140
to detect both sound pressure and particle motion, which has been well established
(e.g., Buwalda 1981 and more recently Coffin et al. 2014). Unfortunately, the phase
models by Schuijf (1975) and Schellart and de Munck (1987) represent only a very
restrictive special case of the broader physical principle, namely point sources (or
plane waves), in the free-field and then only for acoustic sinusoidal signals. In addition, the “directionalization” or the determination of the direction of the vector that
points from the receiver to the source is not always possible, but in most cases the
direction of energy flow (i.e., the acoustic intensity) can be determined and provide
information as to the direction of the sound source. This is true in the free-field for
monopoles and for dipoles, and it is even true for most non-free-field propagation
conditions as well. In the next section, we describe a more general approach to the
problem of directionalization that encompasses the phase model formulated on the
concept of time-average acoustic intensity. It has the advantage of applying to both
sinusoidal and broadband signals.
4 Physics of Directionalization
Directionalization is the determination of the direction of the vector that points from
the receiver to the source. Most significant underwater sound sources are monopole
(omnidirectional) in nature. All fish are thought to be capable of sensing a particle
motion vector such as the acoustic particle velocity (de Vries 1950; Popper and Fay
tor to the source is aligned with the acoustic particle velocity. However, the oscillatory nature of the motion of the fluid makes it impossible for the particle velocity
alone to unambiguously determine the direction to the source since it alternately
points towards then away from the source. If a fish aligns its body axis with the
acoustic particle velocity, it cannot be sure whether it is facing directly towards or
directly away from the source. This ambiguity presents a serious problem since the
appropriate response to a relevant sound is to move towards it (e.g., towards a mate
or prey) or away from it (e.g., away from a predator). In order to resolve this ambiguity, knowledge of a second variable such as acoustic pressure, which is scalar, is
required. The motion vector quantity could be particle displacement, velocity or
acceleration, all of which are aligned in the same direction. The scalar quantity in
principle could be acoustic pressure, density, or temperature. We will assume for
simplicity that the measured quantities are acoustic pressure and acoustic particle
velocity, since all of the motion quantities can be obtained from one another by differentiation or integration, which can be performed by the CNS and the various
scalar quantities are proportional to one another. The (ambiguous) line of bearing to
the source can be determined either by aligning with the maximum particle velocity
or by separately determining the velocity components in orthogonal directions and
utilizing an arctangent algorithm to derive the source direction (Rogers et al. 1988).
J.A. Sisneros and P.H. Rogers
to detect both sound pressure and particle motion, which has been well established
(e.g., Buwalda 1981 and more recently Coffin et al. 2014). Unfortunately, the phase
models by Schuijf (1975) and Schellart and de Munck (1987) represent only a very
restrictive special case of the broader physical principle, namely point sources (or
plane waves), in the free-field and then only for acoustic sinusoidal signals. In addition, the “directionalization” or the determination of the direction of the vector that
points from the receiver to the source is not always possible, but in most cases the
direction of energy flow (i.e., the acoustic intensity) can be determined and provide
information as to the direction of the sound source. This is true in the free-field for
monopoles and for dipoles, and it is even true for most non-free-field propagation
conditions as well. In the next section, we describe a more general approach to the
problem of directionalization that encompasses the phase model formulated on the
concept of time-average acoustic intensity. It has the advantage of applying to both
sinusoidal and broadband signals.
4 Physics of Directionalization
Directionalization is the determination of the direction of the vector that points from
the receiver to the source. Most significant underwater sound sources are monopole
(omnidirectional) in nature. All fish are thought to be capable of sensing a particle
motion vector such as the acoustic particle velocity (de Vries 1950; Popper and Fay
tor to the source is aligned with the acoustic particle velocity. However, the oscillatory nature of the motion of the fluid makes it impossible for the particle velocity
alone to unambiguously determine the direction to the source since it alternately
points towards then away from the source. If a fish aligns its body axis with the
acoustic particle velocity, it cannot be sure whether it is facing directly towards or
directly away from the source. This ambiguity presents a serious problem since the
appropriate response to a relevant sound is to move towards it (e.g., towards a mate
or prey) or away from it (e.g., away from a predator). In order to resolve this ambiguity, knowledge of a second variable such as acoustic pressure, which is scalar, is
required. The motion vector quantity could be particle displacement, velocity or
acceleration, all of which are aligned in the same direction. The scalar quantity in
principle could be acoustic pressure, density, or temperature. We will assume for
simplicity that the measured quantities are acoustic pressure and acoustic particle
velocity, since all of the motion quantities can be obtained from one another by differentiation or integration, which can be performed by the CNS and the various
scalar quantities are proportional to one another. The (ambiguous) line of bearing to
the source can be determined either by aligning with the maximum particle velocity
or by separately determining the velocity components in orthogonal directions and
utilizing an arctangent algorithm to derive the source direction (Rogers et al. 1988).
J.A. Sisneros and P.H. Rogers