A study of numerical linear algebra library in the context of adaptive array beamforming

Sonar system is an embodiment of sensor signal processing, designed to extract information from the real world, which shares much embedded computing application. The complexity of the most modern computing platforms has made it extremely difficult to write numerical code that achieves the best possible performance. In this paper, we discuss Intel Math Kernel Library (MKL) as a mathematical software package for scientific and technical computation for ease of use in Intel-architecture based COTS (commercial off- the-shelf) systems that can vary greatly. Here we discuss how the advances in computing platform can advantage the sonar signal processing system. We have taken adaptive beamforming as the case study for the comparison of improvements from direct implementation to implementation using linear algebra library for COTS systems.