Algorithms--Data processing

This thesis is concerned with adapting a sequential code that calculates the Radar Cross Section (RCS) of an open-ended rectangular waveguide cavity to a massively parallel computational platform. The primary motivation for doing this is to obtain wideband data over a large range of incident angles in order to generate a two-dimensional radar cross section image. Images generated from measured and computed data will be compared to evaluate program performance. The computer used in this implementation is a MasPar MP-1 single instruction, multiple data massively parallel computer consisting of 4,096 processors arranged in a two-dimensional mesh. The algorithm uses the mode matching method of analysis to match fields over the cavity aperture to obtain an expression for the scattered far field.

Alopex is a biologically influenced computation paradigm that uses a stochastic procedure to find the global optimum of linear and nonlinear functions. It maps to a hierarchical SIMD (Single-Instruction-Multiple-Data) architecture with simple neuronal processing elements (PE's), therefore the large amount of interconnects in other types of neural networks are not required and more efficient utilization of chip level and board level "real estate" is realized. In this study, verifications were performed on the use of a simplified Alopex algorithm in handwritten digit recognition with the intent that the verified algorithm be digitally implementable. The inputs to the simulated Alopex hardware are a set of 32 features extracted from the input characters. Although the goal of verifying the algorithm was not achieved, a firm direction for future studies has been established and a flexible software model for these future studies is available.