Discrete nonlinear optical systems exhibit unique properties unknown from wave propagation in bulk materials. Among them are the possibilities to form highly localized discrete solitons and the ability of a wide beam to propagate without diffraction and modulational instability across the array. The interaction between a highly localized discrete soliton and a non-diffracting beam has potential applications for all optical routing and switching. We present our results on the experimental investigation of this kind of beam interactions in a one-dimensional AlGaAs array at a wavelength of 1550 nm. A discrete soliton, almost completely confined to a single waveguide, was excited and the interaction with a wide beam of the same or orthogonal polarization was studied. We confirmed that the wide beam is able to drag the soliton over multiple waveguides towards itself while the soliton is able to maintain its original, highly confined shape. The outcome of the coherent interaction depends on the power of the wide beam and the relative phase between the two beams. This phase-dependence is due to linear interference in the case of co-polarized beams and due to four-wave mixing for orthogonally polarized beams.

Keen interest has been shown in exploiting the transport properties of nonlinear photonic crystals for modulation, switching and routing applications at telecommunication frequencies. This is due to the rich anisotropy evidenced by a highly-textured dispersion surface, and its dependence on the permittivity profile of the structure. We consider a two-dimensional photonic crystal, presume an instantaneous change in profile due to an optical Kerr effect, and show how the beam's refraction angle depends on its intensity. For computational simplicity, we employ a self-consistent approach and the linear eigenvalue equation, which we solve using a finite element method. Previous studies have extracted directions of optical power flow from equifrequency contour gradients. Such gradients, more formally the group velocity, lack local definition and are only meaningful in the context of a spatial average over the unit cell. As a result, the relationship between the local character of the optical transport and the induced permittivity profile is obscured. By contrast, we explicitly consider the spatial dependence of the Poynting's vector, from which we also extract mean transport directions with much greater computational efficiency. We thereby demonstrate the interrelationship between optical transport and nonlinear response at the nanoscale. A consequent analysis of refraction in the context of the superprism effect reveals new aspects to the optical transport in such nonlinear systems.

We demonstrate the generation of high harmonics (up to the 65^th order, λ=12.24 nm) of a Ti:sapphire laser radiation after the propagation of femtosecond laser pulses through the low-excited boron plasma produced by a prepulse radiation on the surface of different targets. High-order harmonics generated from the surface plasma of most targets showed a plateau pattern. The harmonic generation in these conditions assumed to occur due to the interaction of femtosecond pulses with ions. The conversion efficiencies in the plateau region were varied between 10^-7 to 8×10^-5 depending on the target. The main contribution to the limitation of harmonic generation efficiency and cutoff energy was attributed to the free-electrons-induced self-defocusing of main pulse.

Laser scanning nonlinear optical microscopy is used to study structure and dynamics of cellular and sub-cellular structures in vivo. Under tight focusing conditions with a high numerical aperture objective, nonlinear optical signals such as third harmonic generation (THG), second harmonic generation (SHG), and multiphoton excitation fluorescence (MPF) are simultaneously produced. MPF is extensively used in biological imaging. Unfortunately, fluorescence is accompanied by heat dissipation in the sample and photobleaching effects. On the other hand, parametric processes such as SHG and THG are free of photobleaching since they involve only virtual electronic states where there is no transfer of energy into the medium. There are many naturally occurring structures that exhibit harmonic generation effects, and hence, do not require dyes that can potentially disrupt the normal functionality of the system. SHG is efficiently generated in non-centrosymmetric media, such as chiral structures and interfaces. The THG signal is generated due to a break in symmetry at interfaces and can be enhanced by the presence of multilamellar structures, as in the mitochondria or chloroplasts. Many interesting biological processes, such as signal transduction in neurons or ATP synthesis in mitochondria, involve the movement of ions across membranes. THG and SHG are sensitive to changing electric potential gradients, and hence are ideally suited for dynamical investigations of these biological processes. The present work will expose the structural factors and conditions that influence THG and SHG generation efficiencies in biological samples. Examples of visualizing chloroplasts and mitochondria will illustrate the advantages of harmonic generation microscopy for studying structural and functional properties of the in vivo systems.

CARS microscopy has emerged as a powerful tool in imaging of biological matter. In addition to a high-3D spatial resolution, the technique delivers an attractive set of properties such as chemical specificity, high sensitivity, and fast data acquisition rates thus making it very suitable for biomedical applications. However, these advantages come at a cost of complex tunable laser sources, beam guiding and delivery optics. In particular, two high intensity laser pulses, whose carrier frequencies ωp (pump) and ωs (Stokes) are separated by the corresponding Raman shift value, are required to interact with the imaged media. In this paper, we present experimental results corresponding to our first step towards an integrated CARS-microscope. We demonstrate optical fiber delivery of two color picosecond pulses before they interact in the focus of microscope objective in order to produce CARS image. Certain aspects related to the effect of the pump and Stokes pulse parameters on image quality (e.g. contrast, sensitivity) after the pulses' propagation in the fiber will be addressed. The incorporation of fiber delivery feature significantly improves the microscope performance and ease its operation. In addition, we are exploring certain approaches in further development of CARS-microscope as a biomedical tool towards fully functional endoscope for in vivo chemically sensitive imaging.

Nonlinear spectroscopic method based on degenerate four-wave mixing allows unusually sensitive measurement of isotopes and hyperfine profiles with sub-Doppler spectral resolution. Compared to other isotope capable methods including mass spectrometry, wave mixing offers a simple, more cost effective approach with minimal sample preparation steps. Since no two hyperfine profiles are alike, hyperfine-based isotope analysis offers unambiguous isotope information. Unlike mass-based methods such as mass spectrometry, frequency-based wave-mixing spectroscopy offers more information rich spectra. A non-planar 3-D wave-mixing optical setup generates a coherent signal beam in its own space and propagation direction. Hence, virtually 100% of the signal can be collected and detected conveniently. Using compact solid-state lasers and popular atomizers, a wave-mixing setup is relatively compact. Sub-Doppler spectral resolution can be further enhanced by using simple low-pressure atomizers. Ultrasensitive fingerprinting of important isotopes promises many potential applications.

We demonstrate the first observation of significant resonance enhancement of a single high-order harmonic in the extreme ultraviolet region. Such intense harmonics are generated during the interaction of a femtosecond Ti:sapphire laser pulse with low-ionized indium ablation. A strong 13^th harmonic (61.2 nm) with conversion efficiency of 8×10^-5 and output intensity almost two orders of magnitude higher than neighbouring harmonics is demonstrated. Such an approach paves the way for efficient single-harmonic enhancement in the extreme ultraviolet range using different ablation sources.

Second harmonic generation (SHG) in a periodically poled MgO-doped lithium niobate (PPMGLN) waveguide is studied using a tunable pulsed pump source composed of a mode-locked fiber ring laser and two tunable filters. In the experiment, the lasing wavelength can be tuned from 1530 to 1579 nm, and the pulse width can be tuned from 2 to 7 picoseconds at 40 GHz. Second-harmonic pulses are generated when the picosecond pump pulses pass through the PPMGLN waveguide. SHG conversion efficiency versus pump pulse width, pump power, and pump wavelength is investigated experimentally. Propagation behaviors of both pump and SHG pulses are then numerically simulated. Based on the temporal and spectral characteristics of conversion, a quantitative analysis on SHG efficiency is presented. The simulation results are in good agreement with the experimental data.

Quasiphase matched (QPM) wavelength converters have attracted much attention due to their versatile applications, such as unltrashort-pulse generation at short wavelength. The behavior of pulse propagation in QPM waveguides has been an interesting research topic around the world. In this report, numerical study of second harmonic generation (SHG) in QPM waveguides is, for the first time, performed systematically for the fundamental pump pulses from nanosecond to femtosecond. In the proposed theoretical approach, all the dispersion effects are considered. Furthermore, our simulations take into account not only the SHG effect but also the sum frequency generation (SFG) effect on the nonlinear interaction when a pulsed pump light is used. Group-velocity mismatch (GVM) and phase-velocity mismatch are also considered in the study. Therefore, the proposed simulation model is suitable for analyzing the SHG of ultrashort-pulse since the spectral full width at half maximum (FWHM) increases with the decrease of pump pulse width. It is shown that conversion efficiencies are strongly dependent on the fundamental pulse width since the walk-off effect gradually dominates with the decrease of the fundamental pulse width. Furthermore, the temporal FWHM of converted pulses is determined by the fundamental pulse width as well as the effective interaction length related to the GVM. Under the condition of large GVM, large distortion exhibits in converted pulses. The dependence of the conversion efficiency on pump energy is also studied. The results show that the conversion efficiency saturates when the pump energy increases. The simulation results provide a guideline of device design and applications of the QPM wavelength converters in ultrashort-pulse region.

We studied the wavelength-, time-, and intensity-dependence of the 3^rd-order nonlinear optical response of As₂Se₃ chalcogenide glass. Bulk samples were characterized using a wavelength-tunable z-scan system, over the range 1200-1600 nm. Thin film samples were characterized using an ultrafast time-resolved differential optical Kerr effect (DOKE) experiment, fed by 125 fs pulses centered at 1425 nm. The z-scans revealed only slight variation in the optical Kerr coefficient n₂ over the wavelength range studied. The DOKE experiment confirmed that the nonlinear response is predominately electronic, with response time limited by the experimental setup. For the same beam intensity, DOKE and z-scan measurements were in good agreement. The optical Kerr coefficient extracted from DOKE measurements at varying pump beam intensity showed intensity-dependent behavior, which can be attributed to fifth and higher order nonlinearities.

In future all-optical networks pure optical signal processing, such as switching, routing and signal regeneration is going to be essential. Each of these tasks puts different constraints to the chosen solution regarding speed, wavelength range of operation, noise and polarization properties etc. However, a large fraction of these functionalities may be obtained by utilizing optical components with a strong nonlinear refractive index [1]. Silica has a very low nonlinear refractive index. Fortunately, silica also has a very low loss. As a consequence of the latter a significant nonlinear phase shift may be accumulated over a large distance i.e. over tens of kilometers of optical fiber. Because of this long length silica is not a viable material when designing compact nonlinear planare lightwave circuits. Recently, nanostructured materials have been proposed as promising candidates for nonlinear waveguides. More specifically glass based materials doped with nanometer sized clusters of for example metals or semiconductors. In this work we demonstrate processing of waveguides with strong confinement of the electrical field achieved using air trenches and processing of glass doped with germanium nanoclusters. We illustrate how the cluster size may be controlled and we show that realization of nonlinear waveguides may be within reach.

Self-assembled quantum dot (QD) Semiconductor Optical Amplifiers (SOAs) are believed to have faster carrier recovery times than conventional multiple quantum well, or bulk SOAs. It is therefore of interest to study the carrier dynamics of QD SOAs to assess their potential as ultrafast nonlinear devices for switching and signal processing. In this work we report experimental characterization of the ultrafast carrier dynamics of a novel InAs/InGaAsP self-assembled QD SOA with its peak gain in the important 1.55 μm telecommunications wavelength range. The temporal dynamics are measured with a heterodyne pump-probe technique with 150 fs resolution. The measurements show carrier heating dynamics with lifetimes of 0.5-2.5 ps, and a 13.2 ps gain recovery, making the device a promising candidate for ultrafast switching applications. The results are compared to previous reports on QD amplifiers operating in the 1.3 μm and 1.1 μm spectral regions. This report represents the first study of the temporal dynamics of a QD SOA operating at 1.55 μm.

The nonlinear refractive index, n₂, of single crystal ZnSe was characterized in the 1200-1950 nm wavelength region using the z-scan technique with picosecond pulses provided by a widely tunable traveling-wave optical parametric amplifier. We have found that the n₂ values range from ~15.8×10^-6 to ~9.3×10^-6 cm²/GW. The measured spectrum and scaling of the nonlinear refractive index complements previously reported values at shorter wavelengths, and is in good agreement with a theoretical model based on the nonlinear Kramers-Kroenig transformation.

A series of right-hand-side hydroxy functionalized merocyanines containing a powerful cyanodicyanomethylidenedihydrofuran electron acceptor has been designed and synthesized. Using the "build up" approach to synthesis, variations in both the donor moiety and conjugation length of these zwitterionic systems are possible, thereby giving rise to a suite of chromophores. Hyper-Raleigh scattering has confirmed that the highly conjugated chromophores 8c and 10c have large first hyperpolarizabilities (β_ο) - values that are of a similar magnitude to many of those reported for "bench mark" left-hand-side systems. The hydroxy functionalized chromophores were successfully grafted at various loadings onto a series of recently developed carboxylic acid containing polyetherimides.

We demonstrate how ultrashort optical pulses can be used to tune the optical properties of a photonic crystal using the real (Kerr) and imaginary (two photon absorption) parts of a third order optical nonlinearity. We demonstrate this effect by tuning the long (1.6 μm) and short wavelength (1.3 μm) band-edges of a stop-gap in a 2-D silicon photonic crystal. From pump-probe reflectivity experiments using 150-200 fs pulses, we observe that a 2 μm pulse induces optical tuning of the 1.3 μm edge via the Kerr effect whereas a 1.76 μm pulse induces tuning of the 1.6 μm band edge via both Kerr and Drude effects with the latter related to 2-photon induced generation of free carriers with a lifetime of ~ 700ps. In separate experiments we show how the properties of the pump eigenmode can influence the magnitude and temporal dynamics of the tuning behavior. When carriers are injected via a pump eigenmode for which the initial carrier distribution is inhomogeneous, diffusion is responsible for an initially fast (10 ps time scale) component of the recovery of the probe reflectivity with surface recombination accounting for a slower response (700 ps time scale) after the carriers are nearly uniformly distributed within the silicon backbone. When carriers are initially generated homogeneously, surface recombination alone controls the time evolution of the probed mode.

The all-trans to 13-cis isomerization of the retinal chromophore in bacteriorhodopsin (bR) plays an essential role in Nature (e.g., in photosynthesis of halobacteria). bR is a candidate for optical nanodevices driven by laser pulses, and a prospective material for optical memory storage devices and photoswitches. From the viewpoint of possible applications of bR in nanodevices we performed an experimental study of the isomerization yield by excitation with tailored laser pulses, using a coherent control approach. With specially shaped excitation pulses (found in optimization experiments) we are able to manipulate the 13-cis yield in bR over an absolute range of 60% (30% enhancement as well as 30% suppression in comparison to excitation with a transform-limited pulse) while keeping the absorbed excitation energy at a constant level.

This work proposes a complete photosensitive model based on the color center approach for Ge-doped silica fibers. It permits accurately to predict the refractive index growing dynamic in the Fiber Bragg Gratings (FBG) writing process. The model is based on the study and mathematical modeling of the physical processes that takes place when the fiber is irradiated with a high power UV light source. A bibliographic review and discussion is presented, and an initial model is established. As this initial model did not reproduce the experimental results reported in the bibliography, we considered that there were some other processes still not considered. We have introduced important modifications, which are based on interpretation of physically viable and probable photochemical reactions that take place in the FBG writing process. The mathematical equations for the model are introduced and simulation results matching experimental data are presented in order to validate our model.

Einstein explained the experimental characteristics of the photo-electric effect by the concept that the photon is a localized packet of electromagnetic radiation with the mechanism: photon + atom → emitted electron. A classical plane wave theory of light concluding that the photoelectric effect can be explained without localized photons is reexamined and found to have been flawed in predicting no time delay for photoelectron emission; this accords with Millikan's conclusion (based upon energy conservation) that a classical plane wave would entail a delay prior to emission of the order of hours. The quantification of Einstein's localized packet concept as a solitary wave one wavelength in length (one period long in time) predicts a delay of one period of the photon's oscillating field, which accords with the experimentally measured response time being a few femtoseconds.

Fiber lasers have recently received a lot of attention after the dramatic increase in output power achieved from single fibers. In particular, Ytterbium doped fibers offer a very low quantum defect and a very broad emission between 1 and 1.1 μm. Triggered by the progress in high-brightness pump diodes and the availability of large-mode-area (LMA) gain fibers, several fiber lasers with output powers in the 1kW range from a single fiber have been demonstrated [1-4]. While these demonstrations typically employ a length of gain fiber pumped via free-space coupling and free space optics as the high reflector, there are fewer reports of integrated all-fiber laser cavities, e.g. [4]. The availability of high-power fiber-optic components and the assembly thereof is therefore crucial for making this technology accessible for a variety of applications. Fiber lasers and amplifiers are very attractive light sources for applications requiring high power as well as excellent beam quality, because they are much less susceptible to thermo-optic distortions than conventional solid-state lasers. A transform-limited beam quality (M²=1) is possible even at kW level output power. Another advantage is the excellent overlap between the signal light and the pump absorption achievable in properly designed fibers. This allows a very efficient operation and up to 80% of optical conversion efficiency have been demonstrated based on the launched pump power [2]. Once assembled, fiber-optic modules do not require alignment and are therefore inherently robust. The tight confinement of the laser light combined with the long interaction length in fibers also makes them prime candidates for high gain systems.

Temporal Talbot effect is the time domain counterpart of spatial self imaging phenomenon. When a periodic time signal is propagated through a first order dispersive medium, exact replicas of the signal are reproduced at specific distance along the direction of propagation. At other distances, the signal is self imaged with a higher repetition-rate than the original periodic sequence (Fractional Talbot effect). In this paper, the problem of propagation of an ideal periodic optical pulse sequence through a linear dispersive fiber is investigated in the joint time-frequency domain using an optimized representation [i.e. Wigner Ville-Multiresolution spectrograms providing an optimal resolution in both time and frequency domains with reduced cross-term interferences]. Based on these optimized representations, complete numerical simulations were carried out to analyze the evolution of the time-frequency distribution of a periodic signal propagating through a linear dispersive medium, thus providing a deeper insight into the physics of the temporal Talbot problem. Moreover, we have used an elegant ray-matrix approach to describe the signal propagation in phase space and we have showed that for the fractional temporal Talbot effect (repetition rate factor M), each newly generated individual temporal pulse has contributions only from every M^th spectral component of the train's discrete spectrum. This interpretation is in fact in very good agreement with the notion that the fractional temporal Talbot effect can be explained as a result of interference between consecutive, chirped TF patterns. Our numerical simulations have confirmed our heuristic descriptions of the Talbot phenomena.

A variational approach with an arbitrary ansatz is used to derive the governing equations for the characteristic parameters of dispersion-managed solitons. Both Gaussian and super-Gaussian pulse profiles are considered as particular cases. The fundamental dynamics of DM pulses are characterized by their pulse width and frequency chirp. The possibility of soliton propagation when the average dispersion is zero or normal is examined.

Fiber lasers have small size, high conversion rates and excellent thermal properties. On the other hand they generally produce smaller output intensity than semiconductor lasers. Recent experiments reported that a small number of fiber lasers can be synchronized simply by coupling them with an optical waveguide coupler near the output end. As a result the output peak intensity increases as a square of the number of coupled lasers, and well focused beams of high intensities can be produced, while preserving all other properties of fiber lasers. Synchronous behavior arises spontaneously, at constant pumping levels and without any active control. We investigate synchronization properties using a theoretical model, based on an iterated map with a particular symmetry. Our model captures key qualitative features seen in the experiments. We illustrate our results in a simple fiber laser configuration.

An analytical solution for hard-aperturing Kerr lens mode locking of compact three-element resonators is derived using nonlinear ABCD ray matrix method. The dependence of the hard-aperturing Kerr nonlinear strength on aperture position and other cavity parameters is discussed. Selection guidelines on aperture position, and other cavity parameters for effective KLM are provided.

By using two orthogonally-polarized pump beams, an ultrabroad tunable wavelength converter is demonstrated with uniform efficiency and equalized signal-to-noise ratio (SNR) through four-wave mixing (FWM) in an 1500-nm semiconductor optical amplifier (SOA). This device allows the conversion of the input data signal to lower or higher frequencies with nearly-constant conversion efficiency and SNR over a 10.66 THz tuning range. This result is a significant improvement of both the conversion efficiency and the SNR as compared with the conventional FWM-based wavelength converters. We have also investigated the effect of parameters of both input power and wavelength of pump P₂ on conversion efficiency and SNR of the wavelength-converted signals.

We demonstrate that a time lens (quadratic phase temporal modulator) followed by a dispersive device can be used to implement real-time Fourier transformation (RTFT) of temporal optical pulses without introducing any additional temporal phase distortion. In this so-called transform-limited RTFT operation, the time and frequency domains are fully interchanged from the input to the output of the device; in other words, the temporal waveform of the pulse at the output of the device is a replica of the input energy spectrum and at the same time, the output energy spectrum is proportional to the temporal intensity shape of the input signal. As compared with the conventional methods, the proposed configuration does not require the use of an input dispersive device preceding the time lens, thus resulting in a much simpler and more practical alternative for implementing transform-limited RTFT of optical signals. Transform-limited RTFT has enormous application in optical signal processing especially for reconfigurable, ultra-fast pulse filtering in the all-optical domain. Ultrafast optical pulse filtering enables other important optical pulse processing operations, such as all-optical temporal correlations or convolutions. We propose and analyze a novel ultra-fast optical pulse filtering design based on the above-simplified configuration for transform-limited RTFT. In this proposed filtering configuration, the time lens process is implemented using a phase electro-optic modulator driven by a RF tone. Our proposal results in a much more compact and practical design than the conventional 4-f ultra-fast optical pulse filtering system. We further carried out the analytical study of the proposed filtering system and demonstrated its simplicity and feasibility.

The Q-switching elements are based on thiopyrylotricarbocyanine dyes introduced into polyurethane polymer matrix. We report on the spectroscopic and photochemical properties of the dye in the polyurethane matrice and durability resource characteristics of the Q-switching elements. The Q-switching efficiency of the new elements exceeded the efficiency of the elements based on the BDN dyes by 50%. The absorption cross-section of the thiopyrylotricarbocyanine dyes in polyurethane polymer is 6•10^-16 in the spectral range 1.064-1.079 mkm, relaxation time 35-55 picoseconds.

The goals of nanophotonics are to confine light to the nanoscale and to increase the local field intensity. Subwavelength confinement is of interest for nanolithography, photonic integrated circuits, subwavelength imaging and nano-sensors, whereas increased local field intensity is of interest for spectroscopy and nonlinear optics. Nanostructured metals allow us to achieve both of these goals by confining and focusing light below the normal diffraction limit, with the help of surface plasmons. Consequently, the understanding of how surface plasmons behave in nanoholes is important to the field of nanophotonics. In this paper we consider two different hole shapes: the rectangle and the double hole. We show that decreasing the short edge in a rectangular hole increases the wavelength of light that can propagate in the hole. We show that for a 15 nm hole in a real metal, the cut-off wavelength is more than doubled. Next we consider the double hole configuration, where the holes overlap to produce two apexes. We show that the local field in the nanometric vicinity of the apexes is 5 orders of magnitude larger than the field elsewhere. We calculate the optimum centre-to-centre spacing for the double-holes to maximize the cut-off wavelength and we calculate the dispersion of the light propagation near that wavelength.

The further miniaturization of integrated optical devices requires investigating optical elements with dimensions on the nano-scale. Methods are therefore needed for detecting and guiding light on a scale much smaller than the wavelength of the light. It is clear that to investigate light-nanostructure interaction in a spatial extension much less than the optical wavelength, one can not in general have confidence in the macroscopic electrodynamics so far popular in near-field optics and photonic band-structures. Instead a microscopic approach treating rigorously the local-field electrodynamics is highly desirable. In this work we present a microscopic theory of near-field optical effect in nanostructured systems. Our theory is based on the rigorous Lagrangian and Hamiltonian formulation of local-field electrodynamics, where the nanostructure optical response is treated quantum mechanically, while the electromagnetic field can be treated either classically or quantum electrodynamically within a unified space-time picture.

In this paper we consider the effective electric and magnetic properties of a three-dimensional collection of non-magnetic spheres. Polaritonic materials are used, so that the Mie resonances of the spheres are excited in the long-wavelength regime of the surrounding medium. We consider a simple cubic lattice based on LiTaO₃ and find that it is possible to engineer a fundamental resonant magnetic response. The effective media parameters derived by this approach are isotropic, and closely match those obtained by band structure calculations. Frequency ranges with either negative permittivity or negative permeability are found. Within these ranges a negative group velocity is observed. Coated spheres with a negative index of refraction are also presented.

Photonic crystals consisting of semiconductor nanowire arrays grown using a metal catalyzed vapor-liquid-solid (VLS) method are excellent candidates for photonic elements and devices, such as micro-cavities, due to the high dielectric constant contrast and high aspect ratio. In addition, it is easy to control the crystal structure by patterning the metal catalysis, and the versatility of composition of nanowires (including II-VI, III-V and ternary III-V) makes the integration of optical components in diversified wavelength ranges possible. Here we use a Plane-Wave-Expansion (PWE) method and Finite Difference Time Domain (FDTD) technique toinvestigate the optical properties of nanowire based photonic crystals. It is found that arrays consisting of nanowires with radius at or below the edge of the effective single-wire confining range for a stand alone Fabry-Perot cavity can still form a high-Q value cavity with single mode operation. Our results will help to extend the concept of the-state-of-art 1-D distributed bragg reflector (DBR) and distributed feedback (DFB) lasers into 2-D ones with a working range from ultraviolet to near infrared.

Arrays of free-standing ZnSe nanowires of length 8-10 μm and diameter 60-150 nm were fabricated by Au-catalyzed vapor-liquid-solid growth. Electron microscopy showed that these were high quality single crystal nanowires. Photoluminescence (PL) measurements of the as-grown nanowires were characterized by weak near band edge emission and strong defect-related emission. The effect of post-growth annealing on the PL spectra under both Zn-rich and Se-rich conditions were studied. Annealing under a Zn-rich atmosphere was found to significantly enhance the near band edge emission and suppress deep-level emission, resulting in spectra dominated by the near band edge emission. On the other hand, annealing in a Se-rich atmosphere had the reverse effect, resulting in spectra dominated by deep level emission.

Highly oriented gallium arsenide (GaAs) nanowires are grown on GaAs (100), (110) and (111) substrates by molecularbeam epitaxy using vapor-liquid-solid growth. The preferred growth direction of the nanowires is <111>. GaAs nanowire arrays are grown using a number of approaches such as nanochannel alumina template, gold colloid, and patterns fabricated using focused ion beam. Large interwire separation in the range of submicron can be obtained using the later two methods, which is required for applications in photonic devices such as photonic crystals.

We report a novel method of fabricating self-assembled carbon nanotube (CNT) on Si nanocrystals and the photocurrent from this network. Silicon-on-insulator (SOI) substrate with 10nm thin top silicon layer is annealed at elevate temperature in an ultra-high vacuum environment. The Si layer dewets and aggregates into Si nanocrystal islands with dimensions about 90 nm high, 100-150 nm wide, and 200nm apart. 1nm thin Fe film is deposited on the decomposed SOI as catalyst for CNT growth. The growth is done by chemical vapor deposition (CVD) at 900 °C with a flow of CH₄ at 400sccm and H₂ at 20sccm. The CVD grown CNTs show strong preferential growth on the top portion of the Si nanocrystals and form a suspended network connecting the nanocrystals. No photolithographic process is needed to create this self-assembled CNT network. We find that the reason that few CNT are found on the oxide surface is because of the influence of the island topography on the CH₄ gas flow pattern, with feedstock unable to reach the oxide surface when the islands are close to each other. We demonstrate that, by shining a low power 650nm wavelength commercial red laser pointer on this network, it generates photocurrent on the level of 20nA photocurrent under 1 volt bias condition. Since a 100 mW 1.175 μm wavelength IR laser does not generate any distinguishable photocurrent in our measurement setup, we believe the photocurrent generated by 650 nm red laser mainly comes from the Si nanocrystals instead of the CNTs. We demonstrate that a dense, self-assembled CNT network can be formed on the decomposed Si nanocrystals and can be used as conducting media for electric measurement.

A controllable delivery of spins in nanodevices is required for applications in spintronics technologies. A pure spin current, in which oppositely oriented spins move in opposite directions, is a phenomenon that could be used for this purpose. Various optical techniques can efficiently excite such spin currents in bulk semiconductors and nanostructures. We here propose and analyze two new optical infrared-light techniques for the injection of a pure spin current in nanostructures. The techniques are based on the intersubband light absorption (one-photon process) and stimulated Raman scattering (two-photon process). The infrared light absorption deposits approximately 100 meV for each absorption event associated with current injection. In the spin-flip Raman process which is possible due to spin-orbit (SO) coupling, the corresponding energy transfer to the system, is on the order of 1 meV. The stimulated Raman process depends on the electron momentum, and therefore, electrons with different spins can be launched in different directions. The infrared-injected pure spin currents can be engineered by changing the Rashba spin-orbit coupling using an external bias across the quantum well. The injected spin current should be detectable by pump-probe optical spectroscopy, and thus points the way toward the design of full-optical write-and-read spintronics devices.

The perfect matched layer (PML) backed by infinite potential barriers is applied to state analysis of tilted quantum well structure where electron wave can be represented by quasi-bound state. Electroabsorption with exciton effect in quantum well is calculated based on quasi bound state. In comparison with commonly used infinite potential barrier boundary, PML allows the analysis of the structures under strong applied electrical field and/or shallow quantum well which can find applications in electro-absorption based devices such as modulators and wavelength converters.

We propose a new optical data storage system with minute spheres. To avoid a problem of jitter as instability of rotating disk speed and fluctuation of recording bit formation, we have devised a new process. The process involves the use of dye-doped minute spheres arranged upon a surface-relief structure as recording bits. Alternate laminating recordable sphere layer with sensitivity and buffer layer with insensitivity structurally limits recording bits in three dimensions. We can limit a sensitive region within a sphere diameter. A reflection-type confocal optical microscope can read out both bit signals and shape signals from minute spheres at high resolution. Confocal relection of ~10% was measured before and after recording from a single minute sphere with Spiropyran as a recording dye. Also, the shape signal from each minute sphere is utilized as a clock signal in recording and readout. The clock signal can be produced by separating a high-level signal and a low-level signal on the basis of a threshold. In our minute-sphere optical storage system, a shift between positions of the recording bit and the clock signal does not occur because the clock signal is generated based on the shape signal from a minute sphere as the recording bit. This jitter-free technique proved to be extremely effective for disk recording and readout.

Self-organized nanostructures have been recently observed when femtosecond laser pulses were focused inside fused silica glass. We have shown that these nanostructures extend throughout the focal volume and their order is preserved over macroscopic distances when the focus is scanned. We discuss the present understanding of the formation of the nanostructures including a model based on transient nanoplasmonics. The model predicts the periodicity of nanoplanes to scale as λ/2 in the medium. This is experimentally verified at 800 nm and 400 nm light with which we obtain nanoplane spacing of 250 ± 20 nm and 140 ± 20 nm respectively, which scale as predicted. Another requirement of the model is that ionization occurs preferentially at regions that have previously been ionized. This allows an initially inhomogeneous plasma to develop into an ordered nanoplasma array. Using transmission measurements we show that the required "memory" exists in the case of fused silica.

Compact photonic crystal (PhC) microcavity filters in a ridge waveguide format could play a useful role for wavelength division multiplexing (WDM) and de-multiplexing functionality in dense integrated photonic circuits. The microcavity filters are embedded in ridge waveguides with high lateral refractive-index contrast because good lateral confinement and efficient coupling of light into the device can be achieved using this established waveguide technology. However, this configuration leads to significant modal mismatch at the interfaces between the PhC and waveguide sections, contributing to reflection losses and reduced transmission over much of the useful spectrum. By the same token, mode-matching features consisting of two rows of PhC holes with a different filling factor and displaced to mirror-image positions with respect to the outer two rows of the main PhC mirrors have been implemented to enhance the optical transmission by more than a factor of two. Furthermore, an increase in Q-factor (more than 100 %) is achieved by the addition of two further rows of PhC holes on either side of the microcavity. Moreover, Bragg-grating concepts have been applied in several other filter designs using the same hexagonal PhC lattice configuration, in an attempt to control the filter response. This work involves the design, fabrication (using electron-beam lithography and reactive ion etching) and characterization of such hexagonal-lattice PhC microcavity filters embedded in ridge waveguides.

Modal characteristics of the one-dimensional (1D) photonic crystal waveguides (PCWs) are investigated thoroughly. By employing the transfer matrix method, we can put the design parameters related to the general multiplayer structure into a compact analytical expression, which serves as the basis for analysis of the band-gap structure of the general 1D photonic crystals (PCs) and the modal characteristics of the general 1D PCWs. The band structure of 1D PCs and modal properties of 1D PCWs, such as the effective index, the modal field profile, the dispersion, the confinement loss, and the confinement factor, are all calculated and simulated. With the help of the band-gap map of the 1D PCs, four guiding regimes for the 1D PCWs are recognized, in accordance with the index of the guiding core. It is shown that the modal characteristics for each regime behave differently from the point of view of guiding mechanism. Furthermore, some related issues such as PCFs are discussed.

Numerical simulations are performed on the narrowing of transient linewidths of the absorption peaks of an ensemble of four-level atoms doped in a dispersive band-gap material. The atoms do not interact with each other. They are, however, interacting with a reservoir, due to which, the top two levels of the atoms decay to a pair of lower levels. A probe laser field is applied to study the absorption characteristics of the system. The density matrix method is used to calculate the photon yield for the absorption peak in order to observe linewidth narrowing. Numerical simulations are performed for SiC which is a dispersive polaritonic band-gap material. We have observed the occurrence of linewidth narrowing in SiC when the resonance energies lie within the bands. But when the resonance energy lies near the lower band edge of the polaritonic crystal, the narrowing disappears. On the other hand, when the resonance energy lies near the upper band edge, the narrowing of the linewidth remains as pronounced as before. Moreover, we have found that the photon yield is very sensitive to the locations of the resonance energies within the band. These unusual properties can be used to make new types of photonic devices such as switches, gates, etc. Similar results are also found in photonic band-gap materials.

Transmission and coupling mechanisms in photonic crystal waveguides have been extensively studied in the last few years to optimize photonic crystal designs. Previous numerical results, using 2D FDTD methods have shown that the technique of tapering the photonic crystal lattice can improve coupling efficiencies up to 80%, as compared to 30-40% efficiencies in butt-coupled waveguides. However, for the 3D structures such as photonic crystal slabs, 2D calculations do not take into consideration the losses in the vertical direction and hence 3D simulations are necessary to obtain more accurate results which can be better compared with experimental data. In this paper 2D and 3D FDTD calculation results obtained for a finite height hexagonal silicon photonic crystal slab waveguide (air as top and bottom cladding) with air holes embedded in a silicon dielectric matrix are presented. Various coupling design configurations were investigated using 3D FDTD and coupling efficiencies of 78% in the photonic crystal waveguide and 72% through the output conventional waveguide were obtained for a conventional waveguide of width 3μm coupled to a step tapered PC waveguide on the input and output ports. Furthermore some designs which show excellent efficiencies with 2D calculations are clearly shown to have significant losses in the vertical direction in 3D simulations.

In recent years there has been increasing interest in the synthesis of nanocrystalline metal oxides. Nanostructured garnets are of special interest due to their improved properties such as lower temperature sinterability, greater thermal stability, increased hardness, good optical properties etc. In our idea, the sol-gel method is a useful and attractive technique for the preparation of nanocrystalline powders because of its advantages: good stoichiometric control, the production of ultrafine particles and low temperature. In this paper, we reported the preparation and characterization of Yb³⁺ and Ho³⁺ codoped yttrium scandium gallium garnet (YSGG) nanocrystal powders by the sol-gel technique. The phase purity, mass loss, composition and microstructure features of the materials were analyzed by means of X-ray diffraction (XRD), TG analysis, infrared spectroscopy (IR) and transmission electron microscopy (TEM).

Using full wave simulations the behavior of two structures composed of split ring resonators (SRRs) and strip wires (SWs) is examined. In the region where the real parts of the permittivity and permeability are both expected to be negative both structures exhibit a transmission peak, a property which is generally assumed to imply a negative index of refraction. However, through an analysis of the dispersion characteristics and insertion phase of the two structures it is shown that only the index in the first structure, in which the SRRs and SWs are printed on opposite sides of a dielectric substrate, is negative in the passband. In the second structure, in which SRRs and SWs are printed on the same side of the substrate, the index in the passband is positive. Therefore the emergence of transmission peaks does not provide sufficient evidence of the existence of a negative index of refraction. To determine the correct sign of the index two methods are investigated. The first uses the transmission phase for propagation through various lengths of the structure and the second utilizes its dispersion diagrams. The dependence of the sign of the index on the dimensions of the unit cell size is also examined.

Wireless access systems are attractive because of their potential high data rate transmissions. For example, frequency bands in the range from 3.1 GHz up to 10.6 GHz are allocated to Ultra Wide Band IEEE 802.15.3a standards for future Wireless Personal Access Networks (WPAN). The frequency conversion of wireless signals in the optical domain is interesting since it benefits from the huge optical bandwidth for generation and distribution of up-converted sub-bands. This papers explores different techniques for photonics generation of microwave mixing with digital modulation of the microwave sub-carriers, at 1550 nm. The solutions exploit the non-linearity of different devices such as laser diodes, electro-optic modulators, dispersive fiber generating frequency-to-intensity modulation conversion. Comparisons are made regarding their potential applications to low-cost and broadband radio-over fiber systems like Ultra Wide Band over fiber. The mixing conversion gain, available bandwidth, complexity of the system, its applicability to broadband radio over fiber networks, are elements of comparison that are discussed in this paper.

In this paper an all-optical signal processor that performs both microwave mixing and bandpass filtering in a radio-over-fiber link is proposed and demonstrated. The frequency mixing is achieved by applying a local oscillator frequency and a BPSK modulated subcarrier to an electrooptic phase modulator. The mixed signals at the output of the electrooptic phase modulator are then fed to a single mode fiber link, which acts as a dispersive device for bandpass filtering and distributes the mixed signal to a remote site. The combination of the phase modulator, a multiwavelength laser source and the SMF link forms an all-optical microwave bandpass filter to suppress the levels of unwanted mixing products. A subcarrier frequency up-conversion from 3.25 GHz and 3.5 GHz to 11.7 GHz performed over a 25 km fiber link is experimentally demonstrated, in which BPSK modulation formats with data rates of 172 Mb/s and 344 Mb/s are applied. Eye diagrams are measured at the receiver end after demodulation, demonstrating a good up-conversion is achieved.

Photonic frequency down- shifting techniques for millimeter-wave band radio over fiber (RoF) systems are investigated and verified by simulation. The mechanism of frequency shifting is based on subcarrier modulation (SCM). An optical carrier with a subcarrier is injected into the frequency shifter consisting of a Mach-Zehnder modulator (MZM) or electro-absorption modulator (EAM) driven by a radio frequency sinusoidal signal. The frequency-shifted optical carrier with a frequency-shifted subcarrier is thus generated at least by SCM modulation. Optical modulation depth, the power ratio of output optical carrier to its subcarrier, can be adjusted by using the modulation voltage of the MZM or EAM; and thus RoF system performance can be easily optimized.

Subcarrier multiplexed transmission of cellular (900 MHz), personal communications systems (1.8 GHz) and wireless LAN (2.4 GHz) over the fiber has interesting applications. These multi channel radio over fiber links can connect enhanced wireless hot-spots that will support high speed wireless LAN services or low speed cellular services to different customers from the same antenna. Optical pre-filtering of SCM signals allows the use of inexpensive photodetector and increases network flexibility with fiber based optical filters. However, realizing optical demultiplexing at such low frequencies necessitates optical filters with high selectivity and low insertion loss. In this paper, we implemented a fiber wireless access system, where demultiplexing of subcarrier multiplexed cellular and WLAN signals was demonstrated in optical domain using a sub-picometer bandpass filter. Our novel fiber Bragg grating based bandpass filter has a bandwidth of 120 MHz at -3dB, 360 MHz at -10 dB and 1.5 GHz at -20 dB respectively. We experimentally verified that this filter could adequately isolate signals at as low as 900 MHz from 2.4 GHz. Experimental results show that the designed all optical demultiplexer provides about 25 dB isolation between 900 MHz and 2.4 GHz radio signals.

In this work we present a new mode-locked device that can be used for photonic millimetre-wave applications, and more specifically optoelectronic mixing. This device is based on a mode-locked MQW-DFB multisection laser that presents for certain bias conditions a dual longitudinal mode behavior (39.5 GHz separation) that can be used for mm-wave generation and transmission. In this work we focus on the possibility of achieving optoelectronic mixing using this new device through the injection of an intermediate frequency (IF) signal in one of the sections (absorber) while the gain section is used to mode-locked the two longitudinal modes by injecting a signal at 39.5 GHz. Demonstration of the optical up-conversion is carried out through the study of the modulation sidebands for different IF frequencies and a spurious free dynamic range (SFDR) of 65dB-HZ^2/3 has been measured. Transmission of an up-converted NRZ 32Mbps PRBS signal using this device for optical mixing is also demonstrated.

Wireless technology is a cost-effective means to bring broadband communications to both mobile users and home consumers; however, deploying next generation, multi-GHz wireless systems is currently too expensive. For these systems, photonic technologies can bring cost reduction as well as an increase in performance, mainly due to the ultra low-loss property of optical fibers. One approach to signal distribution is to capitalise on the vast fibre-optic distribution networks deployed within and between cities. A microwave carrier can be optically deployed from central offices to remote antenna sites using these optical links. This paper will discuss the generation of such a microwave carrier using a dual-wavelength, external-cavity laser (ECL). Two different dual-wavelength ECL's, constructed with fiber-Bragg-gratings (FBG's), have been investigated. One uses a semiconductor gain chip with a dual-FBG acting as an external reflector. The other uses two similar dual-FBG reflectors on each side of a semiconductor optical amplifier (SOA). In both cases the wavelength separation between the gratings is 0.25 nm. We will demonstrate that a dual-wavelength emission can be temporarily stabilized in the gain-chip ECL if a specific phase relation, between the external feedback from the FBG's and the residual feedback from the gain chip, is satisfied at both lasing wavelengths. The power of the RF beat signal generated by the dual-wavelength optical signal was typically 25 dB above the noise floor. The 3-dB linewidth of the RF signal was approximately 2 MHz and it can be tuned over a frequency range of 200 MHz. The physical mechanisms underlying the observed laser instability will be briefly discussed.

The spur-free dynamic range (SFDR) of a novel microwave-photonic link, using a polarization mode-converter electro-optic modulator in a balanced output configuration, is characterized in this work. Common-mode intensity noise and optical-amplifier-induced beat noise are suppressed using a polarization-selective balanced optical receiver. In addition, third-order predistortion is used to reduce 2-tone intermodulation distortion by up to 20 dB, further increasing the SFDR. Unlike the conventional approach using a dual-output Mach-Zehnder modulator, the complementary output signals are combined naturally as orthogonal polarizations into one transmission fiber.

We report on the first experiments of optical microwave generation by heterodyne beating of two doped-fiber semiconductor external-cavity lasers (DFECL). Our measurements show that minor side-mode beating in DFECL affect the optical heterodyne beat signal. Using the RF methods we study the side-mode suppression in the DFECL.

This paper presents the theoretical results from semi-analytical and numerical simulations of the THz generation in a two-dimensional geometry in a periodically poled material. The Green's function of a propagating line source is utilized to calculate the field radiated into a homogeneous material and generated by a nonlinear polarization which is confined in one transverse direction. A numerical (FDTD) method is taken for the investigation of the radiation from a material-to-air interface parallel to the direction of propagation of the optical wave. The angular dependence of the radiated transverse electric and magnetic fields is shown and explained for different ratios between the optical pulse width and the walk-off time of optical and THz waves. The numerical results are calculated for lithium niobate, a material known for its good suitability for periodical domain inversion.

This paper is focused on the demonstration of an externally modulated analogue fibre-optic link that improves microwave gain with minimal penalty on the output noise. The key device is the optical source composed of parallel wavelength-multiplexed semi-conductor lasers. The microwave gain is proportional to the optical power incident onto the photodetector, while the noise figure is also related to the optical noise parameter ("RIN") of the laser source. The dedicated multi-laser source increases the total optical power well over the maximum values of so-called "power" DFB lasers. Thus it leads to an improved link gain over bandwidths broader than 20 GHz. Simultaneously, this arrangement reduces the equivalent RIN of the multi-laser source (compared to the constituting DFB laser RINs), which keeps the noise figure to a low level. The constituting laser chips are wavelength-multiplexed into a single monomode fibre then into an external modulator. The use of a wavelength multiplexer reduces the insertion losses of each individual DFB laser into the common fibre, in order to increase the total optical power delivered into the link. The wavelength-multiplexing also shifts unwanted heterodyning beatings between optical carriers beyond the photoreceiver bandwidth. We report theoretical and experimental considerations on potential limitations of this device for microwave signal remoting. Experiments show that the modulator efficiency, the signal phase (for each optical carrier) and the microwave distortions have negligible dependency with respect to optical wavelength, while non-linear optical interactions along the fibre are not a stringent penalty. Finally, due to chromatic dispersion limits, we demonstrate the application of broad-bandwidth microwave signal remotings with high gains and low noise figures to a few hundred-metres fibre-link lengths.

In this paper, we present a novel approach for broadband RF photonic signal processing using a femtosecond laser and an Acousto-Optic Tunable Filter (AOTF). We demonstrate that by using spectral filtering in the AOTF, we are able to map each frequency component of the broadband RF signal onto a corresponding set of frequency comb lines of a femtosecond pulse train. The time domain interpretation of this RF to optical frequency mapping yields a femtosecond pulse shaping operation, which is in distinct contrast to the conventional double sideband amplitude modulation of a CW carrier used in RF photonic links. The interaction with a traveling acoustic wave in the AOTF leads to Doppler shift of each frequency comb line by its corresponding RF conterpart, which allows the recovery of the encoded RF signal by heterodyne detection with an unmodulated reference pulse train. As a proof of concept, we experimentally demonstrated mapping of 10 MHz bandwidth RF signals onto 65 nm FWHM optical bandwidth of a 2 GHz repetition rate femtosecond laser using a commercial AOTF and a 40 MHz bandwidth RF signal mapping using a supercontinuum source. Optical processing functions such as RF bandpass/notch filtering can be achieved in the optical domain using optical filters, thereby avoiding the limitations of RF analog filters. Down-conversion naturally occurs based on the laser repetition rate.

By using the phase modulated optical fiber loop mirror, repetition-rate-doubled or -tripled output is realized in the FM mode-locking fiber laser. 80 and 120GHz transform-limited pulses are experimentally demonstrated.

The concept of a traveling-wave photodector based on the integration of semiconductor substrate with high-temperature superconducting (HTS) slabs is proposed for the first time. The photoconductivity effect in semiconducting substrate and the kinetic inductive photoresponse in HTS slabs has been used to obtain a sensitive and broadband optical detection with a large responsivity over a very wide range of electrical frequency spectrum. The analysis of optical wave propagation down the photodetector waveguide is presented in detail by introducing a transcendental equation for optical propagation constant. The photomixing technique is used to excite the photodetector waveguide and the output photo-induced electrical signal is analyzed by means of transmission line theory in frequency domain. The presented analytical and numerical studies of such a photodetector waveguide reveals many interesting device characteristics including high responsivity for high beat electrical frequencies and the possibility of high-power mm-wave signal generation in such a structure.

Broadband RF imaging by spatial Fourier beam-forming suffers from beam-squint. The compensation of this frequency dependent beam-steering requires true-time-delay multiple beam-forming or frequency-channelized beam-forming, substantially increasing system complexity. Real-time imaging using a wide bandwidth antenna array with a large number of elements is inevitably corrupted by beam-squint and is well beyond the capability of current or projected digital approaches. In this paper, we introduce a novel microwave imaging technique by use of the spectral selectivity of inhomogeneously broadened absorber (IBA) materials, which have tens of GHz bandwidth and sub-MHz spectral resolution, allowing real-time, high resolution, beam-squint compensated, broadband RF imaging. Our imager uses a self-calibrated optical Fourier processor for beam-forming, which allows rapid imaging without massive parallel digitization or RF receivers, and generates a squinted broadband image. We correct for the beam squint by capturing independent images at each resolvable spectral frequency in a cryogenically-cooled IBA crystal and then using a chirped laser to sequentially read out each spectral image with a synchronously scanned zoom lens to compensate for the frequency dependent magnification of beam squint. Preliminary experimental results for a 1-D broadband microwave imager are presented.

In a microwave fiber wireless communication system, reducing optical carrier-to-sideband ratio can improve the receiving sensitivity and dynamic range of the system. However, at lower modulation frequencies corresponding to cellular radio (900 MHz), personal communication system (1.8 GHz) and IEEE 802.11g (2.4 GHz), separating the signals from the carrier and reducing carrier power in optic domain is a rather challenging task. In this paper, we have presented a new method to improve the transmission efficiency in microwave fiber-optic link by suppressing the optical carrier using a sub-picometer bandpass filter. The filter we designed and fabricated has a −3dB bandwidth of 120 MHz, capable of even filtering out a microwave signal as low as 900 MHz. We compared the transmission performance between filtered signals and unfiltered signals. The experimental results showed that the receiver sensitivity was improved significantly, 4.4dB for the 900MHz signal, and up to 8.7dB for the 1.8GHz at bit-error-rate of 10^-9, by optical carrier suppression.

A single sideband (SSB) modulation scheme using a fiber-based equivalent phase shifted Bragg grating for a remotely controlled photonic true time-delay (TTD) beamforming module is presented in this paper. Photonic TTD is considered a promising technique for wideband phased array antennas (PAA) as it allows beam steering of the antenna without the beam squint problem. For remotely controlled phased array antennas, the dispersive properties of a single mode fiber induce a power penalty at higher RF frequencies when a double sideband (DSB) modulation scheme is used. The SSB modulation scheme is an effective way to eliminate this power penalty as only one sideband is transmitted and thus there is no cancellation between the two signals generated by beating the upper sideband with the optical carrier and the lower sideband with the optical carrier at a photodetector. This paper presents for the first time experimental results of a SSB modulation scheme using an equivalent phase shifted fiber Bragg grating. The true-time delay system considered utilizes a discrete uniform fiber Bragg grating prism which allows discrete beam steering capabilities for the phased array antenna.

In this paper, we propose a novel millimeter-wave (mm-wave) band radio over fiber (RoF) system with dense wavelength division multiplexing (DWDM) star architecture. Two lasers with a small wavelength difference, phase locked and polarization-aligned, are allocated at a central station (CS) for connecting the CS and each base station (BS); one laser is used for transmitting light and the other for the remote local oscillator. For the conceptual illustration, we consider a DWDM RoF system with a channel spacing of 12.5 GHz and radio frequency (RF) of ~30-GHz mm-wave band. In the downlink system, a single-side band (SSB) subcarrier is used with low RF imposed onto an optical carrier at the CS, and an mm-wave band RF signal is obtained at each BS using direct photo-detection by the SSB subcarrier beat with the remote oscillator. In the uplink system, the received mm-wave band RF signal at each BS is imposed onto the two optical carriers simultaneously, one optical carrier with the closest SSB subcarrier is optically filtered out and fed into in the uplink transmission fiber without frequency interleaving; the electrical signal with a low intermediate frequency can be photo-detected directly at the CS. Such a RoF system has simple, cost-effective and maintenance reduced BS's, and is immune to laser phase noise in principle.

The nonlinear effects of an electro-optic intensity modulator in an optical up-conversion system for millimeter-wave (mm-wave) over fiber applications are investigated in this paper. In the analysis, general nonlinearities caused by the Mach-Zehnder modulator used for optical up-conversion are analyzed and discussed. Electrical fields of the up-converted optical signal are expressed in power series form, which are widely used in electronics in expressing memoryless nonlinearity. Harmonic distortion and inter-modulation distortion generated in an optical up-conversion, a scheme recently proposed for mm-wave over fiber applications, are analyzed in detail. One-tone and two-tone measurements are performed to validate the analyses based on our newly proposed optical up-conversion configuration.

In this paper, we propose a novel approach to implementing an all-optical microwave filter that is suitable for direct deployment in a radio-over-fiber link. The proposed filter consists of a narrow linewidth laser source, two Mach-Zehnder electro-optic intensity modulators (MZM), a local RF signal source and a chirped fiber Bragg grating (CFBG). The light emitted from the laser source is externally intensity modulated by a driving RF signal (RF1), and then injected to a second electro-optic intensity modulator. The second intensity modulator is driven by a strong local RF signal (RF2), to produce a spectrum with a carrier and multiple sidebands. Thanks to the frequency mixing effect of the second intensity modulator, the RF signal carried by the optical source is transferred to the sidebands as well. By using a CFBG as a dispersive device to induce time delays for the carrier and the sidebands, an all-optical microwave filter with only a single light source, but multiple taps is realized. The tunability of the proposed filter can be achieved by adjusting the frequency of RF2. A 2-tap lowpass filter with a free spectral range (FSR) tunable from 2.1 GHz to 4.2 GHz, and a 3-tap lowpass filter with a free spectral range tunable from 4.2 GHz to 4.8 GHz are experimentally demonstrated.

We propose performing ultrawideband RF spectrum analysis using spectral-hole-burning (SHB) crystals, which are crystal hosts lightly doped with rare earth ions such as Tm³⁺ or Er³⁺. Cooling SHB crystals to cryogenic temperatures suppresses phonon broadening, narrowing the ions' homogeneous linewidths to <100 kHz; local inhomogeneities in the crystal lattice shift the individual ionic resonances such that they're distributed over a bandwidth of 20 GHz or in some structurally disordered crystals to up to 200 GHz. Illuminating an SHB crystal with a beam modulated with multiple RF sidebands digs spectral holes in the crystal's absorption profile that persist for the excited state lifetime, about 10 ms. The spectral holes are a negative image of the modulated beam's spectrum. We can determine the location of these spectral holes by probing the crystal with a chirped laser and measuring the transmitted intensity. The transmitted intensity is the double-sideband spectrum of the original illumination blurred by a 100 kHz Lorentzian and mapped into a time-varying signal. Scaling the time series associated with the transmitted intensity by the instantaneous chirp rate yields the spectrum of the original illumination. Postprocessing algorithms undo distortion due to swept laser nonuniformities and ringing induced by fast chirp beams, eliminating the need for long dwell times to resolve narrow spectral features. Because the read and write processes occur simultaneously, SHB spectrum analyzers can operate with unity probability of intercept over a bandwidth limited only by the inhomogeneous linewidth. These capabilities make SHB spectrum analyzers attractive alternates to other approaches to wideband spectrum analysis.

We propose here to apply the technique of the optical control of microwave devices to perform phase-shifting for microstrip phased arrays. The physical principle used is the photoconductive effect which enables the creation of a variable microwave load, when a gap in a line is illuminated with an optical signal of adequate wavelength. This principle is applied on a hybrid microwave ring. We have designed a phase-shifter with continuously varying phase and quasi-constant |S₂₁|, with few return losses. The measurements have given interesting results. A low cost and small size microstrip phased array antenna has been designed. By illuminating the phase-shifter under different optical powers, the main beam can be scanned in the H-plane continuously from -25° to 25°.

In this paper, we propose a novel approach to implementing unipolar-bipolar phase-time encoding/decoding for optical code division multiple access (CDMA) networks. In the proposed approach, an electrooptic phase modulator and two fiber Bragg grating (FBG) arrays are employed to perform En/De coding. At the transmitter, a low-bit-rate data sequence modulates the optical phase and is then mapped to a high-bit-rate optical sequence via the encoder FBG array in a unipolar way. At the receiver, an identical FBG array that functions as a matched filter is used. Bipolar decoding is achieved by locating the optical carriers on either the positive or the negative slopes of the reflection responses of the decoder FBG array. The proposed encoding/decoding scheme is equivalent to a sequence inversion keyed direct sequence CDMA, which can provide an improved performance compared with the conventional incoherent scheme using optical orthogonal codes. In addition, compared with bipolar decoder applying balanced detection, this approach has a simpler architecture. A proof-of-principle experiment is demonstrated.

In this paper, optical phase modulation to intensity modulation by the use of a fiber Bragg grating (FBG) based frequency discriminator is proposed and experimentally demonstrated. In the proposed approach, the optical carrier frequency is placed at the quadrature point of the positive or negative slope of the reflection response of the FBG. The phase modulated light reflected from the two opposite slopes will have a π phase difference, which makes bipolar operation possible in an all-optical microwave signal processor or an optical code division multiple-access system. Both the frequency and phase responses of the FBG are taken into account to build a theoretical model in a frequency domain. Phase modulation to intensity modulation conversion based on a Gaussian apodized FBG is experimentally implemented. The results confirm the theoretical analysis.

The optical control study works on both the optical and the microwave behaviours of the plasma photo-induced in the semiconductor enlightened by a laser beam. The presented study is based on the necessity to be able to foresee the microwave response of CPW microwave devices versus different optical powers and different kinds of optical fibers, single-mode or multimode. The optical part has been achieved analytically by solving the diffusion equation of photo-induced carriers using the Hankel transform in 3-Dimensions. The added value of this technique is its precision and fastness. For the electromagnetic part we have chosen to use CST Microwave Studio software, which solves numerically Maxwell's equations with a Finite Integration Technique (FIT). For this aim we have had to model the photo-induced load using the locally changed conductivity directly depending of the excess carriers distribution. In the final paper, the first part will deal with the analytical computation of the photo-induced excess carrier in silicon substrate using the Hankel transform under permanent enlightening. Then the explanation of the model will be based on the need of a 3-Dimension model that may be described in an electromagnetic software. Finally simulation results of simple CPW devices as stub will be compared to measurements. In conclusion, we will show that the model is suitable for designing more complex devices and that it can be simplified in case of low precision needs.

The applications of optical amplifiers such as erbium-doped fiber amplifiers (EDFAs) and semiconductor optical amplifiers (SOAs) are inevitable in most optical transmission links. These optical amplifiers employed in a transmission link will provide amplification to the optical signals to be transmitted, at the same time the amplifiers will also add amplified spontaneous emission (ASE) noise to the amplified optical signals. In radio-over-fiber systems, optical links are used to distribute high quality radio frequency (RF) signal, microwave signal or millimeter-wave (mm-wave) signal over optical fiber for low loss long-distant transmission. In this paper, the effects of amplified spontaneous emission (ASE) noise of optical amplifiers on the quality of the optically generated electrical signals are theoretically studied.

Software Defined Radio (SDR), a radio that provides software control of a variety of modulation techniques over a broad frequency range, is an emerging technology that offers numerous advantages over conventional radio designs. With SDR, one would implement a common hardware platform and accommodate the different communications standards and technologies via software modules and firmware. This platform must be compatible with the high degree of versatility of SDR-based communication systems. SDR technology is being promoted by the US Department of Defence to replace tens of thousands of single protocol, single use radios with a common platform that could be reprogrammed to ensure interoperability. Military and public safety organisations from around the world are also considering this technology to solve their interoperability problems. Although SDR can be easily implemented below 6 GHz using conventional electronics, it is increasingly difficult to do so at the higher operating frequencies proposed by many new wireless and SATCOM standards. To take full advantage of the SDR concept, a hardware platform is required that is capable of continuous operation from frequencies where electrical sources have difficulty providing continuously tunable operation, up to 60 GHz. In addition, various signal modulation schemes will need to be supported. We present here a prototype for such a transmitter based on optical technology. It can generate a RF carrier tunable from about 18 to more than 40 GHz, which can be modulated using both intensity and phase modulation techniques. Simulations and experimental results are presented.

Using the gain profile of an erbium-doped fiber amplifier (EDFA), it is possible to create groups of wavelength within the C and/or L bands and provide necessary attenuation of the bands to equalize the entire wavelength spectrum. The equalized spectrum of wavelengths using this technique uses one variable optical attenuator (VOA) per band while dynamic channel equalizer (DCE) uses one VOA per wavelength channel. In this paper, modeling of the optimized C+L band EDFA is analyzed. Its integration with the control electronics and 8-channel DCE is proposed for use as a multi-channel dynamically gain controlled optical amplifier with flattened output gain spectra - "Smart Amplifier Solution". In addition, its feasibility for use as a broadband amplified spontaneous emission (ASE) source is discussed. The Optimized Gain Flattened C-band EDFA without gain flattening filters (GFF) and C-band Booster EDFA are presented. The optimization of amplification needs by system design approach is discussed.

In this paper we introduce and analyze a multiple-RF-beam beamformer in receive mode utilizing the principle of space-time delta-sigma modulation. This principle is based on sampling input signals in both time and space and converting the sampled signals into a digital format by delta-sigma conversion. Noise shaping is achieved in 2D frequency domain. We show that the modulator can receive signals of narrow and wide bandwidths with steering capability, can receive multiple beams, and establish tradeoffs between sampling in time and in space. The ability of the modulator to trade off between time and space provides an effective way to sample high frequency RF signals without down conversion. In addition, a space-time delta-sigma modulator has better performance than a solely temporal delta-sigma modulator (for the same filter order), as is typically used in communication systems to digitize the down-converted analog signals.

Three-dimensional vectorial diffraction analysis of phase and amplitude gratings in conical mounting is presented based on Legendre expansion of electromagnetic fields. In the so-called conical mounting, different fields components are coupled and the solution is not separable in terms of independent TE and TM cases. In contrast to conventional RCWA in which the solution is obtained using state variables representation of the coupled wave amplitudes by expanding space harmonic amplitudes of the fields in terms of the eigenfunctions and eigenvectors of the coefficient matrix defined by rigorous coupled wave equations, here the solution of first order coupled Maxwell's equations is expanded in terms of Legendre polynomials. This approach yields well-behaved algebraic equations for deriving diffraction efficiencies and electromagnetic field profiles. It can nicely handle the cases in which conventional methods face the problem of numerical instability and inevitable round off errors; also, it yields accurate results to any desired level of accuracy. The method is applied to phase and amplitude gratings in conical mountings, comparison to other methods already reported in the literature is made, and the presented approach is justified and its usefulness in cases that other methods usually fail is demonstrated. This general method applies well even in such cases as thick gratings, non-Bragg incidence, and cases in which higher diffracted orders are needed to be retained, or evanescent orders corresponding to real eigenvalues have to be included. The efficacy of the proposed method relies on the fact that although Legendre polynomials span a complete space, they are not eigensolutions and hence each polynomial basis function bears a weighted projection of all eigenfunctions. Thus no modal information is completely missed in the ineluctable truncation process. In deriving the formulation, a rigorous approach is followed.

Free electron lasers have been the subject of intensive interest during the recent decades. In this paper, free electron laser having sheet electron beam with arbitrary inhomogeneous profile of transverse distribution of the beam current density is studied in the linear regime, whereas a novel approach based on the Legendre polynomial expansion of eigenfunctions, already used in analyzing optical structures including stratified structures and diffraction gratings, is adapted to find the eigenfunctions and eigenvalues of the structure. As for this method is unconditionally stable, it works pretty well even in those cases in which the conventional transfer matrix method suffers from numerical instability, i.e. the detuning parameter is negative with a large absolute value. Though the used formulation, obtained by solving Maxwell's and Vlasov's equations simultaneously, is limited to the linear regime, it includes the effects of energy spread and space charge fields.