Development of a fine-grained parallel Karhunen-Loeve transform

摘要

A high-performance Karhunen-Loeve transform for multi-spectral imagery suitable for remote sensing applications has been prototyped on a platform FPGA, through hardware compilation onto a PC-based development board. Performance estimates suggest that the design will outperform implementation on a high-end microprocessor, given due attention to I/O (input/output). Detailed analysis of the design steps taken to produce a successful prototype are given. A design that addresses the issue of data bandwidth is included. General conclusions are reached for the utility of this architecture and design method for fine-grained parallel processing.