|
Propagation
Models
ESL has developed several propagation models that
are used within the laboratories and at other research
facilities in the world.
The
normal mode model ORCA was developed by Dr. Evan K.
Westwood while he was a visiting scientist at the
Defense Research Establishment Pacific in Victoria,
British Columbia. It has become widely used in the
underwater acoustics research community, as well as
within ESL. For a given ocean environment, specified
by the sound speed profile in the water column and
a geoacoustic profile of the ocean bottom, ORCA finds
the normal modes and computes the acoustic field.
The model includes the effects of sound speed gradients
in the water and the bottom layers, shear waves in
the bottom layers, steep-angle propagation represented
by leaky modes, and attenuation in the bottom layers.
It may be used to predict narrowband or broadband
propagation. The model is unique among underwater
acoustic propagation codes because it is largely automatic:
the user does not need to guess at any obscure convergence
parameters such as depth- or range-sampling resolutions.
Several graphics from the journal articles describing
ORCA are given below [1,2].
 |
|
Data
Analysis
One of the strengths of ESL is that we combine the
ability to understand and model underwater acoustic
propagation with the ability to perform advanced signal
processing on measured data. Using our modeling and
signal processing tools, we simulate measured data
(usually at the time series level), process the simulated
and measured data in the same manner, and often are
able to demonstrate remarkable agreement in the resulting
data products. Such agreement is extremely valuable
in understanding the performance of sonar systems
and in developing and improving the signal processing
algorithms that are at the heart of those systems.
As an example
of a data/model comparison, we examined the broadband
correlation structure of data measured on two bottom-mounted
hydrophones separated by about 450 m in the shallow
water of the English Channel. As a ship passed nearby
the receivers, the multiple paths of acoustic energy
to the two receivers produced a complex correlation
structure as a function of time delay and time [5].
|
|
Another
area of data analysis involves the technique of
source localization by way of matched field processing
(MFP), in which the field measured at a set of receivers
is matched with a set of fields computed from a propagation
model. The simulated fields are for a number of hypothesized
source positions, and the output of the matched field
processor is an ambiguity surface over hypothesized
source position. Using a vertical array of receivers,
the hypothesized source position is a function of
range and depth only. An example output of a broadband
MFP technique, where the hypothesized source depth
has been fixed, is shown [6,7].
References
1. E. K. Westwood, C. T. Tindle, and N. R. Chapman,
"A normal mode model for acousto-elastic ocean environments,"
J. Acoust. Soc. Am., 100, 3631-3645 (1996).
2. E. K. Westwood
and R. A. Koch, "Elimination of branch cuts from the
normal mode solution using gradient half spaces,"
J. Acoust. Soc. Am., 106, 2513-2523 (1999).
3. E. K. Westwood
and P. J. Vidmar, "Eigenray findingand time series
simulation in a layered-bottom ocean," J. Acoust.
Soc. Am., 81, 912-924 (1987).
4. E. K. Westwood
and C. T. Tindle, "Shallow water time series simulation
using ray theory," J. Acoust. Soc. Am., 81,
1752-1761 (1987).
5. E. K. Westwood
and D. P. Knobles, "Source track localization via
multipath correlation matching," J. Acoust. Soc. Am.,
102, 2645-2654 (1997).
6. D. P. Knobles,
E. K. Westwood, and J. E. LeMond, "Modal time-series
structure in a shallow water environment," IEEE J.
Oceanic Eng., 23, 188-202 (1998).
7. E. K. Westwood, "Broadband matched field source
localization," J. Acoust. Soc. Am., 91, 2777-2789
(1992).
|
 |
