Parallel and Distributed Computing Architectures and Algorithms for Fault-Tolerant Sonar Arrays

摘要

This report summarizes the progress and results of the third year of a three-year study whose goal is the use of fault-tolerant distributed and parallel processing techniques to decrease the cost and improve the performance and reliability of large, battery-powered, autonomous sonar arrays. In FY98, work has been completed for single-aperture conventional parallel beamforming (CBF), recently completed for split-aperture parallel beamforming (SA-CBF), and initiated for more advanced adaptive or optimal processing algorithms. A comprehensive set of performance experiments has been conducted with the new CBF and SA-CBF parallel algorithms on both clustered workstation testbeds and simulated array architectures, and the results have illuminated the strengths and weaknesses of the algorithms. Several of the new CBF and SA-CBF parallel algorithms developed in this project have been shown to have promising performance and scalability characteristics in terms of processing and memory capacity. In addition, to support sensitivity studies for the mapping of parallel algorithms to array architectures, a new rapid virtual prototyping tool for the design and analysis of distributed parallel sonar arrays has been completed and demonstrated. Finally, an 8-node distributed parallel array prototype has been designed, developed, fabricated, and tested, and promising preliminary results have been attained.