Acoustics Seminars

Large-Scale Acoustic Array Processing

Friday, February 13, 2015 4:00pm in ETC 4.150

Dr. Mark Lai
The Institute for Computational Engineering and Sciences
The University of Texas at Austin

Consider a sensing system using a large number of N microphones placed in multiple dimensions to monitor an acoustic field. Using all the microphones at once is impractical because of the amount data generated. Instead, we choose a subset of D microphones to be active. Specifically, we wish to find the D set of microphones that minimizes the largest interference gain at multiple frequencies while monitoring a target of interest. A direct, combinatorial approach testing all N choose D subsets of microphones is impractical because of problem size. Instead, we use a convex optimization technique that induces sparsity. Our work investigates not only the optimal placement (space) of microphones but also how to process the output of each microphone (time/frequency). We explore this problem for both single and multi-frequency sources, optimizing both microphone weights and positions simultaneously. In addition, we explore this problem for random sources where the output of each of the N microphones is processed by an individual multirate filterbank. The N processed filterbank outputs are then combined to form one final signal. In this case, we fix all the analysis filters and optimize over all the synthesis filters. In this random source/multirate filterbank case, we once again optimize over space-time-frequency simultaneously.

Generation and Refraction of the Microbarom Signal by Hurricanes over the Atlantic Ocean

Friday, February 6, 2015 4:00pm in ETC 4.150

Dr. Roger Waxler
National Center for Physical Acoustics
The University of Mississippi
http://www.olemiss.edu/depts/physics_and_astronomy/faculty/waxler.html

It is well known that both acoustic and seismic noise spectra show an increased band of spatially coherent noise around 0.2 Hz, the so-called microbarom and microseism signals. It has been appreciated for over a half of a century that hurricanes over the open ocean are a source of microbaroms and microseisms. While the mechanism for microseism generation has been understood since the early 1960’s, a complete theory of the generation of microbaroms did not appear until 2006. It has been shown that microbaroms and microseisms are two manifestations of the same phenomenon: a radiating harmonic of the ocean surface wave field produced by the head-on collision of ocean surface waves of the same period. In 2008 it was conjectured that, for a deep ocean hurricane, the source of the colliding waves is the interaction of the hurricane generated waves with the background ocean swell. The region in which this interaction takes place, the microbarom source region, is generally several hundred kilometers from the eye of the storm. This source region’s position with respect to the eye of the storm changes slowly as the storm moves across the ocean so that it can be considered to be more or less static with respect to the storm. As the microbarom signal propagates away from the source region, propagation paths that pass through the storm are strongly refracted by the storm winds. Thus, hurricanes carry a sound source with them which might be used to probe the interior of the storm itself. To study this effect, a temporary network of infrasound stations was deployed along the US eastern seaboard during the 2010 and 2011 hurricane seasons. The underlying theory will be presented and results from the deployments will be shown.

Laboratory Experiments on Sound Propagation in a Continuously Stratified Ocean Containing Internal Gravity Waves

Friday, January 30, 2015 4:00pm in ETC 4.150

Dr. Likun Zhang
Center for Nonlinear Dynamics
The University of Texas at Austin
http://chaos.utexas.edu/people/post-docs/likun-zhang

The speed of sound in the ocean varies with temperature, salinity, and pressure over the entire ocean depth. This variation results in a sound speed profile that supports a sound channel for transmitting information over great distances in the ocean. The transmission is strongly affected by local water column oscillations that result from gravity wave motions internal to the ocean. Internal gravity waves can propagate in any density-stratified fluid; in the oceans the increase of density with depth is due to decreasing temperature and increasing salinity. We present results from a laboratory tank experiment that models sound propagation in a stratified ocean containing internal gravity waves. The experiment determines (1) the refraction of acoustic wavefronts due to sound speed gradients and (2) sound speed fluctuations arising from the internal gravity waves. This research provides a data set for comparison with modeling of sound propagation in the oceans.

A Brief History of Sediment Acoustics

Friday, December 5, 2014 4:00 p.m. ETC 4.150

Anthony L. Bonomo
Applied Research Laboratories
The University of Texas at Austin
http://www.arlut.utexas.edu

The acoustic behavior of sediments has been studied extensively. The earliest models were based on the assumption that sediments behaved like fluids. Since sediments generally can support shear stresses, the assumptions made when using fluid models were found to be tenuous at best, and many of these models have been replaced with those treating the sediment as an elastic or viscoelastic medium. However, it has recently been shown that model predications and experimental results are in better accord when the sediment is assumed to behave as a poroelastic medium governed by Biot theory. This talk intends to serve as a primer on the development of fluid, elastic, and poroelastic sediment models and will attempt to give historical justification supporting the uses and limitations of each. Recent attempts at formulating more sophisticated and comprehensive sediment models such as the Viscous Grain Shearing model of Buckingham and the Extended Biot model of Chotiros will be discussed as well.

Analysis of Acoustic Scattering from Large Fish Schools Using Bloch Wave Formalism

Wednesday, December 3, 2014 4:00 p.m. ARL A009

Jason A. Kulpe
George W. Woodruff School of Mechanical Engineering
Georgia Institute of Technology
http://www.me.gatech.edu

In the open ocean, acoustic scattering of SONAR signals in the 1-10 kHz frequency range is dominated by large fish schools, where multiple scattering effects between the air-filled swim-bladders of the fishes within the school are strong. These schools are typically large in comparison to the acoustic wavelength and the fish typically swim in nearly-periodic arrangements with a separation distance of approximately one body length. Hence, the fish school can be studied simply and effectively by invoking the formalism of Bloch waves in periodic media. Analysis of the periodic school is aided through the Bloch theorem, which reduces the study of the entire school to the study of a unit cell containing a single fish’s swim-bladder. Application of the Bloch formalism to the school requires study of acoustic reflection from a semi-infinite half-space composed of an infinite arrangement of air swim bladders in water; this media is denoted a fluid phononic crystal (PC). The reflection is considered, using a finite element discretization of the unit cell, via an expansion of Bloch waves for the transmitted wave field. Next, scattering from a large finite school is studied through the context of the Helmholtz-Kirchhoff integral theorem where the semi-infinite PC pressure, determined by the Bloch wave expansion, is used as the integral’s inputs. A general model using the Bloch formalism and encompasses the internal fish structure, fish biologic properties, and realistic school effects such as varying school geometry and disorder, will be explored. Transient analysis of the frequency dependent scattering, using the proposed model, may assist SONAR operators to better classify large fish schools based on the observed characteristics of the scattered field. Comparison of results is accomplished through a finite element model (two dimensions) and a low frequency analytical multiple scattering model (three dimensions).

Biologically Inspired Microphones

Friday, November 21, 2014 4:00 p.m. ETC 4.150

Professor Neal A. Hall
Department of Electrical and Computer Engineering
The University of Texas at Austin
https://sites.google.com/site/utacousticmems

The parasitoid fly Ormia Ochracea has the remarkable ability to locate crickets using audible sound. This ability is, in fact, remarkable as the fly’s hearing mechanism spans only 1.5 mm, which is 50 times smaller than the wavelength of sound emitted by the cricket. The hearing mechanism is, for all practical purposes, a point in space with no significant interaural time or level differences to draw from. It has been discovered that evolution has empowered the fly with a hearing mechanism that utilizes multiple vibration modes to amplify interaural time and level differences. Here, we present a fully integrated, man-made mimic of the Ormia’s hearing mechanism capable of replicating the remarkable sound localization ability of the special fly. A silicon-micromachined prototype is presented which uses multiple piezoelectric sensing ports to simultaneously transduce two orthogonal vibration modes of the sensing structure, thereby enabling simultaneous measurement of sound pressure and pressure gradient.

Electro-Mechanical Modeling of Piezoelectric Fluid Ejectors for High Viscosity Liquids

Friday, November 14, 2014 4:00 p.m. ETC 4.150

Dr. Drew Loney
George W. Woodruff School of Mechanical Engineering
Georgia Institute of Technology
http://www.me.gatech.edu

The controlled atomization of high viscosity liquids to produce individual droplets of uniform diameter remains an ongoing technological challenge. Traditional atomization techniques for the production of single droplets—inkjet printers—do not extend to high viscosity materials as fluid ejection ceases. Other atomization methods, such as horn-based ultrasonic atomization, have demonstrated the capability to atomize such materials but a detailed understanding of the mechanism that enables fluid ejection is lacking. The underlying physical mechanisms that govern atomization of high viscosity materials (> 10 mPa·s) by piezoelectric transducer-driven fluid ejectors remain poorly understood, requiring an in-depth understanding of the acoustics underlying the fluid ejection process. This presentation focuses on the principle acoustic phenomena that regulate atomization of high viscosity materials—fluid cavity/transducer interactions and waveform propagation within the fluid cavity—to ascertain an upper viscosity bound on the materials that can be atomized by piezoelectric transducer-driven fluid ejectors. A coupled electro-mechanical analytical modeling framework is developed by decomposing fluid ejector geometries into simplified acoustic elements with closed-form solutions joined together through boundary conditions. These simplified acoustic models maintain the dominant sources of viscous dissipation and dispersion within the modeling framework to provide an accurate estimate of the acoustic field amplitude which is the driving mechanism for fluid ejection. By comparing the acoustic pressure amplitude to that required to yield fluid ejection at the aperture, obtained by CFD modeling and scaling analysis, an upper limit on liquid viscosity for atomization can be determined. The electro-mechanical modeling framework is also applied to explore new concepts of piezoelectric transducer-driven fluid atomizers in an effort to achieve the controlled atomization of high viscosity materials.

Design and Construction of Acoustic Test Chambers

Friday, November 7, 2014 4:00 p.m. ETC 4.150

Dr. Douglas F. Winker
ETS-Lindgren
Cedar Park, Texas
http://www.ets-lindgren.com

ETS-Lindgren manufactures acoustic test chambers for a wide variety of clients and applications. This presentation will discuss case studies of test and measurement solutions. The design and construction of ETS-Lindgren’s acoustic lab facility in Cedar Park, Texas will be discussed in detail. ETS-Lindgren’s lab consists of a hemi-anechoic chamber and a reverberation chamber suite. The hemi-anechoic chamber was designed to obtain a noise floor of <0 dBA at frequencies above 100 Hz. Another design goal was a 2 m radius free field above 80 Hz that is compliant with ISO 3745. It features a 200 m3 volume inside the wedges. To achieve the design goals, several aspects were considered including ambient levels, vibration isolation, HVAC noise, free field performance, and future development in the area. The design solutions for each of these areas will be presented. The reverberation chamber suite consists of two reverberation chambers designed to perform transmission loss tests and sound absorption testing. The source chamber has an internal volume of 214 m3 and the receive room has an internal volume of 418 m3. Design considerations will be discussed including vibration isolation, wall panel design, and diffuser placement. Additionally, the design and construction of the transmission loss coupling will be addressed.

Seismic Attenuation, Dispersion and Anisotropy in Porous Rocks: Mechanisms and Models

Monday, November 3, 2014 1:30 p.m. ARL Auditorium

Dr. Boris Gurevich
Curtin University and CSIRO
Perth, Australia
http://www.csiro.au

Understanding and modelling of attenuation of elastic waves in fluid-saturated rocks is important for a range of geophysical technologies that utilise seismic, acoustic or ultrasonic amplitudes. A major cause of elastic wave attenuation is viscous dissipation due to the flow of the pore fluid induced by the passing wave. Wave-induced fluid flow occurs as a passing wave creates local pressure gradients within the fluid phase and the resulting fluid flow is accompanied with internal friction until the pore pressure is equilibrated. The fluid flow can take place on various length scales: for example, from compliant fractures into the equant pores (so-called squirt flow), or between mesoscopic heterogeneities like fluid patches in partially saturated rocks. A common feature of these mechanisms is heterogeneity of the pore space. I will explore how this heterogeneity affects attenuation, dispersion and anisotropy of porous rocks. I give a brief outline of a consistent theoretical approach that gives quantitative estimates of these phenomena and discuss rigorous bounds for attenuation and dispersion, which represent an extension of Hashin-Shtrikman bounds to viscoelastic media.

Boris Gurevich has MSc from Moscow State University (1981) and PhD from Moscow Institute of Geosystems (1988). In 1990s he worked at a number of institutions in Russia, Germany, UK and Israel before joining Curtin University and CSIRO in Perth, Australia in 2001 as Professor of Petroleum Geophysics. From 2010 he has served as head of Curtin’s Department of Exploration Geophysics and Director of the Curtin Reservoir Geophysics Consortium. His research interests include rock physics, poroelasticity, diffraction imaging and time lapse seismic monitoring.

Submarine Sonar

Friday, October 17, 2014 4:00 p.m. ETC 2.136

Dr. F. Michael Pestorius
Applied Research Laboratories
The University of Texas at Austin
http://www.arlut.utexas.edu

Modern submarines use sonar almost exclusively for ship navigation, obstacle avoidance, contact detection and warfare missions. Rudimentary sonars were first developed in World War I and they reached fairly high levels of sophistication in World War II. However, submarines up to about 1960 were basically surface craft that could submerge for relatively short periods of time. The marrying of nuclear power to the submarine created a true undersea capable ship. With this development came a major improvement in submarine sonars. The development of the US submarine force since the advent of nuclear power with emphasis on the continuing development of active and passive sonar will be addressed in this seminar. The Applied Research Laboratories at UT Austin have long been involved in sonar research and development. General information about submarine sonars will be outlined.

Dr. Mike Pestorius, a UT graduate (PhD EE), is a retired submariner who spent close to 27 years in the Navy. He served on several submarines and commanded a ballistic missile submarine, the USS Mariano G. Vallejo (SSBN 658), for 4 years. After retiring from the Navy, he served 12 years as director of Applied Research Laboratories at UT and then 4 years as technical director of the international office in the London office of the Office of Naval Research.

Snapping Acoustic Metamaterials: Enhanced Material Nonlinearity and Absorption of Mechanical Energy

Monday, October 13, 2014 3:00 p.m. ECJ 1.202

Dr. Michael R. Haberman
Applied Research Laboratories and
Department of Mechanical Engineering

The University of Texas at Austin
http://www.arlut.utexas.edu
http://www.me.utexas.edu

Acoustic metamaterials (AMM) are material systems whose overall performance originates from engineered sub-wavelength structure rather than the inherent material properties of their constituents. This relatively new topic in applied physics has garnered attention in the scientific community because of the potential role in realizing exotic behavior such as acoustic cloaking, negative refraction, and one-way sound transmission. This talk discusses a new AMM that was designed to amplify acoustic absorption and nonlinearity for potential use in new acoustical devices and vibro-acoustic coating materials. The AMM consists of a nearly incompressible viscoelastic matrix material containing a low volume fraction of sub-wavelength metamaterial structures (inclusions) that possess a non-monotonic stress-strain response. A nonlinear multiscale material model is presented that captures the strain-dependent evolution of the stiffness of the homogenized medium. That material model is then used to determine the effective quadratic and cubic parameters of nonlinearity of the AMM. Those parameters of nonlinearity are compared with those of conventional materials and examples of one-dimensional wave distortion effects are provided. The forced nonlinear multiscale dynamics in the AMM is then explored using a modified Rayleigh-Plesset model to highlight the influence of pre-stress and inclusion-scale dynamics on macroscopic energy absorbing capabilities for this AMM.

Modeling Sound Propagation Through an Incompressible Flow

Friday, October 3, 2014 4:00 p.m. ETC 4.150

Sumedh M. Joshi
Center for Applied Mathematics
Cornell University
http://www.cam.cornell.edu

Sound propagating in a moving fluid will be advected, refracted, and Doppler shifted by the patterns in the moving flow. For example, a pulse of sound impinging on an inviscid shear layer is advected in the direction of the shear flow, in addition to being reflected from the shear-layer interface. To model such sound-flow interactions, it was assumed that the background flow was known and satisfied the incompressible Navier-Stokes equations. Furthermore, it was also assumed that the sum of the acoustic and hydrodynamic fields satisfied momentum and mass conservation. Finally, the sound propagation was assumed to be linear. These assumptions lead to hyperbolic conservation laws that were discretized and solved with a time-domain finite difference model. To demonstrate, a few example flows and their resulting sound fields will be presented, and I will also discuss a modeling effort that attempts to quantify the degree to which the acoustic scattering from an idealized tornado funnel can be used to infer properties of the tornado. The modeling suggests that although there is a measurable acoustic reflection, practical concerns suggest that acoustics are not a viable method of inferring tornado properties.

Statistical Inference of Seabed Sound-Speed Structure in the Gulf of Oman Basin

Friday, September 26, 2014 4:00 p.m. ETC 4.150

Dr. Jason D. Sagers
Applied Research Laboratories
The University of Texas at Austin
http://www.arlut.utexas.edu

Addressed is the statistical inference of the sound-speed depth profile of a thick soft seabed from broadband sound propagation data recorded in the Gulf of Oman Basin in 1977. The acoustic data are in the form of time series signals recorded on a sparse vertical line array and generated by explosive sources deployed along a 280 km track. The acoustic data offer a unique opportunity to study a deep-water bottom-limited thickly sedimented environment because of the large number of time series measurements, very low seabed attenuation, and auxiliary measurements. A maximum entropy method is employed to obtain a conditional posterior probability distribution (PPD) for the sound-speed ratio and the near-surface sound-speed gradient. The multiple data samples allow for a determination of the average error constraint value required to uniquely specify the PPD for each data sample. Two complicating features of the statistical inference study are addressed: (1) the need to develop an error function that can both utilize the measured multipath arrival structure and mitigate the effects of data errors and (2) the effect of small bathymetric slopes on the structure of the bottom interacting arrivals.

Acoustical Foundations of Scales, Tempered Tuning, and Pitch Perception

Friday, September 19, 2014 4:00 p.m. ETC 4.150

Dr. James M. Gelb
Applied Research Laboratories
The University of Texas at Austin
http://www.arlut.utexas.edu

While seemingly disparate, the musical scales used throughout the world are in fact surprisingly universal. This talk touches on all aspects of this universality, from the perspectives of the acoustical properties of instruments (acoustics), the frequency resolution of the ear (signal processing), and the pattern-matching strategies of the brain (psychoacoustics). The lecture begins with an explanation of the distinction between, and the physical generation of, approximate pure tones (e.g., those produced by Helmholtz resonators and tuning forks) and complex tones (distributions of tones) produced by actual instruments. The science behind bowing and the reasons for the rich harmonic structure of the violin will be explored. The Helmholtz theory of consonance (pleasant-sounding intervals), as well as modern refinements of the theory that take critical bands in the ear into account, are discussed. This segues into an explanation of the non-uniform spacing of notes in the ubiquitous pentatonic and diatonic scales, which in turn leads to the topic of tempered tuning—the compromise between perfectly “just” intervals (frequency ratios of fundamentals involving small integers) and equal-tempered tuning used to freely support modulation (the switching between musical keys on the fly). Interesting aspects of tempered tuning (including the physical cause of the Wolf interval) are singled out for discussion, so as to avoid getting lost in the myriad of historical tuning schemes. The lecture concludes with an investigation of a pattern-matching model to explain pitch perception, drawing from results of a recent experiment conducted at ARL.

Shear Waves in Viscoelastic Wormlike Micellar Fluids

Monday, September 15, 2014 1:00 p.m. RLM 11.204

Professor Joseph R. Gladden
Department of Physics
National Center for Physical Acoustics
The University of Mississippi
ncpa.olemiss.edu

In viscous Newtonian fluids, support of shear waves is limited to the viscous boundary layer. However, non-Newtonian fluids, which have a shear modulus, support shear waves over much longer distances. Wormlike micellar fluids are an interesting class of non-Newtonian fluids in which surfactant molecules, aided by the addition of salts, self-aggregate into long and flexible cylindrical structures. The dimensions and topology of these structures depend on concentration, surfactant/salt ratio, and temperature. We will present studies in which acoustic shear wave propagation is used to better understand various structural phases of this system as a function of concentration and temperature. These studies indicate 2 distinct phase transitions between 0 (water) and 600 mM surfactant. Birefringent properties of this fluid make the acoustic field simple to visualize using crossed polarizers. We will also present neutron scattering results on microstructure, relaxation in static shear strain fields, and rheological studies to help flesh out the story on this complex fluid.

Estimates of Source Range Using Horizontal Multi-path in Continental Shelf Environments

Friday, September 5, 2014 4:00 p.m. ETC 4.150

Dr. Megan S. Ballard
Applied Research Laboratories
The University of Texas at Austin
www.arlut.utexas.edu

A method has been developed to estimate source range in continental shelf environments that exhibit three-dimensional propagation effects. The technique exploits measurements recorded on a horizontal line array of a direct path arrival, which results from sound propagating across the shelf to the receiver array, and a refracted path arrival, which results from sound propagating obliquely upslope and refracting back downslope to the receiver array. A hybrid modeling approach using vertical modes and horizontal rays provides the ranging estimate. According to this approach, rays are traced in the horizontal plane with refraction determined by the modal phase speed. To obtain an estimate of source range, the principle of reciprocity is used such that the rays originate from the center of the array with launch angles equal to the estimated bearing angles of the direct and refracted paths. The location of the source in the horizontal plane is estimated from the point where the rays intersect. In this talk, the technique is applied to data recorded on a horizontal line array located about 12 km east of the southern coast of Florida. The effects of unknown environmental parameters, including the sediment properties and the water-column sound-speed profile, on the source range estimate are quantified. Error resulting from uncertainty in the measured bathymetry and location of the receiver array will also be addressed.

Measuring the Acoustic Parameters of Fish Schools

Friday, April 25, 2014 4:00 p.m. ETC 4.150

Craig Dolder
Department of Mechanical Engineering
Applied Research Laboratories
The University of Texas at Austin
http://www.arlut.utexas.edu

While the literature contains extensive in situ measurements of scattering by fish schools, significant uncertainties exist with respect to characterizing the size, quantity, and distribution of fish within the schools that confound accurate measurement-model comparison. Measurements of the sound speed through collections of live fish (Danio rerio) were conducted in a laboratory setting. The sound speed was investigated using a resonator technique which yielded inferences of the phase speed within the fish school though measurements of the resonances of a one-dimensional waveguide. Fish densities were investigated ranging from 8.6 to 1.7 fish lengths per mean free path. Measurements agree well with a predictive model that is based on shell-free spherical bubbles, which indicates that the phase speed is not significantly affected by the fish flesh or swimbladder morphology for the species studied. The variation in phase speed due to individual fish motion within the model school was measured to be up to ±5.6%. This indicates that precise knowledge of the fish position is required to achieve greater model accuracy. To complement the phase speed measurements, the attenuation through a cloud of encapsulated bubbles was evaluated through insertion loss measurements. Multiple arrangements of balloons of radius 4.68 cm were used to surround a projector. The insertion loss measurements indicated an amplification of around 10 dB at frequencies below the individual balloon resonance frequency and an insertion loss of around 40 dB above the individual balloon resonance frequency. Analytical modeling of the bubble collection predicted both the amplification and loss effect, but failed to accurately predict the level of amplification and insertion loss. Finally, effective medium models and full scattering models (requiring knowledge of bubble size and position) were evaluated for a model fish school. The two models agree for forward scattering for all frequencies except those immediately around the individual bubble resonance frequency. Back scattered results agree at low frequencies, however as soon as the wavelength becomes smaller than four mean free paths between fish the models diverge. Ramifications of these findings and potential future research directions are discussed.

Large-scale Bayesian Inverse Wave Propagation

Friday, April 18, 2014 4:00 p.m. ETC 4.150

Professor Omar Ghattas
Department of Mechanical Engineering and Department of Geological Sciences
Director of the Center for Computational Geosciences
Institute for Computational Engineering and Sciences
The University of Texas at Austin
http://www.jsg.utexas.edu

Inverse problems governed by wave propagation–in which we seek to reconstruct the unknown shape of a scatterer, or the unknown properties of a medium, from observations of waves that are scattered by the shape or medium–play an important role in a number of engineered or natural systems. We formulate the inverse problem in the framework of Bayesian inference. This provides a systematic and coherent treatment of uncertainties in all components of the inverse problem, from observations to prior knowledge to the wave propagation model, yielding the uncertainty in the inferred medium/shape in a systematic and consistent manner. Unfortunately, state-ofthe-art Markov chain Monte Carlo methods for characterizing the solution of Bayesian inverse problems are prohibitive when the forward problem is of large scale (as in our 100-1000 wavelength target problems) and a high-dimensional parameterization is employed to describe the unknown medium (as in our target problems involving infinite-dimensional medium/shape fields, which result in millions of parameters when discretized). We report on recent research aimed at overcoming the mathematical and computational barriers for large-scale Bayesian inverse wave propagation problems. These include:

  • a high order, parallel, adaptive hp-non-conforming discontinuous Galerkin (DG) method for acoustic/elastic/electromagnetic wave propagation
  • infinite-dimensional formulations of Bayesian inverse problems and their consistent finite-dimensional discretizations;
  • a stochastic Newton MCMC method for solution of the statistical inverse problem that reduces the number of samples needed by several orders of magnitude, relative to conventional MCMC;
  • fast low rank randomized SVD approximation of the Hessian based on compactness properties; and
  • applications to Bayesian inverse wave propagation in whole Earth seismology with up to one million earth model parameters, 630 million state variables, on up to 100,000 processors.

This work is joint with George Biros, Tan Bui-Thanh, Carsten Burstedde, James Martin, Georg Stadler, Hari Sundar, and Lucas Wilcox.

More than Imaging: Cancer and Cardiovascular Therapies via Focused Ultrasound

Tuesday, April 8, 2014 11:00 a.m. ETC 2.102

Dr. Linsey Phillips
University of North Carolina at
Chapel Hill & NC State University

Cancer and cardiovascular disease are two of the most common diseases affecting industrialized countries like the United States. Diagnosis and treatment are complex for these diseases, often requiring extensive imaging exams and chronic, expensive therapies. Ultrasound has the capacity for both imaging and therapy, and is less expensive and less invasive than other diagnostic and therapeutic modalities. The thermal and mechanical effects induced by ultrasound have significant therapeutic potential. In the last decade, focused ultrasound has gained popularity as a method to localize therapeutic effects including drug delivery, gene delivery, and blood-brain barrier opening. At high intensities, focused ultrasound can achieve temperatures sufficient to ablate biological tissue, and has the potential to eliminate a variety of cancers non-surgically. Focused ultrasound therapy is currently approved by the FDA for uterine fibroid and bone metastasis ablation. Few methods exist to selectively target chemotherapeutic drugs to cancer, and side effects from accumulation in non-targeted tissues remain a problem. However, focused ultrasound and targeted microbubbles overcome this major limitation of existing therapies by enabling localized, enhanced drug delivery with sub-millimeter precision. I investigate how ultrasound can improve detection and treatment of cancer and cardiovascular diseases by combining “theranostic” agents with ultrasound. Through my future research, I aim to explore new, clinically translatable therapeutic applications of focused ultrasound and acoustically active agents to treat diseases. To reach these objectives, I will use a combination of acoustics and biology, building on my 10 years of experience with ultrasound and microbubbles.

Bond Graph Modeling of Frog Vocal Production

Friday, April 4, 2014 4:00 p.m. ETC 4.150

Professor Nicole M. Kime
Department of Biological Sciences
Edgewood University
http://biology.edgewood.edu

The túngara frog, Physalaemus pustulosus, is a well-known model for investigating the evolution of acoustic communication. All male frogs in the genus Physalaemus produce a species-specific “whine”. Túngara frogs and some populations of its sister species, P. petersi, also produce a second complex call component, a “chuck” or “squawk” that relies upon the presence of an enlarged fibrous mass that extends from the vocal folds. Females generally prefer complex calls to simpler whines. The evolution of laryngeal morphology has thus been influenced by sexual selection. In spite of evidence for a correlation between structure and function, the mechanism by which the fibrous mass interacts with other elements of the vocal system during the production of whines and chucks remains unknown. We have recently been using bond graphs, a lumped element modeling technique, to investigate the biomechanics of vocal production in túngara frogs. These models explore the production of whines and chucks, the mechanics of the fibrous mass, and interactions between different components of the frog vocal system. A bond graph approach to vocal system modeling will be advantageous to understanding the evolution of groups of morphological traits within the integrated vocal systems in a variety of animals.