Beamspace transformation for data reduction using genetic algorithms

Estimation of the incident waveform directions, or direction of arrival (DOA) estimation, is an essential problem in sonar array processing. A sonar array usually consists of a large number of elements, and the associated data processing complexity is extensive. An approach to reduce the computation is to map the data from full-dimension element space into a lower dimension beam space, through a linear transformation prior to signal processing. In this paper, a new algorithm for designing beam space transformation with interference suppression is proposed. This approach optimizes in-sector estimation accuracy and out-of-sector interference rejection capabilities simultaneously and yields superior beam space DOA estimates with significantly reduced computation load.