The ever increasing demand for communication bandwidth and system interconnectivity has been a motivating factor behind the integration of optoelectronics device and conventional data processing circuitry. Over the last two decades, fiber optic components have become the dominant technology in the telecommunications industry. In last 5 years, optical interconnection techniques have been suggested as a solution to the interconnect density and bandwidth problems faced by electrical systems at the cabinet, PC-board and even chip level. Based on the smart pixel architectures in the last decade, the proposed chip monolithically integrates optical sensors with silicon CMOS based circuitry. This project incorporates an instruction fetch unit (IFU), that fetches the instructions from an external host computer, and a 2D-array of one-bit smart pixels called the processing element (PE). Each PE consists of an ALU, control logic, dual port register memory bank, photo-receiver circuit and associated driver circuits. By tiling these smart pixels in 2D, it is possible to form a programmable smart pixel array that is well suited to read optical page-oriented data types. The CASPR chip contains a 4x4 array of PEs connected to a single IFU. Inter PE communication has been established through nearest neighbor communication. Simultaneous communication to all the PEs is possible through global communication. The instruction set for this architecture is 17-bit long. The chip has been successfully fabricated in 0.5μ technology. We present in this paper the design and initial test results from the recent fabrication.

Two semiconductor optical amplifiers (SOAs) are directly coupled by connecting the output of each to the input of the other. This provides positive feedback for an inverted optical threshold; there is negligible output above a specified threshold. The SOA pair is analyzed and simulated to optimize parameters. Two SOA pairs are cross-connected in a similar manner by connecting the output of each pair to the input of the other, to construct an R-S flip-flop. Analysis shows that only two states exist. Simulation is used to demonstrate setting and resetting the R-S flip-flop at up to 5 Gbps.

For efficient optoelectronic implementation of parallel algorithms, a novel two-step digit-set-restricted modified signed-digit (MSD) arithmetic based on content-addressable-memory is presented. With the introduction of the reference digits, carry propagation is avoided by restricting digit sets of the intermediate carry and sum into {-1,0} and {0,1}, respectively. In our new algorithm, without using the complement property for the nonzero outputs, only 12 minterms for all the outputs are required. More significantly, since no complement operation is involved, the optical system needs no additional reflecting unit and mask. An incoherent correlator based optoelectronic shared-content-addressable-memory processor is used to perform the arithmetic operations. A proof-of-the-principle experiment is demonstrated.

The results of experimental investigation of dynamic holograms recording by 20 ns and 300 ps laser pulses in fullerene-containing solid-state matrices are presented in this paper. The spectral analysis in range 300 - 700 nm shows correlation between efficiency of dynamic holograms recording and presence of absorption band at nearly 330 nm. We use two types of solid-state matrices: porous glass and polymethylmethacrylate hosts, which were filled with fullerene C₆₀. From the dynamic holograms recording experiments and from the spectral analysis we obtained data on sensitivity and stability of the media, which can be used later in practical applications like all-optical switching.

Photonic true time delay is very important for phased array antenna applications. The conventional photonic delay line architectures have very high insertion loss. Therefore, the deployment of these photonic delay lines for phased array antennas is not cost effective. In this paper, we will review and compare the performance of different fiber optic photonic delay line architectures. To explore the application of MEMS mirror for photonic true time delay, a multi-base photonic delay line architecture is employed because this architecture has the potential to exhibit ultra-low loss for large binary bit photonic delay. Approaches to implement this photonic delay line architecture based on optical MEMS mirror technology have been proposed and theoretically analyzed. The proposed photonic delay line approaches may provide an ultra-low loss, module, low cost, and deployable photonic delay line for phased array beam forming applications.

In this paper, we propose a new design configuration of a novel multiple outputs and multiple wavelength fiber ring-optical lasers and experimentally measure its characteristics. This novel laser has output wavelength bands in 1310 nm and 1550 nm windows, respectively, at each output port of the fourteen outputs. The CNR, CSO, CTB are calculated and compared for the fourteen outputs. Such a multiple-output and multi-wavelength (MOMW) fiber ring laser can be applied in WDM/SDM/SCM fiber networks.

A specific case a four-order non-collinear light scattering in anisotropic media is presented. Compared to our previous studies, an innovation lies in the fact that now we consider passing just the quartet of incident light waves through a single crystal that is perturbed by a coherent steam of acoustic phonons. The exact and closed analytical model for describing this strongly nonlinear phenomenon is developed. In fact, specially designed regime of a four-order light scattering, when transitions of four input light beams into four output light modes are allowed and electronically controlled, is examined. The feasibility of applying such an effect to perform an all-optical switching is analyzed. An opportunity for arranging the digital 4-bit register is revealed and algorithmically investigated.

This paper presents a method for the calibration of a novel two-way optical component network analyzer (TWOCNA) we designed. The performance of the calibration error box for both electronic and optical interface were simulated and measured. Furthermore, the decreses of the incident power which will introduce deviation of the TWOCNA spectrum range are also presented.

In object recognition, one goal of matched filter design has been to define a matching function that produces an ideal correlation peak when a target object in an image scene precisely matches the pre-defined template object. The benefit of such a function is that it guarantees a precise detection/identification. The ideal correlation-based function that defines the match has been described as a dirac delta in the correlation plane. This paper suggests that if similarity as opposed to precise matching is the goal of the correlation function, then ony using current two-dimensional correlation techniques will result in a non-dirac delta in the correlation plane. This paper suggests basing the design of the function on the object recognition goal. The approach for correlation function design is demonstrated using psychophysical evidence for class differentiation. A function is designed based on psychophysical experimental results for distinguishing between two simple objects and their deformations: a square and a circle.

The automatic target recognition (ATR), often time, is limited by the presence of background clutter and distortions such as scale, translation and rotation (both in-plane and out-of-plane) in both single and multi object cases. Such distortion invariant ATR and image understanding have been the subject of intense research in machine vision. In a previous work, we have demonstrated the usefulness of an amplitude-coupled minimum-average correlation energy (AC-MACE) filter in in-plane rotated SAR image ATR. The AC-MACE filter outperforms the regular MACE filter in rotation-related cases. Motion tracking is also an important task in computer vision, especially, when objects are subjected to certain viewing transformation. There are many problems in which very small objects undergoing motion must be detected and then tracked. For example, one of the most difficult goals of ATR is to spot incoming objects at long range, wherein the motion seems small and the signal to noise ratio (SNR) is poor. The system must be able to track such targets long enough to identify whether the object is a friend or foe. In this work, we are interested in locating both long-range and short-range moving objects in IR images wherein the object may vary from a few pixels in size to a large number of pixels in a sequence of IR images. The targets are submerged in background noise and clutter. Additionally, the tracking problem also involves out-of-plane rotation of the target. Thus, we investigate both MACE and AC-MACE filter for rotation and size invariant target detection and tracking using realistic IR images.

The joint transform correlator (JTC) technique has been found to be suitable for real-time matching and tracking operations. Among the various JTC techniques proposed in the literature, the fringe-adjusted JTC has been found to yield better correlation output for target detection. In this paper, we propose a generalized fringe-adjusted JTC based algorithm for detecting and tracking a target in a video image sequence. The proposed JTC based algorithm has been found to be suitable for near real time recognition and tracking of a static or moving target. The performance of the proposed technique has been verified with real life image sequences.

This paper investigates the role of composite filters in reducing the search time for 3D model based object recognition. When one moves from 2D to 3D, one is faced with a huge amount of information to deal with. The composite filter combined with a multistage scheme is developed for processing this huge information. The design scheme for the composite filters is also elaborated. The procedures discussed in this paper demonstrate how detection of these various model images might help formulate a new metric for recognition performance.

In this paper, we discuss the Passive Optical Network deployment on an arbitrary grid with guaranteed tolerance towards p-1 equipment failure. We show that this problem in general is NP-hard. We propose an algorithm, which guarantees a solution of 4-approximation to the optimal deployment, and further argue that this is the best lower bound achievable in our case. We do comparative studied with randomized layouts, were our proposed algorithm saves 45% - 55% deployment cost (fiber, equipment, etc.) on average.

This article presents an application of Phase Only Filter (POF) to the classification of volatile compound samples with chemical sensor arrays. Sinusoidal temperature modulator excites the chemical sensor array. A system composed of multiple sensors for data acquisition requires the analysis of multiple data signals in order to classify the input data. One such system is the Electronic Nose system (eNose). The eNose data is in fact 1D data from multiple data sources. In this work five samples of three different kinds of coffee are used to build an odor database. An unknown test sample is then classified. In this research data, from such system will be analyzed and classified using the POF and compared to the more conventional K Nearest Neighbors (KNN) classifier. Both classifiers use only the mean and standard deviation as the classification feature space. The difference of the correlation between odor database and the auto correlation of the test sample is the measure of closeness for the POF. For the KNN, the measure of closeness is the difference between the odor database and the test sample.

In this paper we review the current status of transparent networks and in particular of all-optical regenerators and wavelength converters that might be part of next generation telecommunication systems.

The goal of this work is to develop micro-scale noninvasive photonic instrumentation and techniques that will enable real-time imaging and monitoring of microstructures and cells in tissues. We utilize a hyperspectral microscope to explore the differences of native fluorescence and polarized light scattering from cellular components using various excitation wavelengths. The key optical “signature” characteristics that differentiate the various cellular components are used to obtain composite images that highlight their presence and the relative concentration of various tissue chromophores. The sensitivity has been optimized in our specially designed instrumentation so that image acquisition times are very short for real-time application in a clinical setting. This technology is not invasive to the cells and therefore it can be used to monitor their function while they are still alive.

This paper describes a software development environment suitable for advanced signal processing and feedback control algorithms for interferometry. The significance of software fast prototyping tools is illustrated where theoretically challenging algorithms are to be implemented. Specifically, a case is discussed where advanced control is deemed necessary for reliable operation of FT-NIR instruments in an on-line environment. In this particular case, an interactive test bed for new algorithmic approaches is developed, such that advanced theories and methods can be rapidly tested and validated with the real-world interferometer hardware. This development environment also allows new interfacing concepts with mechanical and optical hardware to be tested. Use of this environment is illustrated for the case of a novel Michelson NIR spectrometer operating in the range 800-2200 nm.

A challenging task facing the designers for the next generation of archival storage system is to provide storage capacities several orders of magnitude larger than existing systems while maintaining current data access times. To meet this challenge, a CMOS optoelectronic database filter suitable for large capacity relational database systems that use page-oriented optical storage devices has been developed. Based on a smart-pixel strategy, the filter performs optical-to-electrical data conversion, processes database query operations and only passes the data matching a query to the host computer. The photonic VLSI chip monolithically integrates a smart-pixel array that incorporates page-oriented optical reading, data manipulation logic, data buffering, and filter control circuitry for interfacing the filter chip with an electronic host computer. The database filter can offset the bandwidth mismatch without loss of valid data or significant delay in data access. A software simulator is developed to simulate the behavior of the database filter. The implementation and performance analysis of a 32 x 32-bit database filter fabricated through a 0.35-micron CMOS process is reported.

We present a novel design of an 1 x 2 photonic crystal beam splitter using silicon-on-insulator integrated optical technology. According to the design, we can reduce the size of beam splitter without increasing the propagation loss. In this paper, we study the 1x2 beam splitting ratio by changing the position of the center air hole. We use a finite-difference-time-domain (FDTD) method to simulate the propagation of the light. The 1x2 PBG beam splitter is designed with 180 degree output bending angles. The size of this novel beam splitting device are 8 μm x 3.44 μm.

We describe a new kind of symmetric color separation grating (SCSG), which can be applied to separate the frequency-tripling (3ω)light from the basic (1ω) and frequency-doubling (2ω)lights. The profile of the proposed SCSG is a three-level symmetric structure in one period, thus it is easy to fabricate it with the much improved tolerance of mask misalignment without the decrese of the efficiency of the frequency-tripling light. We have analyzed the principles of the SCSGs by employing the scalar diffraction theory and fabricated them experimentally by means of binary optics technology. The theoretical and experimental results demonstrated that the proposed SCSGs can be made in a comparatively easy way for the symmetric structure that can effectively avoid the effect of the fabrication error due to mask misalignment, thus higher manufacturing yield and lower cost can be achieved.

Integration of optical systems is important in practical applications for enhanced compactness and stability, particularly for optical switch. In the previous reports, 1×N electro-optic switches were constructed by the assembling of discrete E-O elements. The aim of this paper is to develop a 1×N switch integrated in a single slab of LiNbO₃, which consists of an array of electro-optic polarization modulators between the interfaces for total-internal reflection. Due to the double refraction effect, the ordinary ray and the extraordinary ray have different reflections. From different switching states of modulators, the internally double reflection along the sequential interfaces of a crystal can lead to different directions of the output beam. On this basis, a 1×N optical switch is designed, which is a crystal slab of 2 refractive interfaces for beam input and output, 4 reflective interfaces for total-internally double reflection, and N-1 E-O modulators with pairs of electrodes. A voltage on the n pair of electrodes will induce output beam onto the n angular position. The suggested 1×N switch is simple and compact in construction, low in loss and insensitive to environment. An experimental 1×4 switch is demonstrated, too.

Compact, low operating voltage, high-speed bypass-exchange switches are major elements in optical interconnection networks. Conventionally, a bypass-exchange switch consists of a controllable half-wave plate sandwiched between two polarizing beam displacers. In this paper, an integrated optical bypass-exchange switch is designed in a single block of LiNbO₃ crystal. The switch is polarization based and consists of only one birefringent crystal slab with electro-optic property. Based on both the phenomena of double refraction and internally double reflection, and the electro-optic effect in a crystal, an electro-optic modulator, a beam combiner and a beam splitter are integrated in a single block of LiNbO₃ crystal. The polarization states of the two input signals are both linear and perpendicular to each other. First the two signal beams are combined into a single channel with two orthogonal polarization directions by double refraction and internally double reflection. The signal couple propagates along the optical axis of the crystal. When a half-wave voltage is applied on the direction normal to the incident plane, the polarization directions of the two beams in a single channel will be exchanged. Last the signal couple is divided separately with the same deflection as that of the two input beams by double refraction and internally double reflection again. The experimental results show that the switch is low operating voltage, low cross talk, low energy loss, fast-speed, and insensitive to environment disturbance. The switch may be applied to the nodes of various multistage interconnection networks.