Jiyoung Son PhD final thesis defense

UNIVERSITY OF UTAH

ELECTRICAL AND COMPUTER ENGINEERING DEPARTMENT

 

DISSERTATION DEFENSE FOR THE DEGREE OF

DOCTOR OF PHILOSOPHY

 

by

Jiyoung Son

Advisor: Bruce Gale

 

 

Active Sperm Separation Technique Using An Inertial Microfluidic Device

 

Microfluidic technology has the unique potential to separate sperm from unwanted debris while improving the effectiveness of assisted reproductive technologies (ART). Current clinical protocol limitations regarding separation of sperm from other cells and cellular debris can lead to low sperm recovery when the sample contains low concentrations of mostly low motility sperm and a high concentration of unwanted cells or cellular debris, such as occurs with surgical testis dissection samples from non-obstructive azoospermia (NOA) patients who went through microsurgical testicular sperm extraction (mTESE), and semen samples from leukospermia patients (high white blood cell (WBC) semen).

                Over the years, most sperm separation approaches utilizing microfluidics have relied on sperm motility for separation with added features through which only highly motile sperm can pass. Thus, these techniques can separate only progressive motile sperm from semen samples, but they lose a significant number of sperm cells including viable non-progressive motile and non-motile sperm. Therefore, these techniques are not feasible for use with immature and non-motile sperm that may be the only sperm produced by some patients. To help address this population, this dissertation proposes a passive microfluidic approach utilizing inertial effects that can separate sperm (regardless of their motility state) from other unwanted cells/debris and can provide enhancement of sperm samples without losing non-motile and non-progressive motile sperm.

This dissertation demonstrates label-free separation of sperm from challenging sperm samples using inertial microfluidics. The approach does not require any externally applied forces except the movement of the fluid sample through the instrument. In this way, it is possible to recover not only any motile sperm, but also viable less-motile and non-motile sperm with high recovery rates. The preliminary results show the usefulness of inertial microfluidics to significantly reduce unwanted cells/cellular debris (RBC and WBC) concentrations by flow focusing of debris within a spiral channel flow. The majority (~80%) of sperm cells collect to the designated outlet and ~98% of debris goes to the waste outlet. The estimated sample process time is more rapid (~5minutes) and autonomous than conventional methods that take at least 1 hour for the most basic ones and 10-18 hours when the common practice of manually searching for sperm under the microscope is used. This study also performed viability, toxicity, and recovery tests on the proposed sperm separation method for biocompatibility verification. These tests should provide initial validation of clinical usefulness.

The sperm separation data obtained using the spiral channel shows that the flow focusing of sperm is not as sharp as the flow focusing obtained for red blood cells. The likely cause of this difference is the different shapes of the cells, but there is no good model for demonstrating this effect. To address this issue, this study presents an improved model of sperm cell motion in curved channels based on both 2D COMSOL ® simulations and experimental studies.

 

 

Thursday April 27, 2017

2:00 PM

Sorenson Molecular Bioengineering Building (SMBB) room 3750

The public is invited

Md Abid Hossain MS final thesis defense

UNIVERSITY OF UTAH

ELECTRICAL AND COMPUTER ENGINEERING DEPARTMENT

 

THESIS DEFENSE FOR THE DEGREE OF

MASTER OF SCIENCE

 

by

 

Md Abid Hossain

Advisor: Marc Bodson

                                                                                                                                                           

 

Measuring Electric Power System Susceptibility to Cascading Blackouts

 

The purpose of the research is to develop methods and software tools to measure stress, the susceptibility of the bulk electric power systems to cascading blackouts. We define some metrics to evaluate stress, compute those using real data from Western Electricity Coordinating Council (WECC) and relate them to cascading blackouts. The cases analyzed include five for 2016, one from 2012, plus the September 8, 2011 pre-blackout state.

            Our work analyzes properties of a novel network based on line outage distribution factors (LODF or DFAX) instead of the familiar network based on the Y-bus matrix. The traditional Y-bus matrix only reflects the physical connections between the buses. The Y-bus network is excellent for studying how power flows through a network, but for contagion to cascading, the issue is how failures can propagate in a network, and Y-bus is not very helpful.  More details about the DFAX network will be discussed later.

            In the second chapter of this report, definitions of DFAX and metrics that measure stress are presented using an example of a thirteen bus system. For some areas, mostly the southwestern part of the Western Interconnection (WI), these metrics were computed using two software tools, PowerWorld and a program written for this research using MATLAB. We analyze five seasonal cases for 2016, with very different levels of load. PowerWorld was used to read the WECC data and to calculate the DFAX matrix, which is the foundation matrix for our new network. This matrix along with some other data from PowerWorld was processed using programs written in MATLAB to get metrics showing characteristics of the cascading failure system. We compare WECC results to results from earlier studies of other systems.                              In the third chapter, the pre-blackout state of the southwestern part of the WI before the September 08, 2011 cascading blackout is analyzed. The system stress for this case is compared to the five cases of 2016 and to a peak summer case for 2012. The metrics proposed in Chapter 2 were computed for this case. We also sought a tipping point between stressed and not-stressed system, with unexpected results.

            In Chapter 3 we also analyze “hot spots” with stunning results. Monitoring these metrics on December 8, 2011would have identified Area 1 as highly vulnerable, and outages of any of a single series of four lines as critical contingencies. This series includes the line whose failure initiated the blackout and a second outage of the same series which would have had exactly the same effect. Had these results been available on September 8, the substation work which triggered the blackout absolutely should not have been approved without prior action to reduce the vulnerability of Area 1. Chapter 4 (plus a copy of a computer program) contains the effect of errors in the network model on the values of the DFAXes, computed by a standard power flow program. Then the effects of inaccurate DFAXes on the stress metrics are examined. Finally, a description is given of the software developed and submitted to WECC in this study.

 

Monday April 24, 2017

12:00 PM

ECE conference room

Merrill Engineering Building (MEB) room 2109

The public is invited

Pooja Mehta MS thesis final defense

UNIVERSITY OF UTAH

ELECTRICAL AND COMPUTER ENGINEERING DEPARTMENT

 

THESIS DEFENSE FOR THE DEGREE OF

MASTER OF SCIENCE

 

by

 

Pooja Mehta

Advisor: Joel Harley

                                                                                                                                                           

 

Structural Health Monitoring With Large Data Sets

 

 

Carbon fiber-reinforced composite materials have been increasingly used in aerospace and aeronautics industries due to their superior strength over metals, low fatigue life, high corrosion, and temperature resistance. Since most damage, such as delaminations, manifest inside the composite material, we often cannot detect damage through visual inspection. As a replacement for visual inspection, ultrasonic guided waves have been widely researched to remotely detect, locate and characterize damage in structures due to their unique capability to travel long distances and inspect inaccessible locations for damage. Yet, the anisotropic nature of composites makes it difficult to identify the velocity characteristics of the guided waves and utilize them for damage localization.

 

To address this challenge, we use sparse wavenumber analysis to determine the anisotropic frequency-wavenumber characteristics of guided waves. We then use these multimodal and dispersive properties to predict how guided waves propagate in the anisotropic plate through sparse wavenumber synthesis. Finally, these predictions, which form a wave propagation model for the composite, are integrated with matched field processing, a model-based localization framework, to locate damage on the composite.

 

 

 

Tuesday April 18, 2017

11:00 AM

Patel seminar room

Warnock Engineering Building (WEB) room 2460

The public is invited

 

Peter Hillyard PhD final defense

UNIVERSITY OF UTAH

ELECTRICAL AND COMPUTER ENGINEERING DEPARTMENT

 

DISSERTATION DEFENSE FOR THE DEGREE OF

DOCTOR OF PHILOSOPHY

 

by

Peter Hillyard

Advisor: Neal Patwari

 

 

New RF Sensor Network Measurement Models and Methods for Tracking Applications

 

 

Device-free localization (DFL) and tracking services are important components in security, emergency response, home and building automation, and assisted living applications where an action is taken based on a person’s location. In this dissertation, we develop new methods and models to enable and improve DFL in a variety of radio frequency sensor network configurations. 

 

In the first contribution of this work, we develop a linear regression and line stabbing method which use a history of line crossing measurements to estimate the track of a person walking through a wireless network. Our methods provide an alternative approach to DFL in wireless networks where the number of nodes that can communicate with each other in a wireless network is limited and traditional DFL methods are ill-suited.

 

We then present new methods that enable through-wall DFL when nodes in the network are in motion. We demonstrate that we can detect when a person crosses between ultra-wideband radios in motion based on changes in the energy contained in the first few nanoseconds of a measured channel impulse response. Through experimental testing, we show how our methods can localize a person through walls with transceivers in motion.

 

Next, we develop new algorithms to localize boundary crossings when a person crosses between multiple nodes simultaneously. We experimentally evaluate our algorithms with RSS measurements collected from a row of RF nodes placed along a boundary and show that our algorithms achieve orders of magnitude better localization classification than baseline DFL methods.

 

We then present a way to improve the models used in through-wall radio tomographic imaging with E-shaped patch antennas we develop and fabricate which remain tuned even when placed against a dielectric. Through experimentation, we demonstrate the E-shaped patch antennas lower localization error by 44% compared with omnidirectional and microstrip patch antennas.

 

In our final contribution, we develop a new mixture model that relates a link’s RSS as a function of a person’s location in a wireless network. We develop new localization methods that compute the probabilities of a person occupying a location based on our mixture model. Our methods continuously recalibrate the model to achieve a low localization error even in changing environments.

 

 

Monday April 17, 2017

9:30 AM

Merrill Engineering Building (MEB) room 2325

The public is invited

Jonathan Hedstrom PhD final defense 3/24

UNIVERSITY OF UTAH

ELECTRICAL AND COMPUTER ENGINEERING DEPARTMENT

 

DISSERTATION DEFENSE FOR THE DEGREE OF

DOCTOR OF PHILOSOPHY

 

by

Jonathan Hedstrom

Advisor: Behrouz Farhang

 

 

ACHIEVING NEAR MAP PERFORMANCE WITH AN EXCITED MARKOV CHAIN MONTE CARLO MIMO DETECTOR

 

 

The continuous grow of wireless communication use has largely exhausted the limited spectrum available. Methods to improve spectral efficiency are in high demand and will continue to be for the foreseeable future. Several technologies have the potential to make large improvements to spectral efficiency and the total capacity of networks including massive multiple-input multiple-output (MIMO), cognitive radio, and spatial-multiplexing MIMO. Of these, spatial-multiplexing MIMO has the largest near term potential as it has already been adopted in the WiFi, WiMAX, and LTE standards. Although transmitting independent MIMO streams is cheap and easy, with a mere linear increase in cost with streams, receiving MIMO is difficult with the optimal methods having exponentially increasing cost and power consumption. Suboptimal MIMO detectors such as K-Best have a drastically reduced complexity compared to optimal methods but still have an undesirable exponentially increasing cost with data-rate. The Markov Chain Monte Carlo (MCMC) detector has been proposed as a near-optimal method with polynomial cost, but it has a history of unusual performance issues which have hindered its adoption.

 

In this defense, we will introduce a revised bitwise MCMC MIMO detector. The new approach resolves the previously reported high SNR stalling problem of MCMC without the need for hybridization with another detector method or adding heuristic temperature scaling terms. The new excited MCMC (X-MCMC) detector is shown to have near maximum-a-posteriori (MAP) performance even with challenging, realistic, highly-correlated channels at the maximum MIMO sizes and modulation rates supported by the 802.11ac WiFi specification, 8×8 256 QAM.

 

 

 

 

Friday March 24, 2017

9:00 AM, Scott Seminar Room

Warnock Engineering Building (WEB) room 1460

The public is invited

 

 

Anh Luong PhD final defense 3/27

UNIVERSITY OF UTAH

ELECTRICAL AND COMPUTER ENGINEERING DEPARTMENT

 

DISSERTATION DEFENSE FOR THE DEGREE OF

DOCTOR OF PHILOSOPHY

 

by

Anh Luong

Advisor: Neal Patwari

 

 

RF SENSING NETWORKS FOR LOCALIZATION, SYNCHRONIZATION, AND HEALTH MONITORING

 

 

Low-cost wireless embedded systems can make radio channel measurements for the purposes of radio localization, synchronization, and breathing monitoring. Most of those systems measure the radio channel via the received signal strength indicator (RSSI), which is widely available on inexpensive radio transceivers. However, the use of standard RSSI imposes multiple limitations on the accuracy and reliability of such systems; moreover, higher accuracy is only accessible with very high-cost systems, both in bandwidth and device costs. On the other hand, wireless devices also rely on synchronized notion of time to coordinate tasks (transmit, receive, sleep, etc.), especially in time-based localization systems. Existing solutions use multiple message exchanges to estimate time offset and clock skew, which further increases channel utilization.

 

In this dissertation, the design of the systems which use RSSI for device-free localization, device-based localization, and breathing monitoring applications are evaluated. Next, the design and evaluation of novel wireless embedded systems are introduced to enable more fine-grained radio signal measurements to the application. I design and study the effect of increasing the resolution of RSSI beyond the typical 1 dB step size, which is the current standard, with a couple of example applications: breathing monitoring and gesture recognition. Lastly, the \emph{Stitch} architecture is then proposed to allow the frequency and time synchronization of multiple nodes’ clocks. The prototype platform, Chronos, implements radio frequency synchronization (RFS), which accesses complex baseband samples from a low-power low-cost narrowband radio, estimates the carrier frequency offset, and iteratively drives the difference between two nodes’ main local oscillators (LO) to less than 3 parts per billion (ppb). An optimized time synchronization and ranging protocols (EffToF) is designed and implemented to achieve the same timing accuracy as the state-of-the-art but with 59\% less utilization of the UWB channel. Based on this dissertation, I could foresee Stitch and RFS to further improve the robustness of communications infrastructure to GPS jamming, allow exploration of applications such as distributed beamforming and MIMO, and enable new highly-synchronous wireless sensing and actuation systems.

 

 

 

Monday March 27, 2017

9:00 AM, ECE Conference Room

Merrill Engineering Building (MEB) room 2109

The public is invited

 

Spencer Shiveley MS thesis final defense

UNIVERSITY OF UTAH

ELECTRICAL AND COMPUTER ENGINEERING DEPARTMENT

 

THESIS DEFENSE FOR THE DEGREE OF

MASTER OF SCIENCE

 

by

 

Spencer Shiveley

Advisor: Joel Harley

                                                                                                                                                           

 

Structural Health Monitoring With Large Data Sets

 

 

Structural health monitoring systems collect and process large volumes of data taken over many years of a structure’s service. Ultrasonic guided wave systems, in particular, must process an abundance of time-domain waveform data from widely distributed sensors. As few as 8 sensors that transmit and receive ultrasonic waves in pitch-catch mode every 10 minutes can accumulate over one terabyte of data in five to ten years. This number quickly rises as systems grow in size and complexity. As a result, computation and storage efficiency is extremely important, and current guided wave damage detection technologies cannot efficiently process such large data sets. This thesis starts with an introduction and survey of the structural health monitoring and data compression fields. A dimensionality reduction technique using random projections is proposed. The potential for dimensionality reduction method for improving computation time and storage efficiency is discussed. Random projections using sparse matrices is investigated as a tool in implementing a real-time structural health monitoring system with singular value decomposition as a damage detection method. At the end, future directions for research to make this technology more viable in application are suggested.

 

Thursday March 9, 2017

11:30 AM

ECE conference room

Merrill Engineering Building (MEB) room 3235

The public is invited

 

Jingru Zhou PhD final defense

UNIVERSITY OF UTAH

ELECTRICAL AND COMPUTER ENGINEERING DEPARTMENT

 

DISSERTATION DEFENSE FOR THE DEGREE OF

DOCTOR OF PHILOSOPHY

 

by

Jingru Zhou

Advisor: V. John Mathews

 

 

 

Impact location estimation and damage classification for composite structures

 

 

Impacts that occur during service or maintenance are major causes of in-service damage of aerospace structures. Therefore, impact location estimation techniques and damage assessment are key components for structural health monitoring (SHM), especially for large and expensive structures such as aircraft and space vehicles. This dissertation focuses on composite structures, which typically are anisotropic and consequently have more complex wave propagation properties than isotropic structures. The SHM system employs a sensor array that acquires acoustic emission (AE) signals emitted from impact locations. In this dissertation, two impact location estimation algorithms that employ AE signals are proposed for anisotropic composite structures. The first one requires minimum velocity information, while the second one does not require any prior knowledge of the direction and location-dependent wave propagation properties within the structures and it is able to perform fast and accurate impact location estimation. The performance of the algorithms is first assessed by numerical simulations to understand the capabilities of the algorithms and to design the experimental setup. Experimental validation was also performed using different types of composite structures. The results demonstrated the ability of the methods to accurately estimate the impact location in composite structures without prior knowledge of the wave propagation properties of the structures.

 

The second goal of this dissertation is damage assessment from AE signals and focuses on damage classification. Specifically, the goal is to identify the most important features of the AE signals that identify impacts resulting in structural damage and also classify the damage as either delamination only or delamination plus fiber breakage. An efficient machine learning approach based on logistic regression is developed for this purpose. The experimental preparations which included composite structure selection, sensor selection, type of impacting experiments and inspection for damage were carefully and systematically set up for training and testing the method. The most useful features of the AE signals were obtained from the training data. Cross-validation experiments indicated that the methods identified impacts resulting in damage with 100% accuracy. Classification of damage type showed a 74% accuracy.

 

 

 

Wednesday January 4, 2017

1:30 PM, ECE conference room

Merrill Engineering Building (MEB) room 2109

The public is invited

 

Xiaojun Sun PhD final defense

UNIVERSITY OF UTAH

ELECTRICAL AND COMPUTER ENGINEERING DEPARTMENT

 

DISSERTATION DEFENSE FOR THE DEGREE OF

DOCTOR OF PHILOSOPHY

 

by

 

Xiaojun Sun

Advisor: Priyank Kalla

 

 

Word-level Abstractions for Sequential Design Verification using Algebraic Geometry

 

 

Formal verification of hardware designs has become an essential component of the overall system design flow. The designs are generally modeled as finite state machines, on which property and equivalence checking problems are solved for verification. Reachability analysis forms the core of these techniques. However, increasing size and complexity of the circuits causes the state explosion problem. Abstraction is key to tackle the scalability challenge.

This dissertation presents new techniques for word-level abstraction with applications in sequential design verification. By bundling together k bit-level state-variables into one word-level constraint expression, the state-space is construed as solutions (variety) to a set of polynomial constraints (ideal), modeled over the finite (Galois) field of 2^k elements. Subsequently, techniques from algebraic geometry — notable, Groebner basis theory and technology — are researched to perform reachability analysis and verification of sequential circuits. This approach adds a `”word-level dimension” to state-space abstraction and verification to make the process more efficient.

While algebraic geometry provides powerful abstraction and reasoning capabilities, the algorithms exhibit high computational complexity. In the dissertation, we show that by analyzing the constraints, it is possible to obtain more insights about the polynomial ideals, which can be exploited to overcome the complexity. Using our algorithm design and implementations, we demonstrate how to perform reachability analysis of finite-state machines purely at the word-level. Using this concept, we perform scalable verification of sequential arithmetic circuits. As contemporary approaches make use of resolution proofs and unsatisfiable cores for state-space abstraction, we introduce the algebraic geometry analog of unsatisfiable cores, and present algorithms to extract and refine unsatisfiable cores of polynomial ideals. Experiments are performed to demonstrate the efficacy of our approaches.

 

 

 

Friday December 16, 2016

12:30-2:30 PM, ECE conference room

Merrill Engineering Building (MEB) room 3235

The public is invited

Barun Gupta PhD final defense

UNIVERSITY OF UTAH

ELECTRICAL AND COMPUTER ENGINEERING DEPARTMENT

 

DISSERTATION DEFENSE FOR THE DEGREE OF

DOCTOR OF PHILOSOPHY

 

by

Barun Gupta

Advisor: Ajay Nahata

 

 

CONTROLLING PROPAGATION PROPERTIES OF SURFACE PLASMON POLARITON AT TERAHERTZ FREQUENCY

 

Despite great scientific exploration since 1900’s, the terahertz range is one of the least explored regions of electromagnetic spectrum today. In the field of plasmonics, texturing and patterning allows for control over electromagnetic waves bound to the interface between a metal and the adjacent dielectric medium. These surface plasmon-polaritons (SPPs) display unique dispersion characteristics that depend upon the plasma frequency of the medium. In the long wavelength regime, where metals are highly conductive, such texturing can create an effective medium that can be characterized by an effective plasma frequency that is determined by the geometrical parameters of the surface structure. The terahertz (THz) spectral range offers unique opportunities to utilize such materials. While there has been significant work on developing coherent sources and detectors, relatively few other device technologies currently exist. A major issue that has constrained this development is the fact that most conventional dielectrics and semiconductors are highly lossy in this spectral range. Since metals are highly conductive, SPPs experience very low propagation loss when air acts as the adjacent dielectric medium.

                This thesis describes a number of terahertz plasmonic devices, both passive and active, fabricated using different techniques. As an example, inkjet printing is exploited for fabricating two-dimensional plasmonic devices. In this case, we demonstrated the terahertz plasmonic structures in which the conductivity of the metallic film is varied spatially in order to further control the plasmonic response. Using a commercially available inkjet printers, in which one cartridge is filled with conductive silver ink and a second cartridge is filled with resistive carbon ink, computer generated drawings of plasmonic structures are printed in which the individual printed dots can have differing amounts of the two inks, thereby creating a spatial variation in the conductivity. The silver ink has a DC conductivity that is only a factor of six lower than bulk silver, while the carbon ink acts as a lossy dielectric at THz frequencies. Both inks sinter at room temperature immediately after contact with the plastic film. Using a periodic array of subwavelength apertures as a test structure, patterns printed with different fractional amounts of the two inks show dramatically different enhanced optical transmission properties. These differences arise from changes in the propagation loss properties as a function of conductivity. This data is used to design and fabricate aperture arrays in which the conductivity varies spatially. The resulting plasmonic effect is found to dramatically alter the spatial beam profile of the transmitted THz radiation, as measured by THz imaging. These plasmonic devices are passive devices.

The inkjet printing technique is limited to the two-dimensional structurers. In order to expand the capability of printing complex terahertz devices, which cannot otherwise be fabricated using standard fabrication techniques, we employed 3D printing. This technique is an additive manufacturing approach, which allows for the fabrication of complex terahertz devices using polymers. The printed structures are then coated with ~ 1 µm of gold on all sides, in order to make it a plasmonic device. Using this printing methodology, we fabricated both planar and non-planar terahertz waveguide devices, including 3D bends, 3D y-splitters and curved waveguides. For the purposes of comparison, we fabricated terahertz waveguide devices and compared them with the standard waveguide devices fabricated using laser ablation techniques with stainless steel films. We find excellent agreement between these two types of devices. At a stage where THz technology is still maturing, the development of devices where the propagation properties can be easily modulated holds great promise for the development of terahertz optoelectronic devices.

In the realm of active plasmonic devices, a wide range of innovative approaches have been developed utilizing a variety of materials including liquid crystals semiconductors, liquid metals, photochromic and electrochromic molecules and phase-change materials. One of the most heavily studied phase change materials for active plasmonic and metamaterial device implementations is vanadium dioxide, VO2, which undergoes a thermally-driven metal–insulator transition near room temperature associated with a structural change in its crystal symmetry. Phase transitions can lead to a variety of different macroscopic effects, which may be useful for active optical applications. As an example, shape memory alloys (SMAs) can be thermally cycled between different physical geometries. As an example, Nitinol, a nickel-titanium alloy, has been shown to be associated with a transformation between the martensite phase below the transition temperature and the austenite phase above the transition temperature. The two most commonly used approaches to control these transitions are referred to as one-way memory and two-way memory. In the former approach, an SMA that has been deformed returns to its original shape after being heated.  This is most commonly demonstrated using wires, though numerous applications utilizing thin metal foils have also been shown. Two-way memory requires that the SMA undergo specific thermo-mechanical treatments, commonly referred to as training procedures, in order to thermally cycle between two different alloy geometries.

We discuss the use of SMAs for terahertz (THz) plasmonics that allows for switching between different physical geometries corresponding to different electromagnetic responses. We use Nitinol, a metal alloy of nickel and titanium composed of approximately equal atomic percentages, as the SMA medium that is structured to give the desired electromagnetic response. Nitinol has a DC conductivity of ~1.25 x 106 S/m for both phases which is similar to the value for stainless steel, making it well suited for THz plasmonic applications. As an SMA, it undergoes a structural transition between the martensite phase to the austentite phase that is bistable and reproducible at temperatures that are only slightly above room temperature. Using a two-way training protocol that we developed, we created samples that transition between either a one-dimensional (1D) or two-dimensional (2D) sinusoidally corrugated geometry and a flat substrate. In order to observe a plasmonic response, the foils are patterned either with a periodic array of subwavelength apertures or a single aperture and their transmission properties are measured using THz time-domain spectroscopy

Friday December 16, 2016

3-5 PM, ECE conference room

Merrill Engineering Building (MEB) room 2109

The public is invited