A comprehensive analysis of the light induced processes in the bacteriorhodopsin (BR) molecule is. presented based on the rate equations approach. A simplified energy level scheme is proposed to represent the complex photochemical cycle of the BR molecule. Light intensity-induced population densities in various states of the molecule at steady-state are computed and used to derive an analytic expression for the absorption coefficient of the modulation beam passing through the medium containing the BR molecules. This coefficient encompasses all possible photochromic transitions of the BR molecule and can be used to analyse various nonlinear processes exhibited by the BR molecules. The analysis is used to study the spatial light modulation and the mode-locking applications of BR molecules. It is shown that for a probe beam at 412 nm upto 82% modulation can be achieved using a modulating laser of intensity 3.2 W/cm2 at 570 nm. For temperatures ~77 K, the transmission at 610 nm can be switched from the initial 13% to 96% for modulating laser intensity of 11 W/cm2. The analysis demonstrates the feasibility of constructing a molecular spatial light modulator using BR molecules with a dynamic range of 0.12 and sensitivity of 0.66 cm^/W for the read beam at 412 nm. The analysis of mode-locking demonstrates the possibility of producing pulses of 64 fs duration and 7 nJ energy by passively mode-locking a laser in the visible region using the BR molecules.

Bacteriorhodopsin, a natural occuring photochromic protein present in the purple membrane of certain halophilic bacteria has been shown to posses unique optical properties. The ground state has a broad absorption band (bR) centred at 570 nm (light adapted bacteriorhodopsin) and a metastable state (M) centred at 412 nm with a thermal relaxtion time of 10 ms. The metastable can be stimulated to decay to the ground state through light in 200 ns. The thermal relaxation time can, however, be extended by five orders of magnitude through suitable chemical treatment of the purple membrane, without compromising the stimulated decay of 200 ns. At the same time, there is a concomitant change in the dipole moment accompanying trans to cis isomerization and. charge translocation across the purple membrane. Bacteriorhodopsin is also a highly nonlinear optical material with a large third-order nonlinearity. A nonlinear Kerr coefficient of 10-4 cm2 /W has been reported for this material, and new mutants are reported to possess even higher nonlinearities. This high nonlinerity enables the use of even fairly low- powered lasers to investigate nonlinear optical propagation. The time constant of the nonlinearity is on the order of a few milliseconds. We report in this talk applications of bacteriorhodopsin thin films for holographic storage, coherent-to- incoherent conversion, incoherent-to-coherent conversion, and contrast reversal of images.

Much attention has been dedicated in the recent years to the study of suitable methodologies for preparing nonlinear optical composites, with the aim of designing all-optical switches for optoelectronics applications. The nonlinear response of glasses may be enhanced by several orders of magnitude by introducing metal nanoclusters in the glass matrix. Metal nanoclusters have been formed in light waveguides, obtained by Ag+- or Cu+-Na+ exchange process in glass, by either irradiating with low-mass ion beams or by heating in hydrogen atmosphere at temperatures varying in the range 100-250°C. Metal nanocluster-silica composites have been also obtained by the sol-gel technique. Nanocluster modifications induced by pulsed laser irradiation have been investigated and annealing behavior of nanoclusters synthesized by the sol-gel process has been studied. Composites were characterized by Secondary Ion Mass Spectrometry and Rutherford Backscattering Spectrometry, in order to determine concentration depth-profiles and by Transmission Electron Microscopy for the nanocluster detection and size evaluation. Optical analyses were performed to evidence linear and nonlinear properties.

In recent years, many novel materials of interest to the opto-electronics community have emerged. Extensive research of the newly introduced materials in the form of epitaxial thin films and heterostructures requires a rapid prototyping technique. The pulsed laser deposition (PLD) technique is well suited for such rapid and controlled growth of multilayers, demanding lesser parameter optimization effort than the conventional MBE and MOCVD processes. We have used the PLD technique to deposit epitaxial films of TiN, AIN, GaN films and heterostructures on sapphire substrates. The films were characterized by X-ray diffraction (XRD), Rutherford backscattering/channeling (RBS), transmission electron microscopy (TEM), electrical, and spectrophotometric measurements, which indicate a high degree of epitaxy and crystal quality.

The formation of hollow microcylinders, starting from a thin film of sputtered glass, is reported. To begin, a three-inch silicon wafer was coated with 5 µm of SiO2. An RF sputtering system was then used to deposit 1.7 µm film of a multi-component glass. The sputter target was an erbium-doped lanthanum sodium silicate glass, normally used for optical amplification and filtering. Ridge waveguides with rectangular cross-sections (~ 6 µm x 1 µm) were obtained by partially wet-etching die sputtered glass film. The waveguides were then annealed in a thermal oxidation furnace. When heated above the glass transition temperature, the rectangular ridges were re-shaped into hollow cylinders with a diameter of approximately 9 mm. The hollow microcylinders remained structurally intact over distances of up to 1 cm. Similar results were observed with structures fabricated from three different sputter deposition runs, indicating that the process was repeatable. Early results investigating the cause for this transformation will be presented. Attempts to launch light and fluid in this structure will also be discussed. These microcylinders may find application as microchannels for gas/fluid transport, grooves for fiber alignment and microlenses.

An ablation wave from materials based on laser supported detonation and laser supported combustion has been created by exposing them with laser. The purpose of this study is to analyse the mechanism of the ablation wave using a high speed streaked camera and to realize more precise microprocessing by controlling the wave. The laser used in this experiment was an ultraviolet XeCl excimer laser of high photon energy with a wavelength of 308nm, a maximum pulse energy of 500mJ, a pulse width of 30ns and a maximum frequency of 50Hz. The high speed streaked camera had a resolving power of 100ps. Metal-ceramics functionally graded materials (FGMs) were selected as sample materials. An experimental system of laser microprocessing was constructed and the behavior of the ablation wave from FGMs were observed. The experimental results showed that the mechanism of the ablation wave from FGMs was a photochemical reaction with thermal interaction.

Dramatic progress in the study of intersubband transitions in quantum wells in recent years has led to the fabrication of high sensitivity infrared detectors and focal plane arrays. The use of different material systems, using both III-V compounds and silicon-based has allowed a wide wavelength coverage from 3 to 20 /µn. Novel device architectures have also been incorporated to overcome the polarization selection rule and to allow the fabrication of multicolor detectors. For example, random gratings on the detector and strained quantum well structures have been incorporated for normal incident detection. For multicolor detection and the tuning of detection wavelength, asymmetric quantum well structures have been utilized. In this paper, we will first review intersubband transitions in symmetric and asymmetric quantum well structures and then discuss recent progress of normal incident quantum well infrared detectors. Finally, the trends towards the realization of multicolor detectors and large area infrared imagers will be discussed.

Photonic Devices inherently have demanding materials requirements. The demanding needs of photonic devices generally require both large variations direct bandgaps and a large changes in the index of refraction in the materials used in their construction. Unfortunately, nature generally does not accommodate the material needs of these devices. It is generally difficult, if not impossible, to combine materials with the desired bandgaps and indices of refraction in a homogeneous structure formed by epitaxial growth techniques. Bonding has thus become an important technique to combine the required materials to achieve the features needed to create photonic devices.

This paper describes the observation of optical short pulses by gain switching method, using circuit modeling techniques. The models are illustrated with a number of simulated laser response characteristics. The models can be easily used with circuit analysis computer programs such as PSPICE.

Synthesis of InGaAs quantum wells, lattice matched to InP by metalorganic vapour phase epitaxy poses severe problem because arsenic is incorporated into the growing solid much more readily than phosphorus. Therefore growth of InP after growing InGaAs is likely to have a transition layer of InGaAsP/InAsP of uncontrolled composition even though the arsine flow is switched off, because of trace amounts of arsine in the growth reactor. A common technique used to minimise the problem is to introduce a suitable pause in the growth sequence during the transition from InGaAs layer to InP layer in order to minimise the arsine content in the growth chamber, before starting the growth of InP. We find from photoluminescence (PL) measurements that this pause is not sufficient to optimise the well quality. The halfwidth of PL can be improved further by growing an intermediate layer of InGaAsP between InGaAs and InP, such that the quantum well structure consists of 5 layers in the sequence: InP/InGaAsP/InGaAs/InGaAsP/InP. The optimisation of pause sequence is made during the switching steps: (i) pause PI- in the transition from InGaAsP to InGaAs and (ii) pause P2- in the transition from InGaAs to InGaAsP. The nominal composition of the well is In0.53 Ga0.47 As with thickness of 100°A. The nominal composition of the intermediate layer is In0.72Gt0.28gAs0.6P0.4 Two Ga sources have been used during the growth of In0.53 Ga0.47As and In0.72Ga0.28As0.6P0.4, while the In source is kept the same. The PL emmision peak lies at 0.815 eV at 9 K, with halfwidth ~ 6 meV which is close to the state of art. In comparison, the PL measurement from QW samples with three layer sequence InP/InGaAs/InP with pauses at both interfaces of InGaAs gave halfwidth of more than 10 meV. Lattice matching of all the layers used in the above studies is checked separately on thick layers grown under similar growth conditions, by using X-ray diffraction measurements.

Strained layer quantum well structures present an interesting potential for optoelectronic and high speed device applications.We study the effect of strain on the energy levels of InxGa1-xAs /GaAs quantum well structures, in whichthe InxGa1-xAs layers are strained. Strain effects the band structure by increasing the band gap and splitting thevalence band , the heavy hole (HH) band lying at a higher energy than the light hole (LH) band, therby confining the heavy holes in the InxGa1-xAs and light holes in the GaAs layers. We analyse the strain and quantum size effects by considering a multiple finite quantum well potential structure. We vaiy the well width, barrier width, the number of the wells as well as the alloy concentration and calculate the optical transition between the conduction and valence band levels.

New formulas are derived to calculate the tunneling and thermionic dark currents in GaAs/Alx Ga1-xAs quantum well infrared detectors. Variation of the dark current (Id) with barrier width and doping density is systematically studied. It is shown that increasing the barrier width and/or decreasing the doping density in the well do not always reduce the dark current. Theoretical variation of Id with bias is compared with experimental data.

With the active layer photomixing technique, parasitic-free intrinsic modulation frequency responses of semiconductor laser diodes (LD's) can be measured. But the mechanism of the carrier density modulation, and consequently, the region where the carrier density is modulated, depends on the wavelengths of the injected beams. In this paper, the modulation frequency response of a 1.55 µ m distributed feedback LD has been measured with the active layer photomixing technique under two distinct conditions; one is the case when the carrier density is modulated by the band-band absorption, and the other by the stimulated emission. The measured modulation frequency responses are quite different from each other. Especially, the response at the relaxation oscillation frequency in the band-band absorption case is suppressed about 1 OdB compared with that in the stimulated emission one.

We report the experimental and theoretical results for high reflectivity (>99% at 850 nm) AlAs/AlGaAs distributed Bragg reflectors (DBR) for vertical cavity surface emitting semiconductor lasers (VCSEL). It is investigated that up to ± 10% deviations in thickness distribution of DBR layers negligibly affects the reflected bandwidth if the distribution is random. However, ascending/ descending order in thickness distribution beyond ±5% results in rapid shrinking of the reflected bandwidth with >99% reflectivity.

Multilayer and compound structures CdSe/polyethylene terephthalate with the blue shift up to 0.8 eV are obtained by electron-beam evaporation. Influence of forming conditions, concentration and size parameters on optical and photoelectrical characteristics and band structure of the semiconductor phase is investigated. It is shown that the monomolecular polymer layer is sufficient for effective demarcation of growth of semiconductor particles. That is why high concentration of CdSe (up to 80 vol.%) is obtained in the structure.

With the increasing use of the new generation of semiconductor (s/c) optical devices that support weakly diffracting fields it is seen that the conventional analysis, widely used in the context of (longitudinally) uniform structures, is not readily applicable. Prominent examples of weakly non-uniform devices are: i) High Power Semiconductor Optical Sources using adiabatically tapered structures, [1], [2]; ii) VCSELs with Quantum Well layers that are pumped over a small area but have no explicit lateral guiding, [3], Field analysis using eigenfunction expansions seems a suitable approach and this paper presents results of calculations on tapered optical sources that are obtained by using the Hermite-Gauss (HG) functions in a field expansion scheme with the Collocation method for numerical solution.Entirely numerical procedures, e.g., Finite Elements, Beam Propagation Method, or Cellular Automata [4] have been developed for analysing optical fields in non-uniform structures. However, there are advantages in developing quasi-analytic methods since they tend to provide a closer link with the physical model and may lead to very efficient computation schemes. Such features can be invaluable in the modelling of, particularly, active devices where, in general, a self consistent solution is obtained only after several iterations and hence an efficient computation technique is desired. Although the Gaussian Beam method [5] has been applied often in the past to analyse, for example, gas and solid state lasers, and is a quasi-analytic approach, it appears to be less amenable for use with laterally inhomogeneous structures (typical of s/c devices), [6]. Thus the more systematic eigenfunction expansion procedure is described in this paper since it is far better suited to analyse the s/c optical devices of present interest Specifically, considering a two-dimensional (x,z) structure that supports fields propagating predominantly in the z-direction, the total field solution is written in the general form, [7] F(x,z) = X Am(z)\|/m(x) (1)m where yra(x) belong to a complete set of orthonormal basis functions, and Am(z) are the z-dependent expansion coefficients. Whereas any complete function set may be used in eqn. (1), e.g., [8], [9], [10] in the present formulation the set of HG functions has been considered to be the most apt since the complete set of HG functions is discrete, and both the HG functions and the actual, physically sustainable total field of any waveguide decay to zero along the transverse (x) axis at infinity (this is so even in the presence of radiation modes), [11], [12]. Substitution of eqn. (1) into the wave equation (along with the boundary conditions) yields a set of coupledjlordinary) differential equations in the expansion coefficients, Am(z). It has been argued that this set of equations may be numerically solved very efficiently by the Collocation method, [13], and hence this procedure has been used for the present calculations. Test results obtained from the HG-Collocation method compare very favourably with those from analytic and/or purely numerical schemes for both, purely real and complex refractive index media (with small imaginary part). The proper analysis of active devices requires calculations which self-consistently solve for the optical field (intensity) and the carrier (inversion population) density distribution. Several methods exist for solving the non-linear diffusion equation for the carrier density, [14]. However, in the present analysis the HG-Collocation scheme is used and is shown to be a very effective method for solving for the carrier distribution. The talk will present a brief description of the theoretical development of the model and will provide a comparison of several experimental and theoretical results not only to prove the suitability of the HG-Collocation method, but also to confirm the substantial advantages in using the proposed scheme, not least of which is the fast and efficient numerical model that has been achieved.

Tapered geometry semiconductor optical sources yielding high-power and narrow far-field patterns have proven to be extremely successful, not least because of the relatively simple fabrication requirements. Almost all of the publications to date concentrate on tapers where the width increases linearly along the length of the device. A common feature of the output from such devices is a quadratic phase curvature [1]. In order to achieve the desired, narrow far-field profiles this phase curvature is usually removed by using an appropriate lens external to the device. However, such an arrangement is expensive and it is desirable to design device structures which produce high powers and acceptably narrow far-fields without the use of a lens.Two categories of devices have been reported -(i) those without any built-in lateral index step;(ii) those with a built-in lateral index step. This paper will provide persuasive evidence that a parabolic taper, with a built-in lateral index step, is a simple and convenient structure which significantly reduces the quadratic phase curvature so that a nearly diffraction-limited far-field is obtained without the use of a lens. A model has been developed to analyse tapered device structures. The important features of the model will be provided along with computed results for the output characteristics of a travelling-wave amplifier and a single-taper laser. In addition, a comparison of the characteristics of linear and parabolic taper Bow-Tie lasers will be presented to clearly demonstrate the advantages of using the parabolic shape. The model is based on a piece-wise constant (longitudinally segmented, stepped width) and local mode expansion analysis. Since built-in, lateral index-step devices are considered it is justifiably assumed that the refractive index change due to the carrier distribution is small so that the mode shapes remain unchanged. However, of paramount importance are the effects of the carrier distribution on the complex propagation constants of the local modes. In this model the propagation constants are calculated using a perturbation analysis. It is important to note that the total field in the device may still change due to the carriers at any longitudinal position. The model is thus compact and efficient in computation time. Near diffraction-limited lateral far-field profiles, directly from the facet, with FWHM less than 2° are predicted for a 2mm long bow-tie laser 30jnm wide at the facets. An increase in slope efficiency for the parabolic shape compared to the linear shape is predicted for bow-tie lasers due to improved coupling of the field reflected at the facet into the narrow section of the device. These results show that parabolically tapered bow-tie lasers should be useful as high-power semiconductor optical sources producing narrow output fields. A model has been developed to analyse tapered device structures. The important features of the model will be provided along with computed results for the output characteristics of a travelling-wave amplifier and a single-taper laser. In addition, a comparison of the characteristics of linear and parabolic taper Bow-Tie lasers will be presented to clearly demonstrate the advantages of using the parabolic shape. The model is based on a piece-wise constant (longitudinally segmented, stepped width) and local mode expansion analysis. Since built-in, lateral index-step devices are considered it is justifiably assumed that the refractive index change due to the carrier distribution is small so that the mode shapes remain unchanged. However, of paramount importance are the effects of the carrier distribution on the complex propagation constants of the local modes. In this model the propagation constants are calculated using a perturbation analysis. It is important to note that the total field in the device may still change due to the carriers at any longitudinal position. The model is thus compact and efficient in computation time. Near diffraction-limited lateral far-field profiles, directly from the facet, with FWHM less than 2° are predicted for a 2mm long bow-tie laser 30jnm wide at the facets. An increase in slope efficiency for the parabolic shape compared to the linear shape is predicted for bow-tie lasers due to improved coupling of the field reflected at the facet into the narrow section of the device. These results show that parabolically tapered bow-tie lasers should be useful as high-power semiconductor optical sources producing narrow output fields.

A large circuit model of a strained quantum well laser has been developed from rate equations. The electrical and optical properties of the laser are dealt with, and interactions between these quantities are taken into account The model is compatible with a more versatile, universally applicable and readily available circuit simulation package PSPICE, and has added advantage that each of the elements in the model has a physical interpretation. This enables the dynamic operation of the laser to be easily visualized. The application of the model are illustrated with a number of simulated laser response characteristics.

Recently a high level of in-plane emission in VCSELs has been reported [1]. Existence of the problem for vertical cavities has been previously demonstrated in connection with electron beam pumped lasers [2] and LED [3]. The effect results in increasing threshold and a reduction of efficiency of the devices [2]. Prili- menary estimation shows importance of the problem for large diameter VCSEL as well [4]. Thus, suppression of the in-plane emission gives a way of improvement of VCSEL operational characteristics. Features of the spontaneous emission (SE) pattern is a basis for simulation and modelling of the effect. In this paper the SE is considered for both A— and A/2— cavities in the framework of a planar multilayer structure model taking into account the QW optical spectrum. The model is suitable for large diameter VCSELs with weak radial boundaries and allows the consideration of both in-plane and vertical propagating modes. Results of calculation show that, for a A/2— cavity, all in-plane propagating modes have weak coupling to a dipole of the active region (QW), unlike the A— cavity case where there is a mode having a field maximum in the position of QW and thus strong dipole coupling. This leads to a 2 - 4 times reduction of spontaneous emission in the in-plane direction for the A/2— cavity. However, this is still comparable with the SE in ”vertical” modes and therefore can not be neglected, as is usually assumed. The ” vertical” modes in general propagate at different angles to the 55 pure vertical direction” and hence have an in-plane component of the propagation constant. Analysis shows the ’’pure” vertical part of the SE to be extremely small compared to the total. Thus, the in-plane component of the SE should be taken into account to improve VCSEL desig

At 1.3 mm the low group velocity dispersion of standard single mode fibre would appear to make this wavelength attractive for high bit rate transmission applications. The transmissions of optical solitons is one possible technique for reaching the high bit rates required in future transmission systems. A key component in such a system is the soliton laser source. This device must be both a stable and reliable source of optical pulses at a high bit rate. For this purpose a Mode-Locked Extended Cavity Laser (ML-ECL) using a Bragg reflector as wavelength selective feedback component for generation of a > 10 Gbit/s pulse stream has been developed operating at 1.7 GHz in the fundamental mode. A semiconductor laser chip, anti-reflection coated at one facet and high reflection coated at the other, was coupled with a lensed fibre grating. In comparison to bulk optical components the use of fibre based components ensured a better mechanical stability and about 50 times more output power. Additionally, the integration of the device is generally simplified To use this mode-locked laser as a pulse source for a high-bit-rate telecommunications system detailed analysis of the ultrashort optical pulse performance is important. Adjacent pulse interaction in a fibre link may be enhanced or suppressed depending on their non-linear phase behaviour during transmission. We briefly present a recently developed interferometric autocorrelator which is capable of detemiing both, the temporal width (FWHM) and the non-linear phase of optical pulses by assuming a certain intensity shape and extrapolation the phase. By comparing the autocorrelation traces with simulated data on the basis of a Gaussian intensity profile and a quadratic phase behaviour a linear chirp of about 25 ps within the FWHM of the optical pulses was found. The developed ML-ECL is a prototype of a laser source which will be utilised in transmission experiments in the project UPGRADE in the European ACTS programme. The realisation of optical soliton transmission at 1.3 pm over standard single mode optical fibres is a goal of this project. A field trial of 10 Gbit/s, SDH based, soliton transmission will be demonstrated at the CeBIT exhibition in Hannover in 1997. The techniques reported here play an important role in achieving the project objectives

Intermixing of the well and barrier layers in quantum well infrared photodetectors (QWIPs) can be used to realize a broadened spectral response as well as multiple color detectors. We describe die experimental results of both rapid thermal annealing (RTA) and laser annealing (LA) QWIPs operating in the 8-12µm regime. The peak spectral response of the annealed detectors was shifted to longer wavelength as compared to die as-grown detectors. In general, a decrease in detector performance after annealing is also observed which may be attributable to a change in the absorption coefficient caused by the out-diffiision of dopants during annealing. Recent advances in growth technology, complimented by innovative structures should offset any degredation in performance. Thus, the post-growth control of the composition profiles by annealing offers opportunities to fine tune various aspects of a QWIP’s response.

Low-loss high-silica fibers were first reported in 1970 and since then these glass fibers have proven to be the most viable transmission media for lightwave communication. Even though the first low loss fiber was a single-mode fiber, initial systems revolved around multimode fibers from the point of view of practical reasons like easier fiber to fiber splicing and coupling from off-the-shelf available LEDs. Since early 1980s, single-mode fibers became the preferred media for optical transmission and single-mode fibers overtook the multimode fibers for lightwave communication. Networks which operate with single-mode fibers at the 1.31 /mm low-loss wavelength window are now taken for granted. Operations with advanced fiber designs at the silica fiber’s lowest loss window of 1.55 /mm are now gearing to take over the long-haul routes; these are contemplated to be based on erbium doped fiber amplifiers (EDFA) and dispersion shifted fibers. An alternative and promising scheme for this wavelength window involving use of already laid fibers optimized for 1.31 /an requires insertion of a dispersion compensating fiber(DCF) as an additional component in the link. In the immediate future, lightwave systems would operate with EDFA at the OC-48 level, which amounts to a bit rate of — 2.5 Gb/s and are compatible with SDH (STM-16) networks. For WDM transmission with EDFAs, one would require a variety of in-line fiber components like wavelength selective couplers, and bandpass/bandstop filters to add/drop channels. Until recently, optical fibers were considered to be optically linear. However, with the introduction of EDFAs, which provide large and broadband (* 30-35 nm) optical gains, a large number of WDM information channels can be transmitted and simultaneously provided gains across this broad gain spectrum of EDFAs. Furthermore, amplifier spacings could be much longer than the present day regenerator spacings. These two fall outs of EDFA namely, large optical power density and long interaction lengths have elevated optical fiber nonlinearities to an important design issue in lightwave systems. In this presentation we would aim to trace the evolutionary trends in lightwave communications seen along the above lines since the reporting of the first low-loss fibers a quarter century ago. INTRODUCTION

Optical fiber communications have emerged as the leading edge technology of the present and the coming twenty first century. The deployment of optical fibers has seen nearly 20% compound annual growth since early 1980. This growth is accelerating, and the world wide production of optical fiber is expected to reach over 50 million km by the year 2000. Thus optical fibers have emerged as a high technology commodity. This paper will review two decades of advances in optical fiber process technology.

The growing interests in fibre lasers, amplifiers and sensors have led to the need for fabricating preforms and fibres with the given concentration of rare-earth ions and composition of the host matrix. Several techniques have been developed, of which the solution doping process to incorporate rare-earth ions into the core of fibre is now most widely used. In.this technique rare-earth ions are introduced from solutions and a number of modifications to the conventional MCVD process allow the incorporation of dopant ions from solution. In this paper, we report measurements of the R.I. profile, distribution of rare-earth ions in the silica matrix, Photoluminescence spectra of the rare-earth ions in the silica matrix as a function of composition of matrix and absorption spectra for preforms and fibers made through solution doping technique. The soot was deposited in such a way that it forms a porous and adherent layer suitable for solution doping. A detail thermochemical calculation has been worked out to ascertain reaction kinetics including temperature for completion of reaction, efficient deposition and dehydration of soot. Calculation shows higher refractive index is obtained for a given flow rate of GeCl4 in a solution doped preform than the conventional MCVD preform. NJ/OJ has been used in place of Cl2/02 and it gives reasonable dehydration efficiency to remove -OH ions. The effectiveness of the dehydration procedure is confirmed by the very low -OH ion absorption loss around 1390nm. The rare-earth doped preforms and fibres with different core compositions (GeO2-SiO2, GeO2-Al2O3-SiO2, Al2O3-P2O5-SiO2 and P2Os-SiO2) fabricated have been characterised by measuring the R.I. profile, attenuation and fluorescence spectra. The doped fiber shows the change in relative intensity of the absorption peak for different host composition. Strong fluorescence has been observed at around 1536 nm for Er3+ ion in Ge02-Si02 host glass. However, the peak changes its position and shape when A1 put in the host.

We propose a novel dispersion compensating fiber design consisting of two highly asymmetric concentric cores. We show that the fundamental mode of the proposed fiber can have very large negative dispersion values (~ —5100 ps/(nm.km)) with larger mode field diameter (~ 9/µm) relative to the existing dispersion compensating fibers.

A bright, broad-band, optically-pumped fiber fluorescent source is described. Fluorescence data obtained from fiber and bulk samples of rare-earth doped chalcogenide glass are presented.

Since the pioneering work of Enoch in 1960, it has been experimentally well established that photoreceptors (rods and cones in the retina) of the human eye exhibit waveguiding properties. Photoreceptors behave as classical fiberoptic elements. It can be inferred that photoreceptors have specific orientation properties and it is found that they are aligned toward the exit pupil of the eye. This can be studied by non-invasive psychophysics and is known as the Stiles Crawford Effect of the First Kind. This refers to the feet that the eye is sensitive to the direction of incidence of light. Modeling the behavior of single photoreceptors as weakly guiding waveguide to incident light has greatly enhanced our knowledge of the optical properties of the visual system. In more recent work, the spatial impulse response of a single variable-cross-section photoreceptor has been characterized by assuming the incoming radiation to be an initial field propagating under confinement conditions inside the receptor. Using techniques of Fourier analysis the total transfer function is obtained. Both inner and outer segments of the photoreceptor are characterized as low-pass filters. The dependence of the transfer function on the modal parameter has been analyzed. In this article, I will review psychophysical and other evidence for waveguiding in photoreceptors as well as theoretical models of the Stiles-Crawford effect.

The last decade has set the stage for the emergence of epitaxial metal-oxide films as one of the leading Optoelectronic material system of the future. This scenario was primarily stimulated by the advent of high temperature superconductors, a multi-component oxide, but more recently the resurgence of interest in doped oxide thin films has led to renewed interest in epitaxial oxide films in which a number of functionalities of importance to Optoelectronics have been seen. In this article we will make a brief sketch of the status of this technology.

The traditional procedure to synthesize optical devices with specified transmission characteristics is to start from device parameters such as refractive index profile with a known transmission characteristics that is close to the specified characteristics of the device and iteratively change the device parameters until the specified transmission characteristics of the device is achieved. This traditional procedure is time consuming and the choice of initial device parameters that has a transmission characteristics close to the specified characteristics is not always easy. However, techniques based on inverse scattering theory can directly provide the device design parameters given the specified transmission characteristics of the device. The inverse scattering techniques contrasts the traditional design procedure by their ability to yield design parameters in a non-iterative manner. The transmission of electromagnetic radiation in an optical waveguide is governed by vector wave equations which can be simplified to scalar wave equations under the assumption of “weakly guiding” approximation for TE and TM modes. The scalar equations obtained for TE and TM modes can be recast into Schrodinger type of equation that is normally encountered in quantum mechanics. One can draw an analogy between the quantum mechanical problem and the optical device problem. The potential well of the quantum mechanics corresponds to refractive index profile of the optical device problem; the time evolution of wave packets in quantum mechanics corresponds to propagation of optical modes along the axis in an optical device and the existence of discrete bound states in quantum mechanics is analogues to the presence of discrete propagating modes which in the geometrical-optics model corresponds to the modes that satisfy total internal reflection. Now, the refractive index profile; an important device parameter can be reconstructed from an assumed scattering coefficient that characterizes the specified transmission characteristics of the device. An analytical technique based on Gelfand-Levitan-Marchenko integral formulation and an numerical technique that extends the capability of the inverse techniques to provide solutions for generalized reflection coefficients that have been developed by us will be discussed. Design of a class of devices such as intra-chip optical interconnects, all-optical logic devices and efficient guiding structures for integrated optical amplifiers based on the inverse techniques will be discussed. The design of efficient guiding structures for optical amplifiers requires a guiding medium that has same propagation constant for both the signal and the pump. This leads to the inverse theory for reconstruction of energy dependent potentials or wavelength dependent refractive indices. We have solved this problem and we will show you a refractive index profile of a guiding structure obtained by inverse scattering technique that has same propagation constant at 1.55mm and 0.98/mm. The extension of the inverse techniques developed for planar structures to cylindrical structures is not straight forward, rather very cumbersome. However, we have succeeded in developing inverse scattering theory for the design of multimode cylindrical optical waveguides with same propagation constant for all modes and the design of single mode fiber that has the same propagation constants for more than one wavelength. Such structures will find application in image transmission and fiber amplifier design respectively. Our talk will include discussion of these cases.

This paper deals with the spectral properties of optical networks that consist of a single, distributed 3x3 coupler embedded in a feedback circuit. We first specify a set of coupled linear differential equations that characterize the coupler. The solution of this set of equations yields the scattering transfer matrix of the coupler from which, through linear transformation, we obtain its scattering matrix. Subsequently, we determine the scattering matrix of the composite network, i.e., the coupler, embedded in the external circuitry. Numerical experiments illustrate the properties of composite networks of various configuration.

Coherence multiplexing (CM) exploits the coherence properties of optical sources so that many users may be multiplexed onto a single physical channel giving a secure, asynchronous mode of transmission, free of central network control. In this paper, for the first time, the CM system performance limitations owing to the impact of interference from unwanted users is presented and the suitability of LED, single mode Fabry-Perot laser and DFB sources are considered. It is concluded that the single mode Fabry-Perot is preferred, offering readily achievable interferometer length tolerances and near complete RF rejection of interferometric noise at the receiver. However, interferometric noise still limits the network capacity: a linewidth of lnm permits 17 users to be simultaneously offered a single video channel with a BER<10-9.

This paper describes a new pass-band optical filter utilizing a sharp wavelength-selective power-transfer occurring in the directional coupler composed of the K+- and Ag+-ion exchange waveguides. The wavelength- selective power-transfer is achieved by greatly different dispersion relations caused by the large mismatch in the index profile between the ion-exchange waveguides forming the directional coupler. A narrow pass-band width of about 7 nm is measured in the filter fabricated in the soda-lime glass substrate. The fabrication technique with two-step ion-exchange of the K+ and Ag+ ions, is also described.

Thermally tapered optical fiber components such as couplers, wavelength division de/multiplexers require a precise analysis of the relationship between technological and device parameters. Over the past 15 years there have been a lot of studies devoted to the familiar "fused biconical taper coupler" and similar fused bitapered optical components usually fabricated with microbumers. The main drawback with this fabrication method is that the temperature cannot be monitored during the fiber- coupler fabrication and hence the fabrication conditions are not controlled by physical parameters such as the fusion temperature and the fusion time of the fibers. Recently, to solve this problem an interesting method using a microheater has been proposed which allows wide and stable heat regions in order to obtain couplers with smooth and gradual taper shapes for low excess losses. Using this fabrication method, we propose in this paper a precise model for predicting the shape of thermally tapered fibers and for analyzing the influence of tapering on the sintering of fibers during the fabrication of fused bitapered optical components. The model has been developed and verified by preliminary experimental measurements in order to relate technological parameters to the characteristics of fused optical fiber components. Much of the literature published previously about this problem still lacks of generality and often approximates the taper geometry with simple analytic function and considers that tapering and sintering (fusion) phenomena are uncoupled. The basic approach is first to fuse the fibers and then elongate the fused fibers while considering that the degree of fusion is conserved during the tapering step. This hypothesis can be justified if the sintering is executed at high temperature ( above 1500°C) and elongation at a lower temperature (about 1300°C). But if sintering and tapering are executed at a similar temperature, or more generally if fibers are elongated at high temperature (typically 1500°C) it appears necessary to consider both sintering and tapering as coupled phenomena. Given that most glasses can be modeled as Newtonian liquids, we have simplified the basic model of nearly unidirectional viscous dominated extensional flow by using a simple rheological model to describe the process of pulling fibers in the range of their softening temperature. But the salient feature of our model is that it allows also a description of the degree of fusion by taking into account of surface tension forces and clearly demonstrates a coupling between the tapering and sintering phenomena.

Radially graded - refractive index (GRIN) rod lenses have a number of advantages over conventional lenses in manipulating and processing the optical signals in optical fiber communication systems(1). In order to have a device using GRIN rod lenses, design requirements are varied in nature. A system containing one or more than one radial grin lenses or a combination of radial and axial GRIN dements is designed. The next step is to devdop the required GRIN elements with desired material properties. Finally, the system is optimised as a compromise between the ideal and real available GRIN elements. Ion exchange technique is the most widely used method to devdop GRIN profile in glass due to its simple instrumentation and ease of control. We have devdoped a novel method to predict the GRIN profile for a given set of experimental conditions (2). This method helps us in devdoping different types of GRIN profiles in the glass rods with minimum number of experimentation. After the glass rods have gone through the process of ion exchange, refractive index profile are measured in a set - up devdoped by us (3,4). In this paper, we present the development of radial GRIN rod lenses with a diameter of 1 to 2 mm. These GRIN rod lenses can be used for coupling of laser light source to the optical fiber.

Tap couplers with small coupling ratios (1-10%) play a critical role in monitoring optical fiber systems such as optical amplifiers. They are often used in a feedback control loop and hence must exhibit extreme stability. Very small changes in the operating characteristics of the taps can be interpreted as changes in laser power giving rise to instability in the amplifier gain. Tap couplers made by the fused biconical taper (FBT) process are inherently stable with regard to temperature and, thus, this is usually not a concern. Of much greater significance are the wavelength dependence and polarizastion sensitivity of the tap ratio. Wavelength dependence can be minimized by introducing a mismatch in the propagation constants between the two fibers, but it is also influenced to a lesser extent by the coalescence of the fibers. On the other hand, polarization sensitivity is primarily dependent on coalescence and only somewhat dependent on propagation constant mismatch. As a consequence, it is necessary, in practice, to trade off wavelength dependence against polarization sensitivity in order to optimize the overall stability of the tap coupler. We report here the results of such an optimization study, consistently yielding taps with less than 0.5 dB total change in the insertion loss of the tap leg due to all effects.

The stability of fiber-optic solitons is studied by Lyapunov’s direct method. The necessary and sufficient conditions for the stability of the fundamental non-nodal solitons, under the condition that the energy (power) of the soliton is conserved, are established for arbitrary nonlinearity in the model equation. They are then used to determine the stability domains of solitons in the model with saturating nonlinearity. For the last couple of decades optical solitons in fibers have attracted much attention due to their practical utility in various nonlinear guided wave devices [1-3]. It is needless to say that the stability of these structures, including the bistable solitons [4] and light bullets, is of immense importance in connection with their practical applications in all-optical communication systems, switches etc. In the given paper we apply the direct Lyapunov method for the determination of general criteria for the stability of the fiber-optic solitons.

Hie generation and detection of quadrature squeezed states of light with a fiber ring reflector is reviewed. A 0.85µm polarization maintaining (PM) fiber is used as the nonlinear medium, and it is pumped by a 1.064µm mode-locked Nd:YAG laser. By the optical Kerr effect in the PM fiber, the input 1.064 mm light is squeezed. The combination of the smaller core diameter and the shorter source wavelength is effective in obtaining a higher peak power density, and consequently, a larger squeezing parameter at a low pump power. The squeezed vacuum is generated by combining the 50m long PM fiber ring with a 1:1 PM fiber coupler, and is detected by a balanced homodyne detector. The squeezed light noise is measured at 200kHz, and a squeezing parameter of 3.7dB is obtained at the input peak power of 5W. A simple calculation model based on the wave optics is used to estimate the squeezing parameter as a function of the input peak power. The calculated results agree well with the experimental ones. The squeezing in PM fibers is limited by the polarized guided acoustic wave Brillouin scattering (GAWBS), which is the phase noise originated from thermally induced fluctuations in the refractive index of the fiber. To avoid the noise due to GAWBS, a shorter PM fiber is preferable.

The design procedure is described of integrated optical switches to be fabricated by inserting a liquid-ciystal cell into a glass planar waveguide. Two configurations are analysed, corresponding to ON-OFF and OFF-ON operation, respectively. Preliminary experimental data on the waveguide characteristics and device fabrication process are also presented.

An optically activated waveguide phase modulator has been analysed by treating as a four layer model. Theoretical investigation reveals that the required phase modulation/polarization can be obtained by optically illuminating the guide with appropriate wavelength and energy.

In certain fiber optic sensors and gyroscopes using birefringent fibers, it is essential to precisely align the principal axes of the fibers used, when two fibers are spliced or connected. The proper alignment of the linearly polarized input light wave with the birefringence axis of a fiber is crucial for maintaining the state of polarization of the guided wave. In this paper, a new method for the detection of the eigen axes and precise alignment of the principal axes between two birefringent fibers (HiBi) using shear interferometry is presented.

A novel scheme for simultaneous compensation of dispersion and loss is proposed. This should be realizable by doping an optical fiber having a high negative dispersion coefficient at 1.55 /mm with erbium ions. Numerical calculations are presented for a typical design example. Such an erbium doped dispersion compensating fiber (ED-DCF) should be a promising alternative as a more compact unit as compared to the spliced combination of an EDFA and DCF as two distinct components.

A detailed numerical study of soliton switching in nonlinear directional coupler with saturating nonlinearity is carried out. All commonly used models have been studied. Relevant physical characteristics for switching have been determined, compared and discussed.

Recent developments in dispersion compensating fibers and fiber amplifiers have had a significant impact on optical fiber communication These will be briefly discussed in this article.

We describe here the matrix method for determining the propagation characteristics of optical waveguides

It is well known that fibers with a high index difference between the core and the cladding have additional optical losses, which can not be explained by the known sources. In the case of fibers with a germanosilicate core these losses can achieve several dB/kin for a germanium dioxide concentration of 20-30 mol.%. For fibers with an alumosilicate core additional losses amount to 10-30 dB/km. We believe that one of the origins of the loss increase consist in the interpenetrating of the core and cladding materials at their boundary. This phenomenon takes place in preforms produced by inside vapor deposition, if the viscosities of the cladding and the core glasses are different because of a high level of the core doping. During the collapsing process the boundary of two liquids with the different viscosities can lose stability leading to the material interpenetrating. As the result, so called "viscous fingers" arise along the core-cladding boundary. It is easy to observe these formations in preforms with an alumosilicate core, where the characteristic length of "viscous fingers" is comparable with the core diameter study of the core-cladding interface instability. The main point of the theoretical consideration is the eigen oscillation frequencies of the core-cladding boundary. The study is based on the model of contact of two cylindrical layers of liquids with different viscosities. It is shown th In this paper we present the results of a theoretical and experimental at there is a set of eigen oscillation frequencies such that perturbations at these frequencies are not damped out with time. Also, we have found the frequencies resulting in the fastest increase of the perturbations. For the preforms with an alumosilicat core the simulated results and the number of the observed "viscous fingers" are in a satisfactory agreement. It follow from this investigation that the stability of the core-cladding interface can be improved by using special co-dopants. These co-dopants should change the composition of the two contacting glasses (or one of them) so as to match their viscosities.

Experimental study is made on the active stabilization of the microwave subcarrier (w/2 = 6.8 GHz) phase in the anisotropic fiber lightguide (AFL). Both the AFL electrical length thermal shift and the birefringence increment are proportional to the temperature increment (AT). In experiments, AFL with the length 120 m has been used. The fiber was cooled in the temperature range 46÷20 °C. It is shown the system reduces the subcarrier phase thermal instability 20 times when the temperature speed variations are near 30% as compared with the average temperature speed 5 10-3 K/s. It should be noted that the average temperature speed for a fiber cable on ground surface is 5.10-4 K/s. It can hope, therefore, that the residual relative phase instability for the system under consideration will be far less than 5%.

Great hopes currently lie in new all-optical switching devices based on polymer film waveguides. Here, the design of a novel all-optical waveguide Frequency Converter (FC) is reported, the research activity for which was carried out within the framework of the Italian Project on Telecommunications.

Fiber components based on evanescent coupling of a side-polished single-mode fiber to a highly multimode active/passive overlay k waveguide form an important class of a host of in-line fiber components. We report a normal mode analysis to accurately model such fiber components. Our model yields complete transfer function of all such components/devices. The method is extended to analyze certain very recently proposed semiconductor clad dielectric waveguides which find applications in integrated optics

An experimental and numerical investigation of active waveguides realized on lithium niobate crystals using erbium as doping element is presented. A systematic study of Er diffusion in LiNbCO3 is performed with secondary ion-mass spectrometry to determine surface concentration, penetration depth and diffusion coefficient of erbium as function of various process parameters. A software package is implemented to simulate the propagation of electromagnetic field in active waveguides using an Active Vectorial Beam Propagation Method, which permits to calculate the field evolution along the propagation axis. A comparative study of signal gain evolution, optimal device length and threshold pump power as function of erbium depth profile obtained by various doping methods is presented.

Integrated optical lasers and amplifiers are attracting large attention for their use in optical sensing as well in optical communication systems. Here we present preliminary results on the characterisation of optical waveguides fabricated in Er- doped phosphate glasses either by ion-exchange or by pulsed laser deposition. It is shown that spectroscopic features of the waveguides are fully suitable to the implementation of active guided-wave devices.

We analyze by numerical simulation the optical intensity transmission of several optical waveguide devices. We present characteristic transmission curves for optical fiber sensors and selective WDM channel filters. A Computer Aided Engineering program for helping on the design of optical devices is under development.

Optical fibres are used extensively for the transmission of information and have largely replaced copper circuits for all but the shortest distances. They are capable of enormous bandwidth, very low transmission loss and minimal error rates. However whilst a range of passive fibre components is available to help exploit the fibre capability, much of the signal processing is still carried out electrically. This repetitive transformation from the optical to the electrical domains and back is obviously clumsy and inefficient. Fortunately active fibre devices also exist and the range of these is steadily increasing thus reducing the need for electrical intervention in optical systems. A new technology of integrated fibre circuits is emerging.

The photochromic protein bR is receiving much attention as an interesting alternative to conventional inorganic semiconductor based materials for a variety of applications in optical recording and information processing. Recent advances in genetic engineering techniques for tailoring molecular properties without degrading of its inherent mechanical and thermal stability make it an outstanding photonic material. Some of the photonic applications already demonstrated by our group are reviewed. We exploit the parallel processing of optics and the real-time nature of bR films for the implementation of several photonic applications. We reported a convenient method to obtain all-optical light modulation in bacteriorhodopsin films using a degenerate four-wave mixing geometry. Chemically stabilized films of bacteriorhodopsin in a polymer matrix for which the life time of the excited M state is tens of seconds are used to demonstrate all-optical light intensity modulation. Small intensities of order mW/cm2 are used in the modulation experiments. The fast photochemical transition from M to B permits reasonably fast modulation speeds independent of the slow thermal M to B relaxation time. The experimental system can also be viewed as an all-optical switch, that uses molecular states in a bacteriorhodopsin thin film, where a low power blue pulse turns on a signal red beam. All-optical logic gates are implemented with wild- type and chemically stabilized films of bacteriorhodopsin using a two-color backward degenerate four- wave mixing geometry. The sensitivity of each bR film is markedly different due to the large differences in the saturation intensity. Red light is used to form a grating due to the B to M transition and blue light is used to form a grating due to the fast photochemical transition from M to B. Each of the two wavelengths in the experimental system acts as an input to the all-optical gate and the phase conjugate signal beam bears the output of the gate. To establish the origin of the nonlinearity we studied the intensity dependence of the self-focusing and self-defocusing properties of wild-type bR in water solution using the Z-scan technique with low power cw lasers at two wavelengths on either side of the absorption band. Our measurements indicate that the sign of the nonlinearity depends on the wavelength and the magnitude depends on the fluence of the incident laser beam. The observed self-defocusing and focusing is not due to the intrinsic electronic nonlinearity. The observations can be explained in terms of the Kramers-Kronig dispersion relation that relates the real and imaginary parts of the complex index of refraction to the absorption spectrum.We developed an optical pattern recognition system using Optical Fourier Transform with bR films. The photoinduced dichroism and its dependence on light intensity allows one to use the bR film as a real-time self-adaptive spatial filter at the FT plane. We also demonstrate flow disturbance visualization and beam shaping.

Single-mode integrated-optic directional couplers on electro-optic substrates can be used to realize photonic space-division multiplexers/switches with low cross-talk level and switching voltage. This paper describes briefly the methods which have been used by the present authors to design 2x2 and 4x4 switch matrices oir LiNbO3 substrates. A computer-aided analytical model has been developed to obtain the two-dimensional distribution of Ti-concentration and refractive index profiles for arbitrary waveguide geometry, Ti-film thickness and processing conditions. The matrix method which is a simple analytical tool for layered structures has been employed to study the wave propagation through Ti indiffused LiNb03 waveguides and directional couplers. The data for refractive index profile required as the input for running the computer program implementing the matrix method were obtained from the above computer-aided analysis. The computed results of Ti-concentration, refractive index profiles and the critical coupling length of directional couplers were validated with available experimental data. Due attention has also been paid to reduce the radiation loss from waveguide bends and to incorporate the polarization dependence in the design of directional couplers. The electrode structure and its placement on Ti:LiNb03 waveguides have been optimized for a 5V switching voltage. The effects of buffer layer and I/O fiber-to-waveguide coupling have also been considered to estimate the total loss. Necessary data generation in appropriate format for electron beam masks for Ti-waveguides and electrode patterns has been completed, and the fabrication of masks is now in progress. A simplistic design of 8x8 switch matrix has also been included.

While quasi-phase-matching (QPM) has led to remarkable results in certain devices, major problems remain for accomplishing it's fiill promise. In this paper we review the current results and identify remaining problems to be overcome.

Since several years, more and more work about non linear optics is performed in various laboratories. In addition, commercial products begin to appear in various forms: frequency doublers or triplers, multiple frequency mixers and optical parametric oscillators.

This paper reports on theoretical and experimental study of linear and nonlinear optical effects of the silica fibers on high-speed and high-capacity lightwave communications. High-speed all-optical fiber switches, all-optical wavelength and signal conversions, optical clock recovery, mode-locked fiber lasers, supercontinuum pulse generation, and all-optical packet signal detection devices, have been experimentally demonstrated by effectively compensating and utilizing the linear and nonlinear optical effects in optical fibers for high-speed TDM and high-density WDM all-optical communication systems. In addition, theoretical analyses have been performed for the optical chromatic dispersion and nonlinear effects affecting such large capacity optical communication systems provides us a clue on determination of the critical channel power and of the maximum FDM channel numbers limited by the four-wave mixing effect for various lengths of dispersion-shifted fibers, and the maximum TDM bit rate and channel power limited by the chromatic dispersion, cross-phase modulation and stimulated Raman scattering effects for various lengths of normal single-mode fibers.

Upconversion fiber lasers appear as promising candidates of infrared diode pumpable sources of blue and green radiation at moderate output power levels ( ~50 - 200 mW). Despite significant technological advances made by direct doubled-diode sources and doubled diode- pumped solid state lasers, recent improvements in the reliability and cost of fluoride fibers have made upconversion fiber lasers competitive as compact blue and green sources, particularly with regard to choices of specific wavelengths and power levels needed for numerous applications such as full color displays, high performance imaging and printing (e.g. for digital radiography), high capacity optical storage, satellite-to-satellite communications, semiconductor wafer inspection systems, and biotechnology (specifically for flow cytometry and phase fluorometry systems). In this paper, we will review the work on the development of blue and green upconversion fiber lasers that is being pursued at the University of New Mexico.

Some nontrivial features of nonlinear transformation of frequency in the ring fiber resonator of free running YAG:Nd laser have been experimentally studied. Fiber of resonator has been matched with the laser active medium using microobjectives. Fibers of different lengths with the core diameters from 340 to 10 µm have been used. Nonlinear frequency transformation has been observed for fibers with core diameter of 60 pm and investigated lenths of resonator 43, 100 and 226 m. For all these fiber resonators in the antiStokes region spectral lines of 505, 532 and 558 nm have been presented. In the region of the Stokes component generation of the stimulated Raman scattering the row of narrow (with width less then 0.2 nm) spectral lines in the spectral region of 1127. . .1131 nm has been observed. The amount of the lines has been increased from 1 to 12 with increasing of the laser pumping power. In the anti-Stokes region for the long fibers (100 m and 226 m) the third harmonic of the main laser frequency additionally has been generated. Here it is also observed the two groups of lines in the spectral region of 1129 nm and 1137 nm, the amount of which depends from the pumping power of laser.In the same fibers under the excitation by the external YAG:Nd laser with the same power of radiation the spectral lines in the anti-Stokes region have not been observed and in the Stokes region the cascade generation of the stimulated Raman scattering has been taken place. The connection of the observed features with the spectral selective properties of the fiber ring resonator is discussed.

The observation of Second Harmonic Generation (SHG), in centrosymmetric silica optical fibres stimulated great theoretical and experimental interest, in the study of this phenomenon with potential applications. SHG could be observed mainly in Ge doped or Ge-P doped silica fibres, and it has been established that some doping is essential for SHG in optical fibres. Several physical models have been proposed to explain this phenomenon, and almost all the models explain the origin of this x(2) effect through the x(3) effect in silica fibres. In this paper, we present a coupled mode analysis to study SHG in optical fibres. Both ‘seeded’ and ‘unseeded’ SHG have been analysed, taking into consideration the transverse modal distribution in the fibre. Details on our experimental efforts to observe SHG in Ge doped optical fibres are also presented.

A new method for obtaining mode cutoffs for asymmetric planar waveguides is developed. The method is based on defining the wave equation for modes at cutoff and solving it using the Galerkin method. The method has been used to obtain cutoffs of a variety of asymmetric waveguides with arbitrary profiles including exponential, traingular, parabolic and other power law profiles. Comparisons with available results show that very good accuracies can be obtained with moderate matrix sizes.

It is well known that the information capacity of a lightwave communication system is ultimately limited by the nonlinear interactions between the information signals and the fiber medium. These optical non-linear interactions can lead to interference, distortion, and attenuation of the signals, resulting in system degradation. There are four types of non-linear interactions in optical fibers. (1) stimulated Raman scattering (SRS); (2) stimulated Brillouin scattering (SBS); (3) self-and cross phase modulation (SPM and CPM), and (4) four-photon mixing (FPM). These nonlinear effects degrade both the single channel and multichannel systems in different ways. The SBS and FPM effects depend on the signal modulation format and detection schemes (direct or coherent) respectively and whereas the SPM and CPM effects are of concern in angle modulated systems. The effects of SRS, CPM, and FPM induce interchannel crosstalk in wavelength division multiplexing (WDM) systems and affects the system capacity. In this paper, the effects of nonlinearities in lightwave communication systems are reviewed with the emphasis on recent developments. Additionally, techniques to minimize the effects of nonlinearities are discussed.

Using the coupled mode approach we present theoretical modelling of quasi phase-matched parametric amplification in a planar waveguide in which, the pump and signal are guided while the idler is radiated into the substrate. Numerical results on optimum waveguide parameters (which provide maximum signal gain) for a given pump and signal wavelengths are presented for a periodically domain reversed, proton-exchanged LiNbO3 planar waveguide. The proposed configuration is shown to provide large signal gain bandwidth.

Optical fibre sensors have enormous potential as an important technology in structural monitoring. Their particular advantages include the ability to make distributed measurements, immunity from electromagnetic noise and pick-up and the capability of operating over large areas and/or at interrogation distances of many kilometres. This paper will, first of all, examine the requirements for structural monitoring in both civil engineering and composite structures for aerospace and underground transportation. The benefits and disadvantages of fibre optic techniques will then be explored and the general features of important contributory technologies including fibre Bragg gratings, microbend sensing, interferometric and sub-carrier sensors and non-linear optical techniques, e.g. stimulated Brillouin scatter will be highlighted and their applications and properties described. The paper then examines some specific applications in each of these two principal sectors and concludes by speculating on future prospects for fibre optic sensing and structural assessment.

Numerous optical fibre sensing techniques have been evaluated to monitor composite material structures. Most of these sensor systems aim to measure responses to static loads or vibrational spectra. An alternative approach is to monitor the ultrasonic signature over structure and evaluate changes in this signature in terms of modifications to structural properties. This paper describes the principles and implementation of such a system. A simple line integrating interferometric optical fibre sensor is used to detect the propagation of Lamb waves at typically 250kHz within the composite material. The wavelength of these Lamb waves (typically 2cm) exceeds the dimensions of structural defects of interest to the optical fibre sensor is, in effect, detecting changes in the ultrasonic scattering signature of the test structure and relating these changes to the evolution of faults or damage within the structure. The paper will present preliminary results which explore the basic sensitivity mechanisms within the optical fibre sensor and examine the modifications to the ultrasonic signature with the introduction of predetermined damage and deterioration. The eventual aim is a fully integrated system which will enable “plug in” structural testing. These preliminary results indicate that such systems concepts are a medium term prospect.

Biomedical fiber optic sensors are attractive for the measurement of physical chemical and biochemical parameters and for spectral measurements directly performed on the patient. An overview of fiber optic sensors for ih-vivo monitoring is given, with particular attention to the advantages that these sensors are able to offer in different application fields such as cardiovascular and intensive care, angiology, gastroenterology, ophthalmology, oncology, neurology, dermatology and dentistry.

Photoelasticity is one of the oldest methods for experimental stress analysis, but has been overshadowed by the Finite Element Method for engineering applications over the past two/three decades. However, certain new and novel developments and applications have revived the use of photoelasticity. Among them are image processing for fringe analysis, poiarimetric fiber optic sensors, infra-red photoelasticity, low cost dynamic photoelasticity and photoelastic applications in stereolithography. This paper will, after a brief introduction to photoelasticity, highlight some of these advances and developments.

In this paper the behaviour of fiber Bragg gratings as temperature and strain sensors is presented. The formation, using a photolitographic method, and characterization of those narrow-band filter are also described. The filter response, tunable in frequency due to mechanical strain and temperature change, has shown good linearity over a wide range of both strain and temperature variations. Arrays of up to five gratings, with a peak separation of about 1 nm, have been written using the same phase mask.

A high-temperature, high sensitive borehole seismometer with fibre optic interferometric displacement sensors has been constructed and practically tested Its parameters are suited for monitoring of seismic vibrations at working temperatures from 20°C to 300°C. Long term stability of performance characteristics has been confirmed for the seismic receiver and the fibre optic components at elevated temperatures and for several km of fibre optic signal transmission lines. Seismic vibrations could be monitored by coupling the seismic mass via spherical reflectors to a single-mode fibre optic Michelson interferometer. Optical phase processing based on wavelength modulation of tuneable DFB laser source yields <0. 1nm sensitivity of seismic vibrations within 0.1. .30Hz information bandwidth. Laboratory test measurements at elevated temperatures agree well with the results of an electrodynamic seismometer.

An open loop interferometric Fiber Optic Gyro (IFOG) has been designed,developed,tested and demonstrated at LEOS.This Gyro is of all fiber construction,using polarisation maintaining components, an integrated optic phase modulator and a superluminescent diode as light source. The interferometer uses 400m long polarisation maintaining fiber wound in Quadruple configuration on a 100 mm diameter Aluminium former. The coil and former weight 400 gms. An inhouse built processing electronics, consisting of a Lock-in-Amplifier, superluminescent diode driver with photodiode feedback, a Thermoelectric cooler controller for wavelength stabilization, a pulse suppression logic and regulated power supplies have been used for meeting signal processing requirements. The total power consumption of Gyro is 5 watts. Weight (including all electronics and sensor) is 1.5 Kg. Gyro drift was measured to be 0.07 deg./hr. and random walk 0.006hr. Scale factor stability was 200 ppm.

A fiber optic infrared (IR) spectroscopic system to be used with the cone penetrometer has been developed. This system can be used to perform real time, in-situ site characterization and analysis by identifying and quantifying organic contaminants in soil.

In recent years, a significant amount of clinical research has been devoted to establish relationship between variations in fetal heart rate and abnormal prenatal conditions. For example, it has been shown that the progression of fetal asphyxia is closely linked to recognizable changes in fetal heart rate patterns. Currently, fetal heart rate monitoring is done using either an external ultrasound device, or internal fetal electrocardiogram taken from a scalp electrode placed directly on top of the fetus’ head. Both of these techniques are considered to be invasive, and are therefore not acceptable in situations where long term monitoring of the fetal heart rate is required. Further, the sensors used in these techniques are susceptible to electrical noise and interference. Many noninvasive acoustic sensors lack the sensitivity and frequency response for operation at peak pressures of approximately 50 Pascals between 10 and 60 Hz range suitable for fetal heart rate monitoring. We have developed a fiber optic coupler sensor which is passive, and therefore totally non-invasive. A 2x2 biconical coupler is made by fusion in a microfumace and mounted on an elastic membrane. The light input to the coupler produces two outputs which depend on the coupling coefficient at the fused region. The sensor outputs when it is placed at the chest and the neck of an adult are shown as examples. The sensor output closely resembles a normal electrocardiogram. The sensor is shown to be useful to detect a fetus’ heart-beat; however, a proper filtering technique is needed to eliminate acoustic background produced by other fetal activities such as breathing and movement.

The polarization sensors on the singlemode optical fibers are investigated. The segment of the optical fiber clamped between two cylindrical or plane surfaces rotates the plane of polarization of the linear polarized light on the output of the fiber. The angle of rotation is proportional to the strength of clamping of the surfaces. This phenomenon can be used for the creation fiber-optical sensors of pressure or temperature.

A statistical analysis of sensitivity of integrated-optical sensors whose principle of operation is based on the registration of reflected beam parameters in the prism coupling scheme is carried out. Measurements of ammonia concentration in atmosphere are performed.

An ideal fiber optic rotation rate sensor is shot noise limited. In present day technology, the maximum power that can be received at the detector end of the sensor is of the order of 1 to 10 µw. This corresponds to a minimum detectable rotation rate of 0.1 deg/hr to 0.035 deg/hr for a sensing coil length of 500 meters and radius 10 cm, operated at 1.3 µm wavelength. However a practical rotation sensor is limited by thermal noise of the load resistance, back scattering noise in optical fiber, polarization induced noise, source intensity noise and noise in processing electronics. These noise sources cause long term offset at sensor output resulting in drift. We have analyzed and estimated the various noise sources with an objective to achieve optimum performance of a fiber optic rotation sensor. The noise equivalent rotation rate due to various sources has been calculated for different values of optical power, wavelength of operation, length of the fiber and detection bandwidth. Polarization filtering, coherence length of the optical sources, phase modulator amplitude and frequency stability were analyzed to yield optimal system performance. Other effects which cause drift in rotation sensor such as Faraday effect and Kerr effect have also been considered.

The design criteria of an optical-fiber probe for Photon Correlation Spectroscopy are given. A merit factor is defined to evaluate the measurement efficiency, and a miniaturized probe comprising a circular array of optical fibers coupled to graded-index microlens is presented. The probe demonstrated to fit the most stringent optogeometric requirements, and has been used for controlling the production process of polystyrene particles.

The design and architecture of bi-directional optical backplanes with single and multiple bus lines for high performance bus are reviewed and bus systems with one and 2-bus lines at a wavelength of 850 nm are experimentally demonstrated. For optical backplanes with single-bus line, a speed of 1.2 Gbit/sec at a wavelength of 1,3 µm is demonstrated. Arrays of multiplexed holograms are employed, in conjunction with a waveguiding plate within which cascaded fanouts are generated. For the backplane bus system containing multiple bus lines and 9 processor/memory boards, VCSELs (Vertical Cavity Surface Emitting Lasers) and photodetector arrays, such parameters as power budget, misalignment and packaging related issues are discussed. With the introduction of GRIN lenses into the backplane system, it is found that not only can the signal beam from the VCSELs get collimated, but die angular tolerance of the system is greatly enhanced. The optical backplane bus system developed here is transparent to higher level bus protocols, thus can support standard backplane buses such as Futurebus+, Multi-bus II, and VMEbus.

A novel optical computing technique using recoded quaternary signed-digit arithmetic is presented in this paper. An efficient shared content-addressable memory based optical implementation which employs a fixed number of minterms independent of the operand length is developed. This technique requires the minimum number of minterms when compared to the similar previously reported techniques. An alternate algorithm for quaternary signed-digit arithmetic is .also suggested.

The main difficulty in the realization of intelligent processing and computing techniques lies with the requirements of enourmous amount of machine computations and therefore must operate in a parallel processing environment. In this context interference-free parallel processing capability of optics technology can be efficiently exploited. Optical computing system can interconnect and operate on two dimensional input-output data structures and can store data in three dimension. Therefore the implementation of neural network models in optical domain is likely to be more effifient. The optical neural computing techniques with facilities of intelligent processing have become versatile tools for pattern identification, classification and recognition. The paper introduces a representative optical neural computing system developed for intelligent pattern association tasks. The system can associate patterns from noisy and incomplete informations. The space invariant properties can also be introduced during identification of patterns.

A fractional power filter based joint transform correlator (JTC) for multiple target detection is presented and analyzed. The proposed technique can be used for real time multiple target detection using classical matched filtering, inverse filtering and/or phase-only filtering. An efficient JTC architecture is developed for optical implementation of the proposed technique and simulation results axe presented to validate the performance of this technique

Basic concepts and preliminary applications in the generation, propagation, and detection of terahertz radiation with femtosecond optics are discussed. Recent progress in ultrafast free-space THz optoelectronics is reviewed.

A bias coding method is proposed for cellular neural networks which facilitates the representation of analog values into unipolar format. These bias-coded values are spatially arranged on the transparencies using a designed spatial coding scheme. A new compact and efficient optoelectronic architecture for cellular neural networks is also reported which is based on the proposed bias coding method and incoherent correlation.

We consider the following problem. In a wavelength-routing optical network, given a specified logical topology to be embedded on a physical topology (PT) how do we choose the routes on the physical topology so as to minimize the number of distinct wavelengths used in the construction of the logical topology (LT) . The construction of LT may be with or without wavelength changers; both the cases are considered. We formulate an Integer Linear Program (ILP), which is identical to a multicommodity flow problem, to solve the LT construction problem with wavelength changers (WC), the objective being to minimize the distinct wavelengths used on any link of the PT. We note that this is equivalent to solving for minimizing the maximum congestion on any link in a multicommodity flow problem .The solutions of the ILP give an exact solution to the problem with WC. On relaxing the Integer constraints in the above program we get a lower bound on the number of wavelengths required with or without WC. Computational results are presented for the NSFNET , EONNET, ARPANET and randomly generated physical topologies for a complete logical topolgy i,e the logical topology to be embedded is a (N-l) regular directed graph, where the N denotes the number of sources/sinks of traffic. The solutions of the LP relaxation of the above are rounded to give a physical route for every source destination pair under consideration . Given the set of paths we formulate the wavelenght assignment problem, that arise when no WC are used as a graph (vertex) colouring problem and some heuristic algorithms are given to solve for the colouring. We also formulate an ILP to solve the colouring problem once the set of paths are given on which we have route to achieve the desired construction of logical topology.

In earth-satellite-earth laser communication links,laser beam passes through the earth s atmosphere which occasionally contains clouds. The important distortion effects imposed by clouds in the atmosphere are temporal broadening of the signal and attenuation due to multiple scattering. Multiple scattering problem arises when light propagates through an optically thick clouds hence the photons encounter a number of scatterers before emerging from the medium. In the present work, Monte Carlo simulation is used to show the trajectory of photons (rays) in the medium and interaction between the particles and laser beam.Results show that the output beam is spread over time and space depending on the random nature of the medium.

We describe a simple scheme of optical interconnections between M input and N output terminals based on holographic code-division multiple-access (CDMA). Here an MxN multimode fiber coupler is used for the signal beam paths and its mode speckle patterns are for interconnection destination codes. Fully parallel nonblocking interconnections are accomplished by launching the signal beams into the input terminals properly and then by detecting the generated mode speckle patterns separately using M multiplexed holographic filters at each output terminal. We report basic experimental results to show the feasibility of our scheme, in which for simple demonstration the beam launching angles are controlled at the input terminal to generate the different output mode speckle patterns.

In Germany the Deutsche Telekom is realizing worldwide the largest installation of FITL systems under the programme OPAL (Optical Access Line) to build up a modem infrastructure in the eastern part of the country. At the end of 1997 about 1,5 million access lines will be based on fibre systems, mainly realized with FTTC solutions. Whilst in the first period of the installation PON structures have been favoured later on active networks have been implemented which showed much better economics and higher flexibility. The experiences with these systems are used to modernize also the existing telephone network in die western part in combination with a full digitization of all local exchanges. With the FITL systems the switching equipment can be concentrated on fewer locations leading to lower operation and maintenance costs. Deutsche Telekom operates besides the telephone network also worldwide the largest CATV network with about 25 million households passed. Currently work is going on to look for network upgrading techniques which allow the introduction of new broadband services like video-on-demand and broadband online services. FTTH is for all major network operators a long term target, however, due to the high digging costs and the still too high system prices FTTH is not yet economic and mainly the replacement of the first few hundred meters from the customers to the network node is much too costly. Fibre to the cabinet and new techniques like VDSL on copper lines or 64QAM on coax cables in connection with digital video transmission can be the next step to bring the fibre nearer to the home. Scenarios for these upgrading techniques will be presented and compared.

Narrow channel wavelength division multiplexing (WDM) is an attractive technique for increasing the capacity of optical fibre transmission systems. An essential component for WDM systems is a wavelength demultiplexer capable of providing high levels of isolation between adjacent channels. The control of the device performance with regard to low levels of crosstalk between adjacent channels, operating wavelength, polarisation sensitivity and total device loss is essential. These device characteristics are influenced by material properties and fabrication techniques. Silica based planar technology is a key technology for passive optical components, especially dense wavelength demultiplexers, as required for the next generation of fibre optics communications systems. Fabrication of planar waveguide wavelength demultiplexers has been reported by several workers using different oxide deposition techniques such as flame hydrolysis, plasma enhanced chemical vapour deposition (PECVD) and low pressure chemical vapour deposition (LPCVD). In this paper, high throughput deposition techniques for fabricating various planar waveguide devices, potentially required in WDM systems, will be discussed. Multichannel wavelength demultiplexers with high levels of channel isolation (> 30 dB), will be used as an example for comparing various fabrication technologies. The control and reproducibility of optical and physical properties of the deposited oxide films is crucial for fabricating high performance devices with low insertion loss and high yield. All the above mentioned deposition techniques have been reported to be capable of producing low loss waveguide materials (< 0.05 dBcm-1). The device yield for high performance devices, however, will depend on the ability of any particular technique to control the refractive index uniformity, refractive index homogeneity across the film thickness, waveguide dimensions and shape, and uniformity of film composition and refractive index in the narrow gaps between channel waveguides. These factors need to be controlled across a wafer and from wafer to wafer. This paper discusses the above mentioned requirements and their effect on device performance, focusing particularly on work carried out at Nortel.

A high frequency ac power distribution systems for the drop segment of a hybrid fiber/coax architecture is presented. It is shown in the paper that the high frequency distribution system provides self current limiting against a short circuit fault and no voltage glitch across the other converters. A prototype system which distributes 30 Vrms at 128 kHz over coax cable is built and experimental results are given which support the high frequency distribution as an attractive and viable option.

In this paper the realization of high-speed computer networks using incoherent optical CDMA with coding in time- domain is investigated, which allow cost effective realization with optical standard components. Based on a feasible transceiver module we examine the performance taking into account different parameters and discuss methods to improve the performance.

A new Interconnection Multihop Optical Network called EXSTAR was conceived in order to allow an implementation of all- optical routing processor and, at the same time, to improve the characteristics of modularity and performance of other known multihop structures. For this network some routing algorithms, named EXSTRA (Extended STarring Routing Algorithms), based on a header recognition scheme , have been devised.The first one allows to obtain the best performance in conjunction with a suitable network structure; a simplified version offers a notable simplicity of implementation, giving rise to routing processors including only few all-optical gates.

The development of optical amplifiers for signal transmission over transoceanic distances has opened up a new horizon in photonics towards the practical realisation of extremely fast and ultra long haul and wide band transmission systems, particularly so, when the high performance coherent detection systems are used. In this paper, the performance behaviour of coherent optical binary and Multilevel FSK signalling schemes employing optical amplifiers is analysed by accounting for the laser phase noise of the transmitter and local oscillator lasers in addition to the noise of the receiver and the multiple amplifiers. The results are compared with the case when the phase noise is assumed to be absent. Further, the performance of PSK systems are compared with those of their PSK counterparts.

A new approach of the link-level modeling and simulation of fiber-optic communication systems is reported and its benefits are illustrated. The model is used to simulate a cascaded optical-amplifier based long-distance fiber-optic transmission system. Through the simulation the waveform distortions at different stages of the communication path and the pulse-shape improvement by mutual cancellation of chromatic dispersion and SPM-efifect are investigated. Our result shows that significant performance improvement is achievable in a practical dispersion-limited system when the system power is tuned to a value at which the SPM-efifect optimally cancels the anomalous dispersion effect.

Optical multi-wavelength transport networks (MWTN) have generated considerable recent interests because of their inherent capability to achieve higher information capacity, greater flexibility, efficient routing, transparent switching, reconfugarability etc.. The effect of switch cross-talk, adjacent channel cross-talk, filtering effect, accumulated ASE in MWTN has been recently addressed. However, the performance of closely packed WDM systems with dispersion shifted fibers is highly vulnerable to the effect of FWM. The present study investigates the degradation encountered in multi-wavelength switched optical network due to the combined influence of receiver noise, phase noise, accumulated amplified spontaneous emission (ASE), FWM crosstalk and dispersion. For each channel CPFSK modulation is considered with direct detection receiver employing Mach-Zehnder interferometer (MZI). The analysis is carried out to include the combined effect of accumulated optical amplifiers’ spontaneous emission (ASE) noise and the beat noise components viz. signal-spontaneous emission beat noise, ASE-ASE beat noise, adjacent channel-spontaneous beat noise, signal- FWM beat noise, FWM-spontaneous beat noise etc.. An expression for the probability density function (pdf) of the random phase fluctuation due to FWM effect is developed and the bit error rate is estimated at each node for different number of channels. In the present analysis, a node architecture of the optical mesh network is considered in which an interconnection between two consecutive nodes consists of optical in-line amplifiers, optical space-switches, power splitters, wavelength multiplexers and demultiplexers and the fiber protection switch. All the wavelengths are combined in the multiplexer and amplified before feeding to the transmission fiber. The gains of the in-line amplifiers are adjusted to compensate for the losses in the fiber, splitters and other passive components. The demultiplexer separates the desired signal wavelength from the WDM multiplex and is realized by a Febry-Perot filter (FPF) tuned to the desired signal channel. The node uses M transmitters and M receivers if M is the number of wavelength channels. Computations of results are carried out at a bit rate of 2.5 Gb/s for several fiber spans with different sets of system parameters, viz. number of channels, number of nodes, optical SPIE Vol. 3211 • 0277-786X/97/$10.00? wavelength from the WDM multiplex and is realized by a Febry-Perot filter (FPF) tuned to the desired signal channel. The node uses N transmitters and N receivers if N is the number of wavelength channels

The proposed development of fibre-optic communication links at bit rates exceeding 10Gb/s places a demand for photoreceiveis with high bandwidth (BW). Metal-Semiconductor-Metal (MSM) photodetectors (PDs) show very high intrinsic bandwidth (500GHz), are easily integrable with High Electron Mobility Transistors (HEMTs) and HeteroBipolar Transistors (HBTs) and are very attractive for communication and chip-to-chip interconnections. Widespread interest is being shown to these photoreceivers and to optimize their performance. In this paper we have calculated the gain, BW and minimum detectable optical power of a combination of MSM-PD and a HEMT and of a MSM-PD and a HBT and made a comparison. The InGaAs-InP heterojunctions and quantum wells are in the heart of file devices. The intrinsic capacitance of MSM-PD is made very low (6fF). The HEMT or HBT is used in common source (emitter) and cascode configurations. The minimum detectable power is calculated for direct intensity modulation with NRZ coding. Typical values ofPmin for MSM-PD and HEMT combination are -35dBm for 0. 1Gb/s, -30dBm for lGb/s and - 25dBm for 10 Gb/s. Due to less gate current noise this combination gives a better performance than a MSM-PD and HBT combination. However, with inclusion of proper model of gate current noise, there is some degradation of performance. Some possible means to optimize the performance are discussed in this paper.

With the advent of ultra high speed optical communication systems, it has become important to investigate the bit delay associtaed with each channel of a Byte-Wide WDM system. In this paper, the relative bit-delay caused by different transmission wavelengths in a Byte-wide WDM optical system have been analysed and estimated for low loss transmission windows (1.3 &1.5 µm). Byte-rate limitation imposed upon WDM system has been investigated. Further, Scheme for duplex Byte-wide WDM system for an Optical Line Terminal Equipment (OLTE) has been proposed.

PHOTON, a project within the ACTS programme of the European Community, aims at substantiating the concept of a pan-European all-optical transport overlay network by the development of key elements and the implementation of then- key functionalities in a field trial. The theoretical work deals with further refinement of network concepts - like network architecture and network evolution - by studying link and network behaviour using appropriate modelling and simulation tools. The PHOTON field trial will implement a border crossing star network with transmission and cross-connection of optical frequency division multiplexed (OFDM) signals by connecting Munich with Vienna via Passau over already installed monomode fibre, spanning a distance of 524 km. Eight channels in the 1.55 µm window, separated by 400 GHz, for 10 Gbit/s and 2.5 Gbit/s SDH transmission and a leased optical channel service are available without demanding for full optical regeneration. Some key components like suitable OFDM filters, lasers with specific output frequencies, and a Krypton frequency reference are developed within the project. In a parallel project, called MOON, the PHOTON field trial will be used as a basis for the development of management concepts.

The design of the optical link for the downstream (from a head-end or a central office to a fiber node near the customer premises) and upstream (from the customer premises to the head-end or central office) transmissions in a hybrid fiber-coax access system is reviewed in detail. Particularly, the end-to-end bit-error performance of 64-QAM transmission using a hybrid fiber-coax system test bed designed for delivery of 77 AM-VSB and 25 64-QAM video channels covering the frequency band 54 - 702 MHz was studied. Error-free QAM transmission was achieved with the use of modulated analog carriers and Reed-Solomon (204,188) error-correction coding indicating that in hybrid fiber-coax systems transport of AM- VSB and 64-QAM signals can be realized maintaining both analog and digital performance requirements. Results are also presented on the design and choice of optical transmitters for the upstream path. The role of laser clipping in setting system signal-to-noise and dynamic range for error-free QPSK transmission in the upstream direction is described.

In this paper, the importance of wavelength converter modules in all-optical networks is highlighted. Two types of wavelength converters are proposed, static and dynamic wavelength converters. A wavelength converter (also referred to as Frequency Transformer) converts the wavelength of the incoming optical signal to another wavelength in the wavelength spectrum. The goal of this work is to provide some insights into the effect of static versus dynamic wavelength conversion on the performance of optical networks based on Bidirectional de Bruijn (BdB) topology. We study how performance (in terms of blocking probability and the network load) depend on the number of wavelengths per optical link. We also evaluate performance of a given network employing both static and dynamic wavelength converters through simulation and present the simulation results. The motivation behind having static wavelength converters is the relatively low cost compared to dynamic wavelength converters. Also an Error-Correcting Routing Algorithm (ECRA) is proposed that makes use of static wavelength converters which is based on the principle of error-correcting routing.

In this paper we present a novel optical fiber multiple access system based on the combination of Coherence Multiplexing (CM) and Spread Spectrum techniques. By exploiting the coherence properties of the source, a reference signal is transmitted along with an encoded signal to achieve self-homodyne detection, so bipolar codes can be used. At the receiver, the desired signal can be recovered using the same code as the transmitter. This is the CDMA principle. This scheme offers security in communication and is suggested for network applications where security is of interest. Due to combination of the two multiplexing schemes, i.e. CM and CDMA the network can support a larger number of subscribers than each multiplexing scheme individually. This system is described and demonstrated experimentally.

The measurement of small differences in density of liquids has found increasing use in the monitoring of concentration of liquid mixtures. The concentration can be measured by its effect on the refractive index. This paper reports a simple and yet sensitive technique for the measurement of the change in refractive index of liquid mixtures by using the speckle generated in a multimode optical fiber. The results show that changes in refractive index as small as 0. 0001 can be measured.

Transillumination is a diagnostic technique based on the differences on the diffused transmittance in the visible or near- infrared radiations. A system has been developed by combining computed tomography and transillumination to obtain cross- sectional images of human breast models. The light from a 15 mW He-Ne laser ( 632.8 nm) is transmitted through an optical fibre of 1 mm core diameter to a cylindrical lens assembly to obtain a fan beam configuration to illuminate the model. The scanning of the model is carried out by three stepper motor arrangement which carries upward-downward, rotational and pitch movements. A curved array 16 photodiodes measure the transmitted light through the model. At each position along the height 50 projections are taken. Data acquisition and processing are carried out by PC-AT 386 equipped with ADC card. Based on this a tomogram is constructed. The experiments are carried out with foam models with and without embedded abnormalities. By reconstructing the tomograms at various heights, the size and shape of the embedded abnormality in the model are determined.

Optical time domain reflectometry (OTDR) is a versatile technique that enables loss measurement and fault location. Such an information is essential in the design of a good fiber optic communication system. A high resolution OTDR is also necessary in distributed fiber optic sensor systems, where the distance between sensors is limited by the achievable OTDR resolution. Improved resolution in OTDR is generally realized using pulses of shorter width (typically in the picosecond range). However, such systems are more expensive. In this paper, we present a novel signal processing technique based on inverse spiking filter to improve the resolution of an OTDR system. In addition, we present experimental verifications of the resolution enhancement achieved for a nanosecond pulse with an OTDR system configured by us. The minimum length of the second fiber (butt coupled to a parent fiber of length about half a kilometer) that could be resolved using a deconvolution technique was 1.5 metres. This method has great potential not only for detecting splice locations accurately but also for distributed sensor configurations.