FPGA implementation of a real-time filter and sum beamformer for acoustic antenna

Beam forming performs spatial filtering of signals, by a sensor array, allowing, in this way, spatial tracking and localization of acoustic sources. Due to its high computational burden, on large arrays, beam forming is generally used only as a post-processing technique. In this paper, we present the FPGA implementation of a real-time filter and sum beam former, inspired to the Frost one. We have been implemented an highly optimized and flexible algorithm for FPGA architecture. Such a result has been achieved by planning the most efficient design for the targeted hardware, and implementing and optimizing such design. The implemented algorithm has been then thoroughly tested for numerical accuracy and functionality with a large number of different signals, and is going to be used with an acoustic antenna system, for the measurement of environmental noise emitted by motor vehicles in pass-by tests.