m2/W (4 10-20 m2/W) and effective
36

core area of 80 m2 (30 m2). Various residual dispersion between 12 ps/nm is
applied at the receiver to explore the ability of dispersion compensation to fully
correct penalties induced by interaction of chirp, dispersion and nonlinearities. BER
is estimated at the receiver to find power penalties at 10-9.
.

Fig. 5-3. Various basic chirp profiles inside a pulse. Power penalties almost the
same for CPx/CNx pairs.

37

Fig. 5-4. Power penalty at 10-9 BER versus average normalized chirp on the
edges for 8ps/nm dispersion for (a) Pin= -13dBm, 80km SMF (b) Pin= -3dBm,
25×80km SMF.

38

Fig. 5-5. (a) Different chirp peak locations. (b) Low NLs: Pin= -13dBm, 80km
SMF. (c) High NLs: Pin= -3dBm, 25×80km SMF. (BER>10-9 for CRef. @10-9
BER).

39

Fig. 5-6. (a) Different chirp regimes. (b) Low NLs: Pin= -13dBm, 80km SMF. (c)
High NLs: Pin= -3dBm, 25×80km SMF. (BER>10-9 for CRef. Power penalties at
10-9.)

40

5.4 Results and Discussion
The concept of “normalized chirp” is depicted in Fig. 5-2. The “average
normalized chirp” on leading (trailing) edge is defined as the product of chirp with
intensity divided by the maximum peak power of the pulse which helps to compare
various chirp regimes. In Fig. 5-4, each triangle corresponds to a chirp profile in Fig.
5-3. The power penalty at BER=10-9 is plotted at 8 ps/nm dispersion for low and
high power launch power, and shows almost equal power penalties for pairs of chirp
profiles labeled CNx/CPx in Fig. 5-3. Figs. 5-5(b) and 5-6(b) show dispersion
tolerance of chirp profiles at low launch power and Figs. 5-5(c) and 5-6(c) depict
penalties when fiber nonlinearities are high. From Fig. 5-5, central chirp peak, CP1,
has ~0.5dB power penalty at ±10 ps/nm dispersion, while other profiles show ~2dB
penalty. As can be seen in Figs. 5-4(b), 5-5(c) and 5-6(c), when nonlinearities and
chirp interact in fiber, simple dispersion compensation does not fully correct
penalties for a chirp peak at the center of the pulse. Symmetric chirp around center,
C2, is more robust, but has ~1.5 dB penalty, while in a negative chirp, C3, power
penalty is improved by ~2 dB with the optimum choice of residual dispersion. Yet,
±4 ps/nm deviation from the optimum dispersion cancels this improvement.
41

Chapter 6: System performance of DPSK signals
transmitted through broadband SBS-based slow light
element and reduction of slow-light-induced data-pattern
dependence

In this chapter, we carry out a comprehensive study of differential phase
modulated signals when transmitted through a delay element based on stimulated
Brillouin scattering (SBS). Slow light delay based on SBS has recently seen hectic
research activity to explore its viability in a real commercial network. Since phase
modulated signals are increasing becoming the popular choice compared to intensity
modulated format, it is therefore, paramount to investigate the debilitating effects
that a SBS-based slow light element will incur on a phase-modulated signal. It is
shown that the SBS-based delay will affect the duobinary and AMI ports in a
different manner. Moreover, it is shown that by detuning the center frequency of the
input signal from the spectral center of the SBS-delay bandwidth, system
improvement can be enhanced. .
6.1 Introduction
Slow light techniques have enjoyed much recent interest due to the potential
systems applications involving tunable delay lines, such as bit-level synchronizers,
equalizers, and signal processors. In general, slow light is achieved by tailoring an
42

enhanced group-index resonance within a given medium[33]. Unfortunately, a high
group index accompanied by a narrow resonance bandwidth tends to also distort a
high-bit-rate data signal and cause data-pattern dependence[34]. Recent publications
have described the quality of a data stream that has passed through the slow light
element, including results for data-pattern dependence, limited data bandwidth, and
bit-error-rate (BER) measurements. Promising slow light techniques to achieve
tunable delay lines for Gbit/s data include the use of: (i) stimulated Brillouin
scattering (SBS) in fiber, (ii) stimulated Raman scattering (SRS) on silicon chip, (iii)
optical parametric amplification (OPA) in fiber, and (iv) four-wave-mixing (FWM)
in semiconductor optical amplifiers (SOAs).
We emphasize that all previously published slow light system results were for
intensity-modulated signals. However, phase-encoded formats, such as differential-
phase-shift-keying (DPSK), have not been explored in a slow light element before.
DPSK is becoming ever-more important in the optical communications community
due to its potential for increased receiver sensitivity, tolerance to various fiber
impairments, and better spectral efficiency. It is highly desirable to understand how
the phase information of the DPSK signal could be preserved and how much
fractional delay it could experience. Furthermore, it is important to explore how slow
light nonlinearities could affect differently the demodulated two ports of a delay
interferometer-based DPSK receiver. A laudable goal would thus be to examine
43

critical system limitations on Gbit/s DPSK data as it traverses a tunable slow light
element.
In this chapter, we demonstrate experimentally and via simulation slowing down
of a phase-modulated signal. A 10.7-Gb/s NRZ-DPSK signal can be delayed by as
much as 42 ps (45% fractional delay) while still achieving error free via broadband
SBS-based slow-light element.
We further analyze slow-light-induced data-pattern dependence on demodulated
output ports. By detuning the SBS gain profile, 3-dB Q factor improvement is
achieved by reducing the data-pattern dependence. Performance comparison between
2.5-Gb/s and 10-Gb/s with the same fractional delay shows that data-pattern
dependence is bit-rate specific. Finally, system level comparisons of 2.5-Gb/s NRZ-
DPSK with RZ-DPSK under the same 5-GHz SBS bandwidth show different
robustness to slow-light-induced data-pattern dependence.

44

Fig. 6-1. Concept of slow light on phase-modulated optical signals.
6.2 Concept of slow light on phase-encoded optical signals
The concept of slowing down phase-modulated optical signals is shown in Fig. 6-
1 (Left). When a DPSK signal passes through the slow light element, one expects
that its phase patterns get delayed according to the slow light gain and bandwidth.
Meanwhile, phase preservation should also be expected for information integrity.
However, commonly-generated DPSK signals feature unavoidable residual intensity
modulation, which also experiences slow-light nonlinearities. Demodulation of such
delayed DPSK signal encounters the problem of data-pattern dependence on both the
constructive “DB” (Duo-binary) and the destructive “AMI” (Alternate-Mark-
Inversion) ports after the one-bit delay interferometer (DI), as shown in Fig.6-1
(Left). It is thus crucial to analyze critical system limitations on Gbit/s DPSK signals
45

transmitted through a narrowband tunable slow-light element. Figure 6-1 (Right)
shows the simulation result of slow light on the phase patterns of a 10-Gb/s DPSK
signal. The slow-light element is analytically modeled to have a Lorentzian-shaped
imaginary part of the refractive index, with controllable bandwidth and gain.
Kramers–Kronig relationship determines the real part of the refractive index, whose
derivative gives the slow- light delay profile. A 10-Gb/s NRZ-DPSK signal is
simulated using a Mach-Zehnder modulator with proper bias and driving voltage.
The phase patterns of the DPSK signal are shown both before and after an 8-GHz
slow-light element. We show that phase patterns can be delayed by up to 46-ps and
the differential phase relationship preserves quite well. This confirms the
concept of slow light delays on phase information.

46

Fig. 6-2. Slow-light induced data-pattern dependence on demodulated two
output ports. Simulation result of phase patterns of a 10-Gb/s DPSK signal
before and after 8GHz BW slow light element. Phase is delayed by 46 ps.

47

6.3 Experimental results of slow light on 10-Gb/s NRZ-DPSK
signals

We further carry out DPSK slow-light experiment and the setup is shown in Fig.
6-2 (Left). The slow-light mechanism is based on broadband SBS in a piece of
highly nonlinear fiber (HNLF). Broadband SBS pump is used to accommodate
Gbit/s optical signals. We use a Gaussian noise source driven by 400-MHz clock to
modulate the injection current of a commercial directly-modulated laser (DML). The
pump spectral-width is adjusted by an RF attenuator. The broadband pump is then
amplified by a high-power EDFA and enters a 2-km HNLF, with the measured
Brillouin shift to be 10.3-GHz. An NRZ-DPSK probe data stream is generated by
externally modulating the tunable laser source (TLS) using a Mach-Zehnder
modulator (MZM), which is biased at its transmission null and driven by
approximately 2V. A sinusoidally-driven second pulse carver modulator is used to
generate 50% RZ-DPSK signals. The amplified and attenuated DPSK signal with
controllable power counter propagates with the pump in the HNLF. One polarization
controller is used on the signal path to maximize the SBS interaction. The amplified
and delayed DPSK signal is finally demodulated using a one-bit DI and both DB and
AMI ports are detected. An optical attenuator is adjusted accordingly to the SBS gain
so as to keep the input power into the EDFA fixed. BER measurements are taken on
both the DB and AMI demodulated signals.
48

Fig. 6-3. Experimental Setup for DPSK slow-light based on broadband SBS.

Fig. 6-4. Observation of DPSK slow-light: continuous delay of up to 42 ps for a
10.7Gb/s DPSK signal

49

Figure 6-2 shows the measured delay of a 10.7-Gb/s NRZ-DPSK signal with
0dBm power under an 8-GHz SBS gain bandwidth. The measured delay scales fairly
linearly with the increased pump power, demonstrating the ability to continuously
control the delay of the DPSK phase pattern. The detected balanced DPSK eyes are
shown for three different pump powers, with a maximum of 42 ps delay at a pump
power of 800 mW. The achieved 42 ps delay of a 10.7-Gb/s NRZ-DPSK signal
corresponds to a fractional delay of 45%.
6.4 DPSK data-pattern dependence
As shown in Fig. 6-2 (Right), delayed DPSK eyes exhibit severe signal distortion
with the increased slow light delay. In order to assess the signal quality, we analyze
both the constructive and destructive ports of the DI after demodulation individually.
Figure 6-3 shows the 10.7-Gb/s NRZ-DPSK intensity patterns before and after
passing through the slow light element, with positions recorded right before
demodulation (NRZ-DPSK) and right after demodulation (DB and AMI),
respectively.
The typical and well recognized method for generating an NRZ-DPSK signal,
using an MZM, has several advantages: (i) exact phase modulation, (ii)
insignificant frequency chirping, and (iii) increased tolerance to driving voltage
imperfections [7]. However, residual intensity modulation occurs unavoidably during
50

phase transitions. We can categorize these “intensity dips” as isolated “1”s (between
two consecutive dips) and consecutive “1”s (between two long separated dips).
Isolated “1”s occupy higher frequency components compared to consecutive “1”s,
and will therefore experience much less gain after passing through a narrowband
slow-light resonance. This effect can be clearly seen for the distorted NRZ-DPSK
intensity patterns after slow light.

Fig. 6-5. Slow-light-induced data-pattern dependence: 10.7-Gb/s NRZ-DPSK
through an 8-GHz slow light element. Bit patterns before and after
demodulation are shown.

The pattern-dependent gain NRZ-DPSK experiences will translate into two
different types of data-pattern dependence on demodulated two signals. In the DB
port, the peak power is much higher for long “1”s, compared with single “1”s. This
can be explained from the fact that single “1”s are only demodulated from two
consecutive “1”s in NRZ-DPSK which has a much slower rising time due to slow-
light third-order dispersion. This leads to an insufficient constructive interference for
51

the generation of single “1”s. The AMI port exhibits strong pattern dependence
within a group of “1” pulses. Compared with the “1”s in the middle, the leading and
the trailing “1”s always have much higher peak powers in that they both experience
unequal-power constructive interference from the edge pulses in a group of “isolated
dips” in delayed NRZ-DPSK pattern. Both DB and AMI eye diagrams exhibit
vertical data-pattern dependence. Furthermore, the AMI port also features non-
negligible pulse walk-off, which can be attributed to the slower rising and falling
times of the two edge pulses compared with fast-transitioned middle pulses, in a
group of “1” pulses.
BER measurements on the DB port of a demodulated 10.7-Gb/s NRZ-DPSK
signal under different delay conditions are shown in Fig. 6-4 (Left). We emphasize
that we could still achieve error free at a delay of up to 42 ps with a power penalty of
9.5dB. The clear tradeoff between signal fidelity and delay can be explained by the
following two main reasons. Data-pattern dependence due to limited slow-light
bandwidth is one major factor for signal degradation, as confirmed by the vertically
closed eyes. Not only the gain but also the phase (delay) spectrum of the broadband
SBS will affect the delayed PRBS data quality. Spectra in Fig. 6-4 (Left) show that
crosstalk from Rayleigh backscattering of the broadband pump is another contributor
to the power penalty, especially when the bit-rate is comparable to the Brillouin shift.
The performance of the demodulated AMI port from 10.7-Gb/s DPSK signals is
worse than that of the DB port because of severe pulse-walkoff and increased
52

Rayleigh spectral overlapping due to much wider AMI bandwidth. Figure 6-4
(Right) shows the performance comparison of 10.7-Gb/s and 2.5-Gb/s NRZ-DPSK
data with a fixed SBS gain bandwidth of 7-GHz. System performance of 2.5-Gb/s
NRZ-DPSK exhibits 6.5dB better performance at 800 mW pump power, the main
reason being lower bit-rate signals see much less data-pattern dependence and much
smaller Rayleigh crosstalk, as can be confirmed by the two DB eyes.

Fig. 6-6. BER of DB port from 10.7-Gb/s DPSK signals after SBS slow light
element. Data-pattern dependence and Rayleigh crosstalk are the two main
reasons for DPSK signal degradation.
53

Fig. 6-7. Power penalty comparison between 2.5-Gb/s and 10-Gb/s NRZ-DPSK
shows that data-pattern dependence is bit-rate specific.

6.5 Reduction of DPSK data-pattern dependence
Realizing that the slow-light-induced data-pattern dependence mainly comes
from the patterndependent gain, we red-detune the peak of the SBS gain profile by
0.016nm from the channel center, resulting in gain equalization and thus pattern-
dependence reduction between isolated “1”s and consecutive “1”s within NRZ-
DPSK “intensity dips”, shown in Fig. 6-5. Bit-patterns and eye diagrams with and
without detuning for both demodulated DB and AMI ports are also recorded for
comparison. The optimum 3-dB Q factor (determined from BER measurement)
improvement (from 12 to 15dB) for the AMI eyes confirms the effectiveness of this
detuning method. The detuning not only resolves vertical data-pattern dependence,
54

but also reshapes the rising and falling times of the edge pulses in a group of “1”
pulses, such that pulse walk-off is also alleviated, as can be seen from the AMI eye
diagram after detuning.

Fig. 6-8. Reduction of DPSK data-pattern dependence by detuning the SBS
gain peak: 3-dB Q factor improvement on the AMI port is achieved.

6.6 System performance comparison between 2.5-Gb/s NRZ-DPSK
and RZ-DPSK

Motivated by the fact that RZ-DPSK is also another popular modulation format
thanks to the increased tolerance to fiber nonlinearities, we conduct performance
comparison of NRZDPSK with RZ-DPSK at a bit rate of 2.5-Gb/s. The reason we
are not comparing them at 10- Gb/s is that RZ-DPSK bandwidth exceeds the 10-GHz
Brillouin shift. Figure 6-6 shows delay and power penalty comparison as a function
of increased pump power. Under a fixed 5-GHz SBS gain bandwidth, the fractional
55

delay (absolute delay divided by pulse-width) of RZDPSK is comparable with that of
NRZ-DPSK. In terms of signal quality, RZ-DPSK outperforms NRZ-DPSK by as
much as 2dB at 700 mW pump power. The inset AMI eye diagrams show that RZ-
DPSK is much more tolerant than NRZ-DPSK in terms of slow-light induced data-
pattern dependence. The main reason can be understood from the fact that pulse
carver modulator used in RZ-DPSK extracts only the amplitude-modulation-free
center portions of the bits, thus largely eliminating any residual dips, which is the
main cause of data-pattern dependence in NRZ-DPSK.

Fig. 6-9. Delay for 2.5-Gb/s NRZ and RZ-DPSK with the same 5-GHz SBS
BW. The fractional delays for both NRZ and RZ-DPSK are comparable.

56

Fig. 6-10. RZ-DPSK outperforms NRZ-DPSK by as much as 2dB, which
shows its robustness to data-pattern dependence

57

We experimentally demonstrate slow light effect on a phase-encoded optical
signal. By utilizing broadband SBS-base slow light in HNLF, 10.7-Gb/s NRZ-DPSK
signals can be continuously delayed by as much as 42 ps while still achieving error
free. Slow-light-induced DPSK data-pattern dependence on demodulated output
ports are systematically analyzed and reduction of data-pattern dependence is
achieved by detuning the SBS gain peak away from the channel center frequency,
resulting in 3-dB Q factor improvement for the AMI port. Future research directions
as to slow down >10-Gb/s phase-modulated signals would involve the use of narrow
band parametric amplification in optical fibers, which proves to maintain signal
fidelity while still achieving reasonable slow-light delay.

58

Chapter 7: SOA-Assisted Data-Polarization-Insensitive
Wavelength Conversion in a PPLN Waveguide

In this chapter, we demonstrate a novel method to reduce the polarization
sensitivity of a periodically poled Lithium Niobate (PPLN)-based wavelength
converter by introducing a SOA-based wavelength converter prior to it[35]. The first
SOA-based module changes the ON-OFF keyed signal to a polarization-modulated
signal, while the subsequent PPLN-based wavelength converter re-converts the
polarization-modulation back to ON-OFF keying. Two modes of operation (data-
inverting and data-non-inverting) are introduced, and the polarization sensitivity of
the wavelength converter is reduced by more than 30-dB.
7.1 Introduction
An all-optical wavelength converter is an essential device in wavelength-division
multiplexed (WDM) systems. Some of the critical functions in optical systems and
networks it can fulfill include reconfigurable routing, contention resolution,
wavelength reuse, multicasting, and traffic balancing. All-optical wavelength
conversion is usually achieved by use of nonlinear effects in semiconductor devices
or optical fibers, where a local laser interacts with the incoming data signal,
converting it to a new output wavelength. However, a key technical challenge for
59

many wavelength converters is that they are sensitive to the polarization state of the
incoming data signal relative to the nonlinear device. Various methods have been
published that alleviate the problem of polarization sensitivity of wavelength
converters, including i) scrambling the polarization of the incoming data probe
signal, ii) polarization diversity using a polarization beam splitter and two
polarization-sensitive elements in parallel, and iii) using active polarization tracking
and rotation of the incoming data signal. We propose and demonstrate a wavelength
conversion method that is relatively insensitive to the polarization state of the
incoming data signal. A technique involving polarization modulation of a
periodically poled lithium niobate (PPLN) waveguide pump channel is used to
convert the incoming data to a new wavelength. This polarization modulation is
achieved using cross-polarization modulation (XpolM) in a semiconductor optical
amplifier (SOA). Using this configuration, we can reduce the polarization
dependency of the PPLN waveguide from 30 dB to 1 dB. Use of second harmonic
generation (SHG) followed by difference frequency generation (DFG) in the PPLN
waveguide also enables “multicast” operation wherein a single data channel can be
converted to multiple output wavelengths. We demonstrate multicast operation for
three wavelengths simultaneously. Moreover, to demonstrate that polarization
modulation of the PPLN waveguide can also be realized at high speeds, we use
XPolM in highly nonlinear fiber (HNLF) at 10 Gb/s.
60

7.2 Concept
As shown in Fig. 7-1, the generalized concept of our proposed technique can be
divided into two stages.

Fig. 7-1. The polarization of the SOA probe (observed on a polarimeter)
changes as the SOA pump alternates between ON and OFF.

The first stage is a polarization-insensitive amplitude-to-polarization converter.
We utilized cross-polarization modulation (XpolM) in an SOA to convert the
intensity variations of a high power signal into polarization modulation of a low
power probe. The CW probe beam’s polarization is aligned at the input of the SOA
such that it is split between the TE and TM modes of the SOA. A pump pulse causes
61

suppression of the carrier density in the SOA that alters its refractive index leading to
a modulation of the probe wave’s phase. Owing to the structural asymmetry of the
SOA, the phase-shifts on the TE and TM components of the probe are different
leading to a rotation of the state of polarization (SOP) of the probe wave as
illustrated in Fig. 7-2. SOP was observed using a polarimeter.
The second stage is a highly polarization-sensitive wavelength converter, here,
implemented via SHG followed by DFG in a PPLN waveguide. Therefore, if the
output of the first stage acts as the pump for the PPLN waveguide, the changes in its
SOP will translate into changes in the wavelength conversion efficiency of the PPLN
waveguide. This will ultimately lead to amplitude modulation of the other CW
channel coupled into the PPLN waveguide, via the process of DFG.
It should be noted that XpolM is accompanied by cross-gain modulation (XGM),
since both originate from the input-signal induced carrier suppression in the SOA. As
a result, the pump signal entering the PPLN waveguide is not only polarization
modulated, but exhibits power variations as well. As explained later in Section IV,
the interplay of this polarization rotation and simultaneous power variation
determines whether the data polarity is maintained or reversed at the output of the
system.
7.3 SHG:DFG in a PPLN waveguide
PPLN waveguides have been widely used in various applications including
wavelength conversion, all-optical signal processing and all-optical networking
62

applications . PPLN waveguides provide a large bandwidth, do not induce additional
noise and are chirp-free [7].

Fig. 7-2. Second harmonic generation followed by difference frequency
generation in a periodically poled lithium niobate waveguide..

As shown in Fig. 7-3, second harmonic generation followed by difference
frequency generation (SHG:DFG) is a : process. In the first nonlinear process. the
pump wavelength generates second harmonic at . This second harmonic beats with
any other wavelength coupled into the PPLN waveguide and generates an idler
wavelength at , which constitutes the second nonlinear process of DFG. For multi-
wavelength operation, instead of a single input wavelength, a comb of wavelengths
are coupled into the PPLN waveguide which simultaneously generate idler
wavelengths mirrored about the pump wavelength. However, for optimal conversion
efficiency, the pump and input polarizations have to be aligned optimally with
63

respect to the PPLN waveguide. This is usually accomplished by inserting
polarization controllers in the paths of all the individual wavelengths fed to the
PPLN waveguide or by using polarization maintaining fiber.
7.4 Data-inverting and non-inverting modes

Fig. 7-3. (a) Data-inverting mode, where XGM supports XPolM. (b) Data-non-
inverting mode, where XGM opposed XPolM.

In the proposed scheme of deploying an SOA before a PPLN waveguide,
whether XGM and XpolM are in the same direction or opposite, leads to two regimes
of operation for the overall module, conceptually clarified in Fig. 7-4(a) and (b). In
both parts of the figure, the circular symbols with an arrow inside (under each bit)
64

are drawn to signify the state of polarization of the signal, while the polarizer shows
the optimum polarization orientation for the PPLN waveguide.
“Data-Inverting Mode”: The polarization of the pump entering the PPLN
waveguide is adjusted (using a polarization controller) in such a way that its
wavelength conversion efficiency is the highest possible when the signal power
entering the SOA is “low.” When the signal power goes “high,” it causes
polarization rotation of the PPLN waveguide’s pump, away from its optimum,
leading to a loss of conversion efficiency. Concurrently, XGM in the SOA also
reduces the PPLN waveguide’s pump power and causes the wavelength-converted
output to be reduced further. Thus, cross-polarization modulation and cross-gain
modulation assist each other and the final output power reduces when the input
signal power increases. This corresponds to the data-inverting mode of operation.
“Data-Non-Inverting Mode”: If the polarization of the PPLN waveguide’s pump
is initially adjusted to be off its optimum direction, the final wavelength converted
output is “low” when the input signal is “low.” When the input signal power
increases, it causes the PPLN waveguide’s pump polarization to rotate towards its
optimum, leading to an increase in its wavelength conversion efficiency and a higher
output power. However, since XGM in the SOA causes the power in the PPLN
waveguide’s pump to drop, the final converted output power is being reduced. It is
clear that in this regime XPolM and XGM oppose each other, and the final output’s
65

extinction ratio is limited. In this regime, the data polarity is maintained from input
to output.
7.5 Characterization of the polarization sensitivity of PPLN
waveguide

The quasi-phasematching wavelength of the PPLN waveguide we used for this
experiment was 1554.78 nm at 91.5 C. The waveguide was 5 cm long, fiber pigtailed
on both ends, and it had an insertion loss (propagation loss and coupling loss
together) of 4 to 5 dB. In order to characterize the polarization sensitivity of the
PPLN waveguide, we injected a CW pump (1554.78 nm) through a polarization
controller into the PPLN waveguide along with another CW signal and observed the
spectrum of the converted wavelength at the output of the PPLN waveguide on an
optical spectrum analyzer (OSA).

66

Fig. 7-4. Polarization sensitivity of DFG in a PPLN waveguide. The wavelength
converted output varies by >30 dB with changes in the pump polarization.

To quantify the polarization sensitivity of the nonlinear processes, the
polarization of the pump was rotated over a wide range and the power in the output
(both of the pump and the converted wavelength) was measured. As shown in Fig. 7-
5, the PPLN waveguide’s converted output exhibited polarization sensitivity in
excess of 30 dB.
7.6 Reduction of the PPLN waveguide’s polarization
sensitivity using an SOA module

The complete experimental setup is shown in Fig. 7-6. The wavelength of the
CW probe in the first stage has to match the PPLN waveguide’s pump wavelength in
the second stage.
67

Fig. 7-5. Experimental setup. Mod (intensity modulator), EDFA (erbium doped
fiber amplifier), SOA (semiconductor optical amplifier), BPF (bandpass filter),
PPLN (periodically poled lithium niobate waveguide), and RX (optical
receiver).

Thus, the low power probe’s wavelength (into the SOA) was fixed at 1554.78 nm
. The probe’s polarization launched into the SOA was adjusted using a polarization
controller such that it split between the TE and TM modes, leading to a large
signal—induced polarization rotation. The input signal wavelength for the first stage
was 1550 nm, and its power was controlled using an attenuator before the coupler.
After the SOA, we filtered out the probe, amplified it, and fed it to the PPLN
waveguide as its pump. Three other CW wavelengths were coupled into the PPLN
waveguide to generate three converted outputs for “multicast” operation. The static
behavior (i.e., without any signal modulation on ) of the final output was analyzed on
an OSA, while the dynamic behavior (i.e., with modulated with a 2.5-Gb/s signal)
was observed by filtering out individual wavelengths. To quantify the performance
of the technique, we did not modulate the signal in the first stage, and used the
68

attenuator to control the total signal power being injected into the SOA. While the
input signal power was “low,” the polarization controller for the PPLN waveguide’s
pump was adjusted so that the final converted output became “low.” As the input
signal power was increased, it induced rotation of the PPLN waveguide’s pump
towards its optimum, leading to an increase in the final converted output power. This
transition is depicted for the “multicast” operation in Fig. 7-7.

Fig. 7-6. PPLN waveguide output spectrum. Non-inverted mode of “multicast”
wavelength conversion using our proposed technique.

69

Note that the EDFA used between the two stages was operated in constant output
power mode, thereby somewhat suppressing the XGM effect. An overall output
extinction ratio of 9 dB was achieved when the input signal power was changed by
20 dB. Conversely, when the PPLN waveguide’s pump polarization controller was
initially adjusted to provide maximum converted output when the input signal power
was “low,” the inverted mode of operation was achieved. Increasing the signal power
caused the PPLN waveguide’s pump polarization to rotate off its optimum position,
thereby introducing a drop in conversion efficiency. An EDFA operating in constant
power mode between the two stages allowed the XGM as well as XpolM effects to
propagate through to the next stage. Since they assisted each other, an extinction
ratio 23 dB for the same 20-dB change in input signal power could be observed. Fig.
8 shows the experimentally observed variation in the final converted output power
with changing input signal power. The total probe power at the output of the SOA is
also shown to quantify the XGM effect. As is clearly visible, the output variation is
dramatically improved in the inverting mode, since XGM and XpolM assist each
other.
70

Fig. 7-7. Polarization-insensitive operation of our technique. The input signal’s
polarization affects the output by <1 dB.

71

In order to verify the polarization insensitivity of the setup, we changed the
polarization state of the input signal to the SOA using a polarization controller and
measured the output extinction ratio for each changed input polarization. These
results are shown in Fig. 7-9 for both inverted and non-inverted regimes of operation,
demonstrating a polarization sensitivity of 1 dB.
To validate our concept with actual data, we modulated the SOA pump with 2.5
Gb/s pseudo random bit sequence (PRBS) data. The bit rate in our experiment was
limited by the long carrier recovery time of the SOA. This limitation can also be seen
in the output eye diagram (inset of Fig. 10). Polarization-independent performance
was verified in the inverting mode, the BER curve for which is shown in Fig. 7-10.

Fig. 7-8. BER curve for inverted mode of operation, where XGM and XpolM
assist each other (2.5 Gb/s). The output eye shows the slow recovery time of the
SOA.

72

By further optimizing the polarizations of the SOA’s CWprobe and PPLN
waveguide’s pump and using an SOA with a higher power-induced birefringence,
improvement in extinction ratios should be achievable. An SOA with a faster carrier
recovery time will enable higher-bit-rate operation.
7.7 Polarization modulation of PPLN waveguide pump by
cross polarization modulation in HNLF

To show that the PPLN waveguide pump polarization modulation is not limited
with respect to bit rate, we modulate the polarization of the PPLN waveguide pump
by using cross-polarization modulation in a highly nonlinear fiber (HNLF), which
has been shown to allow large bandwidth operation. The zero dispersion wavelength
of the HNLF is 1552 nm, dictating the pump wavelength.

Fig. 7-9. Experimental setup for using HNLF for polarization modulation of
PPLN waveguide pump.

73

As shown in Fig. 7-11, the pump was modulated with 10-Gb/ non-return-to-zero
ON–OFF keying (NRZ OOK) using an external modulator. The dummy wavelength
was kept at 1554.78 nm due to it serving as the pump of the PPLN waveguide.With
the help of the polarizer, the dummy wavelength was oriented at 45 with respect to
the input polarization of the pump. This maximized the cross-polarization
modulation of the dummy wavelength. This dummy wavelength was filtered at the
output of the HNLF, amplified with an EDFA, coupled with another CW wavelength
at 1550 nm (chosen arbitrarily) and launched into the PPLN waveguide. The
converted wavelength at the output of the PPLN waveguide was filtered, amplified
and detected using an optical receiver. As shown in Fig. 7-12 the converted
wavelength at the output of the PPLN waveguide was amplitude modulated.

Fig. 7-10. Conversion from polarization modulation of PPLN input pump to
amplitude modulation of the converted wavelength at output of PPLN
waveguide. (a) Converted wavelength at PPLN output. (b) Error-free eye at
optical receiver.
74

It needs to be clarified that this setup is not polarization-insensitive to the state of
polarization of the signal (i.e., the pump of HNLF), since a 45 orientation has to be
maintained between the dummy wavelength and the HNLF pump. However, it
demonstrates that the speed limitation in our earlier SOA-PPLN setup is caused by
the slow carrier recovery time of the SOA and not the PPLN waveguide.
We demonstrate reduction of polarization sensitivity of PPLN waveguide
wavelength conversion by utilizing cross-polarization modulation in an SOA.We
have shown the difference between the two regimes of operation i.e., the data-
inverting and the data non-inverting regime. In the former regime, where cross-
polarization modulation in the SOA is assisted by cross-gain modulation, we
demonstrate a reduction in the PPLN waveguide polarization sensitivity by more
than 30 dB. BER at 2.5 Gb/s is shown, being limited in speed due to the slow
recovery time of the SOA. To demonstrate that this technique is not speed limited by
the PPLN waveguide, we use cross-polarization modulation in HNLF at 10 Gb/s and
show error-free eye diagrams in this mode of operation.

75

Chapter 8: Optical Data Packet Synchronization and
Multiplexing using a Tunable Optical Delay based on
Wavelength Conversion and Inter-channel Chromatic
Dispersion

In this chapter we experimentally demonstrate the application of a conversion-
dispersion based delay for a time-slot interchange functionality for 10 Gb/s optical
packets. The delay element is based on converting the frequency of an incoming
signal to a desired frequency and transmitting the converted signal through a
dispersive element (dispersion compensating fiber) such that the time of flight of the
signal determines its relative delay. The packet size is 196 bits so that it is
comparable to ATM size packets. A delay of 26.4 ps is utilized to delay the packet
by more than a packet length.
8.1 Introduction
Optical packet switching holds the promise of highly efficient use of the
available bandwidth in an optical network. As with electronic packet switching, a
key enabling technology is the ability to controllably delay/buffer a data packet so as
to synchronize packets and rapidly resolve output port contention. Desirable
characteristics of this delay include large tuning range, continuous tunability, and
rapid reconfiguration. Previously published results for optical delays used in packet
switching have typically been chosen from a finite set of discrete optical path
76

lengths. This methodology produces only a fixed set of delays, whether they are in
fiber, waveguides or free-space. However, a laudable goal for a truly flexible and
efficient packet switch would be to generate any arbitrary delay value. Recently,
tunable delays have been shown using wavelength conversion via self-phase
modulation or four wave mixing coupled with dispersive elements. In this method, a
data stream was converted to a slower-propagating wavelength, followed by
conversion back to the original wavelength. Note that proper dispersion
compensation was necessary. The result is that a data bit arrives at the output,
delayed in time relative to a data bit that had always resided at the original, faster-
propagating wavelength. Bit-level delays were shown, but little was reported relating
to using these delays for time manipulation and synchronization of full data packets.
In this chapter, we experimentally demonstrate synchronizing and multiplexing of
optical data packets using a 26-ns tunable optical delay based on wavelength
conversion and inter-channel chromatic dispersion. We achieved tunability for 10-
Gbit/s non-return-to-zero (NRZ) on-off keyed (OOK) data packets, multiplexed two
packet streams, and measured a 10
-9
bit error rate (BER). The delay reconfiguration
time from one delay value to another was measured to be <300 ps..

77

Fig. 8-1: Packet 3 (P3) passes through the delay module which consists of
periodically-poled lithium-niobate (PPLN) -converters, a dispersion
compensating fiber (DCF) and a chirped fiber Bragg grating (FBG).
Concept of conversion/dispersion based delay module for packet synchronization
Shown in Fig. 8-1 is the packet-synchronizing scheme. Two packet streams on
two different wavelengths are routed such that the packets requiring delay (packet-3
on 1) pass through the delay module. After a packet is delayed for
synchronization, it is then multiplexed with the non-delayed stream ( 2) using a 3-
dB optical coupler.

78

Fig. 8-2: In the two scenarios, 1 is converted to different c s, resulting in
different group velocities due to inter-channel dispersion. The undesired intra-
channel dispersion is compensated by a FBG. (PPLN : periodically poled lithium
niobate waveguide)

As shown in Fig. 8-2, the delay module uses wavelength-dependent chromatic
dispersion generated by a dispersive element, such as dispersion compensating fiber
(DCF). Using sum frequency generation (SFG) followed by difference frequency
generation (DFG) in PPLN-1, we convert 1 to the desired wavelength c. The
optical delay equals the total wavelength shift ( 1- c) multiplied by the total
dispersion (ps/nm) of the DCF. We use a second PPLN to convert the delayed signal
back to the original wavelength, thereby preserving the original wavelength at the
output. Since the DCF also causes data-degrading intra-channel dispersion, we use a
chirped fiber Bragg grating (FBG) centered at 1 and with the opposite dispersion
of the DCF in order to perform output dispersion compensation.
Two stage wavelength conversion in PPLN
The wavelength conversion process is shown in Fig. 8-3. SFG in Fig. 8-3(a) is
produced by using two laser pumps that are spectrally equidistant from the quasi-
79

phase-matched (QPM) PPLN wavelength. In PPLN-1, the two pumps are the input
signal at 1 and a local pump at pump-1. These two pumps mix via the first 2
process of SFG to generate p/2 = {( 1+ pump-1)/2}/2. This p/2 mixes with
another input dummy wavelength dummy-1 to produce a converted output at
c = 2 p- dummy-1 via the second 2 process of DFG. By tuning dummy-1,
we can tune our converted signal c to almost any desired value within the PPLN
bandwidth.

Fig. 8-3: 1
(input signal) and pump-1
constitute the pumps for PPLN-1. By tuning
dummy-1
, the output c
can be tuned.
80

Fig. 8-4. Spectral arrangement of PPLN-2. 1
is the output.

At PPLN-2 in Fig. 8-3(b), c and pump-2 are both equidistant from the QPM
of PPLN-2 and constitute the two pumps. By tuning dummy-2, we can ensure that
the output wavelength is equal to 1 again, thus preserving the original signal
wavelength. Although PPLN devices can have a bandwidth of more than 70 nm, our
bandwidth is limited to ~25 nm since we use erbium-doped fiber amplifiers (EDFAs)
to amplify the signals. We note that the “continuous” delay range has small “gaps”,
such that the converted wavelength cannot be spectrally located at the QPM or the
local pump. However, since the QPM is temperature tunable, we can avoid this
problem by slightly tuning the QPM and the local pump wavelengths such that the
gaps can be removed. Furthermore, another inaccessible is the input signal 1,
and this corresponds to zero delay.
81

8.2 Experimental setup and results
The experimental setup is shown in Fig. 8-4. The 196-bit, NRZ-OOK 10-Gbit/s
data packets are 19.6-ns long, and the packet guard time of empty space is 8 bits
(800 ps). Packets are generated electronically and drive Mach-Zehnder intensity
modulators. We manually programmed the pulse pattern generator (PPG) in order to
generate the stream of 196-bit data packets, and the BERs were measured by
programming the PPG and error detector accordingly.

Fig. 8-5: Experimental setup: LD (laser diode), Mod (modulator), PPG (pulse
pattern generator), PPLN (periodically-poled lithium-niobate), DCF (dispersion
compensating fiber), FBG (fiber Bragg grating), PC (polarization controller), Circ
(circulator), EDFA (erbium doped fiber amplifier) and Rx (receiver). Note that
ovals are simple passive couplers.

The 19.74-km DCF has a total dispersion of -1742 ps/nm at 1550 nm, loss of
10.7 dB, and dispersion slope of -0.22 ps/nm2/km. The chirped FBG has a positive
dispersion of +2020 ps/nm, a 0.456-nm bandwidth at 1546.4 nm, and peak
reflectivity of 89%. PPLN-1 and PPLN-2 have QPM wavelengths of 1550.1 and
1554.7 nm, respectively, at 91.5 °C. Laser diode (LD) pump-1 and LD pump-2
82

are fixed at 1553.8 and 1562.8 nm, respectively. LD dummy-1 and LD dummy-
2 are tuned according to the desired converted wavelength, c. Polarization
controllers (PCs) are inserted in the input path of each PPLN since the converters are
polarization dependent. The input powers are: (i) for PPLN-1, 1 and pump-1 are
each 14 dBm and dummy-1 is 10 dBm, and (ii) for PPLN-2, pump-2 and
dummy-2 are each 12 dBm and c is 5 dBm. The converted wavelength is -13
dBm for PPLN-1 and -19 dBm for PPLN-2. The filters at the output of PPLN-1 and
PPLN-2 have 3-dB bandwidths of 0.8 and 1.2 nm, respectively. The receiver is a 10-
Gbit/s p-i-n device.

Fig. 8-6: Packet delay shown at 10 and 26.4 ns. Final output signal is 1546.4 nm.

The packets on 1 (1546.4 nm) are routed through the delay module. Figure 8-5
shows three example delay scenarios of 0, 10 and 26.4 ns, corresponding,
83

respectively, to c at 1556.7, 1551.92 and 1542.5 nm. The inset eye diagram is for
the packet stream that has been delayed by 26.4 ns.

Fig. 8-7: Delay as a function of converted wavelength.

The two packet streams at 1 and 2 are initially offset by more than a packet
length. By changing c from 1556.76 to 1542.5 nm, we introduce a delay of 26.4 ns
that results in aligning the 1 packet into the vacant slot between packets 1 and 2 on
2 (1552 nm). Figure 8-7 shows the synchronized packets that are multiplexed
together using a 3-dB optical coupler. The inset eye diagram of Fig. 8-7 shows that
the delayed data packet is slightly more noisy than the non-delayed packet. We
believe that this is due to the added ASE arising from the non optimally- filtered
high-power EDFAs that were needed to overcome the PPLNs conversion efficiency
of -25 dB. Note that any distortion can be minimized by employing PPLNs with
84

higher conversion efficiency, low noise EDFAs and matched dispersion
compensation. Note that the maximum residual dispersion between the DCF and
FBG is around 310 ps/nm when the converted wavelength is 1542.5 nm.

Fig. 8-8: Packet streams 2 (non-delayed) and 1 (delayed by 26.4 ns)
synchronized and multiplexed. MUX = multiplexer. Our multiplexer is a simple
3-dB passive coupler.

We measured the BER of a manually-programmed bit sequence for the back-to-
back (non-delayed) 1 packet stream, delayed 1 packet stream, and multiplexed
1+ 2 packet stream. Figure 8-8 shows a ~2.5-dB power penalty at a BER=10-9
for the multiplexed signal. This penalty is due to the non-idealities of the delay
module. Since higher bit rates are much more sensitive to residual chromatic
dispersion, finer control on dispersion compensation would be required.
85

Fig. 8-9: Measured bit error rate (BER) for back-to-back, delayed single packet
stream and multiplexed data stream. Power penalty of 2.5 dB is observed.

8.3 Reconfiguration time of optical delay
The delay can be tuned by either changing the input wavelength or the converted
wavelength. For our demonstration to dynamically reconfigure the optical delay, we
keep c constant and change 1 between the two values of 1548.5 and 1551.76
nm. We use this approach for this section because we want to remove from our
measurement the additional frequency dependent time-of-flight of different
wavelengths inside the DCF. The modified approach and experimental setup is
shown in Figs. 8-9 and 8-10.

86

Fig. 8-10: (a) Output spectra of the two PPLNs when the switch is in the OFF
position. (b) Output spectra when the 2x2 switch is turned ON.

Fig. 8-11: Experimental setup for measuring the reconfiguration time of the delay
scheme. The inset shows that the reconfiguration time is 276 ps.

87

Two input lasers (1548.5 and 1551.76 nm) that are equidistant from the QPM of
PPLN-1 are input to a multi-GHz 2x2 prism-based electro-optic switch. The output
port-1 of the switch is connected to a 3-dB coupler. The continuous-wave signal on
one output of the coupler is fed to PPLN-1 as pump-1, while the other coupler
output becomes dummy-2 of PPLN-2. Output from port-2 of the switch is routed
through the modulator and input to PPLN-1 as 1. We keep dummy-1 and
pump-2 fixed at 1553.8 and 1562.8 nm, respectively. The final output from PPLN-
2 is at two different wavelengths, 1560.2 and 1557.5 nm, depending on whether the
case is the “switch-off” or “switch-on” position (see Fig. 8-9). We split the output of
PPLN-2, put optical filters with center wavelengths located at the two possible final
output values, and observed both of the outputs on a dual-channel optical
oscilloscope. The path lengths for both of these outputs are kept equal. As shown in
the inset of Fig. 8-10, the observed elapsed reconfiguration time is ~276 ps.
We note that the reconfiguration time in our experiment includes the inherent
delay of the switch. Furthermore, we chose the values of the switched lasers at
1548.5 nm and 1551.76 nm due to the particular QPM settings of the available
PPLNs.

88

Chapter 9: All-optical time domain 160 Gb/s ADD/DROP
based on pump depletion and nonlinearities in a single
PPLN waveguide

We demonstrate a technique for ultra-high speed all-optical Add/Drop
functionality base on pump depletion and nonlinearities in a single PPLN device.
The SFG+DFG process accomplishes the Drop functionality, while pump depletion
is used to do the Add function. Utilizing the very high speed of response of the
PPLN waveguide the add/drop function is demonstrated at data-rate of 160Gb/s[36].
9.1 Introduction
Future generation optical networks require ultra-fast operations able to reduce the
latency time at the node as well as to increase the available bandwidth, without
affecting the traffic performance. Optical signal processing overcomes the electronic
bandwidth limitations with advantages in terms of transparency and scalability.
Single channel extraction and clearing from time-interleaved optical signals and new
single channel insertion in the time domain, are key feature for networking operation
in wavelength-division-multiplexed (WDM)/optical-time-division-multiplexed
(OTDM) hybrid transmission systems. Up to now optical fiber, semiconductor
optical amplifiers and electro absorption modulators have been used to perform these
operations up to 160Gb\s.
89

Recently, periodically-poled lithium-niobate (PPLN) waveguides have been
considered for all-optical signal processing, due to their ultra-fast dynamics, high
efficiency and compactness. PPLN waveguides have been used to demonstrate
several nonlinear functions like a demultiplexing (essentially an AND gate) at
160Gb/s, and an optical sampling at 1THz. All reported implementations are based
on combinations of sum frequency generation (SFG) and difference frequency
generation (DFG) nonlinear processes. More recently, there was a simulation result
subsequently experimentally validated at 160Gb/s [37] that additionally exploited the
pump-depletion phenomenon in a single PPLN waveguide to simultaneously obtain
the complex logic functions of half-adder, half-subtractor, and OR[38].
In this chapter we exploit such combination of SFG+DFG and pump depletion in
order to make the PPLN waveguide suitable for add/drop operations at 160Gb/s.
BER measurements show error-free performance at 160Gbit/s for all the dropped,
survived and added channels.
.As shown in Fig. 9-1, the PPLN waveguide is operating in a two pump
configuration. The two pumps correspond to the 160Gb/s OTDM signal and to a
10GHz clock synchronized with the 10Gb/s tributary channel to be dropped. Such
pumps can nonlinearly interact and SFG occurs under the quasi-phase matching
condition (QPM). The generated signal simultaneously interacts with a CW light to
produce an idle signal in C-band through DFG process. Looking at the idle signals at
the output of the PPLN waveguide, we can observe that it is present only when the
90

clock is present and the signal is at high levels. In this case the process SFG +DFG
can occur. Then the idle signal represents the demultiplexed tributary channel at
different wavelength that can be tuned opportunely choosing the CW wavelength. At
the same time looking at the 160Gb/s OTDM signal at the output of the PPLN we
can observe that it is depleted by SFG interaction in correspondence of the
demultiplexed channel, while the other channels can survive. Therefore a new
channel can be added through an optical coupler in the OTDM frame at the same
position of the dropped one.

Fig. 9-1: Add/drop working principle.

9.2 Experimental setup and results
As shown in Fig. 9-2, a 10GHz Mode Locking Laser (MLL) producing 2-ps
pulses at c= 1549.4nm is used as 10GHz clock and to generate a modulated
10Gb/s supercontinuum spectrum through propagation in a 500m-long highly
nonlinear fiber (HNLF). The 10Gb/s data signal is obtained by filtering such
supercontinuum spectrum at S = 1553.2nm. Then a 10-to-160Gb/s optical fiber-
91

based multiplexer produces the 160Gb/s OTDM signal. Since the multiplexer is not
polarization maintaining, therefore a polarization controller at the multiplexer input
and an optical polarizer at the multiplexer output are required so that the 16 tributary
channels in the 160 Gb/s data frame have aligned polarization. In addition
polarization aligned signals at the input of the PPLN waveguide are required. The
input mean power is 25dBm 16dBm and 12dBm for the CW, the clock and the
OTDM signal respectively. At the output of the PPLN an optical slitter and two
optical filters allow to separate the dropped channel to the survived ones, and an
optical coupler carries out the ADD operation. The channel to be added is obtained
splitting the original 10Gb/s data signal. Finally the 160-to-10Gb/s optical
demultiplexer used to test the performance of the new 160Gb/s OTDM signal,
exploits four wave mixing (FWM) effect in a 3km-long HNLF span. The 2.6ps
demux pump is obtained by the MLL and the FWM component generated at 1557nm
is extracted.

Fig. 9-2: Add/drop experimental set up.

The pulsewidth measured by an autocorrelator is 2.5, 3.1, 3.3, and 4.1ps for input
clock, input and output OTDM signal and idle signal respectively. Fig. 9-3 (a) shows
92

the eye diagrams of the PPLN waveguide input and output signals at the bit rate of
40Gb/s. 27-1 PRBS sequence is considered. Such sequence length is due to the MUX
limitation to maintain PRBS sequence. Open eye diagrams are obtained for all PPLN
waveguide output signals like dropped channel and survived channel, as well as for
the new aggregated 160Gb/s frame.
Fig. 9-3 (b) shows the optical spectrum after the PPLN waveguide (top), and the
max-old trace of such spectrum changing the CW wavelength (center). This way we
can notice that the efficiency of the SFG+DFG process producing the idle signal is
kept constant for a large range of detuning between CW and pumps. Such detuning is
limited to 15 nm by the amplifier bandwidth, but it allows to tune the dropped
channel wavelength on the whole C-band. In addition the position of the pumps can
be tuned opportunely setting the QPM condition through the thermal control of the
PPLN waveguide. Fig 9-3 (b) also shows the optical spectrum at the output of the
optical demultiplexer (bottom).
Fig. 9-3 (c) shows BER measurements in the case of 160Gb/s input signals. BER
measurements are made on the original 10Gb/s signals, on the 10Gb/s demultiplexed
tributary channels of the input 160Gb/s signal, on the 10Gb/s dropped channel and
on 10Gb/s demultiplexed survived and added channels. Note that, for the sake of
clarity, we report here the BER curve concerning just one of the 15 survived
channels. We verified that penalty variations for all 15 survived channels are within
1.2dB, partially due to the non-perfect channel equalization during the optical
93

multiplexing operation. From Fig. 9-3 (c) we note that the demultiplexing operation
introduces less than 2dB penalty on the input tributary channel with respect to the
back-to-back case, while the add/drop process present a penalty of 3.5dB, for the
dropped channel with respect to the back-to-back case, and a penalty of 0.3dB and
2.8dB for the survived and added channel respectively with respect to the input
demultiplexed channel.
The insets of Fig. 9-3 (c) report in clockwise order the eye diagram of the
original 10Gb/s signal, of the demultiplexed 10Gb/s input channel, and of the
demultiplexed 10Gb/s survived, added, and dropped channel. Finally in the center
the survived frame at the bit rate of 160Gb/s is reported.

Fig. 9-3: Eye-diagrams of the input, survived and new 40Gb/s frame, and dropped
channel. Clear eye opening can be observed on all channels.

94

Fig. 9-4: Optical spectrum at the PPLN output (top), max hold for different pump
positions to tune the drop channel wavelength (center), optical spectrum at the
HNLF out in the optical demultiplexer.

95

Fig. 9-5: BER measurements. The results about a sample channel are reported.
The penalty difference is lower than 1.5dB.

We successfully demonstrated 160Gb/s add/drop operations exploiting a
combination of SFG+DFG and pump depletion processes in a single PPLN
waveguide. This represents an improvement in using such kind of nonlinear element
making possible new key applications for all-optical high bandwidth networks.

96

References

[1] Agrawal, G. P. Fiber-Optic Communication Systems. 3rd. New York:
Wiley, 2002.
[2] Agrawal, G. P. Nonlinear Fiber Optics. Springer Berlin / Heidelberg,
2000.
[3] Akulova, Y. A. et al. "Widely-tunable electroabsorption-modulated
sampled grating DBR laser transmitter." J. Sel. Top. Quantum
Electron (IEEE) 8 (2002): 1349–1357.
[4] Alexeyev, C. N. et al. "Optical vortices and the flow of their angular
momentum in a multimode fiber." Semiconductor Physics, Quantum
Electronics & Optoelectronics 1, no. 1 (1998): 82-89.
[5] Allen, L. et al. Optical Angular Momentum. Bristol: Institute of
Physics Publishing, Bristol and Philadelphia, 2003.
[6] Allen, L. et al. "Orbital angular momentum of light and the
transformation of Laguerre–Gaussian laser modes." Phys. Rev. A 45
(1992): 8185–8189.
[7] Anguita, J. A. et al. "Turbulence-induced channel crosstalk in an
orbital angular momentum-multiplexed free-space optical link." Appl.
Opt 47, no. 13 (2008): 2414–2429.
[8] Berntson, A. et al. "Self-phase modulation in dispersion compensated
optical fibre transmission systems." Optics Communications 130, no.
1 (1996): 153-162.
[9] Bogoni, A. et al. "160 Gb/s time-domain channel extraction/insertion
and all-optical logic operations exploiting a single PPLN waveguide."
Lightwave Technology, Journal of (IEEE) 27, no. 19 (2009): 4221-
4227.
[10] Capmany, J. et al. "1 Tb/s x km multimode fiber link combining
WDM transmission and low-linewidth lasers." 16, no. 11 (2008):
8033–8038.
97

[11] Chang-Hasnain, et al. "Slow-Light Optical Buffers: Capabilities and
Fundamental Limitations." J. Lightwave Technol. (IEEE) 23 (2005):
4046 – 4066.
[12] Chen, Z. et al. "Experimental demonstration of slow light via four
wave mixing in semiconductor optical amplifiers." proceedings of
OFC. Anaheim: OSA, 2006.
[13] Chou, H. F. et al. "Compact 160-Gb/s Add–Drop Multiplexer With a
40-Gb/s Base Rate Using Electroabsorption Modulators." Photon.
Technol. Lett. (IEEE) 16, no. 6 (2004).
[14] Claassen, M. et al. "Two-section electroabsorption modulator with
negative chirp at low insertion loss." Electronic Letters 32 (1996):
2121–2122.
[15] Cvijetic, M. et al. Digital Optical Communications. Naučna Knjiga,
1984.
[16] Cvijetic, M. "Dual-mode optical fibers with zero intermodal
dispersion ." Opt. Quantum Electron 16, no. 4 (1984): 307–317 .
[17] Dagens, B. "SOA-based devices for all-optical signal processing."
Tech. Dig. Cong. Optical Fiber Communication. 2003.
[18] Dagens, B. et al. "Design optimization of all-active Mach-Zehnder
wavelength converters." Photon. Technol. Lett 11 (1999): 424-426.
[19] Djordjevic, I. B. et al. "Four-dimensional optical multiband-OFDM
for beyond 1.4 Tb/s serial optical transmission." Optics Express 19,
no. 2 (2011): 876–882.
[20] Djordjevic, I. B. et al. "Next generation FEC for high-capacity
communication in optical transport networks." J. Lightwave Technol.
27, no. 16 (2009): 3518–3530.
[21] Djordjevic, I. B. et al. "Spatial-domain-based multidimensional
modulation for multi-Tb/s serial optical transmission." Optics Express
(OSA) 19, no. 7 (March 2011).
[22] Doverspike, R. et al. Entering the Zettabyte Era. Cisco, Cisco
Networks, 2011.
98

[23] Eisenstein, G. et al. "Gain recovery time of traveling-wave
semiconductor optical amplifiers." App. Phys. Lett 54 (1989): 454-
456.
[24] Fazal, I. et al. "Demonstration of 2-Tbit/s Data Link using Orthogonal
Orbital-Angular-Momentum Modes and WDM." Frontier in Optics.
San Jose: OSA, 2011.
[25] Fazal, I. et al. "Optical data packet synchronization and multiplexing
using a tunable optical delay based on wavelength conversion and
inter-channel chromatic dispersion." Optics Express (OSA) 15, no. 17
(2007): 10492-10497.
[26] Fazal, I. et al. "Optimizing RZ 40-Gbit/s fiber transmission system
performance when utilizing SOA-MZI λ-converters." Lasers and
Electro-Optics (CLEO) and Quantum Electronics and Laser Science
Conference (QELS), 2010 Conference on. OSA, 2010.
[27] Fazal, I. et al. "SOA-assisted data-polarization-insensitive wavelength
conversion in a PPLN waveguide." Journal of Lightwave Technology
(IEEE) 26, no. 12 (2008): 1690-1695.
[28] Fazal, I. et al. "Optical data packet synchronization and multiplexing
using a tunable optical delay based on wavelength conversion and
inter-channel chromatic dispersion." Opt. Express 15 (2007): 10492-
10497.
[29] Furukawa, H. et al. "Tunable All-Optical Wavelength Conversion of
160-Gb/s RZ Optical Signals by Cascaded SFG-DFG Generation in
PPLN Waveguide." Photon. Technol. Lett. (IEEE) 19, no. 6 (2007):
384-386.
[30] Gauthier, D. J. et al. "Slow and fast light." Progress in Optics, 2002:
497–530.
[31] Gibson, G. et al. "Free-space information transfer using light beams
carrying orbital angular momentum." Optics Express (OSA) 12, no.
22 (Nov. 2004): 5448-5456.
[32] Ginovart, F. et al. "Gain recovery dynamics in semiconductor optical
amplifier." Opt. Commun 199 (2001): 111-115.
[33] Gnauck, A. H. et al. "Optical Phase-Shift-Keyed Transmission."
Journal of Lightwave Technology 23, no. 1 (2005).
99

[34] Henry, K. Y. et al. "All-Optical Packet Switching by Pulsed-Pump
Wavelength Exchange in a Highly Nonlinear Dispersion-Shifted
Fiber." proceedings of OFC. OSA, 2007.
[35] Hess, R. et al. "All-optical demultiplexing of 80 to 10 Gb/s signals
with monolithic integrated high-performance Mach–Zehnder
interferometer." Photon. Technol. Lett 10 (1998): 165-157.
[36] Hunter, D. K. et al. "SLOB: A Switch with Large Optical Buffers for
Packet Switching." J. Lightwave Technol. 16 (1998): 1725 – 1736.
[37] Ip, E. et al. "Carrier synchronization for 3- and 4-bit-per-symbol
optical transmission." Journal of Lightwave Technologies 23 (2005):
4110–4124.
[38] Ito, T. et al. "Extremely low power consumption semiconductor
optical amplifier gate for WDM applications." Electronic Letters 33
(1997): 1791-1792.
[39] Jungerman, R. L. et al. "1-THz bandwidth C- and L-band optical
sampling with a bit rate agile timebase." Photon. Technol. Lett.
(IEEE) 14, no. 8 (2002): 1148-1150.
[40] Kahn, J. M. et al. "Spectral efficiency limits and modulation/detection
techniques for DWDM Systems." Journal of Selected Topics in
Quantum Electronics 10 (2004): 259–271.
[41] Kang, I. et al. "Characterization of the dynamic processes in all-
optical signal processing using semiconductor optical amplifiers."
Journal of Sel. Top. in Quant. Elect. (IEEE) 14, no. 3 (2008): 758-
769.
[42] Khaleghi, S. et al. "System design guidelines when utilizing chirp-
inducing wavelength converters in a fiber transmission system."
Lasers and Electro-Optics (CLEO) and Quantum Electronics and
Laser Science Conference (QELS), 2010 Conference on. 2010.
[43] Kikuchi, K. "Coherent detection of phase-shift keying signals using
digital carrier-phase estimation." Conference on Optical Fiber
Communications. Anaheim, 2006. Paper OTuI4.

100

[44] Koh, P.C. et al. "Correlation between dispersion penalty and
timeresolved chirp for widely tunable electroabsorption-modulated
SGDBR laser across the EDFA gain bandwidth." Photonics Technol.
Lett (IEEE) 15 (2003): 1011–1013.
[45] Langrock, C. et al. "All-Optical Signal Processing using χ (2)
Nonlinearities in Guided-Wave Devices." J. Lightwave Technol.
(IEEE) 24 (2006): 2579 – 2592.
[46] Langrock, C. et al. "All-Optical Signal Processing Using χ(2)
Nonlinearities in Guided-Wave Devices." Journal of Lightwave
Technology (IEEE) 24, no. 7 (July 2006).
[47] Leuthold, J. et al. "100 Gbit/s all-optical wavelength conversion with
integrated SOA delayed-interference configuration." Electronic
Letters 36: 1129-1130.
[48] Li, A. et al. "Reception of mode and polarization multiplexed 107-
Gb/s CO-OFDM signal over a two-mode fiber." Optical Fiber
Communication Conference/National Fiber Optic Engineers
Conference (OFC/NFOEC). Optical Society of America, 2011.
[49] Liu, Y. et al. "Error-free 320Gb/s SOA-based wavelength conversion
using optical filtering." Proceedings of OFC. OSA, 2006.
[50] Liu, Y. et al. "Error-free all-optical wavelength conversion at
160Gbit/s using a semiconductor optical amplifier and an optical
bandpass filter." J. Lightwave. Technol. (IEEE) 24 (2006): 230-236.
[51] Loh, W. H. et al. "10 cm Chirped Fibre Bragg Grating for Dispersion
Compensation at 10 Gbit/s over 400 km of Non-Dispersion Shifted
Fibre." Electron. Lett. 31 (1995): 2203 – 2204.
[52] Makovejs, S. et al. "Experimental investigation of PDM-QAM16
transmission at 112 Gbit/s over 2400 km." Optical Fiber Conference.
San Diego, 2010.
[53] Marcuse, D. "Pulse propagation in two-mode waveguide." Bell Syst.
Tech. J., 1972: 1785–1791.
[54] Mork, J. et al. "Theory of the ultrafast optical response of active
semiconductor waveguides." J. Opt. Soc. Am. B 13 (1996): 1803-
1816.
101

[55] Murshid, S. et al. "Simultaneous transmission of two channels
operating at the same wavelength in standard multimode fibers."
Conference on Lasers and Electro-Optics/Quantum Electronics and
Laser Science Conference and Photonic Applications Systems
Technologies. Optical Society of America, 2008.
[56] Murshid, S. et al. "Spatial combination of optical channels in a
multimode waveguide." Frontiers in Optics. OSA, 2010.
[57] Murshid, S. et al. "Spatial domain multiplexing: a new dimension in
fiber optic multiplexing." Opt. Laser Technol. 40, no. 8 (2008): 1030–
1036.
[58] Okawachi, Y. et al. "All-optical slow-light on a photonic chip." Optics
Express 14 (2006): 2317-2322.
[59] Okawachi, Y. et al. "Large Tunable Optical Delays via Self-Phase
Modulation and Dispersion." Opt. Express (OSA) 14 (2006): 12022-
12027.
[60] Reid, D. et al. "Characterization of a Turbo-switch SOA wavelength
converter using spectrographic pulse measurement." Journal of Sel.
Top. in Quant. Elect. (IEEE) 14, no. 3 (2008): 841-848.
[61] Ren, J. et al. "12.47 ns Continuously-Tunable Two-Pump Parametric
Delay." European Conference on Optical Communication. 2006.
[62] Ryf, R. et al. "Space-division multiplexing over 10km of three-mode
fiber using coherent 6x6 MIMO processing." Optical Fiber
Conference (OFC), Post deadline paper. Los Angeles: OSA, 2011.
[63] Sakaguchi, J. et al. "109-Tb/s (7x97x172-Gb/s SDM/WDM/PDM)
QPSK transmission though 16.8-km homogeneous multi-core fiber."
Optical Fiber Communication Conference/National Fiber Optic
Engineers Conference (OFC/NFOEC), postdeadline papers. OSA,
2011.
[64] Sakai, J. et al. "Design considerations of broadband dual-mode optical
fibers." IEEE Trans. Microw. Theory Tech. (IEEE) 26, no. 9 (1978).
102

[65] Salsi, M. et al. "Tranmission at 2x100Gb/s, over two modes of 40km-
long prototype few-mode fiber, using LCOS-based mode multiplexer
and demultiplexer." Optical Fiber Conference (Post deadline paper).
Los Angeles: OSA, 2011.
[66] Savchenkov, A. A. et al. "Optical vortices with large orbital
momentum: generation and interference." Optics Express 14, no. 7
(2006): 2888–2897.
[67] Schares, L. et al. "Phase dynamics of semiconductor optical
amplifiers at 10-40 GHz." J. Quantum. Electron. 39 (2003): 1394-
1408.
[68] Schubert, C. et al. "Error-free all-optical add-drop multiplexing at
16OGbit/s." Electron Lett. 39, no. 14 (2003).
[69] Serati, S. et al. "Spatial light modulator considerations for beam
control in optical manipulation applications." Proceedings of SPIE -
The International Society for Optical Engineering. 2006.
[70] Sharping, J. et al. "All-Optical, Wavelength and Bandwidth
Preserving, Pulse Delay based on Parametric Wavelength Conversion
and Dispersion." Opt. Express (OSA) 13 (2005): 7872-7877.
[71] Shen, S. et al. "Effects of Self-Phase Modulation on Sub-500 fs Pulse
Transmission over Dispersion Compensated Fiber Links." Journal of
Lightwave Technology 17, no. 2 (1999): 452-458.
[72] Shumakher, E. et al. "On the balance between delay, bandwidth and
signal distortion in slow light systems based on stimulated Brillouin
scattering in optical fibers." Optics Express (OSA) 14 (2006): 5877-
5884.
[73] Simon, J. C. et al. "Semiconductor optical amplifier length effects on
gain dynamics." J. Phys. D: Appl. Phys. 36 (2003): 1473-1476.
[74] Skogen, E. J. et al. "A quantum-well-intermixing process for
wavelength-agile photonic integrated circuits." J. Sel. Top. Quantum
Electron (IEEE) 8 (2002): 863–869.
103

[75] Sokoloff, J. P. et al. "A terahertz optical asymmetric demultiplexer
(TOAD)." Photon. Technol. Lett. 5 (1993): 787-790.
[76] Su, Y. et al. "System performance of a slow-light delay line for 10-
Gb/s data packets." proceedings of Slow and fast light topical
meeting, 2006.
[77] Takiguchi, K. et al. "Optical-Signal-Processing Device Based on
Waveguide-Type Variable Delay Lines and Optical Gates." J.
Lightwave Tech. (IEEE) 24 (2006): 2593 – 2601.
[78] Talli, G. et al. "Gain dynamics of semiconductor optical amplifiers
and three-wavelength devices." J. Quantum. Electron. 39 (2003):
1305-1313.
[79] Tong, Z. et al. "21.4 Gb/s coherent optical OFDM transmission over
multimode fiber." Opto-Electronics and Communications Conference
and the Australian Conference on Optical Fibre Technology.
Engineers Australia and the Australian Optical Society, 2008.
[80] Tucker, R.S. et al. "A New Wavelength-Routed Photonic Packet
Buffer combining Traveling Delay Lines with Delay-Line Loops." J.
Lightwave Technol. (IEEE) 19 (2001): 1085 – 1092.
[81] Verdurmen, E. J. M. et al. "Error-free all-optical add-drop
multiplexing using HNLF in a NOLM at 160 Gbit/s." Electron Lett.
41, no. 6 (2005).
[82] Wang, J. et al. "25.6-bit/s/Hz Spectral Efficiency using 16-QAM
Signals over Pol-Muxed Multiple Orbital-Angular-Momentum
Modes." IEEE Photonics Conference. Arlington: IEEE, 2011.
[83] Wang, J. et al. "Modes, 25.6-bit/s/Hz Spectral Efficiency using 16-
QAM Signals over Pol-Muxed Multiple Orbital-Angular-
Momentum." IEEE Photonics Conference. Arlington, 2011.
[84] Wang, J. et al. "Single-PPLN-based simultaneous half-adder, helf-
subtracter, and OR logic gate: proposal and simulation." Optics
Express (OSA) 15, no. 24 (2007).
104

[85] Wang, P. et al. "A performance optimization method for SOA-MZI
devices." Tech. Dig. Cong. Optical Fiber Communication, 2006.
[86] Wang, Y. et al. "44-ns Continuously Tunable Dispersionless Optical
Delay Element Using a PPLN Waveguide with Two-Pump
Configuration, DCF, and a Dispersion Compensator." Photon. Tech.
Lett. 19 (2007): 861–863.
[87] Warnky, C. M. et al. "Demonstration of a Quartic Cell, a Free- Space
True-Time-Delay Device Based on the White Cell." J. Lightwave
Technol. (IEEE) 24 (2006): 3849 – 3855.
[88] Willner, A. E. "Stable and Reconfigurable Optical Networks." IEEE
Lasers and Electro-Optics Society Annual Meeting 2007. Piscataway,
NJ, USA, 2007. paper WFF1.
[89] Winzer, P. J. et al. "Optical phase-shift-keyed transmission." Jounal
of Lightwave Technology 23 (2005): 115- 130.
[90] Winzer, P. J. "Beyond 100G ethernet." IEEE Commun. Mag 48, no. 7
(2010): 26–30.
[91] Winzer, P. J. et al. "107-Gbit/s optical ETDM transmitter for 100G
Ethernet transport." Proceedings of European Conference on Optical
Communication. Glasgow, 2005. 1-2.
[92] Yariv, A., and P. Yeh. Optical Waves in Crystals: Propagation and
Control of Laser Radiation. Wiley-Interscience, 2002.
[93] Yoo, S. J. B. "Wavelength conversion technologies for WDM
network applications." Journal of Lightwave Technology 14 (June
1996): 955-966.
105

[94] Zhang, L. et al. "Optimizing operating conditions to reduce data
pattern dependence induced by slow light elements." proceedings of
OFC. 2006. OFP7.
[95] Zhang, L. et al. "Reduced recovery time semiconductor optical
amplifier using p-type-doped multiple quantum wells." Photon.
Technol. Lett. 18 (2006): 2323-2325.
[96] Zhu, B. et al. "Space-, wavelength-, polarization-division multiplexed
transmission of 56-Tb/s over a 76.8-km seven-core fiber." Optical
Fiber Communication Conference/National Fiber Optic Engineers
Conference (OFC/NFOEC). Optical Society of America, 2011.

Abstract (if available)

Abstract The unabated demand for more capacity due to the ever-increasing internet traffic dictates that the boundaries of the state of the art maybe pushed to send more data through the network. Traditionally, this need has been satisfied by multiple wavelengths (wavelength division multiplexing), higher order modulation formats and coherent communication (either individually or combined together). WDM has the ability to reduce cost by using multiple channels within the same physical fiber, and with EDFA amplifiers, the need for O-E-O regenerators is eliminated. Moreover the availability of multiple colors allows for wavelength-based routing and network planning. Higher order modulation formats increases the capacity of the link by their ability to encode data in both the phase and amplitude of light, thereby increasing the bits/sec/Hz as compared to simple on-off keyed format. Coherent communications has also emerged as a primary means of transmitting and receiving optical data due to its support of formats that utilize both phase and amplitude to further increase the spectral efficiency of the optical channel, including quadrature amplitude modulation (QAM) and quadrature phase shift keying (QPSK). Polarization multiplexing of channels can double capacity by allowing two channels to share the same wavelength by propagating on orthogonal polarization axis and is easily supported in coherent systems where the polarization tracking can be performed in the digital domain. Furthermore, the forthcoming IEEE 100 Gbit/s Ethernet Standard, 802.3ba, provides greater bandwidth, higher data rates, and supports a mixture of modulation formats. In particular, Pol-MUX QPSK is increasingly becoming the industry’s format of choice as the high spectral efficiency allows for 100 Gbit/s transmission while still occupying the current 50 GHz/channel allocation of current 10 Gbit/s OOK fiber systems. In this manner, 100 Gbit/s transfer speeds using current fiber links, amplifiers, and filters may be possible. ❧ Recently, interest has increased in exploring the spatial dimension of light to increase capacity, both in fiber as well as free-space communication channels. The orbital angular momentum (OAM) of light, carried by Laguerre-Gaussian (LG) beams have the interesting property that, in theory, an infinite number of OAMs can be transmitted

Linked assets

University of Southern California Dissertations and Theses

Conceptually similar

PDF

All-optical signal processing toward reconfigurable optical networks

PDF

Reconfigurable high‐speed optical signal processing and high‐capacity optical transmitter

PDF

High-speed and reconfigurable all-optical signal processing for phase and amplitude modulated signals

PDF

Reconfigurable and flexible high-speed optical signal processing and spectrally shared optical subsystems

PDF

High capacity optical and RF/mm-wave communications utilizing orbital angular momentum multiplexing

PDF

Optical signal processing for high-speed, reconfigurable fiber optic networks

PDF

Applications of all optical signal processing for advanced optical modulation formats

PDF

Reconfigurable optical signal processing for efficient spectrum utilization in high-speed optical communication systems

PDF

Integrated silicon waveguides and specialty optical fibers for optical communications system applications

PDF

Nonlinear optical signal processing for high-speed, spectrally efficient fiber optic systems and networks

PDF

Orbital angular momentum based spatially multiplexed optical and millimeter-wave communications

PDF

Reconfigurable high speed optical signal processing for optical communication and modulation format manipulation

PDF

Optical wave mixing for tunable delays and high‐speed signal processing

PDF

Reconfigurable high-speed processing and noise mitigation of optical data

PDF

optical communication systems and space-time wave packet generation using frequency combs and spatial modes

PDF

Detection and optical signal processing using phase based optical modulation formats

PDF

Using beams carrying orbital angular momentum for communications and remote sensing

PDF

Optical communications, optical ranging, and optical signal processing using tunable and reconfigurable technologies

PDF

Multiplexing techniques and reconfigurable networking functions for optical communications using orbital angular momentum beams

PDF

Microresonator-based Kerr frequency comb for high-speed optical communications and signal processing

Asset Metadata

Creator Fazal, Muhammad Irfan (author)

Core Title Optical signal processing for enabling high-speed, highly spectrally efficient and high capacity optical systems

School Viterbi School of Engineering

Degree Doctor of Philosophy

Degree Program Electrical Engineering

Publication Date 04/04/2012

Defense Date 07/15/2011

Publisher University of Southern California (original), University of Southern California. Libraries (digital)

Tag fiber optic communication,OAI-PMH Harvest,optical signal processing,spatial division multiplexing

Language English

Contributor Electronically uploaded by the author (provenance)

Advisor Willner, Alan E. (committee chair), Armani, Andrea M. (committee member), Povinelli, Michelle L. (committee member)

Creator Email iffi_fazal@yahoo.com,irfanfazal@gmail.com

Permanent Link (DOI) https://doi.org/10.25549/usctheses-c3-272

Unique identifier UC11287873

Identifier usctheses-c3-272 (legacy record id)

Legacy Identifier etd-FazalMuham-563.pdf

Dmrecord 272

Document Type Dissertation

Rights Fazal, Muhammad Irfan

Type texts

Source University of Southern California (contributing entity), University of Southern California Dissertations and Theses (collection)

Access Conditions The author retains rights to his/her dissertation, thesis or other graduate work according to U.S. copyright law. Electronic access is being provided by the USC Libraries in agreement with the a...

Repository Name University of Southern California Digital Library

Repository Location USC Digital Library, University of Southern California, University Park Campus MC 2810, 3434 South Grand Avenue, 2nd Floor, Los Angeles, California 90089-2810, USA