PERSONAL Sign in with your SPIE account to access your personal subscriptions or to use specific features such as save to my library, sign up for alerts, save searches, etc.
Integrated optics technology can be considered now as a mature technology. As a matter
of fact, several type of circuits are now commercially available from different companies
either as standart products or on a foundry facility basis. This is true for integrated
optical devices based on LiNb03 crystals while semiconductor or other type of technologies
are still in the laboratories.
Access to the requested content is limited to institutions that have purchased or subscribe to SPIE eBooks.
You are receiving this notice because your organization may not have SPIE eBooks access.*
*Shibboleth/Open Athens users─please
sign in
to access your institution's subscriptions.
To obtain this item, you may purchase the complete book in print or electronic format on
SPIE.org.
The development of semiconductor optoelectronics devices will have great impact on optical computing. Progress in semiconductor lasers and photodetectors has revolutionized communication systems and realized new methods of memory, printing and other processes. Further advances in optoelectronic devices, which are now taking a new integrated form, optoelectronics integrated circuit (OEIC). The OEIC will make possible highly versatile optical computing systems in the next century. Recent developments in OEIC technology and future prospects for its application to optical computing are presented.
Access to the requested content is limited to institutions that have purchased or subscribe to SPIE eBooks.
You are receiving this notice because your organization may not have SPIE eBooks access.*
*Shibboleth/Open Athens users─please
sign in
to access your institution's subscriptions.
To obtain this item, you may purchase the complete book in print or electronic format on
SPIE.org.
Optical logic and memory components for use in cellular image processing are discussed. The application of the nonlinear optical responses of the components to programmable logic units and numeric units leads to novel architectural features. A tolerance methodology for the units is described.
Access to the requested content is limited to institutions that have purchased or subscribe to SPIE eBooks.
You are receiving this notice because your organization may not have SPIE eBooks access.*
*Shibboleth/Open Athens users─please
sign in
to access your institution's subscriptions.
To obtain this item, you may purchase the complete book in print or electronic format on
SPIE.org.
Basic parameters of DL, writing and reconstruction of holograms, holographic memory on DL, generation of picosecond pulses, bistability and logic gates, integration of DL with electronic circuits and an example of optical interconnection by OEIC and waveguiding holograms are presented.
Access to the requested content is limited to institutions that have purchased or subscribe to SPIE eBooks.
You are receiving this notice because your organization may not have SPIE eBooks access.*
*Shibboleth/Open Athens users─please
sign in
to access your institution's subscriptions.
To obtain this item, you may purchase the complete book in print or electronic format on
SPIE.org.
A nonlinear interference filter has been incorporated into a CRT and e-beam controlled switching of individual elements in a 2 x 2 array of optically bistable pixels demonstrated.
Access to the requested content is limited to institutions that have purchased or subscribe to SPIE eBooks.
You are receiving this notice because your organization may not have SPIE eBooks access.*
*Shibboleth/Open Athens users─please
sign in
to access your institution's subscriptions.
To obtain this item, you may purchase the complete book in print or electronic format on
SPIE.org.
The results of a fresh look at holographic data storage in photorefractive materials is reported. This work concludes that advances in photorefractive material, particularly SBN, AO deflector technology, CCD technology, and laser technology could make a very interesting low latency mass storage device possible. Two advances in this technology, a technique for a highly asymmetric write/read cycle and a method for building artificial crystal shcipes, are described.
Access to the requested content is limited to institutions that have purchased or subscribe to SPIE eBooks.
You are receiving this notice because your organization may not have SPIE eBooks access.*
*Shibboleth/Open Athens users─please
sign in
to access your institution's subscriptions.
To obtain this item, you may purchase the complete book in print or electronic format on
SPIE.org.
There are a number of applications for optically addressed SLM's which benefit from grey scale capability. We are particularly interested in input devices for optical correlators; programmable beam deflectors; and soft threshold units for neural net architectures. The requirements for each application have a different emphasis. For optical correlators, a high photosensitivity and good dark state are important. For beam deflection, a linear transfer characteristic will assist the formation of a replicate sine wave grating and optimal diffraction efficiency. Finally, soft threshold units require a nonlinear transfer characteristic. It is the suggestion of this paper that a single electro-optic effect in the light modulating layer of the light valve might be useful for each of these several applications. Then we present experimental measurements on two examples of light valves, with particular relevance to the third application.
Access to the requested content is limited to institutions that have purchased or subscribe to SPIE eBooks.
You are receiving this notice because your organization may not have SPIE eBooks access.*
*Shibboleth/Open Athens users─please
sign in
to access your institution's subscriptions.
To obtain this item, you may purchase the complete book in print or electronic format on
SPIE.org.
Applications of a multiple quantum well (MQW) nonlinear directional coupler to optical logic/computing are discussed. Very encouraging performance parameters are presented based on theoretical and experimental analysis of the element.
Access to the requested content is limited to institutions that have purchased or subscribe to SPIE eBooks.
You are receiving this notice because your organization may not have SPIE eBooks access.*
*Shibboleth/Open Athens users─please
sign in
to access your institution's subscriptions.
To obtain this item, you may purchase the complete book in print or electronic format on
SPIE.org.
An experimental technique for estimation of information capacity of optically addressed spatial light modulators (SUB) in coherent optical systems is suggested. The technique can be used for comparison of different types and a proper choice of SLMs.
Access to the requested content is limited to institutions that have purchased or subscribe to SPIE eBooks.
You are receiving this notice because your organization may not have SPIE eBooks access.*
*Shibboleth/Open Athens users─please
sign in
to access your institution's subscriptions.
To obtain this item, you may purchase the complete book in print or electronic format on
SPIE.org.
Considerable attention is being focused on the development of high speed optoelectronic devices not only due to their inherent speed and parallelism but also because of the ease with which these devices can be interfaced with existing electronic devices. This paper presents a novel optoelectronic device called the Optical Distributed Arithmetic Unit (ODAU). The ODAU is comprised of a laser, lens and semiconductor components which are integrated to perform high speed binary arithmetic based on the theory of distributed arithmetic. A simple carry unit is responsible for performing the necessary carry operations in order to make the ODAU a binary device and also function as a data regeneration unit. A description of the ODAU as well as a mathematical model and statistical equations are presented. Simulation as well as analytically derived performance measures and a simple application involving the implementation of a filter are also given.
Access to the requested content is limited to institutions that have purchased or subscribe to SPIE eBooks.
You are receiving this notice because your organization may not have SPIE eBooks access.*
*Shibboleth/Open Athens users─please
sign in
to access your institution's subscriptions.
To obtain this item, you may purchase the complete book in print or electronic format on
SPIE.org.
New studies have been undertaken with a view to use the TITUS spatial light modulator for optical processing applications. Main analysed applications were : real time SAR data processing, passive SONAR data processing, and correction of atmospheric wave front distor-tion. Regarding the TITUS light-valve, several improvements have been tested, leading to increased resolution and speed, and better flatness. At last, a test bench has been defined to characterize rims for its use in the optical processing field, in particular with coherent light.
Access to the requested content is limited to institutions that have purchased or subscribe to SPIE eBooks.
You are receiving this notice because your organization may not have SPIE eBooks access.*
*Shibboleth/Open Athens users─please
sign in
to access your institution's subscriptions.
To obtain this item, you may purchase the complete book in print or electronic format on
SPIE.org.
Several picosecond NOR gates operating without Fabry-Perot cavity are described in GaAs, CdTe and CdSe, based on the induced picosecond absorption by electron-hole plasma. The realization of picosecond optical inverter gates is discussed.
Access to the requested content is limited to institutions that have purchased or subscribe to SPIE eBooks.
You are receiving this notice because your organization may not have SPIE eBooks access.*
*Shibboleth/Open Athens users─please
sign in
to access your institution's subscriptions.
To obtain this item, you may purchase the complete book in print or electronic format on
SPIE.org.
We discuss the principles of development of optical logic elements using stimulated scattering in fibers.,Theoretical estimates show that it ip possilAe to movide a pulse repetition rate of 1012 pule/s, pump pulse energy of 10-l2 - 10-13 J/μ2, and signal pulse energy of 10-15 J/)μm2. Attention is focussed on the SRS-invertor and possibility of its usage for development of generators of optical pulse sequencies and fiber-optic memories.
Access to the requested content is limited to institutions that have purchased or subscribe to SPIE eBooks.
You are receiving this notice because your organization may not have SPIE eBooks access.*
*Shibboleth/Open Athens users─please
sign in
to access your institution's subscriptions.
To obtain this item, you may purchase the complete book in print or electronic format on
SPIE.org.
Different types of memories on the base of fiber optic round circuits of the data recirculation are presented. The main physical aspects of their creating necessary for the construction of various systems for the signal processing and optical computing are con-sidered. The first system for the analog and parallel digital signals processing on the base of a multichannel digital fiber optic memory and a microcomputer has been constructed. For the ultrafast data processing we have suggested and studied a block scheme of the optical memory operating on ultrashort light pulses - solitons. It is shown that in the fiber memory devices operating on solitons with Raman amplification the nonlinear filtering of the noise and soliton components of the field as well as the long-time data storage without noise accumulation can be realized.
Access to the requested content is limited to institutions that have purchased or subscribe to SPIE eBooks.
You are receiving this notice because your organization may not have SPIE eBooks access.*
*Shibboleth/Open Athens users─please
sign in
to access your institution's subscriptions.
To obtain this item, you may purchase the complete book in print or electronic format on
SPIE.org.
In nearly degenerated four-wave mixing the application of running gratings in the crystal produces an enhancement of the four-wave mixing efficiency compared to the case of degenerated four-wave mixing with stationary gratings'. In diffusion dominated photorefractive crystals that enhancement is obtained only if the coupling strength rises above a critical value. In our analytical study we examine that critical coupling strength for weak input signals as a function of the pump beam intensity ratio. Moreover, we discuss the dependence of the reflectivity on the (temporal) grating frequency.
Access to the requested content is limited to institutions that have purchased or subscribe to SPIE eBooks.
You are receiving this notice because your organization may not have SPIE eBooks access.*
*Shibboleth/Open Athens users─please
sign in
to access your institution's subscriptions.
To obtain this item, you may purchase the complete book in print or electronic format on
SPIE.org.
We report on optical bistability in thin films of evaporated CdS at 514 nm. The main mechanism is traced back to thermally induced increasing absorption. Besides all optical devices, externally controlled SEED structures are realized 'by integrated electrical con-tacts. Experimental results on the steady-state and the dynamic behavior of both types of elements are presented.
Access to the requested content is limited to institutions that have purchased or subscribe to SPIE eBooks.
You are receiving this notice because your organization may not have SPIE eBooks access.*
*Shibboleth/Open Athens users─please
sign in
to access your institution's subscriptions.
To obtain this item, you may purchase the complete book in print or electronic format on
SPIE.org.
We demonstrated optical bistability in air-spaced Fabry-Perot etalons caused by thermal expansion of the mirror substrates and a concomitant reduction in mirror separation. The theory of this type of bistability is presented. Experimental results for metal mirror coat-ings of gold or aluminium on glass and polymethylmethacrylate (PMMA) at the He-Ne laser wavelength λ=632.8 nm are given.
Access to the requested content is limited to institutions that have purchased or subscribe to SPIE eBooks.
You are receiving this notice because your organization may not have SPIE eBooks access.*
*Shibboleth/Open Athens users─please
sign in
to access your institution's subscriptions.
To obtain this item, you may purchase the complete book in print or electronic format on
SPIE.org.
Surface plasmons on metal films or guided modes in planar waveguides are excited with laser light by prism coupling. Optical bistability induced by the photo-thermal displacement effect is predicted and demonstrated.
Access to the requested content is limited to institutions that have purchased or subscribe to SPIE eBooks.
You are receiving this notice because your organization may not have SPIE eBooks access.*
*Shibboleth/Open Athens users─please
sign in
to access your institution's subscriptions.
To obtain this item, you may purchase the complete book in print or electronic format on
SPIE.org.
Observation of the nonlinear prism coupling in a glass thin-film optical waveguide with liquid crystalline nonlinear prism coupler is reported. The effect was observed for the case of the maximum coupling efficiency at low input intensities.
Access to the requested content is limited to institutions that have purchased or subscribe to SPIE eBooks.
You are receiving this notice because your organization may not have SPIE eBooks access.*
*Shibboleth/Open Athens users─please
sign in
to access your institution's subscriptions.
To obtain this item, you may purchase the complete book in print or electronic format on
SPIE.org.
An optical clock, temporal characteristics of which are determined by geometrical parameters, works by stationary pulsed emission of a CT-pumped Brillouin fiber ring laser. Modelization has been achieved through coherent equations including optical Kerr effect. Feasibility is demonstrated and a compact device is envisaged.
Access to the requested content is limited to institutions that have purchased or subscribe to SPIE eBooks.
You are receiving this notice because your organization may not have SPIE eBooks access.*
*Shibboleth/Open Athens users─please
sign in
to access your institution's subscriptions.
To obtain this item, you may purchase the complete book in print or electronic format on
SPIE.org.
We consider the potential application of two crossed Fabry-Perot resonators for optical logic operations, such as for a photonic switch or an optical transistor. The crossed resonator is a three-port device that allows the signal to be separated from the control inputs and can yield cascadable gain. An examination of the required material properties reveals that the linear absorption is a crucial quantity, and that the absorption per pass must be limited to 0.05 if gain is required.
Access to the requested content is limited to institutions that have purchased or subscribe to SPIE eBooks.
You are receiving this notice because your organization may not have SPIE eBooks access.*
*Shibboleth/Open Athens users─please
sign in
to access your institution's subscriptions.
To obtain this item, you may purchase the complete book in print or electronic format on
SPIE.org.
An electrooptic spatial light modulator, used as an intensity spatial filter, makes real-time inspection of periodic patterns possible by making the adjustment of object attitude unnecessary. The ways for preparing the filter and detecting defects in periodic pattern in real time are described. Results of detection of defects are presented.
Access to the requested content is limited to institutions that have purchased or subscribe to SPIE eBooks.
You are receiving this notice because your organization may not have SPIE eBooks access.*
*Shibboleth/Open Athens users─please
sign in
to access your institution's subscriptions.
To obtain this item, you may purchase the complete book in print or electronic format on
SPIE.org.
Polarization-modulation or plasma-enhanced reflection change the finesse of a non-linear resonator and consequently bring forward a new commutation mechanism and new tools to study the dynamic behaviour of a non-linear Fabry-Perot.
Access to the requested content is limited to institutions that have purchased or subscribe to SPIE eBooks.
You are receiving this notice because your organization may not have SPIE eBooks access.*
*Shibboleth/Open Athens users─please
sign in
to access your institution's subscriptions.
To obtain this item, you may purchase the complete book in print or electronic format on
SPIE.org.
The anisotropic spatial resolution of the photorefractive BSO crystal in the transversal configuration is analyzed and showed. Such behavior provides a spatially variant filtering operation on the signal to be processed. Experimental results that show such behavior are presented.
Access to the requested content is limited to institutions that have purchased or subscribe to SPIE eBooks.
You are receiving this notice because your organization may not have SPIE eBooks access.*
*Shibboleth/Open Athens users─please
sign in
to access your institution's subscriptions.
To obtain this item, you may purchase the complete book in print or electronic format on
SPIE.org.
The promise of organic and polymeric materials for electro-optic and nonlinear optic applications has been advanced a great deal in recent years. This interest arises from the fact that successful molecular design has produced materials with useful nonlinear susceptibilities while maintaining compliant structural, mechanical, and processing properties" 5. This paper will discuss the application of such materials in the construction of electro-optic channel-waveguide devices. In section 2 we will discuss a technique for fabricating channel-waveguide devices in polymer materials. Section 3 will briefly cover the interest in high-speed electro-optic devices in these materials, and in section 4 we will describe a device architectures that can take advantage of the processing properties used to make polymer waveguides.
Access to the requested content is limited to institutions that have purchased or subscribe to SPIE eBooks.
You are receiving this notice because your organization may not have SPIE eBooks access.*
*Shibboleth/Open Athens users─please
sign in
to access your institution's subscriptions.
To obtain this item, you may purchase the complete book in print or electronic format on
SPIE.org.
The photorefractive behavior of semiconductor materials is receiving much attention for applications involving high nonlinearities at wavelengths compatible with solid state lasers. Bulk GaAs and InP semiconductors are currently examined with regards to achieving efficient interactions in wave mixing experiments. In this paper, we present the characteristics of photorefractive crystals with applications to image amplification, phase conjugation and self-induced oscillations.(1-4) The two crystals InP:Fe and GaAs:Cr are oriented for maximum beam coupling and under uniform illumination their conductivities varie as σInP (0-1CM-1) = (18,5+0,6810)10-9 and aGaAs 41-1CM-1) = (2,8+1,41.) 10-9 (with I. in mW.cm-2). Their measured intrinsic absorption coefficients are : ainp = 2,5 cm-1 and aG,A. = 1,65 cm-1. The experimental set-up for two beam coupling is shown in Fig.1. A low power cw diode pumped YAG laser (40 mW, 1 = 1.06μm) provides a signal beam I.. and a pump beam Ipo whose frequency is Doppler shifted by reflection on the piezomiror M (moving grating recording mode). The gain is measured through the parameter y., defined as (5)
Access to the requested content is limited to institutions that have purchased or subscribe to SPIE eBooks.
You are receiving this notice because your organization may not have SPIE eBooks access.*
*Shibboleth/Open Athens users─please
sign in
to access your institution's subscriptions.
To obtain this item, you may purchase the complete book in print or electronic format on
SPIE.org.
The physical background of persistent storage and playback of temporally, spatially and polarizationally modulated ultrashort (up to 100 fs) light signals is presented and corresponding experimental results are demonstrated. Possible applications of the generalized holography for ultrafast parallel optical data-processing are discussed.
Access to the requested content is limited to institutions that have purchased or subscribe to SPIE eBooks.
You are receiving this notice because your organization may not have SPIE eBooks access.*
*Shibboleth/Open Athens users─please
sign in
to access your institution's subscriptions.
To obtain this item, you may purchase the complete book in print or electronic format on
SPIE.org.
The pnpn optical switch is an optoelectronic device that emits light in response to an optical input while an electrical bias is applied. A clocked bias insures synchronization. Other modes of operation are possible. pnpn structures have been fabricated by different techniques. Turn-on and turn-off times of less than 10 nsec have been obtained. High optical gain is possible. Some logic functions can be accomplished within this threshold device.
Access to the requested content is limited to institutions that have purchased or subscribe to SPIE eBooks.
You are receiving this notice because your organization may not have SPIE eBooks access.*
*Shibboleth/Open Athens users─please
sign in
to access your institution's subscriptions.
To obtain this item, you may purchase the complete book in print or electronic format on
SPIE.org.
This paper discusses progress in device concepts for rapid access to data stored in optical memories. The paper reviews laser beamsteering and control techniques, optical memory configurations, and promising combinations for devices.
Access to the requested content is limited to institutions that have purchased or subscribe to SPIE eBooks.
You are receiving this notice because your organization may not have SPIE eBooks access.*
*Shibboleth/Open Athens users─please
sign in
to access your institution's subscriptions.
To obtain this item, you may purchase the complete book in print or electronic format on
SPIE.org.
The crossover network is introduced as a new interconnection network for optical digital computers. It belongs to the class of multistage interconnection networks such as the perfect shuffle and the Banyan network. For connecting N = 2k input ports to N output ports (where k is an arbitrary integer) log2(N) stages are required. The crossover network holds an isomorphic relationship to the perfect shuffle and the Banyan network. However, using a Michelson-based optical setup the crossover network can be implemented with a higher efficiency in terms of the number of connections and conservation of light. Experimental results are presented where more than 1000 interconnections are implemented.
Access to the requested content is limited to institutions that have purchased or subscribe to SPIE eBooks.
You are receiving this notice because your organization may not have SPIE eBooks access.*
*Shibboleth/Open Athens users─please
sign in
to access your institution's subscriptions.
To obtain this item, you may purchase the complete book in print or electronic format on
SPIE.org.
As an alternative to electronic networks, large-scale (1000x1000) optical crossbar switching networks for use in parallel processing supercomputers may employ freespace optical transmission and known communications protocols in either a beam spreading/masking model or in a beam steering model, with either single-or double-sided nodal arrangements. The particular form of optical crossbar network chosen determines to some extent the class of admissable communications protocols, which in turn affects such performance parameters as connection establishment latency and network throughput.
Access to the requested content is limited to institutions that have purchased or subscribe to SPIE eBooks.
You are receiving this notice because your organization may not have SPIE eBooks access.*
*Shibboleth/Open Athens users─please
sign in
to access your institution's subscriptions.
To obtain this item, you may purchase the complete book in print or electronic format on
SPIE.org.
We present here the MILORD project developed by the CERT for three years. MILORD is, multiprocessor architecture organized around a 34 x 34 LCLV optical crossbar network, interconnecting 8 INMOS T414 microprocessors with 4 bidirectional links each and a ninth T414 host processor with 2 links on the network. This project intends to demonstrate the ability for optics to reach a higher degree of connectivity in parallel architectures.
Access to the requested content is limited to institutions that have purchased or subscribe to SPIE eBooks.
You are receiving this notice because your organization may not have SPIE eBooks access.*
*Shibboleth/Open Athens users─please
sign in
to access your institution's subscriptions.
To obtain this item, you may purchase the complete book in print or electronic format on
SPIE.org.
An array illuminator provides the holding beam for an array of optical switches. The transformation of an incoming light beam can be brought about by several optical arrangements as are: telescopic arrays of lenses or holograms, phase contrast, Fraunhofer or Fresnel diffraction at periodic phase masks. The phase contrast illuminator will be discussed in more detail.
Access to the requested content is limited to institutions that have purchased or subscribe to SPIE eBooks.
You are receiving this notice because your organization may not have SPIE eBooks access.*
*Shibboleth/Open Athens users─please
sign in
to access your institution's subscriptions.
To obtain this item, you may purchase the complete book in print or electronic format on
SPIE.org.
We present performance improvement in thick hologram optical interconnects. The improvement is achieved by resonant mirrors reflecting light back and forth through the hologram. Theoretical predictions are given which are corroborated by experimental results.
Access to the requested content is limited to institutions that have purchased or subscribe to SPIE eBooks.
You are receiving this notice because your organization may not have SPIE eBooks access.*
*Shibboleth/Open Athens users─please
sign in
to access your institution's subscriptions.
To obtain this item, you may purchase the complete book in print or electronic format on
SPIE.org.
A new optical interconnection is proposed, using one dimensional electro-photonic semiconductor arrays (vertical-to-surface transmission electro-photonic devices). It realizes variable interconnection, signal summation and thresholding functions for optical neural networks and parallel processing. Experimental results showed 8 Mbps data rate operations and certified the interconnection feasibility.
Access to the requested content is limited to institutions that have purchased or subscribe to SPIE eBooks.
You are receiving this notice because your organization may not have SPIE eBooks access.*
*Shibboleth/Open Athens users─please
sign in
to access your institution's subscriptions.
To obtain this item, you may purchase the complete book in print or electronic format on
SPIE.org.
This paper describes the architecture of a hybrid switching system for interconnection of high-performance processors. The switching system consists of two distinct switching planes, one an electronic crossbar and the second an all-optical crossbar. Data can be switched through any one of the planes. The electronic plane provides fast reconfiguration while the optical plane provides slower reconfiguration but higher bandwidth. Control of both planes is through a central electronic controller. Communication through the system can be performed using either message switching or circuit switching. The choice of the protocol can be made to minimize the overall transfer time.
Access to the requested content is limited to institutions that have purchased or subscribe to SPIE eBooks.
You are receiving this notice because your organization may not have SPIE eBooks access.*
*Shibboleth/Open Athens users─please
sign in
to access your institution's subscriptions.
To obtain this item, you may purchase the complete book in print or electronic format on
SPIE.org.
Optical interconnections may be an alternative to metallic lines in very large and fast circuits. In this field, integrated optics could be very attractive because the basic approach is similar to the one of microelectronics. From this point of view, the silicon based integrated optics technology developed at LETI is described and expected performances are analysed.
Access to the requested content is limited to institutions that have purchased or subscribe to SPIE eBooks.
You are receiving this notice because your organization may not have SPIE eBooks access.*
*Shibboleth/Open Athens users─please
sign in
to access your institution's subscriptions.
To obtain this item, you may purchase the complete book in print or electronic format on
SPIE.org.
his paper describes optical techniques for implementing volume, dynamic interconnection networks based on perfect shuffles and general shuffles. A mathematical description of 1-D and 2-D general shuffle permutations is given, including a discussion of separable and folded shuffles. An experimental optical system that implements the 2-D separable perfect shuffle, the 2-D folded perfect shuffle, and the 2-D separable 3-shuffle is described. The combination of shuffle permutation stages with optoelectronic dynamic switching elements can implement multistage networks and more general parallel signal processing architectures.
Access to the requested content is limited to institutions that have purchased or subscribe to SPIE eBooks.
You are receiving this notice because your organization may not have SPIE eBooks access.*
*Shibboleth/Open Athens users─please
sign in
to access your institution's subscriptions.
To obtain this item, you may purchase the complete book in print or electronic format on
SPIE.org.
Practical difficulties to be encountered in the implementation of massive optical interconnects are outlined and some improved architectures are proposed. The long term optimum design uses currently unavailable large arrays of laser diodes. An interim solution using available spatial light modulators is shown to allow roughly 1011 interconnections per second.
Access to the requested content is limited to institutions that have purchased or subscribe to SPIE eBooks.
You are receiving this notice because your organization may not have SPIE eBooks access.*
*Shibboleth/Open Athens users─please
sign in
to access your institution's subscriptions.
To obtain this item, you may purchase the complete book in print or electronic format on
SPIE.org.
We suggest the use of a pinhole imaging hologram as a new approach for optical interconnects. It can avoid the formation of intermodulation gratings and has some other advantages. On recording, the object beam passes through a pinhole thus forming an image plane hologram, but when reconstructing the pinhole can be removed. The reconstructed image can be programmed and processed by using optical information processing elements placed in the position of the reconstructed pinhole or pinholes. We present theoretical considerations and experimental results for pinhole holograms produced for holographic optical interconnects.
Access to the requested content is limited to institutions that have purchased or subscribe to SPIE eBooks.
You are receiving this notice because your organization may not have SPIE eBooks access.*
*Shibboleth/Open Athens users─please
sign in
to access your institution's subscriptions.
To obtain this item, you may purchase the complete book in print or electronic format on
SPIE.org.
Composite matched filters, and Circular harmonic composite filters in particular, are associated with high sidelobes that degrade the recognition and discrimination performance of the filters. It is proposed to reduce the sidelobes of Circular harmonic composite filters by using coefficients that minimize the energy in the correlation plane, while still taking into account the constraints imposed on the filter by discrimination requirements. Experiments show that the results are much improved when a three-component CHC filter is used.
Access to the requested content is limited to institutions that have purchased or subscribe to SPIE eBooks.
You are receiving this notice because your organization may not have SPIE eBooks access.*
*Shibboleth/Open Athens users─please
sign in
to access your institution's subscriptions.
To obtain this item, you may purchase the complete book in print or electronic format on
SPIE.org.
Design of matched filters containing a single radial harmonic of an object are shown to provide shift and scale invariant correlations. The theory of the expansion, using Mellin radial harmonics, is analyzed and tested experimentally for some binary optical targets. Cross-correlation pattern recognition in the presence of scale and shift invariances has been successfully demonstrated. Two approaches for improving t he correlation peak are presented.
Access to the requested content is limited to institutions that have purchased or subscribe to SPIE eBooks.
You are receiving this notice because your organization may not have SPIE eBooks access.*
*Shibboleth/Open Athens users─please
sign in
to access your institution's subscriptions.
To obtain this item, you may purchase the complete book in print or electronic format on
SPIE.org.
In this paper we are proposing an incoherent optical processor which can perform Abet inversion. Construction of the processor is based on Tchebycheff and Zernike orthogonal polynomial series expansion. It is shown that series expansion coefficients together with series summation, can be easily computed optically (using cylindrical optics). Preliminary experimental results are presented.
Access to the requested content is limited to institutions that have purchased or subscribe to SPIE eBooks.
You are receiving this notice because your organization may not have SPIE eBooks access.*
*Shibboleth/Open Athens users─please
sign in
to access your institution's subscriptions.
To obtain this item, you may purchase the complete book in print or electronic format on
SPIE.org.
Radio wave holography is useful for imaging and diagnostics but it has time delays. Data acquisition is slow if an antenna is mechanically scanned; this delay can be eliminated with a receptor array. Additional delays arise in acquiring and processing data. For example, visible light reconstruction requires producing scale-reduced replicas of holograms. Digital reconstruction has delays in analog-to-digital conversion and in computation. These delays are undesirable for real-time imaging, say of transient events or scenes. This paper demonstrates a method for real-time wavefront reconstruction and presents two implementations. We start with previously formed microwave holograms, and reconstruct first with an infrared radiation and then with millimeter waves. Two reconstruction wavelengths were utilized because of our interest in scaling between formation and reconstruction.
Access to the requested content is limited to institutions that have purchased or subscribe to SPIE eBooks.
You are receiving this notice because your organization may not have SPIE eBooks access.*
*Shibboleth/Open Athens users─please
sign in
to access your institution's subscriptions.
To obtain this item, you may purchase the complete book in print or electronic format on
SPIE.org.
A programmable real-time optical joint transform correlator that generates sharper autocorrelation peaks is presented. In the system, a microchannel spatial light modulator (MSLM) is utilized as the threshold hard-clipping square-law device. A liquid crystal television (LCTV) is used to display input target and reference image at the input plane with a microcomputer. By using the feed-back loop, as depicted in the system schematic, the system would, in principle, offer adaptive processing capability. Basic principle of the system and a preliminary experimental result are given.
Access to the requested content is limited to institutions that have purchased or subscribe to SPIE eBooks.
You are receiving this notice because your organization may not have SPIE eBooks access.*
*Shibboleth/Open Athens users─please
sign in
to access your institution's subscriptions.
To obtain this item, you may purchase the complete book in print or electronic format on
SPIE.org.
Fibre optic architectures based on systolic array concepts and capable of performing a variety of signal processing functions are presented. Analogue processors with increased accuracy have been developed to implement matrix - matrix operations.
Access to the requested content is limited to institutions that have purchased or subscribe to SPIE eBooks.
You are receiving this notice because your organization may not have SPIE eBooks access.*
*Shibboleth/Open Athens users─please
sign in
to access your institution's subscriptions.
To obtain this item, you may purchase the complete book in print or electronic format on
SPIE.org.
An hybrid electronic/optical processor for large scale phased arrays is described . The optical stage , based on a non coherent lensless shadow-casting architecture , is detailed . This system is applied to RIAS air coverage radar data processing (RIAS = Synthetic Aperture Pulse Radar), and experimental results are provided .
Access to the requested content is limited to institutions that have purchased or subscribe to SPIE eBooks.
You are receiving this notice because your organization may not have SPIE eBooks access.*
*Shibboleth/Open Athens users─please
sign in
to access your institution's subscriptions.
To obtain this item, you may purchase the complete book in print or electronic format on
SPIE.org.
The design of a bit-serial optical computer is described. The central components are LiNb03 directional couplers as logic elements, and fiber optic delay lines as memory elements. The key device characteristics are discussed, including the special properties and limitations that are important when designing with them. Following this the key design issues are addressed. These include computation in a realm where propagation delays are much greater than logic delays, and implementation of circuits without flip-flops. Design principles are exemplified by the design of a n-bit counter. Following this, the design for a simple bit-serial computer is described.
Access to the requested content is limited to institutions that have purchased or subscribe to SPIE eBooks.
You are receiving this notice because your organization may not have SPIE eBooks access.*
*Shibboleth/Open Athens users─please
sign in
to access your institution's subscriptions.
To obtain this item, you may purchase the complete book in print or electronic format on
SPIE.org.
The possibility of spatial frequency filtering and correlation analysis of images by double transformation of spatial coherence of light is considered. Experimental setups and results on image processing with partially coherent light are described.
Access to the requested content is limited to institutions that have purchased or subscribe to SPIE eBooks.
You are receiving this notice because your organization may not have SPIE eBooks access.*
*Shibboleth/Open Athens users─please
sign in
to access your institution's subscriptions.
To obtain this item, you may purchase the complete book in print or electronic format on
SPIE.org.
Conoscopic holography is a new principle which permits to record incoherent holograms in a simple and attractive way. The basic principle was first presented by G.SIRAT and Demetri PSALTIS.1,2 The research is now pursued at ENST and a three dimensionnal (3D) machine vision based on this concept will be developped and industrialized by "Le CONOSCOPE". The paper is organized as follow : first, we will briefly review the basic theory of conoscopic holography and explain its main originalities. In the second part, we describe the numerical algorithms under development which permit to retrieve the three dimensionnal image from its conoscopic hologram. In the last part we present the 3D machine vision that we are currently developing and give the main applications of such a system.
Access to the requested content is limited to institutions that have purchased or subscribe to SPIE eBooks.
You are receiving this notice because your organization may not have SPIE eBooks access.*
*Shibboleth/Open Athens users─please
sign in
to access your institution's subscriptions.
To obtain this item, you may purchase the complete book in print or electronic format on
SPIE.org.
Phase-shifting interferometry provides a fast and accurate mean of computing the geometric moments of an image irradiance distribution. The technique is based on a mathematical relationship between the moments and image Fourier transform. The optical system includes a Mach-Zehnder interferometer and a Fourier transforming lens to generate the irradiance spectrum of the interferogram. Its sampling is achieved with a CCD detector arrays. A digital processor interfaced to an image processing system is used to compute the moments of the input function. The accuracy of the method is discussed, and the experimental results are shown.
Access to the requested content is limited to institutions that have purchased or subscribe to SPIE eBooks.
You are receiving this notice because your organization may not have SPIE eBooks access.*
*Shibboleth/Open Athens users─please
sign in
to access your institution's subscriptions.
To obtain this item, you may purchase the complete book in print or electronic format on
SPIE.org.
The conventional falling-raster/folded-spectrum relationship is shown to be a special case of a more general mapping of a 1-D signal and its spectrum into two dimensions. The well-known falling-raster/folded-spectrum relationship [1,2] allows the full two-dimensional (2-D) parallel processing capabilities of a coherent optical spectrum analyzer to be applied to the spectrum analysis of time waveforms of extremely large (106) time-bandwidth product. The basic relationship has been exploited with considerable success in space-integrating, time-integrating, and hybrid space- and time-integrating optical processors. We show here that the conventional falling-raster/folded-spectrum relationship is a special case of a more general mapping of a 1-D signal and its spectrum into two dimensions. This generalized relationship can also be exploited for optical implementation. The conventional falling raster recording format is shown in Fig. 1. The numbers indicate the order in which the lines are recorded. The fall angle 0 is given by 0 = tan-1( NW ) , (1) where I NI is the number of raster lines and W and H are the raster width and height, identified in the figure. The generalized falling raster is obtained by allowing the fall angle to assume more general values given by = tan-1( NW where ' (2) where M and N are relatively prime nonzero integers. A total of 'MI + I NI - 1 raster lines is recorded in a raster of width W and height H. As illustrated in the example of Fig. 2, the raster record is laid down modulo-W in the horizontal direction and modulo-H in the vertical direction. Thus, if the raster line disappears at the right margin it reappears at the left; if it disappears at the bottom it reappears at the top. The conventional falling raster corresponds to the case where I MI = 1.
Access to the requested content is limited to institutions that have purchased or subscribe to SPIE eBooks.
You are receiving this notice because your organization may not have SPIE eBooks access.*
*Shibboleth/Open Athens users─please
sign in
to access your institution's subscriptions.
To obtain this item, you may purchase the complete book in print or electronic format on
SPIE.org.
The basic algebraic properties of Parity Function and the relevant relations are summarized. Two basic Parity Function Modules, PFM2 and PFM3, respectively for two and three variables, are proposed as the building blocks for optical realization of general multi-level parity functions. The performance and the number of key active components, and the levels of logic and/or functional operations required for each of the proposed configurations are given. A unique dual purpose Combined Full Adder-Subtractor, optically cascadable for simultaneous parallel addition and subtraction of multi-bits is presented as an example of simple application involving Parity Functions.
Access to the requested content is limited to institutions that have purchased or subscribe to SPIE eBooks.
You are receiving this notice because your organization may not have SPIE eBooks access.*
*Shibboleth/Open Athens users─please
sign in
to access your institution's subscriptions.
To obtain this item, you may purchase the complete book in print or electronic format on
SPIE.org.
The design of an optical computer must be based on the characteristics of optics and optical technology, and not on those of electronic technology. The property of optical superposition is considered and the implications it has in the design of computing systems is discussed. It can be exploited in the implementation of optical gates, interconnections, and shared memory.
Access to the requested content is limited to institutions that have purchased or subscribe to SPIE eBooks.
You are receiving this notice because your organization may not have SPIE eBooks access.*
*Shibboleth/Open Athens users─please
sign in
to access your institution's subscriptions.
To obtain this item, you may purchase the complete book in print or electronic format on
SPIE.org.
The essential requirements of all-optical digital computing systems are high diffraction efficiency optical interconnects and low switch energy logic elements. In this article we discuss our results in design and construction of holographic optical elements with > 90% diffraction efficiency and switching elements which require < 25 pJ of switching energy.
Access to the requested content is limited to institutions that have purchased or subscribe to SPIE eBooks.
You are receiving this notice because your organization may not have SPIE eBooks access.*
*Shibboleth/Open Athens users─please
sign in
to access your institution's subscriptions.
To obtain this item, you may purchase the complete book in print or electronic format on
SPIE.org.
The results of the experimental study of several interferometric spectrum analysers with CCD time integration are presented. The input was performed either by the direct time modulation of the semiconductor laser emission or by the modulation carried out with an acoustooptic modulator. A multichannel spectrum analysis has been also performed. The frequency resolution at 10 Hz was achieved with the 100 ms time accumulation in the 1 kHz frequency range. To increase the SNR an additional analogue processing was carried out by CCD.
Access to the requested content is limited to institutions that have purchased or subscribe to SPIE eBooks.
You are receiving this notice because your organization may not have SPIE eBooks access.*
*Shibboleth/Open Athens users─please
sign in
to access your institution's subscriptions.
To obtain this item, you may purchase the complete book in print or electronic format on
SPIE.org.
A hybrid optical-digital processor has been developed to generate, filter and reverse the Wegner Distribution Function. The properties of this distribution have allowed us to recover filtered images after performing space variant filtering. Here, we present some experimental results for the case of real images with one-dimensional variation. The same set-up can be also applied to two-dimensional images.
Access to the requested content is limited to institutions that have purchased or subscribe to SPIE eBooks.
You are receiving this notice because your organization may not have SPIE eBooks access.*
*Shibboleth/Open Athens users─please
sign in
to access your institution's subscriptions.
To obtain this item, you may purchase the complete book in print or electronic format on
SPIE.org.
We demonstrate a flexible high-speed correlator that consists of an incoherent-optical system and a fast electronic digital-image-processor.
Access to the requested content is limited to institutions that have purchased or subscribe to SPIE eBooks.
You are receiving this notice because your organization may not have SPIE eBooks access.*
*Shibboleth/Open Athens users─please
sign in
to access your institution's subscriptions.
To obtain this item, you may purchase the complete book in print or electronic format on
SPIE.org.
A real time optical processor for SAR data, using incoherent light, based on "Add and Shift" correlation process, is presented. Problems relative to the use of CCD detectors and a CRT as input of the system are detailed. Early experimental results are produced.
Access to the requested content is limited to institutions that have purchased or subscribe to SPIE eBooks.
You are receiving this notice because your organization may not have SPIE eBooks access.*
*Shibboleth/Open Athens users─please
sign in
to access your institution's subscriptions.
To obtain this item, you may purchase the complete book in print or electronic format on
SPIE.org.
A hybrid optical processor, utlizing three programmable spatial light modulators and a CCD detector, is presented. Two digital optical architectures based on a binary number encoding technique for multiple matrix multiplication have been implemented in the system. Applications in bilinear transformations are also described. Several experimental demonstrations are provided.
Access to the requested content is limited to institutions that have purchased or subscribe to SPIE eBooks.
You are receiving this notice because your organization may not have SPIE eBooks access.*
*Shibboleth/Open Athens users─please
sign in
to access your institution's subscriptions.
To obtain this item, you may purchase the complete book in print or electronic format on
SPIE.org.
Artificial Intelligence problems are often formulated as search trees. These problems suffer from exponential growth of the solution space with problem size. Consequently, solutions based on exhaustive search are generally impractical. To overcome the combinatorial explosion, forward-checking techniques are employed to prune the search space down to a manageable size before or during the actual search procedure. Boolean matrices are a natural data representation for those problems in which the initial problem information is given in the form of a set of binary constraints. In this paper optical implementations of Boolean matrix operations are described for manipulating the constraint matrices to perform forward-checking and thereby increase the search efficiency. Architectural issues affecting the speed of these techniques are discussed.
Access to the requested content is limited to institutions that have purchased or subscribe to SPIE eBooks.
You are receiving this notice because your organization may not have SPIE eBooks access.*
*Shibboleth/Open Athens users─please
sign in
to access your institution's subscriptions.
To obtain this item, you may purchase the complete book in print or electronic format on
SPIE.org.
The optical implementation of weighted interconnections is investigated and basic relationships are derived between the number of neurons, the number of connections and the size of the optical system that is used to perform the connections. Specific methods for selecting the positions of the neurons to achieve the maximum density of independent connections are presented.
Access to the requested content is limited to institutions that have purchased or subscribe to SPIE eBooks.
You are receiving this notice because your organization may not have SPIE eBooks access.*
*Shibboleth/Open Athens users─please
sign in
to access your institution's subscriptions.
To obtain this item, you may purchase the complete book in print or electronic format on
SPIE.org.
System and device technologies for optical neural networks are described, with focusing on the Hopfield associative memories. The optical neural network with a stochastic process is first demonstrated to reduce the undesirable effects of local minima. The new architectures for large-scale neural networks using the time- and frequency-division multiplexing techniques are also proposed, together with their computer simulation results. Finally, several novel optical devices for all-optical neural networks, which include surface-emitting semiconductor laser arrays and ultrafast all-optical switching devices, are briefly described.
Access to the requested content is limited to institutions that have purchased or subscribe to SPIE eBooks.
You are receiving this notice because your organization may not have SPIE eBooks access.*
*Shibboleth/Open Athens users─please
sign in
to access your institution's subscriptions.
To obtain this item, you may purchase the complete book in print or electronic format on
SPIE.org.
A novel optical neural network architecture based on photorefractive technology is presented. Learning with hidden units is handled with a Mean Field Theory algorithm. Beam depletion and rescattering together with decay and absorbtion effects are simulated with encouraging results.
Access to the requested content is limited to institutions that have purchased or subscribe to SPIE eBooks.
You are receiving this notice because your organization may not have SPIE eBooks access.*
*Shibboleth/Open Athens users─please
sign in
to access your institution's subscriptions.
To obtain this item, you may purchase the complete book in print or electronic format on
SPIE.org.
An integrated optics implementation of content addressable associative memory based on the Hopfield Neural Network Model is proposed. The device uses frequency multiplexed surface acoustic waves (FM-SAW) to input the synaptic matrix, which may be stored as a volume hologram using photorefractive memory. Stored information may then be recovered in real time by applying a partial or distorted input to the system. This input is presented in the form of coherent optical pulses, which may be modulated by an external CCD camera. Phase conjugate mirrors are used to obtain the nonlinear thresholding inherent in the Hopfield model, and to provide feedback and gain.
Access to the requested content is limited to institutions that have purchased or subscribe to SPIE eBooks.
You are receiving this notice because your organization may not have SPIE eBooks access.*
*Shibboleth/Open Athens users─please
sign in
to access your institution's subscriptions.
To obtain this item, you may purchase the complete book in print or electronic format on
SPIE.org.
Introducing and optimizing bit-significance to the Hopfield model, ten highly correlated binary images, i.e., numbers "0" to "9", are successfully stored and retrieved in a 6x8 node system. Unlike many other neural networks models, this model has stronger error correction capability for correlated images such as "6", "8", "3", and "9". The bit-significance optimization is regarded as an adaptive learning process based on least-mean-square error algorithm, and may be implemented with another neural nets optimizer. A design for electro-optic implementation including the adaptive optimization networks is also introduced.
Access to the requested content is limited to institutions that have purchased or subscribe to SPIE eBooks.
You are receiving this notice because your organization may not have SPIE eBooks access.*
*Shibboleth/Open Athens users─please
sign in
to access your institution's subscriptions.
To obtain this item, you may purchase the complete book in print or electronic format on
SPIE.org.
Simulations of massively parallel connectionist systems are very time consuming on serial computers . Vector-matrix multiplications, which are a common calculation of many popular connectionist architectures, require 0(n2) operations on serial machines. A simple optical architecture can perform this calculation in nearly constant time, neglecting the machine loading time. The parallelism in such optical designs makes them attractive for implementation of connectionist models. We have demonstrated an optoelectronic connectionist module with modifiable inputs and weight matrices that performs associative memory operations. Low cost commercial liquid crystal television sets are used to encode unit activities as intensity of linearly polarized light, and signed multiplication as rotation of this light. Integrated computer control will allow the extension to many connectionist models.
Access to the requested content is limited to institutions that have purchased or subscribe to SPIE eBooks.
You are receiving this notice because your organization may not have SPIE eBooks access.*
*Shibboleth/Open Athens users─please
sign in
to access your institution's subscriptions.
To obtain this item, you may purchase the complete book in print or electronic format on
SPIE.org.
We develop and optically implement a Spatial Raster scheme for image recognition with neural nets. With this coding, two dimensional optical elements can be used throughout the system.
Access to the requested content is limited to institutions that have purchased or subscribe to SPIE eBooks.
You are receiving this notice because your organization may not have SPIE eBooks access.*
*Shibboleth/Open Athens users─please
sign in
to access your institution's subscriptions.
To obtain this item, you may purchase the complete book in print or electronic format on
SPIE.org.
A new experimental system of optical neurocomputing, called optical associatron, is shown. The experimental system has a simple architecture for optical associative memory. Especially, the system has versatile adaptive learning capabilities implemented by using Microchannel Spatial Light Modulators (MSLMs). In this paper, various experimental results of learning and recalling on the experimental system are shown. Based on the results, behavior of learning on the optical associative memory is discussed and a modified method of learning is proposed and verified on the experimental system. In addition, a method of implementation of error back propagation is proposed.
Access to the requested content is limited to institutions that have purchased or subscribe to SPIE eBooks.
You are receiving this notice because your organization may not have SPIE eBooks access.*
*Shibboleth/Open Athens users─please
sign in
to access your institution's subscriptions.
To obtain this item, you may purchase the complete book in print or electronic format on
SPIE.org.
Optical neural networks for multi-target tracking, inference processors, imaging spectrometer data and matrix inversion are described, and initial results are presented.
Access to the requested content is limited to institutions that have purchased or subscribe to SPIE eBooks.
You are receiving this notice because your organization may not have SPIE eBooks access.*
*Shibboleth/Open Athens users─please
sign in
to access your institution's subscriptions.
To obtain this item, you may purchase the complete book in print or electronic format on
SPIE.org.
The effects of vector concatenation as well as of eliminating the zero diagonal restriction of associative memory neural network matrices are analyzed. Extensive computer simulations seem to indicate that the use of concatenated vectors increases the storage capacity of the association matrices.
Access to the requested content is limited to institutions that have purchased or subscribe to SPIE eBooks.
You are receiving this notice because your organization may not have SPIE eBooks access.*
*Shibboleth/Open Athens users─please
sign in
to access your institution's subscriptions.
To obtain this item, you may purchase the complete book in print or electronic format on
SPIE.org.
We describe a volume holographic associative memory using photorefractive material and conventional planar mirror. Multiple hologram is generated with two angular multiplexed writing beams and Fourier transformed object beam in BaTiO3 crystal at 0.6328 μm. Complete image can be recalled successfully by partial input of original stored image. It is proved that our system is useful for optical implementation of real-time associative memory and location addressable memory.
Access to the requested content is limited to institutions that have purchased or subscribe to SPIE eBooks.
You are receiving this notice because your organization may not have SPIE eBooks access.*
*Shibboleth/Open Athens users─please
sign in
to access your institution's subscriptions.
To obtain this item, you may purchase the complete book in print or electronic format on
SPIE.org.
A new criterion to evaluate the similarity between two vectors in associative memory is presented. According to it, an experimental research about optical associative memory model with threshold modification using complementary vector is carried out. This model is capable of eliminating the posibility to recall erroneously. Therefore the accuracy of reading out is improved.
Access to the requested content is limited to institutions that have purchased or subscribe to SPIE eBooks.
You are receiving this notice because your organization may not have SPIE eBooks access.*
*Shibboleth/Open Athens users─please
sign in
to access your institution's subscriptions.
To obtain this item, you may purchase the complete book in print or electronic format on
SPIE.org.
We show that full interconnect of an NXN input array to an NXN output array with a rank 1 N4 component matrix is quite straightforward for essentially all values of N. A number of applications and implementations are then described.
Access to the requested content is limited to institutions that have purchased or subscribe to SPIE eBooks.
You are receiving this notice because your organization may not have SPIE eBooks access.*
*Shibboleth/Open Athens users─please
sign in
to access your institution's subscriptions.
To obtain this item, you may purchase the complete book in print or electronic format on
SPIE.org.
Here we present results from an optical implementation of a simple neural network. An optically addressed spatial light modulator (SLM) is used to perform thresholding and thin amplitude computer generated holograms perform the weighted interconnections. Two 6-bit vectors and their complements were stored as the memories and the network was found to always converge to the memory closest in terms of Hamming distance to the input vector. This performance is better than that of an electronic simulation of the model and this difference has been found to result from the temporal characteristics of the SLM (a Hughes Liquid Crystal Light Valve (LCLV)). This temporal response behaviour gives greater stability to the memories and as such allows a larger number of memories to be stored on a given size of network. It is believed that this temporal re-sponse can be used to advantage in future neural network implementations using fast nonlinear optical devices.
Access to the requested content is limited to institutions that have purchased or subscribe to SPIE eBooks.
You are receiving this notice because your organization may not have SPIE eBooks access.*
*Shibboleth/Open Athens users─please
sign in
to access your institution's subscriptions.
To obtain this item, you may purchase the complete book in print or electronic format on
SPIE.org.
Symbolic substitution is a spatial logic which is adapted well to optical processing. Unlike Boolean logic it also includes spatial information in the coding. It requires only regular interconnections and nonlinear devices with limited fan-out. Nevertheless the potential of this kind of logic ranges from parallel binary arithmetic to a programmable MIMD processor. Fundamentals of symbolic substitution and its applications are presented.
Access to the requested content is limited to institutions that have purchased or subscribe to SPIE eBooks.
You are receiving this notice because your organization may not have SPIE eBooks access.*
*Shibboleth/Open Athens users─please
sign in
to access your institution's subscriptions.
To obtain this item, you may purchase the complete book in print or electronic format on
SPIE.org.
Symbolic substitution is a method of computing based on parallel binary pattern replacement, that can be implemented with simple optical components and regular free-space interconnection schemes. A two-dimensional pattern is searched for in parallel in an array and is replaced with another pattern. Pattern transformation rules can be applied sequentially or in parallel to realize complex functions. When the substitution space is modified to be loge SIT connected for N binary spots, and masks are allowed to customize the system, then optical digital circuits using symbolic substitution for network interconnects can be made nearly as efficient in terms of gate count and circuit depth as conventional arbitrary interconnection schemes allow. We describe an optical setup that requires no more than a fanin and fanout of two using optically nonlinear logic devices and a free space interconnection scheme based on symbolic substitution.
Access to the requested content is limited to institutions that have purchased or subscribe to SPIE eBooks.
You are receiving this notice because your organization may not have SPIE eBooks access.*
*Shibboleth/Open Athens users─please
sign in
to access your institution's subscriptions.
To obtain this item, you may purchase the complete book in print or electronic format on
SPIE.org.
The Optical Parallel Array Logic System (OPALS) is a parallel optical digital computing system using optical array logic as its operational basis. Its salient features are the capability of implementing fully 2-D parallel logical operation, parallel neighborhood operation, iterative processing, and programmability. In this paper, we represent optical array logic and realizable versions of OPALS's.1-4 Then we briefly describe programming in optical array logic and the concept of pattern logic useful for space-variant processing with optical array logic.
Access to the requested content is limited to institutions that have purchased or subscribe to SPIE eBooks.
You are receiving this notice because your organization may not have SPIE eBooks access.*
*Shibboleth/Open Athens users─please
sign in
to access your institution's subscriptions.
To obtain this item, you may purchase the complete book in print or electronic format on
SPIE.org.
Previous optical computing schemes offered analog or quasi-digital accuracies with a single fixed primitive. This paper describes how programmable, arbitrary bit length all digital Central Processing Unit (CPU) computations are now possible. In addition, the current state-of-the-art in optical computer subsystem devices such as acousto-optic modulators, detector and source arrays, posture this architecture as a revolutionary technology in and of itself, as it may be applied to an implementation plethora.
Access to the requested content is limited to institutions that have purchased or subscribe to SPIE eBooks.
You are receiving this notice because your organization may not have SPIE eBooks access.*
*Shibboleth/Open Athens users─please
sign in
to access your institution's subscriptions.
To obtain this item, you may purchase the complete book in print or electronic format on
SPIE.org.
Sequential Optical Computing seems to be promising for implementing various algorithms with a unique architecture. Algorithmic implementation may be performed via a sequence of operations on two components : synthetic holograms and a logic gate array. Using this architecture, we discuss herein the use of a Micro-channel Spatial Light Modulator (MSLM) as the 2D logic gate, and demonstrate optical arithmetics in a basic experiment.
Access to the requested content is limited to institutions that have purchased or subscribe to SPIE eBooks.
You are receiving this notice because your organization may not have SPIE eBooks access.*
*Shibboleth/Open Athens users─please
sign in
to access your institution's subscriptions.
To obtain this item, you may purchase the complete book in print or electronic format on
SPIE.org.
An optical space-variant technique based on spatial encoding and superposition of a operation mask with two encoded input patterns have been proposed. A simple method of spatial encoding and the experimental results are described. Experimentally, XOR operation is made between input patterns and code patterns to encode automatically input patterns. This automatic generation of the encoded input patterns and a superposition of the encoded patterns are performed by using a micro-channel plate spatial light modulator. Results of parallel logic operation are obtained by superimposing on two encoded input patterns and the operation mask.
Access to the requested content is limited to institutions that have purchased or subscribe to SPIE eBooks.
You are receiving this notice because your organization may not have SPIE eBooks access.*
*Shibboleth/Open Athens users─please
sign in
to access your institution's subscriptions.
To obtain this item, you may purchase the complete book in print or electronic format on
SPIE.org.
Optical parallel processor which performs cascade logic operations is newly constructed. The encoding concept using polarization state of light and the separation of the encoded light together are the key to the programmability for any logic operation. The operation kernel determines the logic operation to be executed only by spatially filtering spatially the encoded light. Real-time programmable cascade logic operations for three binary patterns have been demonstrated experimentally with the optical processor using a liquid-crystal spatial light modulator as the operation kernel. Up to five-step cascede operation is anticipated by using microchannel spatial light modulator having photomultifying capability as the polarization encoder.
Access to the requested content is limited to institutions that have purchased or subscribe to SPIE eBooks.
You are receiving this notice because your organization may not have SPIE eBooks access.*
*Shibboleth/Open Athens users─please
sign in
to access your institution's subscriptions.
To obtain this item, you may purchase the complete book in print or electronic format on
SPIE.org.
Polychromatism in optical computers (OC) will increase the functional abilities and operation speed of the OC structural blocks, OC as a whole and of OC nets. Polychromatic processor, bichromatic logic and polychromatic multiplier are described.
Access to the requested content is limited to institutions that have purchased or subscribe to SPIE eBooks.
You are receiving this notice because your organization may not have SPIE eBooks access.*
*Shibboleth/Open Athens users─please
sign in
to access your institution's subscriptions.
To obtain this item, you may purchase the complete book in print or electronic format on
SPIE.org.
Optical parallel logic device using twisted nematic liquid crystal cells doped with dichroic dyes and dichroic filters are proposed. Operation of this device is expressed by combination of transfer functions of simple 2X2 matrices. This device can perform several logic operations simultaneously by different wavelength and/or polarized lights. An iterative system can also be composed by using this device and template matching is examined by this system.
Access to the requested content is limited to institutions that have purchased or subscribe to SPIE eBooks.
You are receiving this notice because your organization may not have SPIE eBooks access.*
*Shibboleth/Open Athens users─please
sign in
to access your institution's subscriptions.
To obtain this item, you may purchase the complete book in print or electronic format on
SPIE.org.
We describe the principle of an optical digital processor for mathematical morphology. The proposed central processing unit is based on binary logic using states of polarization and spatial data shift using geometric optics. The system is shown to be able to perform iterative operations. Two possible optical implementations are then discussed.
Access to the requested content is limited to institutions that have purchased or subscribe to SPIE eBooks.
You are receiving this notice because your organization may not have SPIE eBooks access.*
*Shibboleth/Open Athens users─please
sign in
to access your institution's subscriptions.
To obtain this item, you may purchase the complete book in print or electronic format on
SPIE.org.
We propose an optical arrangement for symbolic substitution based on diffraction gratings and Fourier filtering. It features: handling symbols with many pixels, multiple-channel and dual-rail (true/false) substitution, programmability. Experimental results are presented.
Access to the requested content is limited to institutions that have purchased or subscribe to SPIE eBooks.
You are receiving this notice because your organization may not have SPIE eBooks access.*
*Shibboleth/Open Athens users─please
sign in
to access your institution's subscriptions.
To obtain this item, you may purchase the complete book in print or electronic format on
SPIE.org.
The Lau effect is applied to implement the whole set of binary logic operations optically. Our technique works with spatially incoherent light and does not require lenses or any other optical accessory.
Access to the requested content is limited to institutions that have purchased or subscribe to SPIE eBooks.
You are receiving this notice because your organization may not have SPIE eBooks access.*
*Shibboleth/Open Athens users─please
sign in
to access your institution's subscriptions.
To obtain this item, you may purchase the complete book in print or electronic format on
SPIE.org.
Symbolic substitution is a method for manipulating binary data that depends on both the value of the data and its spatial location to realize logical operations [1,2]. A substitution system requires only a pattern recognizer, a nonlinear device, and a pattern scriber. The operation of both the recognition and scribing subsystems is based on the replication of an input object to produce several output images; the replicated images are then translated and overlayed. For optical implementation of symbolic substitution the recognizer and scriber systems can be constructed using classical and holographic optical elements; single-channel and dual-channel systems have been proposed [3]. In a single-channel system holographic elements are used either to replicate an input object or combine several shifted images. The holgraphic elements in a dual-channel system perform both replication and translation. A zero-based representation of binary-phase gratings developed by Dammann [4] has been reported as one method for designing beam splitters and combiners [5,6]. Dammann's method assumes a symmetric one-dimensional display of the replicated images and determines the locations of phase changes based on the number of desired replicas. The application of this method to the design of systems for symbolic substitution requires an extension of Dammann's method to asymmetric displays [3] and an extension to two dimensions. These extensions are discussed herein and results are also presented. A brief review of the requirements for image replication and combination in an optical symbolic substitution system is presented in Section 2. Section 3 discusses Dammann's method and its extensions. Practical considerations with respect to design are discussed in Section 4 and results are presented. Some limitations of the implementation of Dammann's method as well as other concluding remarks are discussed in Section 5.
Access to the requested content is limited to institutions that have purchased or subscribe to SPIE eBooks.
You are receiving this notice because your organization may not have SPIE eBooks access.*
*Shibboleth/Open Athens users─please
sign in
to access your institution's subscriptions.
To obtain this item, you may purchase the complete book in print or electronic format on
SPIE.org.
A new optical system for symbolic substitution is proposed. The system is composed of the lensless shadow-casting system combined with optical logical array and optical image storage elements. The LED's (light-emitting diode) are used as both a light source and a control element in the operation, so that symbolic substitution is implemented by all-photoelectric manipulation. Experimental principle, procedure and result are given.
Access to the requested content is limited to institutions that have purchased or subscribe to SPIE eBooks.
You are receiving this notice because your organization may not have SPIE eBooks access.*
*Shibboleth/Open Athens users─please
sign in
to access your institution's subscriptions.
To obtain this item, you may purchase the complete book in print or electronic format on
SPIE.org.
The theory of mathematical morphology is well established and used in digital electronic image processing. The required operations are highly matched to the capabilities of space invariant optical processors. Here we describe two optical architectures for implementing these basic morphological operations: A symbolic substitution (SS) processing architecture and a new SIMD architecture based on Dual Input Array Logic (DIAL).
Access to the requested content is limited to institutions that have purchased or subscribe to SPIE eBooks.
You are receiving this notice because your organization may not have SPIE eBooks access.*
*Shibboleth/Open Athens users─please
sign in
to access your institution's subscriptions.
To obtain this item, you may purchase the complete book in print or electronic format on
SPIE.org.
The architecture of an electronic mesh array for image processing by simulated annealing is described. On the algorithm side, it is well suited to cases where the input data is a grey-level picture, the output data is a binary picture, and the energy model is a generalization of the Ising model with external field. The example of digital halftoning is covered. On the implementation side, this architecture is intended for VLSI monolithic integration. Its PE uses both analog and digital devices, and is specifically designed to take advantage of an optical random number generator. These features result in a compact processor array with near real time performances for binary pictures stochastic relaxation.
Access to the requested content is limited to institutions that have purchased or subscribe to SPIE eBooks.
You are receiving this notice because your organization may not have SPIE eBooks access.*
*Shibboleth/Open Athens users─please
sign in
to access your institution's subscriptions.
To obtain this item, you may purchase the complete book in print or electronic format on
SPIE.org.
In this paper we consider the computation of combinational boolean equations on a regularly interconnected digital optical computer. First, we provide a new technique by, which any boolean circuit can be mapped on to this architecture maintaining a favorable space-time product. Then we present a systematic method of realizing higher level boolean functions in terms of lower level functions.
Access to the requested content is limited to institutions that have purchased or subscribe to SPIE eBooks.
You are receiving this notice because your organization may not have SPIE eBooks access.*
*Shibboleth/Open Athens users─please
sign in
to access your institution's subscriptions.
To obtain this item, you may purchase the complete book in print or electronic format on
SPIE.org.
A real-time encoding and programming method based on array logic has been proposed by using a polarization rotation in an electro-optic effect and stripe-structured analyzers. We have demonstrated a programmable image processor with stacked liquid crystal panels which performs several parallel operations at video rates.
Access to the requested content is limited to institutions that have purchased or subscribe to SPIE eBooks.
You are receiving this notice because your organization may not have SPIE eBooks access.*
*Shibboleth/Open Athens users─please
sign in
to access your institution's subscriptions.
To obtain this item, you may purchase the complete book in print or electronic format on
SPIE.org.
We describe an optical cellular automaton implementing by symbolic substitution the hexagonal lattice gas automata using the "Optical Shadow Casting" method.
Access to the requested content is limited to institutions that have purchased or subscribe to SPIE eBooks.
You are receiving this notice because your organization may not have SPIE eBooks access.*
*Shibboleth/Open Athens users─please
sign in
to access your institution's subscriptions.
To obtain this item, you may purchase the complete book in print or electronic format on
SPIE.org.
A processing element of a prototype digital optical cellular image processor (DOCIP) is implemented to demonstrate a particular parallel computing and interconnection architecture. This experimental digital optical computing system consists of a 2-D array of 54 optical logic gates, a 2-D array of 53 subholograms to provide interconnections between gates, and electronic input/output interfaces. The multi-facet interconnection hologram used in this system is fabricated by a computer-controlled optical system to offer very flexible interconnections.
Access to the requested content is limited to institutions that have purchased or subscribe to SPIE eBooks.
You are receiving this notice because your organization may not have SPIE eBooks access.*
*Shibboleth/Open Athens users─please
sign in
to access your institution's subscriptions.
To obtain this item, you may purchase the complete book in print or electronic format on
SPIE.org.
We derive the computational limits of optical technology using a proposed model, and present parallel architectures for implementing it. We show the superiority of the unit delay optical interconnection network of these designs by studying efficient algorithms for many problems in image processing and AI.
Access to the requested content is limited to institutions that have purchased or subscribe to SPIE eBooks.
You are receiving this notice because your organization may not have SPIE eBooks access.*
*Shibboleth/Open Athens users─please
sign in
to access your institution's subscriptions.
To obtain this item, you may purchase the complete book in print or electronic format on
SPIE.org.
Recent work on optical table look-up devices has been generalized to include control functions. These functions introduce programmability to this approach to optical digital computing. A general formalism for performing logic functions using positionally encoded look-up tables is developed and applied to a programmable Boolean gate and a residue arithmetic module. Hardware implementations of the tables are described.
Access to the requested content is limited to institutions that have purchased or subscribe to SPIE eBooks.
You are receiving this notice because your organization may not have SPIE eBooks access.*
*Shibboleth/Open Athens users─please
sign in
to access your institution's subscriptions.
To obtain this item, you may purchase the complete book in print or electronic format on
SPIE.org.
The algebraic properties of binary image algebra (BIA) are used to mathematically describe and compare symbolic substitution and pattern recognition algorithms for implementation on optical processors.
Access to the requested content is limited to institutions that have purchased or subscribe to SPIE eBooks.
You are receiving this notice because your organization may not have SPIE eBooks access.*
*Shibboleth/Open Athens users─please
sign in
to access your institution's subscriptions.
To obtain this item, you may purchase the complete book in print or electronic format on
SPIE.org.
An optical symbolic processor that performs logic, numeric, morphological and propositional calculus operations is described and initial laboratory results are presented.
Access to the requested content is limited to institutions that have purchased or subscribe to SPIE eBooks.
You are receiving this notice because your organization may not have SPIE eBooks access.*
*Shibboleth/Open Athens users─please
sign in
to access your institution's subscriptions.
To obtain this item, you may purchase the complete book in print or electronic format on
SPIE.org.