Grant: Engage Grants Program
Dalhousie University
$18,750.00
Evaluating the use of graphics processor units in analyzing and filtering of flow noise towed arrays
EF 2011-12
Résumé
Towed Arrays are an important tool in Anti-Submarine Warfare. They are long linear antennas used to receive
acoustic signals which have reflected or emitted from underwater targets. Towed arrays are towed behind large
ships and are used to scan the depths for underwater threats. One of the major preoccupations of designers is
noise and in particular flow noise, which is generated by the towed array as it moves through the water.
A computationally intensive operation known as k-w noise filtering can be used to nullify or greatly attenuate
all signal components outside the acoustic cone in order to remove structural and flow noise components
associated with non-acoustic propagation speeds. A real time implementation of such a filter would remove a
large part of flow noise on the fly.
The objective of this proposal is to develop and implement software to calculate k-w filtering based upon a 2D
FFT in CUDA on an NVIDIA Tesla, in real time. We will begin by developing k-w filtering software in
Matlab as a gold standard. The Matlab code will be tested against simulated data provided by Ultra. The
Matlab software will then be translated and optimized in C on a best in class workstation and benchmarks tests
will be developed, numerical results will be compared against the Matlab package. Once this work is complete,
we will implement the k-w filtering software in CUDA on an NVIDIA Tesla.
Programme
Engage Grants Program
Bénéficiaire
Emplacement
Nova Scotia, Canada
Type de bénéficiaire
Academia
Type
Grant
Source
NSERC's Awards Data
Tel que publié
Application id
413071-2011
Name nom
Leon, Josh
Department département
Engineering, Faculty of
Organization id
79
Institution établissement
Dalhousie University
Province en
Nova Scotia
Province fr
Nouvelle-Écosse
Country en
CANADA
Country fr
CANADA
Fiscal year exercice financier
2011
Competition year année de concours
2011
Award amount
18750
Program id
EGP
Program nane en
Engage Grants Program
Program name fr
Programme de subventions d'engagement partenariat
Group en
Research partnerships programs
Group fr
Programmes de partenariat
Committee code
1551
Committee name en
Atlantic Internal Decision Committee
Committee name fr
Comité de décision interne pour l'Atlantique
Area of application code
515
Area of application group en
Manufacturing processes and products
Area of application group fr
Procédés et produits industriels
Area of application en
Instrumentation technology
Area of application fr
Technologie de l'instrumentation
Research subject code
2500
Research subject group en
Electrical and electronic engineering
Research subject group fr
Génie électrique et électronique
Research subject en
Electrical and electronic engineering
Research subject fr
Génie électrique et électronique
Partie
1
Nb partie
1
Application title
Evaluating the use of graphics processor units in analyzing and filtering of flow noise towed arrays
Keyword
NSERC
Application summary
Towed Arrays are an important tool in Anti-Submarine Warfare. They are long linear antennas used to receive
acoustic signals which have reflected or emitted from underwater targets. Towed arrays are towed behind large
ships and are used to scan the depths for underwater threats. One of the major preoccupations of designers is
noise and in particular flow noise, which is generated by the towed array as it moves through the water.
A computationally intensive operation known as k-w noise filtering can be used to nullify or greatly attenuate
all signal components outside the acoustic cone in order to remove structural and flow noise components
associated with non-acoustic propagation speeds. A real time implementation of such a filter would remove a
large part of flow noise on the fly.
The objective of this proposal is to develop and implement software to calculate k-w filtering based upon a 2D
FFT in CUDA on an NVIDIA Tesla, in real time. We will begin by developing k-w filtering software in
Matlab as a gold standard. The Matlab code will be tested against simulated data provided by Ultra. The
Matlab software will then be translated and optimized in C on a best in class workstation and benchmarks tests
will be developed, numerical results will be compared against the Matlab package. Once this work is complete,
we will implement the k-w filtering software in CUDA on an NVIDIA Tesla.