The angled grating laser has been successful in achieving high-power diffraction-limited beams. The laser cavity utilizes a grating inclined at an angle to the facet to filter out the filamented beams. As a result, all of the filamented beams except the primary beam will be transmitted through the grating. In this paper we present the cavity resonance equation, and study temperature sensitivity of this device.

A tunable laser operating near 944 nm is important for micro LIDAR systems. The high power laser would consist of a tunable (and possibly Q-switched) low power fiber laser, the output of which would be amplified using a double clad fiber amplifier. A model of a Nd doped laser using a Gaussian beam approximation for both the propagating pump and the laser mode has been developed. The threshold power and output power for various parameters such as pump absorption and reflectivity have been calculated. The spectral width (linewidth) of a fiber laser under CW operation has been calculated. The laser linewidth varies inversely as the square of the fiber length and inversely as the output power. Shorter pulse width and higher peak power is obtained using a small fiber core under Q-switched operation.

Modern trends in design of semiconductor lasers are addressed. Nanoscale coherent inclusions of narrower bandgap semiconductor in a wider gap semiconductor matrix, or quantum dots (QDs), are designed as a new type of active medium of injection lasers. QDs give a possibility to extend the wavelength range of heterostructure lasers on GaAs substrates to 1.3 micrometers and beyond and improve their device performance. 330 mW CW 1.3 micrometers single mode continuous wave (CW) edge emitters and 1.2 mW CW vertical-cavity surface- emitting lasers (VCSELs) are realized. Long operation lifetimes and other competitive device parameters are demonstrated. Novel device designs are proposed. In one concept high-order mode filtering in structures with periodically modulated refractive index containing a defect allows realization of stable narrow beam divergence fundamental model lasing both in edge-emitters and in VCSELs. In a different novel design, light propagates at some angle with respect to multilayer interference mirrors (MIR), and the MIRs and the cavity are calculated for the tilted photon incidence. Tilted cavity laser (TCL) gives wavelength-stabilized operation in edge and (or) in surface direction and does not require materials having high refractive index difference. New generations of semiconductor optical amplifiers, photodetectors, optical fibers, etc. may become a reality.

We present the experimental demonstration of a new fiber laser configuration based on the nonlinear optical loop mirror with a symmetrical coupler, and highly twisted low- birefringent fiber in the loop. The nonlinear optical loop mirror configuration operates by nonlinear polarization rotation.

We report the measurement of electric field correlations of a hybridly modelocked external linear cavity semiconductor laser as a function pulse delay. We also report the measurement of residual phase noise corner frequency and longitudinal mode linewidth as a function of laser cavity length. We find that the pulses in the modelocked pulsetrain are correlated at only multiples of the cavity roundtrip time. Excellent agreement between residual phase noise corner frequency and longitudinal mode linewidth measurements suggest that the corner frequency is the average longitudinal mode linewidth. This relationship leads to a fundamental limit in the timing jitter of modelocked lasers.

Pulses with approximately 200 fs pulse width were generated by using a two stage pulse compression mechanism of gain- switched DFB laser pulses which were originally approximately 10 ps long. The pulse compression mechanism utilizes fiber nonlinearities and it involves propagation through specific lengths of various types of fiber and a nonlinear loop mirror. Simulation of the whole compression scheme by solving the nonlinear Schrodinger equation using the split step method shows good agreement with the experiment results.

Optical processing techniques are expected to play a key role in the next generation of advanced high-speed analog- to-digital converters (ADCs). These techniques will alleviate the current limitations inherent in conventional electronic ADCs. We are currently developing a novel photonic ADC module that incorporates the use of semiconductor saturable absorbers to perform the data quantization at speeds in the tens of GHz regime. Results will be presented for the experimental material characterization of the semiconductor saturable absorbers used in the data conversion process. Enhancement of the contrast ratio of the saturable absorber between the 'on' state and the 'off' state can also be greatly enhanced by the use of an asymmetric Fabry-Perot etalon. Initial experimental results for a saturable absorber contained within an etalon will also be presented.

A unique optical A/D converter topology takes advantage of high-speed optical components while generating signal samples in the electrical domain. The topology uses optical clocks and photo-diode switches to sample the electrical input. The electrical samples are demultiplexed using the same optical components and quantized via a sampling card. The sampling rate is increased to 10 GSPS using a flash architecture.

The stepped conical zoned lens antenna has better overall efficiency than a true lens, and provides an excellent antenna pattern. It also exhibits somewhat different bandwidth characteristics than the Fresnel zone plate antenna. This paper examines the frequency behavior in detail, particularly for microwave and millimeter-wave applications. For the usual zone plate antenna employed at microwave or millimeter wavelengths, path length adjustment (i.e., phase correction) is accomplished by cutting different depths (grooves) in a dielectric plate or by using two or more dielectrics having different dielectric constants. The new design uses a tilted cut in a flat dielectric plate, which more accurately matches the shape of a true lens and produces much lower phase error. The construction is still linear (i.e. spherical or hyperboloidal curves do not have to be cut), and can be made, for example, by a milling machine with a tilted bit. For a circular zone plate, the lens is a stepped conical shape. The phase correction steps are small, usually a few degrees, which is much smaller than for the typical Fresnel zone plate. The bandwidth characteristics are calculated for specific cases.

Industrial, NASA, and DOD spacecraft designers have recognized the advantages of using fiber optic components and networks for their internal satellite data handling needs. Among the benefits are the total elimination of cable-to-cable and box-to-box EMI; significant size, weight and power reduction; greater on-orbit flexibility, simplified integration and test (I&T), and significantly lower I&T costs. Additionally, intra-satellite data rates of 1 to 10 Gbps appear to be an absolute requirement for a number of advanced systems planned for development in the next few years. The only practical way to support these data rates is with fiber optics. Space Photonics and the University of Arkansas have developed fiber optic components (FireFiber^TM) and networks that are designed specifically to meet these on-board, high data rate needs using NASA approved materials, packaging processes, and approved radiation tolerant devices. This paper discusses issues relevant to these components and networks.

Presented here is the second set of testing conducted by the Technology Validation Laboratory for Photonics at NASA Goddard Space Flight Center on the twelve optical fiber ribbon cable with MTP array connector for space flight environments. In the first set of testing the commercial 62.5/125 cable assembly was characterized using space flight parameters (published in SPIE Vol. 3440). The testing showed that the cable assembly would survive a typical space flight mission with the exception of a vacuum environment. Two enhancements were conducted to the existing technology to better suit the vacuum environment as well as the existing optoelectronics and increase the reliability of the assembly during vibration. The MTP assembly characterized here has a 100/140 optical commercial fiber and non outgassing connector and cable components. The characterization for this enhanced fiber optic cable assembly involved vibration, thermal and radiation testing. The data and results of this characterization study are presented which include optical in-situ testing.

I review a few of the most famous quantum algorithms, and discuss them from a practical perspective. What should a quantum system be able to do, in principle, to qualify as a quantum computer for which these algorithms work? The role of superposition, parallelism, and entanglement is discussed.

There are a number of systems that are currently being considered as candidates for the construction of qubits, quantum logic gates and quantum computers. Some of the systems, notably atoms in magnetic traps and nuclear magnetic resonance (NMR) systems, have had some success in performing the elementary operations that would be required in large-scale quantum computer. However, these systems are not necessarily seen as viable technologies for quantum computing in the longer term. The recent demonstration of macroscopic coherence in a superconducting ring (consisting of a thick superconducting ring containing one or more Josephson weak link devices) has added significant weight to the idea of using superconducting persistent current devices (SQUIDs) in quantum logic systems. In this paper, we consider one aspect of the quantum mechanical SQUID, the nonlinear effect of SQUID on the classical control parameters, and we discuss how it may influence the construction and design of quantum logic gates based on SQUID devices. In particular, we look at problems associated with fixing the classical magnetic flux bias for a quantum mechanical SQUID at, or near, a quantum mechanical transition or resonance.

An experimentally realized Bose-Einstein spinor condensate is analyzed. The condensate of sodium atoms in their F equals 1 hyperfine state is optically trapped by a focused, near- infrared laser beam which eliminates many limitations associated with magnetic traps. A minimal theoretical description of the system is formulated and solved for motion along one dimension. The significance of the results of the solution, for possible use in quantum computing, is discussed.

A quantum computer is an information processing systems that utilizes invertible logic elements to perform reversible logic operations. It can also perform irreversible logic operations. Thus, in principle, it can compute the information transformation rule of any physical system for which the rules are known. A quantum computer can be designed to perform parallel computation on a massive scale, and therefore can have fast execution time. The basic structure of a quantum computer includes 1) a quantum register, or a configuration of quantum-mechanical energy states to store quantum bits (qubits) of information, 2) a quantum processor designed to induce quantum mechanical interactions between any arbitrary pair of qubits to perform a specified logic operations. 3) a quantum bus or network for transferring qubits from the registers to the processor and vice versa, and 4) a physical mechanisms for non-destructive measurements on the qubits. Quantum mechanical systems with the potential for realizing quantum computers include ion-trap quantum computers, cavity quantum electrodynamics quantum computers, photon-trap quantum computers, nuclear magnetic resonance quantum computers, and superconducting quantum computers. Decoherence [1 ,2] possesses the greatest challenge to the practical realization of quantum computers, because it causes the collapse of the quantum superposition states that contain the vital results of qubit manipulations. Decoherence processes are measurement-induced, noise-induced, and environment-induced [3] . Schemes to correct or circumvent the detrimental effects of decoherence in quantum processors fall into active error-correction schemes, and passive error-correction schemes [4]. Active error-correction schemes are designed to detect and correct for errors during computation. Algorithmic fault-tolerant computing [5, 6] is the most promising active-error correction scheme. In fault-tolerant computing, ancillary qubits are encoded into data blocks of the quantum registers. They are then used to detect errors. Once detected appropriate error-correction algorithms are then employed to rectify the errors. The implementation of fault-tolerant quantum computing has been hampered by the lack of practical universal set of quantum gates that can process an encoded data without propagating or introducing additional errors. Passive-error correction schemes [7] attempt to finding Hilbert subspaces that are intrinsically free of decoherence effects [8]. The key to quantum computing in such a decoherence-free subspace is the identification of mechanisms that can effect quantum mechanical interactions between the decoherence-free states without taking the system out of the decoherence-free environment. The challenge is to identify and employ physical processes that can achieve these goals.

Following a review of the probe optimization of Slutsky, Sun, Rao, and Fainman [Phys. Rev. A 57, 2383 (1998)] for the standard four-state protocol of quantum key distribution, I generalize the optimization to variable angle between the signal bases. I calculate the corresponding maximum Renyi information gain by the probe, and determine the optimum probe parameters. A larger set of optimum probe parameters is found for the standard protocol than was known previously.

In order to deal with a general pattern recognition problem by quantum computing, this paper proposes an embedding method that corresponds a feature vector of n-dimensional space Vⁿ to a vector of the surface of Riemann sphere in the n+1 dimensional space Vⁿ⁺¹, keeping the topology among feature vectors. Because the radius of this Riemann sphere is one, the feature vectors are mapped into normalized vectors. This paper shows that multiple linear discriminant functions can be defined to separate two arbitrary clusters that are mapped onto the Riemann sphere in the space Vⁿ⁺¹. This paper also shows that we can define a unitary transformation that computes the signs of values of those multiple linear discriminant functions, in parallel by quantum computing.

A real quantum computer theoretically can perform a large number of parallel processing at the same time by superposition states and unitary transformations in a Hilbert space. We show that arbitrary unitary transformation can be operated in a framework of classic physics such as optics. We also show that, using the optics, the same processing as what quantum computing logic does can be implemented and that image recognition can be done by that logic. Although our system currently does not utilize entanglement physics like a quantum computer using electronic spin system, we can construct a high dimensional Hilbert space (e.g. one thousand dimension). We consider that our system may be another approach to realize quantum computing paradigm.

The observed science data generated by all the instruments of the Space Solar Telescope (SST) is about 1728 Gbyte per day, after preprocessing the data volume is about 50 Gbyte, the data then access into the Mass Memory (MM) where the offline compression will reduce the data size advancely to about 10 Gbyte and the compressed data will be transmitted down once the SST contacts with the ground station. Because the MM takes the core position in the data stream transmission, it is the key to the design of the onboard data handling system. In addition, the bit error detection and correction and redundancy are needed for increasing the reliability in the special outer space environment. A ground model MM has been made which performs the main functions for the system. The desired goals are attained by the experiment.

The perspective of neural networks equivalental models (EM) base on vector-matrix procedure with basic operations of continuous and neuro-fuzzy logic (equivalence, absolute difference) are shown. Capacity on base EMs exceeded the amount of neurons in 2.5 times. This is larger than others neural networks paradigms. Amount neurons of this neural networks on base EMs may be 10 - 20 thousands. The base operations in EMs are normalized equivalency operations. The family of new operations equivalency and non-equivalency of neuro-fuzzy logic's, which we have elaborated on the based of such generalized operations of fuzzy-logic's as fuzzy negation, t-norm and s-norm are shown. Generalized rules of construction of new functions (operations) equivalency which uses relations of t-norm and s-norm to fuzzy negation are proposed. Among these elements the following should be underlined: (1) the element which fulfills the operation of limited difference; (2) the element which algebraic product (intensifier with controlled coefficient of transmission or multiplier of analog signals); (3) the element which fulfills a sample summarizing (uniting) of signals (including the one during normalizing). Synthesized structures which realize on the basic of these elements the whole spectrum of required operations: t-norm, s-norm and new operations equivalency are shown. These realization on the basic of new multifunctional optoelectronical BISPIN- devices (MOEBD) represent the circuit with constant and pulse optical input signals. They are modeling the operation of limited difference. These circuits realize frequency- dynamic neuron models and neural networks. Experimental results of these MOEBD and equivalency circuits, which fulfill the limited difference operation are discussed. For effective realization of neural networks on the basic of EMs as it is shown in report, picture elements are required as main nodes to implement element operations equivalence ('non-equivalence') of neuro-fuzzy logic's.

Photothermoplastic media (PTPM) is a non-silver two-layer, semiconductor-thermoplastic, structure for optical data recording which is radiative resistant. The PTPM allows to record photographic, holographic and other kinds of optical data without any wet chemical development that makes them maximal useful for remote sensing applications. Optical data of photothermoplastic recording are characterized by simplicity of developing and erasing processes. We elaborated the PTPM which is photosensitive in different spectral ranges, stable to radiation light mark, being able to record halftones and increasing the resolution of lens- PTPM system at the frequencies close to limiting characteristics of the lenses. Such a medium can work in a circular scheme providing the multiple write-erasure mode of recording. The image obtained on PTPM can be transmitted to the observation station instantly or with a necessary delay. The resolution of PTPM is a function of a thermoplastic layer thickness and varies from 200 to 1800 mm. PTPM may secure information recording in a broad spectral range from X-rays to IR. Camera is a single unit, which includes special light intensive eight-lens objective with visual resolution of 960 mm^-1 and weight of 7 kg. Precise tape-drawing mechanism secures movement of the PTPM with the speed of 1.5 to 2.5 mm/s what is sufficient for compensation of displacement of the Earth surface. The ground tests using stroke optical focusing pattern showed resolution of 240 mm^-1 by simultaneous movement of the apparatus and film relative to the optical focusing pattern.

We have studied holographic grating recording in Fe-doped photorefractive crystals LiNbO₃ (LN) and in doped As-S- Se chalcogenide glassy semiconductor (CGS) films. Transmission gratings in (alpha) -cut crystals of LN are efficient enough to demonstrate effect of optical channeling. Volume gratings recorded in LN crystals may be used as a parallel array of the planar waveguides. For the CGS slanted grating geometry was tested, aiming on creation of asymmetric blazed gratings. Asymmetry of non-Bragg diffraction orders was observed and its was 12 times enhanced by a gold coating. For the first time reflection volume grating were recorded in 2 mm thick CGS by green solid state laser ((lambda) equals 532 nm, P equals 100 mW).