Berlin Ha

Ring-array processor distribution topology for optical interconnects

The existing linear and rectangular processor distribution topologies for optical interconnects, although promising in many respects, cannot solve problems such as clock skews, the lack of supporting elements for efficient optical implementation, etc. The use of a ring-array processor distribution topology, however, can overcome these problems. Here, a study of the ring-array topology is conducted with an aim of implementing various fast clock rate, high-performance, compact optical networks for digital electronic multiprocessor computers. Practical design issues are addressed. Some proof-of-principle experimental results are included.

Parallel optical pyramidal image processing

Pyramidal processing is a form of multiresolution image processing in which the image is decomposed into a sequence of images at different resolutions. Pyramidal processing aims to extract and interpret significant features of an image at different resolutions. Digital pyramidal image processing, because of the large number of convolution- type operations, is time consuming. On the other hand, optical pyramidal processors, described here, are preferable in real-time image-understanding applications because of their ease in performing convolution operations. Preliminary experimental results for optical Gaussian and Laplacian pyramidal image processing are presented.

Free-space optical collinear crossover interconnects

Two new optical free-space collinear cross-over interconnect schemes are suggested. The first optical implementation uses mirrors and beam splitters, while the second uses a Fresnel zone plate and lens combination. Some proof-of-principle experimental results are also presented.

How Large Can an Optical Matrix-Matrix Multiplier be Constructed ?

In this talk, based on the key operations governing all these matrix-matrix multipliers, we propose a statistical model to quantitatively analyze the numerical accuracy that these multipliers could deliver. Our study indicates that no matter what particular scheme is used, statistically the numerical accuracy limit caused by the use of analog computation mechanism is much more serious than the dynamic range and other limits to a matrix-matrix multiplier. Our study also shows that when these multipliers are used for, instead of generating algebraic results, the single threshold operations, such as the operations required to implement a programmable logic array, a content- addressable memory, and a neural network, a better accuracy can be expected depending on the used threshold positions.

Fast digital optical multiplication using an array of binary symmetric logic counters

Because of the lack of fast, accurate, and large dynamic range analog-to-digital converters (ADCs), optical implementation of the digital multiplication through analog convolution (DMAC) algorithm yields a slow digital multiplier. By replacing both the optical adder and ADC arrays by an optical combinatorial logic counter array, a new optical fast digital multiplication method is proposed. Compared to the existing optical DMAC scheme, the new method promises both higher processing speed and accuracy. A comparison of this and some of the other optical and electronic fast digital multiplication schemes is also presented.

Optical position-coded multiple-valued logic and arithmetic using liquid-crystal TVs and holograms

Abstract An optical implementation of multiple-valued logic (MVL) operators that employs an optical holographic look-up table addressed by a pair of position-encoded liquid-crystal TVs is proposed. Realizations of optical modified signed-digit (MSD) as well as Post logic elements are included. Some first-order experimental results are presented.

Optical Binary Symmetric Logic Functions And Their Applications

The binary symmetric logic function (BSLF), because of its invariance under the permutation of its input variables, is an important class of Boolean logic functions. The diversified BSLF applications include the synthesis of a full adder and subtractor, a text comparator, a median filter, a parity checker, and various threshold elements. The classic BSLF realization uses an array of triangularly interconnected slow-speed electric contact switches. Optical switches, because of their fast switching capability, are excellent candidates for an optical BSLF (OBSLF) implementation. In this paper, various OBSLF architectures are described, together with their applications to optical digital and symbolic computing, data communication, image processing, and neural networks.

Optical implementation of binary symmetric logic functions

Optical implementations of binary multiple-variable symmetric logic functions are proposed. By using a triangular array of lossless beam splitters together with optical on-off switches, an optical binary symmetric logic module (OBSLM) is experimentally implemented. The applications of the OBSLM to optical digital, symbolic, and neural computing are discussed.

Reflective optical ring-array interconnects: an optical system design study.

We present a reflective optical ring-array interconnect architecture for handling data routings under various single-instruction-multiple-data array processing environments. The proposed architecture can perform clock-skew-free optical data communications for either a fixed-degree model, such as a nearest-neighbor network, or a variable-degree model, such as a plus-minus-2(i) network. It is found that space-variant routings, which are difficult to perform in a rectangular-array opto-electronic integrated circuit, can easily be mapped into rotation-invariant routines for optical implementation by a ring opto-electronic integrated-circuit array. Our system study also shows that the design of the optical imaging system for interconnecting a ring array of nodes is much easier than that for interconnecting a conventional rectangular-array topology. Design principles for both the individual optical components and the entire optical system are described. The optical network performance parameters, such as the diffraction- and aberration-related processing capabilities, the optical transmitter coupling efficiency, the optical free-space power distribution loss, and the power-dependent element bit rate, are analyzed.

Optical interconnect for a ring array of single-instruction–multiple-data processors

A novel optical interconnect scheme for a ring of single-instruction-multiple-data processing elements is proposed. This unexplored optical interconnect topology promises several important advantages over the conventional rectangular array topology, such as the use of space-invariant elements and identical interconnect latency for implementation.