Real-Time Implementation Of An Adaptive Bayesian Beamformer

Ail implementation of the adaptive Bayesian beamformer is examined for execution on field-programmable-gate-array (FPGA) devices. Using multiple sensor inputs, the Bayesian beamformer can estimate the direction-of-arrival (DOA) of a low-power signal in an environment that is simultaneously populated by high-power interference of limited DOA knowledge. A weighted sum of a discrete set of beamformers with known associated DOAs forms the Bavesian beamformer. Previously observed data provides the basis for the calculation of the a posteriori probability distribution function that renders the sum Weights. This paper incorporates further approximations to the derivations to allow for its implementation on FPGA devices. in particular those with lesser g-ate counts. The feasibility of an all-FGPA implementation versus a heterogeneous implementation is explored.