The USC Andrew and Erna Viterbi School of Engineering USC Signal and Image Processing Institute USC Ming Hsieh Department of Electrical Engineering University of Southern California

Technical Report USC-SIPI-291

“Optical Shared Memory Computing and Multiple Access Protocols for Photonic Networks”

by Kuang-Yu Li

December 1995

An optical implementation of a shared memory uses a single photorefractive crystal and can realize the set of memory modules in a digital shared memory computer. A complete instruction set consists of Read, Write, Selective Erase, and Refresh, which can be applied to any memory module independent of (and in parallel with) instructions to the other memory modules. In addition, a memory module can execute a sequence of Read operations simultaneously with the execution of a Write operation to accommodate differences in optical recording and readout times common to optical volume storage media. An experimental shared memory system is demonstrated and its projected performance is analyzed.

A multiplexing technique is presented to significantly reduce both grating-and beam-degeneracy crosstalk in volume holographic systems, by incorporating space, angle, and wavelength as the multiplexing parameters. In this approach, each hologram, which results from the interference between a single input node and an object array, partially overlaps with the other holograms in its neighborhood. This technique can offer improved interconnection density, optical throughput, signal fidelity, and space-bandwidth product utilization. Design principles and numerical simulation results are presented.

A free-space photonic cellular hypercube parallel computer, with emphasis on the design of a collisionless multiple access protocol, is presented. This design incorporates wavelength-, space-, and time-multiplexing to achieve multiple access, wavelength reuse, dense connectivity, collisionless communications, and a simple control mechanism. Analytic models based on semi-Markov processes are employed to analyze this protocol. The performance of the proposed protocol is evaluated in terms of network throughput and data packet delay with variations in data packet length, buffer size, and system size; it is then compared with two other protocols.

To download the report in PDF format click here: USC-SIPI-291.pdf (24.3Mb)