Experimental and theoretical performance of a particle velocity vector sensor in a hybrid acoustic beamformer

Abstract

Acoustic measurements have traditionally relied exclusively on sound pressure sensors. This research investigated the performance of Microflown 3D hybrid pressure and acoustic particle velocity sensors in a linear array. Each Microflown sensor has three output channels proportional to the acoustic particle velocity in the three, nominally orthogonal, directions in addition to an output from an omnidirectional pressure microphone. The linear array was formed with a conventional omnidirectional microphone as the center element and two Microflown sensors located 17.2cm away on either side. The Microflown acoustic particle velocity channels were characterized first by the amplitude and phase relationship of their transfer functions relative to their co-located pressure microphone. The transfer function between the Microflown pressure hydrophones and the conventional center microphone was also measured. This enabled the amplitude and phase of all channels to be expressed relative to the center microphone signal. Beamforming was carried out in the frequency domain by applying the appropriate weight and phase delay for the desired steer angle. The bandwidth of the beamformer was limited from about 300Hz to 1.5kHz. At lower frequencies, insufficient signal to noise limited the coherence required to establish the transfer functions while at higher frequencies the phase of the particle velocity transfer functions grew increasingly sensitive to orientation angle. Experiments carried out in the Naval Postgraduate School Anechoic Chamber using single and multiple acoustic sources compared extremely well to the theoretical performance. The addition of hybrid pressure and particle velocity sensors proved successful in eliminating the bearing ambiguity inherent in a linear array of omnidirectional sensors with no change in orientation, no complicated post-processing and no additional time expended.