Quiet submarine threats and high clutter in the littoral,undersea:environme
nt increase the processing demands on beamforming arrays, particularly for
applications which require in-array autonomous operation. Whereas tradition
al single-aperture beamforming approaches may falter, the Split-Aperture Co
nventional Beamforming (SA-CBF) algorithm can be used to meet stringent req
uirements for more precise bearing estimation. Moreover, by coupling each t
ransducer node with a microprocessor, parallel processing of the split-aper
ture beamformer on a distributed system can glean advantages in execution s
peed, fault tolerance, scalability, and cost. In this paper, parallel algor
ithms for SA-CBF are introduced using coarse-grained and medium-grained for
ms of decomposition. Performance results from parallel and sequential algor
ithms are presented using a distributed system testbed comprised of a clust
er of workstations connected by a high-speed network. The execution times,
parallel efficiencies, and memory requirements of each parallel algorithm a
re presented and analyzed. The results of these analyses demonstrate that p
arallel in-array processing holds the potential to meet the needs of future
advanced sonar beamforming algorithms in a scalable fashion.