Accession Number : ADA270427

Title :   Development of Parallel Architectures for Sensor Array Processing. Volume 2. A Parallel Architecture for Broad-Band Direction-Of-Arrival Estimation.

Descriptive Note : Final rept.,

Corporate Author : TOLEDO UNIV OH DEPT OF ELECTRICAL ENGINEERING

Personal Author(s) : Jamali, M. M. ; Kwatra, S. C. ; Tabar, Raja F.

Report Date : AUG 1993

Pagination or Media Count : 138

Abstract : One of today's problems in signal processing is the identification of direction-of-arrival (DOA) for multiple broad-band sources. Many algorithms have been proposed in the literature, however, the algorithms are highly complex and require a lot of computing power. New technologies in the past two decades have decreased the cost of digital hardware, and its speed has increased to such an extent that digital signal processing has replaced a great deal of analog signal processing. This thesis presents a digital-domain parallel-pipelined architecture capable of computing the DOA of multiple broad-band sources in real-time. Broad-Band Signal-Subspace Spatial-Spectrum (BASS-ALE) Estimation algorithm has been selected for DOA estimation and has been simulated and verified through computer software programs. The BASS-ALE algorithm has been modified and parallelized. The parallel algorithm has been mapped on an architecture which is suitable for real-time computation.

Descriptors :   *COMPUTER ARCHITECTURE, *PARALLEL PROCESSING, *COMPUTER APPLICATIONS, *SONAR ARRAYS, ALGORITHMS, ANALOG SIGNALS, ANALOGS, ARRIVAL, COMPUTATIONS, COMPUTERS, COSTS, IDENTIFICATION, POWER, REAL TIME, SIGNAL PROCESSING, SIGNALS, THESES, VELOCITY, SOURCES, POSITION(LOCATION), BROADBAND, INTEGRATED CIRCUITS, SYSTEMS APPROACH.

Subject Categories : Computer Programming and Software
      Computer Hardware
      Acoustic Detection and Detectors

Distribution Statement : APPROVED FOR PUBLIC RELEASE