The noise characteristics of the photon echo memory have been investigated. The photon echo memory has the ability to store many bits of information in a diffraction-limited spot, thereby dramatically increasing the storage density. The temporal Fourier transform of the input data sequence is written into the inhomogeneous absorption profile of the recording medium. Data are encoded by means of temporal modulation onto the waveform of a finite duration data beam. Individual bits are not localized to a specific spectral channel; instead, they are stored throughout a region of spectral-addressing space, In order to store and recall the input data accurately, the Fourier transform of the input data sequence must be narrower than the inhomogeneous bandwidth. In the photon echo memory mechanism, there are several factors affecting the system bit error rate such as finite-width write/read pulses, echo efficiency, shot noise, thermal noise, etc. The accuracy of the echo output depends on those system factors. In this paper we formulate a simple model of the photon echo system, and by analyzing this model we derive the relationship between the characteristics of the echo output signal and several factors such as the bandwidth of the system, echo efficiency, atom excited state population fluctuations, shot noise, and thermal noise.
The possibility of reducing information loss in holographic memory systems in which different wavelengths are used for data recording and reconstruction has been investigated. We suggest replacing a thick recording medium which provides selective reconstruction of the holographic data with a multilayer recording structure. It has been shown that such a structure has the selectivity which significantly exceeds the selectivity of a single layer. At the same time it permits compensation for the Bragg mismatch along the thickness of the recording medium when 2-wavelength reconstruction is implemented. We consider various methods for such compensation and the possibility of their realizations.
We present a multiple-input, single-output, weakly nonlinear model of a liquid crystal light valve using a second-order Volterra series and describe an experimental method to measure the nonlinear transfer functions using sinusoidal perturbation and synchronous detection with a lock-in amplifier. Experimentally measured and estimated nonlinear transfer functions are presented. We next discuss the response of the liquid crystal light valve to random inputs and present experimental noise measurements.
The application of volume M-type holograms for building multichannel geometries in pattern recognition system is considered. The results of theoretical and experimental investigations of the hologram's parameters as a function of their recording parameters and their use as filters in correlation setups are presented. Multichannel correlation schemes where the processed signals have either different or the same wavelength are proposed. We have shown that all the correlation schemes proposed allow one to increase data throughput several times over single- channel BR-based correlators.
The degree of protection for common types of security holograms is reviewed. The possibility of hologram recording directly from the holographic film being used for protection is demonstrated. We suggest the structure of security hologram allows one to prevent such unwanted copying. We also propose a detection system which provides for fast testing of a hologram's authenticity.
Development of optical information processing systems requires accurate models of the spatial light modulators employed in these systems. These models must recognize the multiport and non-linear nature of these devices. Here we describe the temporal characteristics of spatial light modulators by a quadratic multiport model. We use spectrum analysis and sinusoidal perturbation with synchronous detection to observe the nonlinear characteristics of a Hughes Model 4050 liquid crystal light valve. We present experimental results demonstrating some of the nonlinear characteristics.
The application of multiple error-correction codes to optical matrix-vector multipliers (OMVMs) can improve the computational accuracy level of these processors. A binary Bose Ray-Chaudhuri (BCH) code was applied to a simulated mod-2 OMVM. Based on the results obtained from the simulations, the conditions under which the use of such error-correction coding is feasible in OMVMs are discussed.
Holographic data storage in BR (bacteriorhodopsin) films has been investigated. The BR film has a fast response time, with high quantum efficiency, high sensitivity, and high spatial resolution. Because of these properties, BR film is attractive as an optical storage medium. Moreover, due to the high spatial resolution, a multiplexing storage scheme can be applied to store a large amount of data. We have demonstrated the possibility of using a phase-multiplexing technique to store the data in BR films. We also show the possibility of using phase masks for spectrum spreading and encoding.
Within this paper we discuss the precision of a fixed-point discrete numeric vector-matrix processor. The concept of precision relates to the quantity of numerically exact information in the processor's calculated result. This analysis is based on a signal space formulation which allows for the determination of precision from the uncertainty in the output signal space. Characteristics of the ideal signal space are explored and then the implications of simple nonideal temporal and spatial effects are considered. The resulting precision limitations are discussed in terms of numerical roundoff and significance errors.
A simple complex-amplitude model is developed and used to establish the behavior of amplitude an polarization state noise. A simple experiment based upon this model shows that the dominate source of noise in a twisted nematic liquid crystal cell results from polarization state fluctuations. A Stokes polarimeter is described and then used to measure the polarization state dynamics of a commercially available liquid crystal modulator.
A generic three-plane optical processor is investigated from a statistical viewpoint. The means and the mutual coherence functions of the output field amplitude and intensity are derived. The photodetection process is then studied, and the mean and the autocorrelation function of the output current are found, thus establishing the functional form of the signal dependence of noise at the processor output. An integral equation and a series expression are also presented for the probability density function of the output signal. A special case is then analyzed, and the use of these expressions is demonstrated. Finally, device models to be used within this framework are summarized.
Analog optical matrix-vector multipliers (OMVMs) compute matrix-vector products rapidly due to the parallelism and high speed of optics. However, their low analog accuracy hinders their widespread application. Digital partitioning is a promising new technique for achieving high accuracy computations on analog OMVMs. One potential drawback of digital partitioning is its sensitivity to errors. The results presented here show how error-correcting codes can reduce this sensitivity to errors. Various computer simulation results are presented to show that a significant reduction of error can be obtained.
Weighted outer product processing, which is a generalization of the bilinear transform, has applications in combinational logic design. An optical implementation of such a processor for multiple-input, multiple-data logic function operation using Texas Instruments' deformable mirror device (DMD) for real-time input is presented. The DMD, a one-dimensional spatial light modulator, allows real-time operation of such a logic function implementation by providing controllable optical encoding ofthe two input vectors that are required forthe outer product. Applications of liquid crystal television and magneto-optic spatial light modulators as real-time weighting matrices are also presented.
An optical quadratic neural network utilizing four-wave mixing in barium titanate (BaTiO3) has been developed. This network implements a feedback loop using a CCD camera, a microcomputer, two monochrome liquid crystal televisions, and various optical elements. For training, the network employs the supervised quadratic perceptron algorithm to associate binary-valued input vectors with specified training vectors. Using a spatial multiplexing scheme for two bipolar neurons, the quadratic network was able to associate an input vector with various target vectors. In addition, the network successfully associated two input vectors with two corresponding target vectors in the same training session. Both analytical and experimental results are presented.
Proc. SPIE. 1564, Optical Information Processing Systems and Architectures III
KEYWORDS: Optical signal processing, Data storage, Error analysis, Interference (communication), Computer programming, Computer simulations, Forward error correction, Error control coding, Signal detection, Binary data
Optical algebraic processors can perform complex calculations in parallel and at high speeds. However, they commonly suffer from a low analog accuracy which hinders their widespread application. Error detection and correction codes provide one technique for improving the accuracy of optical algebraic processors. The use of these codes would allow some of the errors that may occur during a computation to be detected and possibly corrected. This paper describes the results of various computer simulations of optical matrix-vector multipliers employing error-correction codes. It discusses the application of convolutional codes to optical matrix-vector multipliers along with several block codes. Both binary and nonbinary codes are considered. The results indicate that a significant improvement in performance can be obtained when compared with processors not employing error-correction codes. Also, the type of noise, whether signal-independent or signal-dependent noise, has a significant effect on the performance of a matrix-vector multiplier employing an error code. The encoding and decoding operations required for the error codes can be performed optically.
The results of two electro-optical implementations of the alternating projection neural network (APNN) are presented. Both configurations use optical matrix-vector multipliers with electronic feedback, but different spatial light modulators are employed. The first implementation is passive, employing a photographic transparency that intensity-modulates the incident incoherent light. In the second implementation, two spatial light modulators actively modulate the light, permitting real-time updating of the interconnection matrix elements. To the authors' knowledge, the implementations described are the first of the APNN.
Upper arid lower bounds on the number of bits of accuracy achievable are determined by applying a seconth-ortler statistical model to the linear algebra processor. The use of bounds was found necessary due to the strong signal-dependence of the noise at the output of the optical linear algebra processor (OLAP). 1 1. ACCURACY BOUNDS One of the limiting factors in applying OLAPs to real world problems has been the poor achievable accuracy of these processors. Little previous research has been done on determining noise sources from a systems perspective which would include noise generated in the multiplication ard addition operations spatial variations across arrays and crosstalk. We have previously examined these noise sources and determined a general model for the output noise mean and variance. The model demonstrates a strony signaldependency in the noise at the output of the processor which has been confirmed by our experiments. 1 We define accuracy similar to its definition for an analog signal input to an analog-to-digital (ND) converter. The number of bits of accuracy achievable is related to the log (base 2) of the number of separable levels at the P/D converter output. The number of separable levels is fouri by dividing the dynamic range by m times the standard deviation of the signal a. 2 Here m determines the error rate in the P/D conversion. The dynamic range can be expressed as the