UACE: Comparison of passive acoustic beamforming techniques at low-frequency using real data from an acoustic vector sensor

    Back


    • Session: 02. Advances in acoustic measurement systems: Technologies and applications
      Organiser(s): Alessandra Tesei and Purnima Ratilal-Makris
    • Lecture: Comparison of passive acoustic beamforming techniques at low-frequency using real data from an acoustic vector sensor [invited]
      Paper ID: 1975
      Author(s): Rokke Eirik, Stinco Pietro, Tesei Alessandra, Been Robert
      Presenter: Rokke Eirik
      Abstract: The paper will compare additive and multiplicative beamforming for passive detection of low-frequency noise sources and estimation of the direction of arrival using real data collected at sea from an acoustic vector sensor mounted on a bottom node. \nThe theory behind beamforming on acoustic vector sensors is well established in literature, for both single sensors, arrays, and spatially distributed sensors. In theory additive beamforming techniques are optimal for signal detection at low signal to noise ratio, and multiplicative processing will always be suboptimal. There are, however, advantages to multiplicative processing most notably lower processing requirements. The use of multiplicative processing is, in that sense, preferable on an autonomous system with limited processing capability, like a small bottom node.\nWe have implemented an additive Bartlett beamformer and a multiplicative intensity based beamformer. Both methods were tested using real data from an acoustic vector sensor, the 3D GeoSpectrum M20-040, mounted on a small bottom node developed at the Centre for Maritime Research and Experimentation (CMRE). For comparison between the methods, we used data from two different scenarios: A simple case with a single strong narrowband target moving around the bottom node within a distance of approximately 6 km and a more complex case with several targets both narrow- and broadband and both moving and stationary also within a distance of approximately 6 km.\nThe paper will describe and analyze the results from both scenarios and make a quantitative comparison between the two processing methods.
        Download the full paper
    • Corresponding author: Dr Alessandra Tesei
      Affiliation: NATO STO CMRE
      Country: Italy
      e-mail:

    Back