Research Day

The College of Engineering and Computer Science offers six Ph.D. programs and eleven M.S. programs. The programs offer graduate students opportunities for high-quality research.  The research office provides resources for the professional and social life of our graduate students.

To learn about our latest groundbreaking research, visit our research news and events page.

Research Day

  • Research Day
  • We celebrate graduate student research in the College of Engineering and Computer Science on our annual Research Day. Each Research Day, industry representatives, faculty, and students from a wide range of disciplines learn about novel approaches to solving challenging research problems. Through poster presentations and research pitches, Engineering & Computer Science graduate students communicate the intellectual merit and broader impacts of their research in six signature areas:

    • Health and Wellbeing
    • Unmanned Systems
    • Energy Sources Conversion, and Conservation
    • Sustainable and Built Systems
    • Intelligent Systems
    • Security

    Research Day 2020

  • Research Day 2020
  • True to a long-standing tradition at the College of Engineering and Computer Science, we plan to hold the ECS Research Day on Friday, November 6, 2020. The event this year will be virtual.


    • Poster and Presentation Competition Winners

      Poster Competition Awards

      Overall College Poster Prize

      Lin Zhang. EECS. Advisor: Dr. Fanxin Kong

      Departmental awards (1-2 per dept. depending on the number of posters)


      Francielli Genier. Advisor: Dr. Ian Hossein

      Katie Piston. Advisor: Dr. Shikha Nangia


      Libin Yang. Advisor: Dr. Zhao Qin.


      Shengmin Jin. Advisor: Dr. Reza Zafarani

      Zeinab Saghati Jalali. Advisor: Dr. Sucheta Soundarajan


      Jinwoo Song. Advisor: Dr. Young Moon


      Research Pitch Competition Awards

      First Prize

      Katie Piston. Advisor: Dr. Shikha Nangia

      Second Prize

      Thomas Welles. Advisor: Dr. Jeongmin Ahn

      Third prize

      Francielli Genier. Advisor: Dr. Ian Hosein

      Alexander Hartwell. Advisor: Dr. Jeongmin Ahn

    • Keynote address: Dr. Mohammad Khawer ’95 MS, ’15 PhD

      Dr. Mohammad R. Khawer, ’95 MS, ’15 PhD

      Research Day Website

      Dr. Khawer is Bell Labs Fellow & Head of Technology Innovation – Enterprise Digital Automation & Digital Health. An inventor and disruptive technology innovation leader, he is currently leading the development of breakthrough technologies in healthcare, specifically for rare diseases. In his role at Nokia, he also spearheads the creation of disruptive solutions through partnership with startups and other strategic alliances.

      Dr. Khawer holds 73 granted patents, with several dozen more filed patents, including in the Digital Health application domain. He was bestowed with the prestigious Bell Labs Fellow award; granted to only 320 individuals, including several Nobel Laureates. Since 2016, he has spearheaded an internal startup within Nokia, that has resulted in three commercial products.

      Some of the many awards he received include Nokia Business Excellence Award, Nokia Top Inventors Recognition Award, Bell Labs Fellow, and Bell Labs Inventor of the Year Award. Dr. Khawer completed his MS in Computer Science and PhD in Computer & Information Science and Engineering from Syracuse University.

      Disruptive Technology Innovations for the Vertical Market Segment
      Mohammad R. Khawer, Ph.D.
      Bell Labs Fellow & Head of Technology Innovation, Nokia Enterprise Digital Automation

      New vertical market segment that includes Industrial Automation, Private Enterprise Network (Nuclear Power Plants, Fortresses including the Federal Prison system, Package Delivery Companies, Wind Farms, Ports, Mines, Hospitals etc.) have often sought ability to have high speed cellular (LTE for now and 5G in near future) access capabilities that offer more reliable and superior Quality of Service (QoS) in comparison to Wi-Fi networks. Until recently, such capabilities were not possible as the spectrum needed to run such private cellular networks has been in the form of statically licensed spectrum owned by Mobile Network Operators (MNOs) and could not be sub-leased on the sites of interest. With emergence of shared spectrum, wherein highly under-utilized spectrum owned and used by e.g., US federal entities could be shared among commercial users and small market entrants, the issue of availability of spectrum for the vertical market segment is no longer a hurdle. Availability of the shared spectrum obviates the dependence on the licensed spectrum that is owned by the MNOs for deploying cellular network for private enterprises. Specifically, newly formed FCC rules in 47 Code of Federal Regulations (CFR) Part 96 allows sharing of the 3550-3700 MHz Citizens Broadband Radio Service (CBRS) Spectrum band.
      The CBRS band has thus opened-up the possibility as an innovation band for new small entrants such as the digital automation verticals to deploy their own private cellular (LTE/5G) Enterprise Network without any dependency to acquire the LTE/5G service from their regional wireless providers, and architect a private cellular (LTE/5G) enterprise network that meets their own specific mission critical needs for service and coverage.
      For most of the vertical market segment, data isolation/localization is a critical requirement from a legal and security perspective, where the data must remain localized/isolated within the premises of the private enterprise network and should not traverse the MNOs core network that is hosted outside the customer premises. This requirement alone prevents the use of an MNO provided enterprise network deployments for such vertical market segment deployments.
      Nokia Digital Automation Cloud (NDAc) using Nokia Spectrum Controller (NSC) offers a disruptive market differentiating and innovative cloud based scalable plug and play private enterprise network solution for vertical market segment using appropriate mix of various available shared, unlicensed, and opportunistic licensed spectrum options, accounting for the size of the customer site, required network capacity, criticality and mix of applications that need to be supported.

    • Competition Prizes and Criteria

      Poster Competition


      • $300 overall College Prize
      • $250 Department Prize (we will award one or two prizes per department, depending on the number of posters submitted).


      1. scientific contribution of the research
      2. potential applications or impacts of the research
      3. clarity of the presentation, especially for a non-specialist in the field.

      Research Pitch Competition


        • $300 First Place
        • $200 Second Place
        • $100 Third Place


      1. The research addresses a challenging problem and/or uses a novel approach
      2. The research has significant potential applications or impacts
      3. The presentation is clear, especially for student/faculty/industry audience that represents a wide range of disciplines.
      4. Information on the slide is readable from a distance AND includes text or graphics that illustrate the key points the presenter makes during the pitch.
    • Agenda

      9:30—9:45 am


      9:45—11:00 am

      Poster Competition—even numbers

      11:00 a.m. —12:15 pm

      Poster Competition—odd numbers

      12:15—1:00 pm

      Lunch Break

      1:00—2:00 pm
      Keynote address: Dr. Mohammad Khawer ’95 MS, ’15 PhD
      Opening remarks: Dean Peter A. Vanable, Dean of the Graduate School, Associate Provost for Graduate Studies

      2:00—3:00 pm
      Research Pitch Competition

      3:00—3:15 pm

      3:15—3:30 pm
      Poster and research pitch winners are announced. Award Ceremony

      3:30 pm
      2020 ECS Research Day Adjourns

    • Biomedical and Chemical Engineering Posters
    • Civil and Environmental Engineering Posters
    • Computer and Information Science Posters
    • Electrical and Computer Engineering Posters
    • Mechanical and Aerospace Engineering Posters
    • Poster and Presentation Competition Winners


      • Eradication of bacterial persister cells by targeting membrane potential, Sweta Roy
      • 3D Multiscale/Multimaterial Laser Printing Platform for Shaping Hydrogel, Zheng Xiong
      • Reinforced Concrete Columns Damaged by Fire and Retrofitted with CFRP and Steel Jackets, Jia Xu
      • Nested Bigrams: A Novel Approach in Python Source Code De-Anonymization, Pegah Hozhabrierdi
      • Human Supervision for Machine Automated Decision Making, Baocheng Geng
      • Fusion of Deep Neural Networks for Activity Recognition: A Regular Vine Based Approach, Shan Zhang
      • A Secure Cyber-Manufacturing System Augmented by the Blockchain, Jinwoo Song
      • Intrusion Detection and Correlation for Cyber-Physical Attacks in Cyber-Manufacturing System, Mingtao Wu


      • Machine Learning and Fluid Dynamics, Andrew Tenney
      • Genes and Environment: Digging into the Whole Story of Heart Disease, Chenyan Wang
      • New Bioprinting Technology to Print Large-scale Organs, Zheng Xiong
    • Keynote Lecture by Jason Gomez ’92, Ph.D.

      Jason Gomez

      Celebrating its 150th year, the Naval Undersea Warfare Center (NUWC), Division Newport, is the Navy’s full-spectrum research, development, test and evaluation, engineering, and fleet support center for submarine warfare systems and many other systems associated with the undersea battlespace.

      NUWC Division Newport provides the technical foundation that enables the conceptualization, research, development, fielding, modernization, and maintenance of systems that ensure our Navy’s undersea superiority.

      NUWC Division Newport is responsible, cradle to grave, for all aspects of systems under its charter, and is engaged in efforts ranging from participation in fundamental research to the support of evolving operational capabilities in the U.S. Navy fleet. The major thrust of NUWC Division Newport’s activities is in applied research and system development.

      With headquarters in Rhode Island, NUWC Division Newport operates detachments at West Palm Beach, Florida, and Andros Island in the Bahamas. Remote test facilities are located at Seneca Lake, Fisher’s Island in New York, and Dodge Pond Connecticut.

      NUWC Division Newport, with its sister Undersea Warfare center in Keyport Washington, and the nine Surface Warfare Centers across the country, have approximately 24,000 civilian scientist, engineers, and technicians supporting our Fleet.  NUWC’s Vision: Undersea Superiority: Today and Tomorrow.

      In this talk we will introduce the Naval Undersea Warfare Center, and the role it plays in developing warfighting capability for our sailors.  We will follow an example of bio-inspired technology from its basic research inception, thru early prototyping, and its application to a Navy need.  We’ll also explore the wide breath of technologies required to develop an underwater weapon and how they all interact to establish a complete system capability.

      About Jason Gomez ’92. Ph.D.

      Jason Gomez is the chief scientist of the Undersea Warfare Weapons, Vehicles, and Defensive Systems Department at NUWC Division Newport. He is charged with developing innovative solutions to emerging undersea warfighting needs, and coordinating the efforts of multiple organizations to ensure they stay focused on the future Fleet.  Gomez graduated from Syracuse University with a B.S. in aerospace engineering in 1992. In that same year he started working at NUWC Division Newport in the weapons development department. He continued his education while working, and was able to complete both a master’s and Ph.D. in mechanical engineering from the University of Rhode Island. Throughout his 26 year career, Gomez has worked to advance all areas of technology associated with underwater weapons including acoustics, drag reduction, power and energy, propulsion, and control. He has authored multiple refereed journal articles, has four patents, and was awarded the Navy Meritorious Civilian Service Award.

    • Competition Prizes and Criteria

      Poster Competition


      • $300 overall College Prize
      • $250 Department Prize (we will award one or two prizes per department, depending on the number of posters submitted).


      1. scientific contribution of the research
      2. potential applications or impacts of the research
      3. clarity of the presentation, especially for a non-specialist in the field.

      Research Pitch Competition


      • $300 First Place
      • $200 Second Place
      • $100 Third Place


      1. The research addresses a challenging problem and/or uses a novel approach
      2. The research has significant potential applications or impacts
      3. The presentation is clear, especially for student/faculty/industry audience that represents a wide range of disciplines.
      4. Information on the slide is readable from a distance AND includes text or graphics that illustrate the key points the presenter makes during the pitch.
    • Biomedical and Chemical Engineering Posters

      Engineering Nonuniform Mechanical Environment to 3D Human Cardiac Microtissues

      Presented by: Chenyan Wang, Advisor: Zhen Ma

      Poster number: 1

      Nonuniform mechanical environment contributes to the development of heart dysfunctions while cardiac tissues can adjust their contractile functions to adapt to nonuniform tissue mechanics in the heart. Current in vitro cardiac tissue models with uniform mechanical environment fails to recapitulate the mechanical heterogenicity during the development of heart disease, on the other hands, the complexity of in vivo models always makes it difficult to assess the cardiac response. We developed an in vitro 3D cardiac microtissue model where cardiac tissue experienced heterogenous mechanical load provided by artificial hybrid matrices composed of fibers with different stiffness. We found cardiac microtissues exhibited a higher degree of adaptivity to double hybrid matrices while the limited ability of self-adjustments under triple hybrid matrices. This novel hybrid system facilitated the establishment of pathological-inspired cardiac microtissue model for the deeper understanding of heart pathology due to nonuniformity of the mechanical environment experienced by the heart muscle.


      New transdermal drug delivery agent: a multiscale characterization

      Presented by: Kathryn Piston, Advisor: Shikha Nangia

      Poster number: 2

      Dermatological diseases like psoriasis and eczema respectively affect 7.5 million in the United States and 5-20% of children worldwide. Transdermal drug delivery is the ideal treatment but is challenging due to low skin permeability. A unique class of ionic liquid namely Choline And Geranic acid (CAGE) developed by the Mitragotri laboratory at Harvard University, represents a promising new transdermal drug delivery agent as it holds desirable antimicrobial properties. Understanding of CAGE in the presence of water is critical for transdermal applications as both atmospheric and physiological water contact is inevitable. It was found that under atmospheric conditions, CAGE like a typical ionic liquid contained up to 20% water by mole. Using a multiscale simulation approach in collaboration with the experimental work performed in Mitragotri lab, we characterized the properties of water-CAGE mixtures for varying concentrations. Simulations suggest that based on the water content, CAGE ions organize into nanostructural domains to minimize contact of their hydrophobic tails with water; however, under the standard atmospheric conditions, CAGE could be used without pre-drying in most applications. This collaborative work pioneers the understanding of CAGE and its potential use for transdermal applications and commercialization in the future.


      A new inexpensive computational method to study membrane protein association

      Presented by: Nandhini Rajagopal, Advisor: Shikha Nangia

      Poster number: 3

      Integral membrane proteins play a key role in crucial biological events including neurotransmission, the progression of signaling cascade, triggering an immune response and cancer metastasis among others. Not surprisingly, these membrane proteins represent 20-30% of the proteome of most organisms. Despite their significance, experimental study of membrane protein association is highly challenging due to their hydrophobic membrane environment. Although numerous computational methods have been put forth to overcome the limitations, a comprehensive method capturing details of protein-protein assembly remains a challenge. In this work, we present the development of a new method that provides qualitative and quantitative estimates of membrane protein association using their interaction energies to generate a comprehensive Protein Association Energy Landscape (PANEL). This is a computationally inexpensive approach to generating energy profiles for a wide range of proteins belonging to a diverse set of species in variable membrane environments. We have used our approach to show the association of membrane proteins responsible for the selectivity of the blood-brain tight junctions.


      Eradication of bacterial persister cells by targeting membrane potential

      Presented by: Sweta Roy, Advisor: Dacheng Ren

      Poster number: 4

      Bacteria are well known to form persister cells; a subpopulation that is often slow-growing or undergoing growth-arrest but can resume growth after the lethal stress is removed. This feature allows persister cells to be highly tolerant to conventional antibiotics and lead to reoccurring infections. Here, we show that minocycline, a second-generation semi-synthetic antibiotic of the tetracycline class, is more effective in killing persister cells than normal cells. For example, it killed Escherichia coli persister cells by 70.8 ± 5.9%, while it only killed normal cells by 10.3 ± 3.7%. The effects were attributed to lower membrane potential and thus reduced efflux activities of persister cells. Consistently, persister cells were found to accumulate ~3.8 times more minocycline per cell than exponentially growing cells. These findings shed new light on the physiology of persister cells and may help develop new therapies for chronic infections, e.g., those associated implanted medical devices and biomaterials.


      Dynamic Change of Human Stem Cell-Derived Cardiomyocytes on the Programmable Biomaterial Substrate

      Presented by: Shiyang Sun, Advisor: Zhen Ma

      Poster number: 5

      Cardiomyocytes are the main functional cell type in heart and support heart function and normally align in the same direction with an attachment on the extracellular matrix in vivo. Alignment is important in cell shape formation, further cardiac myofibril organization and efficient contractile function of the sarcomere. In previous studies, researchers successfully show cardiomyocytes with higher elongation, closer orientation, enhance contraction and more formation of focal adhesion on the topographic surface. However, although researches proved high improvement of alignment to a cardiomyocyte, there is less evidence to show the dynamic cardiomyocytes behavior respond to topography change and intracellular myofibril remolding. For tracking the reorganization process of cardiomyocytes, we developed shape memory polymer with PEM coating as a dynamic topographic surface system and culture the cardiomyocytes on the surface, drive form IPSCs. Using this system, we investigated the reorganization of the cardiomyocyte.


      Mineral deposition within cell-Laden hydrogels influence their electrical impedance

      Presented by: Vaikhari Marathe, Advisor: Pranav Soman

      Poster number: 6

      Cell-laden hydrogels have been widely used in bone tissue engineering to model cell-mediated mineral deposition that occurs within native bone tissue. Although the composition and morphology of cell-mediated mineral deposits have been well characterized, little is known about the electrical properties of the deposited mineral. Considering the importance of bioelectricity within bone tissue, this study aims to characterize the changes in the electrical impedance properties of the cell-laden hydrogel constructs after mineral deposition from osteoblast-like cells. Materials & Methods. Saos-2 cells encapsulated within gelatin methacrylate (GelMA) hydrogels were chemically stimulated in osteogenic media for 4 weeks. The morphology, composition, and mechanical properties of the mineralized constructs were characterized using brightfield imaging, SEM EDS, FTIR, NMR, microCT, immunostaining, and mechanical compression tests. In parallel, a custom-made device was used to measure the electrical impedance of mineralized constructs. Results. Results demonstrate that an increase in cell-mediated mineral deposition causes a drop in the electrical impedance of the hydrogels. Conclusions. This work provides new information about the electrical nature of deposited mineral and could potentially serve as a first step toward understanding the origins of bioelectricity within mineralized in vitro constructs.


      3D Multiscale/Multimaterial Laser Printing Platform for Shaping Hydrogel

      Presented by: Zheng Xiong, Advisor: Pranav Soman

      Poster number: 7

      One of the requirements for the success of engineered tissues is creating 3D multiscale/multimaterial structures that would recapitulate complex biological environment, for instance, generating the hierarchical network of blood vessels. Although there is a multitude of fabrication techniques and printing platforms, current methods are unable to meet these key requirements such as multiscale, multi-materials, fabrication flexibility, etc. Thus, we have designed and built a versatile new manufacturing technology, hybrid laser printing (HLP) that combines the key advantages of additive projection lithography and optically mediated laser direct writing that offer high-resolution and superior design flexibility. Our new HLP technology can rapidly print user-defined structures at resolutions ranging from several millimeters to sub-micron scale, using multiple model materials. The unique capability of the system to print-on-demand 3D multiscale/multimaterial structures can be exploited a range of applications in biomedical sciences.


      Synthesis of Polymer Waveguides of Increasing Conversions in Solar Cells

      Presented by: Fu-Hao Chen, Advisor: Ian Hosein

      Poster number: 8

      Increasing the power output and efficiency of solar cells through an encapsulant, leaving the native solar cell material unchanged, is the most practical solution that is closest to commercial integration. We have conducted an optical lithography measure to fabricate polymer waveguides and mounted it on a solar cell. The efficiency improves 2.54 % and achieves 14.73%, which means 5.89 GW power is saved based on the global solar power 232 GW. More importantly, the capture angle is up to 70º, largely increased from our previous work, showing a 40º capture angle.


      Safe and Abundant Energy: A Path to Sodium Solid-State Batteries

      Presented by: Francielli Silva Genier, Advisor: Ian Hosein

      Poster number: 9

      An area of intense research related to batteries has been the replacement of the liquid or gel electrolyte with a solid polymer material, namely a solid polymer electrolyte (SPE), because of its potential to double the energy density and further address concerns over the volatility and flammability of currently used organic solvents. The criteria for a solid material to feasibly replace either a liquid or gel electrolyte are (1) suitable ionic conductivity over the batteries operating temperature range (room temperature to 60 °C, usually) and (2) towards safety concerns, thermally stability not only within the operating temperature range, but also well above it. Studies on SPEs do not only focus on different polymer systems, but also on improving the interaction of existing and new polymer networks to ions other than lithium-ion to achieve even less expensive and more efficient rechargeable batteries. Among the univalent ions, sodium can be highlighted due to its wide availability and low cost comparing to lithium. We present a novel crosslinked SPEs for sodium conduction. The SPEs were crosslinked through the copolymerization of polytetrahydrofuran (PTHF) and an epoxy via the active monomer (AM) mechanism, thus without the need of end-functionalization. Sodium ion conductivity and thermal stability were examined, to contribute to the development of cheaper rechargeable all solid state batteries.


      Classification of drug-induced arrhythmic behavior in stem cell-derived cardiomyocytes using nonlinear signal reconstruction

      Presented by: Plansky Hoang, Advisor: Zhen Ma

      Poster number: 10

      Drug-induced cardiotoxicity is a leading cause of drug failure and withdrawal due to unpredictable changes to heart rhythm. These changes are traditionally quantified using linear analysis of amplitude and frequency. However, this cannot objectively

      classify complexities that arrhythmias produce. These complexities represent a signal’s chaotic nature and manifest as changes in contraction rhythm and the presence of signal aberrations that can only be quantified using nonlinear computational analysis. We have implemented human induced pluripotent stem cell technology to derive cardiomyocytes from integrating with phase space analysis for nonlinear analysis of drug-induced arrhythmias. Using this approach, researchers and physicians have at their disposal a variety of quantitative measures to classify arrhythmic behavior to develop personalized treatment and diagnostics.


      Calcium-Ion Rechargeable Battery Comprising Highly Conductive and Thermally Stable Ionic Liquid Gel Electrolyte as an Alternative to Lithium Batteries

      Presented by: Saeid Biria, Advisor: Ian Hosein

      Poster number: 11

      Calcium ion batteries are a promising choice for next-generation rechargeable batteries with high capacity and rate capability as an alternative to lithium batteries. However, the lack of conductive and thermally stable electrolytes limits their use in high demand applications. Herein, we report a rechargeable calcium ion battery operating at room temperature that employs an ionic-liquid gel electrolyte as the separator with high thermal stability up to 300 °C and ion conductivities between 10-4 to 10-3 S/cm at room temperature. Hence this research represents a significant step forward, looking further into the future, to make batteries that are higher performing, safer, more cost-effective, and environmentally friendly.


      Kinetic Study of Ag Containing Adsorbents Aging in Nuclear Fuel Reprocessing Off-Gases

      Presented by: Seungrag Choi, Advisor: Lawrence Tavlarides

      Poster number: 12

      Iodine-129 (129I) is one of the volatile radioactive species in the off-gas streams generated from spent nuclear fuel reprocessing processes. It is very important to capture 129I from off-gas streams due to its long half-life, high mobility, and bioaccumulation. Reduced silver mordenite (Ag0Z) and silver functionalized silica aerogel (Ag0-aerogel), are well-known materials to effectively capture iodine owing to the strong affinity of silver for iodine. However, long-term exposure to off-gas streams including H2O and NOx has a severe impact on the performance of Ag containing adsorbents by causing the deactivation of Ag. To address this problem, the mechanism of aging processes of adsorbents in off-gas streams should be well understood. For better understanding the performance of the adsorbents in off-gas streams, aging and I2 adsorption experiments were executed, and the structural changes to the absorbents through the aging processes were analyzed by scanning electron microscopy (SEM), X-ray diffraction (XRD), X-ray photoelectron spectroscopy (XPS) and X-ray absorption fine structure (XAFS) to determine the mechanisms of the aging processes. Also, kinetic models were developed and evaluated to describe aging processes.


      Physiological Changes in Pseudomonas aeruginosa Cells Induced by Shape Recovery Triggered Biofilm Dispersion

      Presented by: Sang Won Lee, Advisor: Dacheng Ren

      Poster number: 13

      Microbial biofilms are a leading cause of chronic infections in humans and persistent biofouling in industries due to the extremely high-level tolerance of biofilm cells to antimicrobial agents. Eradicating mature biofilms is especially challenging because of the protection of the extracellular matrix and slow growth of biofilm cells. Recently, we reported that established biofilms could be effectively removed (e.g., 99.9% dispersion of 48h Pseudomonas aeruginosa PAO1 biofilms) by shape memory polymer-based dynamic changes in surface topography. In this study, complementary approaches were used to investigate the effects of shape recovery triggered biofilm dispersion on bacterial physiology. Such biofilm dispersion was found to sensitize biofilm cells to conventional antibiotics. For example, shape recovery in the presence of 50 µg/mL tobramycin reduced biofilm cell counts by more than 3 logs (2,479-fold) compared to the static flat control. These results were corroborated with Live/Dead staining and scanning electron microscopy. The observed effects were attributed to the disruption of biofilm structure and increase in cellular activities as evidenced by an 11.8-fold increase in the intracellular level of adenosine triphosphate (ATP), and 4.1-fold increase in expression of the rrnB gene in detached cells. These results can help guide the design of new control methods to better combat biofilm-associated antibiotic-resistant infections.


      Using Visible Light to Develop Advanced Battery Electrodes

      Presented by: Shreyas Pathreeker, Advisor: Ian Hosein

      Poster number: 14

      Conventional Li-ion batteries continue to power a majority of portable electronics and hybrid electric vehicles but are unable to deliver high energy density. Silicon is a candidate material to reversibly store Lithium ions but undergo an enormous expansion in volume upon cycling, consequently leading to mechanical fracture and ultimately, battery failure. To accommodate this volume change, we successfully leverage visible light induced photopolymerization to develop tailored micrometer scale composite pillars which can safely allow the inherent volume expansion of Silicon Nanoparticles. Our approach is scalable and safe, and future work is directed toward tuning the morphology and size of the pillar-like arrays.


      High-Throughput Screening of Small-Molecule Transport Through Pseudomonas aeruginosa’s OccD3 Protein

      Presented by: Yinghui Dai, Advisor: Shikha Nangia

      Poster number: 15

      Development of a new class of therapeutics to treat bacterial infections has gained momentum due to rising antimicrobial resistance. Infections due to multidrug-resistant pathogens such as Pseudomonas aeruginosa have become a real concern in hospital-acquired infections. P. aeruginosa demonstrates intrinsic resistance to antibiotics due to its low outer membrane permeability. Understanding the mechanistic and kinetic aspects of small molecule penetration through the bacterial cell wall could provide insights into increasing the drugs’ uptake and permeability. Our current research focuses on using our newly developed ComputationaL Antibiotic Screening Platform (CLASP) to study the transport behavior through outer membrane protein channels or porins. In his work, we elucidate the mechanism and transport kinetics of known antibiotic molecules through P. aeruginosa’s OccD3 porin as a proof-of-concept. We envision expanding the CLASP platform to new small-molecule libraries to accelerate the drug-discovery pipeline.

    • Civil and Environmental Engineering Posters

      Airport Building Information Modeling Implementation Framework for Smart Airport Life Cycle Management

      Presented by: Basak Keskin, Advisor: BarisSalman

      Poster number: 16

      The research topic -Airport Building Information Modeling (ABIM) Implementation Framework for Smart Airport Lifecycle Management- is fundamentally based upon the significant need of digital disruption in the infrastructure sector. The new era of smart infrastructure has started to be recognized by both academia and industry as the demand for infrastructure upgrade increases at a fast pace. The McKinsey Global Institute estimates that the world will need to spend $57 trillion on infrastructure by 2030 to keep up with global GDP growth. This estimate can also be projected on airports since they are assigned a grade of D in ASCE 2017 Infrastructure Report Card. One of the reasons behind this result can be using a reactive approach instead of a proactive approach for maintenance and renovation processes, that leads to disturbance in regular airport operations, and as well as passenger journey experience.  Accordingly, having been one of the most important economic engines, design-build-operate life cycle of airports should be digitized to meet ever-evolving demands and needs of today’s society. World Economic Forum also articulates that digitalization is central to the required transformation, and digital tools like Building Information Modeling will support renovation; data platform for condition monitoring and predictive maintenance; a data repository for facility and asset management; and platform for virtual handover and commissioning in operation phase.


      Dewatering Performance of Coal Ash

      Presented by: Chinthoory Ganesalingam, Advisor: Shobha Bhatia

      Poster number: 17

      Coal ash is a potentially harmful massive industrial waste generated in the United States. The unused coal ash disposed to the environment has several risks like groundwater contamination, emission of air-borne particulate and catastrophic failure of coal ash ponds. Therefore, sound recycling and safe disposal of coal ash and the closure of unsafe coal ash ponds have become crucial in the country. Geotextile tube dewatering is developing as a promising method in coal ash waste management. Although there are different types of performance tests and indices are used in practice to assess the dewatering behavior, a direct comparison cannot be made between those tests due to missing common platform for comparison. Also, current performance tests are limited in capacity to provide concrete information about large scale tube dewatering. With increasing demand for geotextile tube dewatering, it is necessary to develop a cost and time effective approach to simulate the field conditions without excess manipulation and thus understand the effect of sediment properties and chemical conditioning in dewatering behavior. The major goal of this research is to investigate the dewatering behavior of coal ash sediments using various laboratory tests and to establish a framework for predicting the large-scale geotextile tube performance.


      Reinforced Concrete Columns Damaged by Fire and Retrofitted with CFRP and Steel Jackets

      Presented by: Jia Xu, Advisor: RiyadAboutaha

      Poster number: 18

      Along with the increasing number of high-rise buildings and long-span bridges, high strength concrete (HSC) is occupying a large share of the construction market as normal strength concrete (NSC). In general, fire-exposed reinforced concrete (RC) structure leads to serious structural problems due to the loss of strength and stiffness for both concrete and steel bars. Carbon fiber reinforced polymer composites (CFRP) and steel jacketing are two common methods applied to strengthen RC structure columns. This research is to investigate the structural behavior of fire exposed and retrofitted RC members with CFRP and steel jackets. It presents the compressive behavior of both NSC and HSC RC columns exposed to ISO834 standard fire, cooled by air and retrofitted with CFRP or steel jackets. Cross-sectional temperature distribution, load-deflection response, and strain distribution were investigated in this study. It was observed that the temperature gradient was developed in each specimen during fire exposure. The ultimate capacity and stiffness of fire-damaged RC columns were decreased due to the deterioration of concrete at elevated temperatures. CFRP and steel jacket can effectively improve the ultimate capacity and improve the stiffness for fire damaged NSC and HSC RC columns.


      Flow Behavior of Reactive and Non-Reactive Materials

      Presented by: Nuzhath Fatema, Advisor: Shobha Bhatia

      Poster number: 19

      My research focuses on the metastability and flow behavior of fly ash (reactive) and tailing (non-reactive) materials. Fly ash is a coal combustion residual (CCR) and a reactive material with a range of pH~ 4 – 11. In the USA, approximately 1600 fly ash slurry impoundments are currently in operation, and an additional 670 inactive impoundments exist. The failure of two fly ash impoundments at Kingston, Tennessee in 2008 and Eden, North Carolina in 2014, triggered a great concern about the safety of these impoundments. The failure incident of the Kingston Fossil Plant highlighted ponded fly ash’s “flow like failure” behavior due to containment loss. The United States Environmental Protection Agency (the Coal Ash Management Act, 2014) passed a bill for the fly ash slurry impoundments dewatering and capping. There are other non-reactive materials such as mining residue products (tailings), stored in a large number of impoundments in slurry form. The failure of tailing impoundments has also induced “flow like a failure” which is also not well understood. The main research objective of my Ph.D. work is to investigate the flow like the behavior of fly ash and tailing materials. I believe that I will be able to make an important contribution not only for the sake of the research itself but also for the industries that find these issues are very challenging.


      Effect of Green Roof Aging on Physical Properties and Hydrologic Performance

      Presented by: Yige Yang, Advisor: Cliff Davidson

      Poster number: 20

      Green roofs have become popular as a stormwater management method. Many studies focus on the hydraulic performance of green roofs, but the aging effect remains unknown due to the scarcity of long-term monitored data. This study aims to fill the research gap by characterizing virgin and aged growth media and assessing the impact of observed changes on hydraulic performance. The physical properties of virgin and 7-year-old growth media have been evaluated using the same methods. Significant differences in structure (particle size distribution, porosity, organic content, density) and hydrologic properties (Saturated hydraulic conductivity and Maximum water holding capacity) were observed. The aged sample experienced a shift to more fine particles and smaller pores. Doubled organic content supports this change because organic contents often act as fine particles. The increase in water-filled porosity (micropore) indicates more water can be stored, resulting in a higher MWHC. Green roof hydrological performance is a function of the combined effect of interacting physical processes. Based on those physical property changes, better retention and detention performance are expected. The 5th and 7th year monitored data is used to assess the hydraulic performance. Indeed both retention and detention improve. As the aging effect of green roofs is relatively little understood, this finding can offer useful insights to urban planners and stormwater engineers.

    • Computer and Information Science Posters

      Geographically Relevant Community and Influence extraction on Twitter

      Presented by: Aleksey Panasyuk, Advisor: Kishan Mehrotra

      Poster number: 21

      Given an online social network (OSN) (example Twitter) and a location (example Syracuse, NY), need to identify geographically relevant followers and the corresponding influencers. The messages and profile information of influencers and their followers can be used to study variables underpinning each geographic community. As an example, these variables can help explain why cities differ in levels of crime, innovation, happiness, and other factors. Our work proposes a novel method for capturing and refining city-level communities from an initial set of influencers. The influencers and the locations they serve are stored in a catalog that will be shared with other researchers in the near future.


      Inferences from Interactions with Smart Devices: Security Leaks and Defenses

      Presented by: Diksha Shukla, Advisor: Vir Phoha

      Poster number: 22

      In this work, we study the users’ hand movements while they interact with their devices. The work shows that the observed hand movement patterns can be exploited to learn the user’s corresponding brain wave patterns and hence exposes a serious security threat to brain wave controlled IOT applications. Also, we propose a novel authentication system that utilizes the learned hand gestures and provides enhanced security.

      Unlike conventional biometrics, electroencephalograph (EEG) biometrics are hard to spoof using standard presentation attack methods, given the intrinsic liveness resulting from the bounded randomness of EEG signals specific to an individual. We propose a novel attack on the EEG-based authentication systems by investigating and leveraging the strong correlation between hand movements and brain signals captured through the motion sensors on a smartwatch and the wearable EEG headset respectively. Our method can successfully estimate the user’s EEG signals from the stolen hand movements data while the user types on the keyboard.

      Also, we introduce a novel authentication mechanism that authenticates a user based on the learned user-specific gestures in the form of hand movements captured through the built-in motion sensors. The methods require a user to create specific hand gestures in order to authenticate and hence provide a light-weight, low-cost, and easy-to-use secure authentication system that requires minimal efforts and offers satisfactory usability.


      Exploratory Study of the Application of Anomaly Detection Methodology to the Analysis of Fatigue Data

      Presented by: Jakob Zeitler, Advisor: ChilukuriMohan

      Poster number: 23

      A machine learning evaluation of a dataset resulting from Three-Point bending fatigue tests on glass-epoxy composites is presented. The test results are contained in a publication by Robert Haynes et al. of the US Army Research Laboratory. Displacement versus number of cycles was recorded for four load ranges on the test-machine for 222 specimens. A small number of these were subjected to novel anomaly analysis that aims to support the theoretical understanding of the material fatigue process. The results show that the compliance, displacement over cycling load, increases to failure in three phases with a minimum compliance increase rate over the major portion of the life. This increase consists of at least two characteristics; a steady state monotonous increase interrupted at random intervals by short periods of significantly higher displacement increase steps. Classification algorithms, i.e., neural networks, support vector machines and random forests, show the prediction feasibility of failure via features from data of the last 50 cycles available. The lower displacement and the range of displacement are identified as most informative, leading to a classification AUC of 98%. Oversampling is used to reduce the rate of false negatives. More work, including the use of sensor fusion with other damage indicators, (e.g., acoustic emission) is suggested.


      Unet Laplacian Medical Image Segmentation Net

      Presented by: Jiyang Wang, Advisor: SenemValipasalar

      Poster number: 24

      Adding a laplacian filter layer to original Unet to segment medical images. From results, the additional laplacian layer can help the model to reduce noise from the output. But in all the performance is still relay on the Unet part. This model can be used in diagnosis.


      Nested Bigrams: A Novel Approach in Python Source Code De-Anonymization

      Presented by: Pegah Hozhabrierdi, Advisor: Chilukuri Mohan

      Poster number: 25

      An important issue in cybersecurity is the insertion or modification of code by individuals other than the original authors of the code. This motivates research on authorship attribution of unknown source code. We have addressed the deficiencies of previously used feature extraction methods and propose a novel approach: Nested Bigrams. Such features are easy to extract and carry substantial information about the interconnections between the nodes of the abstract syntax tree. We also show that for a large number of authors, a Strongly Regularized Feed-forward Neural Network outperforms the Random Forest Classifier used in many code stylometric studies. A new ranking system for reducing the number of features is also proposed, and experiments show that this approach can reduce the feature set to 98 nested bigrams while maintaining a classification accuracy above 90 percent.


      Criminal Network Data Collection Game

      Presented by: Pivithuru Wijegunawardana, Advisor: SuchetaSoundarajan

      Poster number: 26

      Criminals and covet entities usually hide in civilian social networks to make it harder for law enforcement to identify them apart from civilians. Collecting and analyzing data about such criminal networks pose many challenges due to adversaries themselves actively trying to hide their information. For example, consider there is a data collector who investigates people in a large social network with the objective of uncovering members of a criminal organization. The data collector can inquire about information such as who are their friends in the social network, whether they are criminals and also whether their friends are criminals. Individuals may provide false information in response to all these questions.

      The objective of this work is understanding what factors would motivate a criminal to become adversarial towards a data collector. We formulate the Network Data Collection (NDC) game to simulate interactions between a data collector and members of a criminal organization. We conduct a series of behavior experiments using recruited participants from Amazon Mechanical Turk (mTurk), where a participant would act as a member of a criminal network in the NDC game. We answer research questions such as 1) What level of incentive should the data collector provide to obtain more accurate data? 2) Does the loyalty of individuals to the criminal group play a role in deceptive behavior?


      Crawling the Community Structure of

      Multiplex Networks

      Presented by: Ricky Laishram, Advisor: SuchetaSoundarajan

      Poster number: 27

      We examine the problem of crawling the community structure of a multiplex network containing multiple layers of edge relationships. While there has been a great deal of work examining community structure in general, and some work on the problem of sampling a network to preserve its community structure, to the best of our knowledge, this is the first work to consider this problem on multiplex networks. We consider the specific case in which the layers of a multiplex network have different query (collection) costs and reliabilities, and a data collector is interested in identifying the community structure of the most expensive layer. We propose multisample (MCS), a novel algorithm for crawling a multiplex network. MCS uses multiple levels of multi-armed bandits to determine the best layers, communities, pillar-like, and node roles for selecting nodes to query. We test MCS against six baseline algorithms on real-world multiplex networks and achieved large gains in performance. For example, after consuming a budget equivalent to sampling 20% of the nodes in the expensive layer, we observe that MCS outperforms the best baseline by up to 49%.


      Representing Networks with 3D Shapes

      Presented by: Shengmin Jin, Advisor: RezaZafarani

      Poster number: 28

      There has been a surge of interest in machine learning in graphs, as graphs and networks are ubiquitous across the globe and within science and engineering: road networks, power grids, protein-protein interaction networks, scientific collaboration networks, social networks, to name a few. Recent machine learning research has focused on efficient and effective ways to represent graph structure. We propose to represent a network with a 3-dimensional shape: the network shape. We introduce the first network shape, which represents a network as a 3D convex polyhedron using stochastic Kronecker graphs. We present a linear time algorithm to build a network shape. Network shapes provide a compact representation of networks that is easy to visualize and interpret. They capture various properties of not only the network but also its subgraphs. For instance, they can provide the distribution of subgraphs within a network, e.g., what proportion of subgraphs are structurally similar to the whole network? Using experiments on real-world networks, we show how network shapes can be used in various applications, from computing similarity between two graphs (using the overlap between network shapes of two networks) to graph compression, where a graph with millions of nodes can be represented with a convex hull with less than 40 boundary points.


      Robust And Scalable Image Tampering Detection

      Presented by: Ziyue Xiang, Advisor: DanielAcuna

      Poster number: 29

      Because the use of digital images becomes more and more ubiquitous, the ability to detect tampered images in an automatic fashion has become increasingly significant for publishers, government agencies and many other organizations. The existing tampering detection methods lack scalability, that is, they are either based on unsupervised learning, which does not require any training at all, or on deep learning, which demands an enormous amount of training data.  We would like to propose a novel image tampering detection method that is applicable for both small datasets with dozens of images and big ones with thousands or even more. It can make better use of limited training data and computation resources.


      Asynchronous Deep Reinforcement Learning  for Robot Navigation

      Presented by: Zilong Jiao, Advisor: JaeOh

      Poster number: 30

      We propose a deep reinforcement learning (DRL) method for collision avoidance in robot navigation. Tackling the computational complexity of DRL, we divide a navigation problem into n representative environments, each containing various obstacles.  Then, n robots, each with a deep neural network, learn their associated environments and update a global neural network policy using lock-free asynchronous stochastic gradient accent. Compared to methods based on the standard Deep Deterministic Policy Gradient algorithm (DDPG), the proposed method can learn a better optimal policy in navigation tasks with less training time.  In our experiments, a policy learned by the proposed method can safely navigate a robot through an environment with various obstacles, including the ones that were unseen in training sessions.


      The Effect of Link Prediction on Homophily,

       Echo Chambers, and Information Fairness

      Presented by: Zeinab Saghati Jalali, Advisor: SuchetaSoundarajan

      Poster number: 31

      Friendship recommendation has become a popular service on many online social media platforms, but little attention has been paid to the societal effects of such recommendations.  In this work, we consider various categories of link prediction algorithms, including methods that use only the topology of the network as well as those that use both topology and attribute properties, and argue that links added according to such algorithms can have several side effects on the underlying network.  In particular, we show that in certain cases, link prediction algorithms can (1) increase the assortativity of the network, which, depending on the attribute under study, can indicate harmful segregation, (2) increase echo chambers within the network, and (3) decrease information fairness, a new criterion that measures the availability of information to diverse nodes.

    • Electrical and Computer Engineering Posters

      Human Supervision for Machine Automated Decision Making

      Presented by: Baocheng Geng, Advisor: Pramod Varshney

      Poster number: 32

      Utilizing human expertise in addition to machine observations can be advantageous in making high-quality signal detection decisions, especially when the physical sensors are not able to collect informative data or in a critical environment that demands a high level of accuracy. In conventional detection theory, it is typically assumed that observations are collected by machines (or physical sensors) and a decision is made automatically according to the predesigned decision rule. In critical environments, the machine judgment may not be reliable and incorporating human expertise along with machine observations can be beneficial in making high-quality decisions. This study of supervision in machine automation provides insights into the design of user-focused intelligent systems and is the first step towards the development of large scale human-machine collaboration network.


      Obstacle Detection and Classification with Portable Uncalibrated Patterned Projected Light

      Presented by: Burak Kakillioglu, Advisor: SenemVelipasalar

      Poster number: 33

      Autonomous obstacle avoidance systems are instrumental for applications including assisting visually impaired people, guiding autonomous robots, and assisted driving. In this paper, we propose a novel method for autonomous obstacle detection and recognition, which employs a different type of sensor incorporating a patterned light with a camera, and provides a lightweight, cost-conscious, energy-efficient, reliable and portable solution. Experimental results indicate that grids, projected by the patterned light source, are apparent and differentiable as the sensing system is hand-carried in both indoor and outdoor environments over and towards different types of obstacles. We propose methods for exploiting these patterns for obstacle detection and classification without requiring calibration. We demonstrate that obstacles can be classified by using our novel patterned light and camera setup and employing a Convolutional Neural Network (CNN)-based approach. The proposed method provides very promising results with overall detection and classification accuracies of 93.11% and 86.06% respectively.


      Actor-Critic Deep Reinforcement Learning for Dynamic Multichannel Access

      Presented by: Chen Zhong, Advisor: M. CenkGursoy

      Poster number: 34

      We consider the dynamic multichannel access problem, which can be formulated as a partially observable Markov decision process (POMDP). We first propose a model-free actor-critic deep reinforcement learning based framework to explore the accessing policy. To evaluate the performance of the proposed accessing policy and the framework’s tolerance against uncertainty, we test the framework in scenarios with different channel switching patterns and consider different switching probabilities. Then, we consider a time-varying environment to identify the adaptive ability of the proposed framework. Additionally, we provide comparisons with the Deep-Q network (DQN) based framework, in terms of both average reward and the time efficiency.


      Real-time Onboard UAV Localization without Relying on GPS Data

      Presented by: Burak Kakillioglu, Advisor: SenemVelipasalar

      Poster number: 35

      Autonomous vehicles often benefit from the Global Positioning System (GPS) for navigational guidance as people do with their mobile phones or automobile radios. However, since GPS is not always available or reliable everywhere, autonomous vehicles need more reliable systems to understand where they are and where they should head to. Moreover, even though GPS is reliable, autonomous vehicles usually need extra sensors for more precise position estimation. In this work, we propose a localization method for autonomous Unmanned Aerial Vehicles (UAVs) for infrastructure health monitoring without relying on GPS data. The proposed method only depends on depth image frames from a 3D camera (Structure Sensor) and the 3D map of the structure. Captured 3D scenes are projected onto 2D binary images as templates, and matched with the 2D projection of relevant facade of the structure. Back-projections of matching regions are then used to calculate 3D translation (shift) as estimated position relative to the structure. Our method estimates position for each frame independently from others at a rate of 200Hz. Thus, the error does not accumulate with the traveled distance. The proposed approach provides promising results with mean Euclidean distance error of 13.4 cm and a standard deviation of 8.4cm.


      Energy Efficient Occupancy Detection

      Presented by: Fatih Altay, Advisor: SenemVelipasalar

      Poster number: 36

      This research aims to provide higher person detection rates for energy efficient vision systems.


      Object Classification from 3D Volumetric Data with 3D Capsule Networks

      Presented by: Burak Kakillioglu, Advisor: SenemVelipasalar

      Poster number: 37

      The proliferation of 3D sensors induced 3D computer vision research for many application areas including virtual reality, autonomous navigation, and surveillance. Recently, different methods have been proposed for 3D object classification. Many of the existing 2D and 3D classification methods rely on convolutional neural networks (CNNs), which are very successful in extracting features from the data. However, CNNs cannot sufficiently address the spatial relationship between features due to the max-pooling layers, and they require a vast amount of training data. In this work, we propose a model architecture for 3D object classification, which is an extension of Capsule Networks (CapsNets) to 3D data. Our proposed architecture, referred to as 3D CapsNet, takes advantage of the fact that a CapsNet preserves the orientation and spatial relationship of the extracted features, and thus requires fewer data to train the network. We compare our approach with ShapeNet on the ModelNet database and show that our method provides performance improvement especially when training data size gets smaller.


      FPGA-based Neuromorphic System for Sensor Data Processing

      Presented by: Haowen Fang, Advisor: Qinru Qiu

      Poster number: 38

      Human brain’s computational power, energy efficiency, and its ability to process real-time sensor data are attractive features for the evolving trend of the Internet of Things. Thus, drawing inspiration from the architecture of the human brain, brain-inspired computing is based on spiking neural networks (SNN). SNN consists of networks of homogeneous computing elements, i.e., neurons communicating with spikes. The sparsely distributed asynchronous events, i.e., spikes allow for a highly parallel and distributed computing architecture. Thus, neuromorphic hardware is highly desirable for embedded self-contained applications requiring real-time processing sensor data. However, there are a few challenges that limit the application of neuromorphic hardware: limited weight precision, I/O precision, training difficulty caused by the discrete nature of SNN. We developed a general workflow to mitigate these issues. To demonstrate its effectiveness, we built an FPGA-based neuromorphic system and trained a neural network to recognize sign language, and then used the proposed workflow to map it to the neuromorphic system.


      Recovering secrets without reverse engineering via side-channels

      Presented by: Ju Chen, Advisor: Yuzhe Tang

      Poster number: 39

      The information leakage through side-channels is an old topic but has been extensively re-studied recently, due to the emerging of the commercially available trusted execution environments (e.g., Intel’s SGX and ARM’s TrustZone). Since 2015, researchers have found various types of side-channels (via page tables, caches, and branch predictions) in this new setting. In 2018, researchers found another type of side-channel through the timing variance for program execution at a CPU instruction level [5].

      The discovery of this new type of side-channel poses another research question, can we automatically correlate the leaky information with the secret program inputs? In the previous works, the authors used reverse engineering and elementary statical analysis to manually recover the protected secrets the noisy leaking messages. In this project, I proposed to use a machine learning technique to automatically recover secrets and tried to understand what the machine learning solution means for the security under the side-channel attacks


      Broadband In-Phase 4-Way Equal Split Planar Power Divider/Combiner Capable of High Power Applications

      Presented by: Jeremy Furgal, Advisor: Jay Lee

      Poster number: 40

      Phased array antenna systems are widely used in radar applications and are being researched for use in with 5G.  One of the limitations of these antenna systems is the bandwidth at which they can operate.  Wideband components, such as power dividers, are key in the implementation of feed networks needed for wideband phased arrays. This power divider addresses the wideband concern and also is capable of handling high power. This makes it suitable for the transmit side of the antenna network where the power is typically much higher.


      Low-Profile Plasma-based Tunable Absorber for Stealth Application

      Presented by: Komlan Payne, Advisor: Jay K.Lee

      Poster number: 41

      With the proliferation of modern frequency hopping radar tracking systems, adaptive/tunable absorbing systems are crucial for stealth technology. Microwave absorbers design is used to mitigate the radar cross section (RCS) of a target in a designated spectrum range from hostile radar signals for military and aerospace applications. The ultimate goal of this project is to investigate the feasibility of devising large scale tunable/adaptive absorbers that can maintain stable operation in harsh and dynamic electromagnetic battlefield surveillance. The use of existing mainstream technologies including PIN diodes, varactors, MEMs devices, liquid crystal polymers, and graphene require tradeoffs in terms of tuning range, cost, reliability, and linearity. In this project, we propose a low-profile tunable absorber enabled by discrete plasma-shells. Some of the advantages using discrete plasma-shells as tuning component include polarization insensitive, simple biasing scheme, wide tuning range and easy integration in many systems.


      Attribute-based Object Localization

      Presented by: Krittaphat Pugdeethosapol, Advisor: Qinru Qiu

      Poster number: 42

      Despite the recent advances in object detection, it is still a challenging task to localize a free-form textual phrase in an image. Unlike locating objects over a deterministic number of classes, localizing textual phrases involves a massively larger search space. Thus, along with learning from the visual cues, it is necessary to develop an understanding of these textual phrases and its relation to the visual cues to reliably reason about locations of described by the phrases. Spatial attention networks are known to learn this relationship and enable the language-encoding recurrent networks to focus its gaze on salient objects in the image. Thus, we propose to utilize spatial attention networks to refine region proposals for the phrases from a Region Proposal Network (RPN) and localize them through reconstruction. Utilizing in-network RPN and attention allows for an independent/self-sufficient model and interpretable results respectively.


      3D Space-to-Frequency Mapping Antenna using Via-Less Composite Right/Left Handed Striplines

      Presented by: Michael Enders, Advisor: Jay Lee

      Poster number: 43

      This work uses Composite Right/Left Handed transmission lines to reduce feed network dimensions necessary for frequency scanning antennas. Frequency scanning is implemented in a 2D planar antenna array to provide full 3D scanning of upper hemisphere. Stripline technology with inherent shielding is used to prevent electromagnetic leakage of feed network affecting radiation pattern. Through a via-less fabrication process, all this scanning capability is achieved in an integrated prototype that has been built entirely on campus. The simplicity of beam steering along with integrated passive design makes this sort of system of interest in cost-driven applications such as in automotive radars.


      Learning-Based Antenna Splitting and User Scheduling in Full Duplex Massive MIMO Systems

      Presented by: Mangqing Guo, Advisor: M. Cenk Gursoy

      Poster number: 44

      We focus on the joint antenna splitting and user scheduling problem in full duplex massive MIMO systems. By converting the original problem into a Kullback-Leibler (KL) divergence minimization problem, and solving it through a related dynamical system via a stochastic gradient descent method, we can approach the optimal solution of the original problem while the complexity is much lower, compared with the naive exhaustive search method. Our proposed algorithm can also be used to solve some other combinatorial optimization problems.


      Critical Component Identification under Load Uncertainty for Cascading Failure Analysis

      Presented by: Mirjavad Hashemi, Advisor: SaraEftekharnejad

      Poster number: 45

      As electric power grids are increasingly growing in scale and complexity, modeling and analysis of cascading failures become more critical. In this work, a new method based on distributed slack power flow is proposed for modeling and analysis of failures under load uncertainty. The developed model differs from state-of-the-art cascading failure models from two aspects: it does not require an optimal power flow solution between subsequent failures, and it alleviates the unrealistic assignment of criticality to the transmission lines that are in the vicinity of the slack bus.  The proposed scheme for cascading failure analysis is implemented on a forecasted operating point of the power grid and a set of uncertainty scenarios that can occur at that operating point. The results from distributed slack cascading failure model with load uncertainty are analyzed using indices that rank components based on their appearance in cascading failure patterns. The proposed model and analysis are tested on Illinois 200-bus system, and results are discussed.


      Online Design of Optimal Precoders for High Dimensional Signal Detection

      Presented by: Prashant Khanduri, Advisor: Pramod Varshney

      Poster number: 46

      In this work, we propose a novel methodology to design optimal compression strategies (precoders) for distributed detection of high dimensional signals. We consider a wireless sensor network (WSN) which consists of multiple sensors that are spatially distributed in a region of interest (RoI) and a fusion center (FC). The sensors observe an unknown high dimensional signal and forward their observations to the FC after precoding. The sensors collect data over both temporal and spatial domains. The FC performs a binary hypothesis test based on the data received from the sensors over noisy channels. We present a technique to design optimal online linear precoding strategies in this setup in the presence of transmitting power constraints. We show analytically that the error exponents achieved by the proposed precoders are independent of the signal dimension in contrast to the state-of-the-art precoding strategies, where the error exponents decay with the signal dimension. Thus, the proposed precoder strategy performs very well even when the signal to be detected a high-dimensional signal.


      On Self-localization and Tracking with an Unknown Number of Targets

      Presented by: Pranay Sharma, Advisor: Pramod Varshney

      Poster number: 47

      We propose a scalable algorithm for cooperative self-localization and multi-target tracking. Mobile agents localize themselves and track an unknown number of targets in the presence of data-association uncertainty. We accomplish this task by modeling the probabilistic relationship between agent and target measurements alongside their state estimates using a suitably designed factor graph, which is solved using loopy belief propagation. By coupling the self-localization and multitarget tracking approaches together, our approach is capable of leveraging the statistical correlation between the agent and target state uncertainty in order to provide improved localization accuracy over time for both agents and targets. Simulation results show improvement in target and agent state estimation error, compared to the conventional approach of first localizing agents, and then using this information for target tracking.


      Risk Impact Assessment of Coordinated Cyber Attacks on PMU-based State Estimator

      Presented by: Sagnik Basumallik, Advisor: SaraEftekharnejad

      Poster number: 48

      Lack of situational awareness and improper modeling of interconnected power systems were among the most important causes of major blackouts in the past decades. Deployment of Energy Management Systems (EMS) improved situational awareness, yet additional cyber security problems were introduced. Substantial research has since been dedicated to the feasibility and consequences of False Data Injection Attacks (FDIA) and Coordinated Cyber Attacks (CCA) against power system State Estimators (SE). However, the full extent of their impacts contributing to cascading failures are not widely explored.

      Moreover, attacks against Remedial Action Schemes (RAS) have not been extensively considered. In this paper, we carry out a risk impact assessment of FDIA-CCA against a PMU-based SE. A Semi-Markov Process (SMP) is used to model the probability of cyber attacks targeting Phasor Data Concentrators (PDC). To estimate the physical impacts of FDIA-CCA when RAS trigger signals are also targeted, a Distributed Slack (DS) cascading failure algorithm is developed. Extensive simulations are carried on synthetic 200 bus and 500 bus systems.


      Noisy 1-Bit Compressed Sensing With Side-Information

      Presented by: Swatantra Kafle, Advisor: Pramod K. Varshney

      Poster number: 49

      My research deals with the sparse signal reconstruction from 1-bit compressed measurements in a noisy setting. I further, improved the reconstruction performance when the receiver has access to a signal similar to the sparse signal called as side-information. This side-information is assumed to be dependent with the signal that we want to reconstruct form noisy 1-bit measurements. I looked into two different kinds of side-information: homogeneous side-information, and heterogeneous side-information. Homogeneous side-information is useful in the sequential reconstruction problem like video frame reconstruction, dynamic MRI reconstruction. Heterogeneous side-information is useful when the receiver has side-information of a different modality. We use copula functions to model the heterogeneous dependence between the side-information and the signal. To the best of the author’s knowledge, this works first to use copula-based functions capture dependence between the side-information and signal for 1-bit compressed sensing problem.


      Fusion of Deep Neural Networks for Activity Recognition: A Regular Vine Based Approach

      Presented by: Shan Zhang, Advisor: Pramod Varshney

      Poster number: 50

      The recent development of small-size, low-power, multifunctional sensors and revolutionary advances in communication and computing technologies have resulted in many applications of sensor/information fusion. One such application is human activity recognition (HAR) that can be used for health care, personal fitness, and border surveillance, etc.. We proposed a regular vine copula-based fusion of multiple deep neural network classifiers for the problem of multi-sensor based human activity recognition. We took the cross-modal dependence into account by employing regular vine copulas that are extremely flexible and powerful graphical models to characterize complex dependence among multiple modalities. Multiple deep neural networks were used to extract high-level features from multi-sensing modalities, with each deep neural network processing the data collected from a single sensor. The extracted high-level features were then combined using a regular vine copula model.


      Wind Energy Intermittency Mitigation by Customer Response to Real-Time Pricing of Electricity

      Presented by: Wolf Peter Jean Philippe, Advisor: Sara Eftekharnejad

      Poster number: 51

      In the effort to produce cleaner energy, more renewable resources are added worldwide to existing power systems in recent years. Among those, wind energy resources are becoming more viable as alternatives to fossil fuels power plants. Wind energy continues to become more affordable and cost-efficient due to technological advancements in recent decades. However, wind energy suffers a critical drawback: intermittency. The intermittency makes meeting real-time consumer demands a challenging task. Hence, utilities are forced to have standby power resources known as spinning reserves. While expensive, those reserves are often not utilized.

      This research focuses on reducing the cost of spinning reserves in the power system while considering the intermittency of wind energy resources. To achieve this objective, a demand response (DR) program is proposed. The DR model is based on real-time pricing (RTP) of electricity, where customers receive incentive payments when reducing their energy consumption in a period of high electricity price, or shifting some of their consumption from high price to low price periods.

      The results obtained prove that the proposed DR helps reduce the cost of spinning reserves in power systems where high wind energy penetration is considered. Therefore, the wind energy penetration can be increased, while power system utilities can securely meet real-time demand.



      Coverage analysis for UAV-assisted mmWave IoT Cellular Networks

      Presented by: XUEYUAN WANG, Advisor: Mustafa Gursoy

      Poster number: 52


      In this work, we jointly consider the downlink simultaneous information (SWIPT) and energy transfer and uplink information transmission in unmanned aerial vehicle (UAV)-assisted millimeter Wave (mmWave) Internet of Things (IoT) cellular networks, in which the UE locations are modeled as a Thomas cluster process or Mat\’ern cluster process. Through numerical results, we investigate the impact of key system parameters on the performance, which gives us insight on how to design the system parameters to obtain optimal performance of this practical system model.


      Autonomous Waypoints Planning and Trajectory Generation with Deep Reinforcement Learning

      Presented by: Yilan Li, Advisor: QinruQiu

      Poster number: 53

      Safe and effective operations for multi-rotor unmanned aerial vehicles (UAVs) demand obstacle avoidance strategies and advanced trajectory planning and control schemes for stability and energy efficiency. However, existing work does not consider the relationship between optimal control and the surrounding environment. To solve those problems in one framework analytically is extremely challenging when the UAV needs to fly large distance in a complex environment. To address this challenge, we purpose a two-level strategy that ensures a globally optimal solution can be obtained. At the higher-level, deep reinforcement learning is adopted to select a sequence of waypoints which lead the UAV from its current position to the destination. At the lower-level, an optimal trajectory is generated between each pair of adjacent waypoints. While the goal of trajectory generation is to maintain the stability of the UAV, the goal of the waypoints planning is to select waypoints with the lowest control thrust consumption throughout the entire trip while avoiding collisions with obstacles.


      Energy Efficiency Optimization in UAV-Assisted Communications and Edge Computing

      Presented by: Yang Yang, Advisor: M. Cenk Gursoy

      Poster number: 54

      We introduce uplink NOMA technique for offloading computation tasks in a UAV-based communication network.

      By guaranteeing the computation latency constraints, we formulate a total energy consumption minimization problem while jointly optimizing the power allocation and the trajectory of UAV.

      Based on a three-step optimization approach, we first obtain the results of the optimization problem under a giving trajectory of UAV; then we optimize the trajectory according to the previous conclusions.


      Efficient Human Activity Classification from Egocentric Videos Incorporating Actor-Critic Reinforcement Learning

      Presented by: Yantao Lu, Advisor: Senem Velipasalar

      Poster number: 55

      In this work, we introduce a novel framework to significantly reduce the computational cost of human temporal activity recognition from egocentric videos while maintaining the accuracy at the same level. We propose to apply the actor-critic model of reinforcement learning to optical flow data to locate a bounding box around the region of interest, which is then used for clipping a sub-image from a video frame. We also propose to use one shallow and one deeper 3D convolutional neural network to process the original image and the clipped image region, respectively. We compared our proposed method with another approach using 3D convolutional networks on the recently released Dataset of Multimodal Semantic Egocentric Video. Experimental results show that the proposed method reduces the processing time by 36.4% while providing comparable accuracy at the same time.


      Distributed Resource Allocation in Wireless Networks via Deep Reinforcement Learning

      Presented by: ZIYANG LU, Advisor: M. Cenk Gursoy

      Poster number: 56

      In wireless communication networks, the spectrum is a scarce resource that needs to be effectively utilized as mobile data traffic over wireless networks grows at an unprecedented rate. Prior research efforts have addressed efficient spectrum allocation via optimization algorithms. However, optimization methods usually require complete knowledge of the network and are difficult to be implemented in dynamic environments.

      Recently, deep reinforcement learning (DRL) has been widely applied to solving diverse, dynamic control problems. Here, we propose a DRL-based method of allocating limited resources (spectrum and power) to maximize specific wireless network utility objectives such as the sum transmission rates or the fairness among the users in the network. It is worth noting that there is no information exchange between the users once the deep neural network is well trained, which makes this algorithm scalable becomes practical in large-scale networks.


      Human Activity Classification Egocentric Video and Inertial Measurement Unit Data

      Presented by: Yantao Lu, Advisor: Senem Velipasalar

      Poster number: 57

      Many methods have been proposed for human activity classification, which rely either on the Inertial Measurement Unit (IMU) data or data from static cameras watching subjects. There has been relatively less work using egocentric videos, and even fewer approaches combining egocentric video and IMU data. Systems relying only on IMU data are limited in the complexity of the activities that they can detect. In this work, we present a robust and autonomous method, for fine-grained activity classification, that leverages data from multiple wearable sensor modalities to differentiate between activities, which are similar in nature, with a level of accuracy that would be impossible by each sensor alone. We use both egocentric videos and IMU sensors on the body. We employ Capsule Networks together with Convolutional Long Short Term Memory (LSTM) to analyze egocentric videos, and an LSTM framework to analyze IMU data, and capture temporal aspect of actions. We performed experiments on the CMU-MMAC dataset achieving overall recall and precision rates of 85.8% and 86.2%, respectively. We also present results of using each sensor modality alone, which show that the proposed approach provides 19.47% and 39.34% increase in accuracy compared to using only ego-vision data and only IMU data, respectively.


      A Simulation Framework for Fast Design Space Exploration of Unmanned Air System Traffic Management

      Presented by: Ziyi Zhao, Advisor: Qinru Qiu

      Poster number: 58

      With the rapid increase of the UAV applications in delivery, surveillance, and rescue mission, there is an urgent need for UAV traffic management that ensures the safety and timeliness of the missions. Accurate and fast UAV traffic prediction and resource usage estimation is a key technique in the traffic management system. It allows us to 1) evaluate different mission schedules or 4G/5G resource allocation schemes in a short period; and2) adjust mission control policy in real-time. A good prediction model should not only consider the UAV mission information, but also the environment information such as weather, 4G/5Gbase station distribution, etc. Its complexity is much beyond the traditional analytical approach. We propose to solve this problem by using machine learning. The model has a combined of convolutional neural network (CNN) and recurrent neural network (RNN). The inputs are multi-channel time-varying streams, such as weather map, cellular network usage map, geographical constraints, and UAV launching/landing information, etc. The outputs will be predicted UAV conflict probability map and 4G/5G channel congestion map. Compression and acceleration of the model will also be studied for real-time prediction.

    • Mechanical and Aerospace Engineering Posters

      Development of a Molten Antimony Solid Oxide Fuel Cell

      Presented by: Alexander Hartwell, Advisor: Jeongmin Ahn

      Poster number: 59

      Solid oxide fuel cells (SOFCs) have a limited range of usable fuels and therefore applications. Carbonaceous fuels which currently are used in inefficient, environmentally harmful combustion processes are abundant and inexpensive, but unusable in SOFCs due to carbon coking.

      Similarly, JP8 (and similar jet fuels) contain impurities that would poison SOFCs preventing their use on military bases where power generation from already present fuel is desirable. Molten antimony SOFCs (MA-SOFCs), however, can use an immense variety of fuels including coal, JP8, sugar char, rice starch, and even graphite due to direct fuel oxidation. Additionally, power generation is only dependent on the presence of antimony and antimony oxide, so even in the absence of fuel, the cell can operate in a “battery mode” where it slowly discharges over time. These combined characteristics make molten MA-SOFCs a very effective stationary power generation system. There are, however, issues to overcome. Initial MA-SOFCs were supported by a thick electrolyte, but due to the ohmic losses this introduces, we are instead introducing a cathode supported cell which could produce more power. Cathode support causes issues with gas permeation, but this can be remedied by either addition of pore-former or use of a freeze-casting manufacturing technique. Overall this is a very promising technology, and the cathode supported MA-SOFC shows tremendous potential in new application fields previously untouched by SOFCs.


      Switched Motorized and Functional Electrical Stimulation Cycling Controllers for Power Tracking

      Presented by: Chen-Hao Chang, Advisor: Victor Duenas

      Poster number: 60

      Motorized functional electrical stimulation (FES) induced cycling is a rehabilitation technique where lower limb muscle groups are artificially activated, and assistance is provided by an electric motor to achieve cadence (speed) and torque tracking objectives. In this paper, switched controllers to track cadence and torque are designed based on a cycle rider model that includes the switching effects of activating multiple muscle groups based on a state-dependent stimulation pattern that exploits the kinematic effectiveness of the rider. Cadence tracking is accomplished by switching between the control inputs of lower-limb muscles via FES (within kinematic efficient regions of the crank cycle) and an electric motor (within regions where the muscles are not activated due to low torque efficiency). The position and cadence reference tracking trajectories are generated by a target impedance model.

      Moreover, while muscles are stimulated within the kinematic efficient muscle regions, a robust sliding-mode torque controller is designed for the electric motor to track the desired interaction torque. A passivity-based analysis is developed to ensure the stability of the closed-loop torque subsystem and a Lyapunov-based stability analysis ensures exponential cadence tracking. The muscle and electric motor controllers were implemented in real-time during a single cycling experiment to illustrate the developed methods.


      Numerical Study of Diesel Spray and Auto-Ignition Behavior

      Presented by: Chenwei Zheng, Advisor: Ben Akih

      Poster number: 61

      Advanced combustion systems such as Diesel, jet and rocket engines continuously target higher pressures for better thermal efficiency; therefore, less fuel consumption. For reductions in regulated pollutant emissions, low-temperature combustion is preferred.  Combustion at these high-density conditions needs further investigation to enable predictive simulation and control. This research investigated several physical phenomena associated with high-density spray combustion. Through combinations of various physical models, numerical results provide accurate agreement with carefully measured combustion flow properties.


      Investigation of accelerating delta wing using Lagrangian coherent structures

      Presented by: Han Tu, Advisor: Melissa Green

      Poster number: 62

      A proper understanding of vorticity production, reorientation, and annihilation around and in the wake of complex three-dimensional bodies such as unmanned combat air vehicles (UCAVs) would provide critical insight for effective flow-control development in unsteady environments. Force measurement, surface pressure measurement, and time-resolved planar flow visualization at high Reynolds number for steady and unsteady translations have been carried out on a NACA 0012 airfoil wing with triangular planform geometry. At high angles of attack, the flow field exhibits sensitivity to axial acceleration when initially separated across the midspan. Flow field data analysis included a finite-time Lyapunov exponent analysis shows a topological change in the coherent structure during different axial accelerations.


      Modular-based Green Design Studio for Sustainable Building

      Presented by: Jialei Shen, Advisor: Jensen Zhang

      Poster number: 63

      Green building design is becoming more and more popular to meet higher requirements for energy saving and indoor environment quality (IEQ) improvement. Nowadays, a great number of green building technologies and strategies have been developed and implemented in buildings. This project aims to build a green building database covering various green technologies and strategies in buildings, and develop a modular-based platform called Green Design Studio (GDS). The GDS platform can use three models to assess building performance and even control the building technology operation using IoT technique, i.e., data-driven model, reduced-order model, and physics-based model. The application of GDS platform would help to improve the energy and IEQ performance of new and existing buildings to create a better living environment for human occupants.


      A Secure Cyber-Manufacturing System Augmented by the Blockchain

      Presented by: Jinwoo Song, Advisor: Young B. Moon

      Poster number: 64

      Cyber-Manufacturing System (CMS) is an innovative factory of the future, where manufacturing information is being constantly exchanged among many entities through various computer networks including the Internet. However, CMS ushers in the unique security challenges from the sheer volume and pervasiveness of exchanged data, and increased accessibility of the system by its outsiders and insiders. Also, the connections among the physical components can reveal accessible routes into the system. This research presents a unique approach to secure CMS augmented by a Blockchain. The secure Cyber-Manufacturing System with the blockchain consists of four layers: a User layer, Provider layer, Service layer, and Blockchain layer. The user in the user layer is the main actor who implements the system. The provider in the provider layer-who is verified and selected by the user-manufactures, designs, or prints the product based on the user’s need. The service layer-that enables communications between the user and the provider via the blockchain layer-can be customized by the user and the provider’s need. The blockchain layer is storing data in a distributed way and mitigating the security risk. Data is transferred from the user layer to the blockchain layer through the service layer. A transaction hash corresponding to the data is returned while the provider uses the transaction hash to assess the data.


      Topology optimization of ducted flows using a multi-fidelity approach

      Presented by: Jack Rossetti, Advisor: John F. Dannenhoffer III

      Poster number: 65

      In current engineering fluid flow systems, space is restricted, meaning that turning a fluid while trying to minimize pressure loss and maintain flow uniformity can be a challenge. Attempts to solve this problem have mainly involved the use of a trial and error methodology, both experimentally and numerically, which invokes some experiential knowledge or intuition. With recent developments in design optimization research, methods exist that can now be used to remove the qualitative and imprecise nature of intuition and experience. This problem involves finding the shape of the turning vanes, as well as their number and distribution. Since the number of vanes is unknown beforehand, it seems appropriate to investigate the problem using topology optimization techniques rather than shape optimization. Current methods for topology optimization of these types of flows fail to accurately capture the boundary layer flow physics, resulting in designs that are used as guidelines for improvement instead of the final product. Furthermore, like other optimization techniques, the final design is dependent on the initial guess. My work aims to improve on these methods by implementing a multi-fidelity approach in which a low-fidelity model is used to produce a promising initial topology, and then a high-fidelity flow solver is used to refine further and change the design topology while capturing the boundary layer physics.


      Experiments on the effects of trailing edge shape and pitching amplitude on the wakes of bio-inspired pitching panels

      Presented by: Justin King, Advisor: Melissa Green

      Poster number: 66

      Stereoscopic particle image velocimetry (PIV) was used to characterize the three-dimensional wake created by a series of bio-inspired, trapezoidal pitching panels with various trailing edge geometries that were sinusoidally pitched about their leading edges with multiple pitching amplitudes. PIV data were collected in 27 planes across the bottom half of the spanwise extent of the panel’s wake at a single flow speed to investigate the wake dynamics and behavior. Results focus on the three-dimensional wakes produced by panels with straight, forked, and pointed trailing edges pitched at Strouhal numbers (St) between 0.09 and 0.66. Results show that geometry and St influence wake behavior and dynamics. Portions of the phase-averaged wakes are often comprised of linked vortex rings, formed by greater amounts of vorticity as St number is increased.


      Developing a Visual Analytics Tool for Engineering Tasks, Assignment for Small and Medium-sized Manufacturing Firm

      Presented by: Kai Sun, Advisor: Utpal Roy

      Poster number: 67

      My research work is focusing on enabling new data analytics and modeling techniques in smart product development under the new model-based system engineering concept. The goal is to develop a modeling and reuse framework for data-driven analytics models in smart-product development using the sPLM (smart product lifecycle management) system which is under developing in our Knowledge Engineering Laboratory at Syracuse University.


      Flame-assisted Fuel Cell Boiler for Combined Heating and micro-Power

      Presented by: Mengyuan Chu, Advisor: Jeongmin Ahn

      Poster number: 68

      Previous study has proved the possibility of off-grid power generation application by integrating fuel cell stack with domestic facilities. A concept for flame-assisted fuel cell boiler for combined heating and micro-power is proposed. Traditional tubular cell manufacturing is anode supported with the cathode on the outside. Comparing to the traditional design, a newly designed inside-out tubular fuel cell has been investigated for the boiler combustion environment.


      Intrusion Detection and Correlation for Cyber-Physical Attacks in Cyber-Manufacturing System

      Presented by: Mingtao Wu, Advisor: Young Moon

      Poster number: 69

      Mingtao Wu’s research focus is on intrusion detection of cyber-physical attacks in Cyber-Manufacturing Systems (CMS). The problem is changeling and critical because (i) the chances of being under cyber-physical attack is enlarged in CMS due to the Internet connection through product development and manufacturing life-cycle; (ii) currently, it takes time for an IDS to reveal true alarms, sometimes over months; (iii) CMS production life-cycle is shorter than a detection period, which increases the chances of physical consequence in production and the consumer market; (iv) the increasing complexity of networks will take an even longer detection period.

      The research methods include (i) apply machine learning in manufacturing data for security, (ii) establish Cyber-Manufacturing Systems security testbed, (iii) develop and apply similarity-based alert correlation theory on cyber and physical alert in CMS.

      The research has broad impact in creating a similarity-based alert correlation method for cyber-physical attack; providing taxonomies for cyber-physical attacks in manufacturing systems; comparing different machine learning algorithms for intrusion detection in manufacturing; establishing Cyber-Manufacturing System Security Testbed (CSST).


      Interactions Between Upstream Turbulent Flow and Quadrotor Thruster Controller Performance

      Presented by: Ningshan Wang, Advisor: Mark Glauser

      Poster number: 70

      The Unmanned Aerial Vehicles (UAV) industry has experienced large prosperity in recent years, especially on the application of multirotor systems. With wider applications, the reliability of the UAVs, especially its controller reliability under turbulence/gust environment has became a rising problem. To solve this problem, it is vital to understand how the turbulent income flow will interact dynamically with the thrusters and the body frame. Therefore, the controller for the turbulent environment can be designed with such knowledge. The goal of this project is to obtain such knowledge and design such a controller with the knowledge obtained.


      Wake characteristics of a bio-inspired propulsor behind a streamlined body

      Presented by: Seth Brooks, Advisor: Melissa Green

      Poster number: 71

      Current underwater vehicles are typically designed to perform a single task to the exclusion of all others. On one extreme, there are torpedoes, which are designed to move very quickly in a straight line but are not maneuverable. On the other extreme, there are box-like inspection vehicles which are highly maneuverable but are slow and susceptible to flow disturbances. Alternatively, nature has many examples that fall between these two extremes. Many different fish species use similar structures (i.e., body and dorsal, ventral, pectoral, and caudal fins) to produce a wide variety of performance characteristics. For example, yellowfin tuna are capable of swimming fast and efficiently while sunfish are not fast but are maneuverable. These differences are a direct result of their morphological differences as well as kinematics.

      Our objective is to understand how fish take advantage of their morphology and kinematics to perform varying tasks such as being fast, maneuverable, and efficient. This knowledge can be used to guide the design of future small-scale underwater vehicles. An ideal morphology can be designed to accomplish the primary functions along with other objectives. Additionally, once a morphology has been selected the control system can take advantage of this knowledge to best utilize the morphology to perform other tasks at optimal efficiency, range, maneuverability, or other performance criteria.


      Closed-Loop Feedback Control for Lower-Limb Exoskeletons

      Presented by: Siqi Wang, Advisor: Victor H. Duenas

      Poster number: 72

      Lower-limb exoskeletons are rehabilitation devices that provide assistance to people with neurological conditions to recover mobility and replace body functions. Moreover, exoskeletons can be coupled with functional electrical stimulation (FES) to provide therapeutic benefits. This project involves the design and development of a cable-driven lower-limb exoskeleton, which actuates 8 DOFs (knee, hip, and ankle joints for both legs) through available position and force feedback. The electric motors that actuate the joints are placed outside of the body to reduce the weight of the device. The lower-limb exoskeleton is used along with a body-weight support system and a harness to prevent falls during over-ground locomotion or while walking on a treadmill. FES is applied to the leg muscles such as the quadriceps and hamstrings to exploit the added physiological benefits and evoke maximum effort produced by the user. Outstanding challenges in the field involve the design and analysis of adaptive controllers with stability guarantees to improve the performance and use of exoskeletons, which are modeled as hybrid systems since continuous and discrete dynamics occur during gait. Future work will include the real-time implementation of adaptive algorithms with able-bodied individuals during stepping in a treadmill before transitioning to over-ground walking.


      Effects of high-density packaging in servers on fan performance

      Presented by: Tong Lin, Advisor: Thong Dang

      Poster number: 73

      High power-density servers require efficient ways of heat removal and flow management. Fans can perform very differently in highly compact servers when com- pared to their performance in open domain (or fan alone) due to the presence of inlet/outlet flow distortions and flow interactions with surrounding components, e.g., CPU cooling fins or memory DIMMs are placed close to the fan. In this study, a combination of experimental and computational research is proposed to analyze the effects of high-density packaging in servers on fan performance.


      Solid Oxide Fuel Cells Replacement of Catalytic Converters in Automotive Exhaust’

      Presented by: Thomas Welles, Advisor: Jeongmin Ahn

      Poster number: 74

      The auto industry, utilizing internal combustion engines, has long been a target for increasing efficiency and decreasing emissions. These internal combustion engines rely on the use of a catalytic converter to reduce emissions. Catalytic converters help to react exhaust gases to rid the flow primarily of hydrocarbons, carbon monoxide, and NOx. Current catalytic converters cannot operate in lean conditions and maintain sufficient reduction of NOx emissions. Therefore, auto manufacturers must limit lean operation time, even though the lean operation has the potential for fuel savings. A Solid Oxide Fuel Cell (SOFC) stack is therefore investigated as a potential emission reduction system. The engine exhaust will pass through the tubular anodes of the fuel cells and act as the primary fuel source for the SOFC. At the anode, any unburned hydrocarbons, H2, and CO will be reacted into water and carbon dioxide, while generating electrical energy. The exhaust gas, upon exiting the tubular anode of the SOFC, will be recycled to the cathode side of the SOFC. The SOFC cathode layer can react NOx in lean conditions. The cathode layer will then decompose NOx into N2, allowing the oxygen ions to travel through the electrolyte layer to react with particulate carbon, CO, H2, or hydrocarbons present on the anode layer. SOFCs have demonstrated significant performance indicating a future emission reduction system. Also, this paper presents a potential reduction in coking concerns.


      Composite Gel-Polymer Electrolytes Utilizing Natural Plant Based Polymers for Rechargeable Li-Ion Batteries

      Presented by: Vincent DeBiase, Advisor: Jeongmin Ahn

      Poster number: 75

      Composite gel polymer electrolytes (GPE) based on the natural polymer lignocellulose (LC) with the inclusion of other common polymers (PEG, PVdF-HFP) are produced for application in lithium-ion batteries for electric vehicles (EVs). Each GPE combination of LC(-PEG, -PVdF-HFP) is tested for liquid electrolyte uptake, ionic conductivity, and thermal stability. Coin cell batteries are then produced using industry standard electrodes typically used in the Tesla Model S and the Nissan Leaf. The coin cells produced are tested for typical cell performance parameters and compared against current EV battery technology. The GPEs produced seek to improve the safety and decrease the cost for EVs allowing for more widespread use of EVs and to cut emissions.


      Non-uniform Curvature and Anisotropic Deformation Control Wrinkling Patterns on Tori

      Presented by: Xiaoxiao Zhang, Advisor: Teng Zhang

      Poster number: 76

      Wrinkling patterns in soft materials have been extensively studied due to their important roles in determining morphologies in biological structures and developing multifunctional devices. Most existing work focuses on relatively simple geometries, such as flat structures or curved structures with constant curvatures, such as the cylinder and 2-sphere. Inspired by pattern formation in developmental biologies, such as follicle pattern formation during the development of chicken embryos, we study the wrinkling patterns on a torus, the Gaussian and mean curvatures of which vary along the poloidal direction. We find that the non-uniform curvature and anisotropic deformation play critical roles in determining the formation and evolution of wrinkling patterns. Our results show that global deformations of a torus lead to strong coupling between elasticity and curvature which may enlarge the design space while affording dynamic control of wrinkling patterns.


      Automatic Generation of Near-Body Structured Grids

      Presented by: Yuyan Hao, Advisor: John Dannenhoffer

      Poster number: 77

      Overset grid is used in simulating complex fluid flow problems. It decomposes a complex geometry into near-body and far-field grids. Hyperbolic Grid Generation (HYPGEN) and Grids about Airfoil and other shapes by the use of Poisson Equations (GRAPE) are the two-main near-body grid generation techniques, which have their strengths and weaknesses. My research is to generate a new scheme which combines these two methods exploits their advantages and reduces the number of user-inputs.

      In HYPGEN, a mesh is generated by propagating in the normal direction from a known level of points to a new level. There are a bunch of mechanisms to make HYPGEN works in different cases. However, it will become unstable in the far field.

      On the other hand, GRAPE requires a user-defined outer boundary and generate the entire mesh simultaneously by iteration. The scheme works well but requires much more time to solve.

      In my combined scheme, the mesh is generated by HYPGEN at first with some “bad” points. Then, we find out those “bad”points by checking the length and skewness of the grids and cut those parts off. After that, we patch those parts by the GRAPE method with boundary conditions given from the rest of HYPGEN. The combined scheme generates grids fast, get a useful grid without overlapping, and only use a few user interactive parameters.


      Antiplane Shear of Cylinders and Layered Systems: Cohesive Fracture and Instability

      Presented by: Yueming Song, Advisor: Alan Levy

      Poster number: 78

      Mode-III fracture is one of the three fundamental fracture modes which is commonly observed in the torsional failure of shafts and failure of layered composites subject to anti-plane loading. An analytical solution is valuable to interpreting mixed-mode and mode-III testing result due to the difficulties in pure mode-III testing. The research integrates the analytical solutions of simple cylinder and slab system subjected to anti-plane loading with nonlinear cohesive interface model which can be used to model interfacial crack and predict crack propagation process. The solution proves to be consistent with existing static crack solution when load/parameter is consistent with stack crack behavior. The characterization of the interface enables analysis on complicated problems such as ductile interface, subsurface crack of a coating system, array of cracks and the interfacial crack between dissimilar media.


      Flends: Generalized Fillets via B-splines

      Presented by: Zachary Eager, Advisor: John Dannenhoffer III

      Poster number: 79

      A common desire in solid modeling is the ability to create a transition (blend) between two surfaces or adjacent faces of a solid. Currently existing approaches include fillets (or rounds) and chamfers.

      The issue with fillets and chamfers is that they are, by definition or implementation, highly restrictive. We wish to create a method using open-source tools that allow the creation of transition surfaces where the curvature of the resulting surface is an input parameter. Furthermore, we wish to provide the user with the ability to choose where the transition surface begins/ends on the original surfaces.

      A comprehensive method that achieves this would greatly increase the flexibility of any such feature in a CAD system, furthermore allowing a designer to more freely express their creativity while greatly reducing the amount of time required to do so.


    • Keynote Speaker, Rakesh “Teddy” Kumar, Ph.D

      Keynote Abstract
      In this talk we will present AI based techniques and applications based on measurement of human physical, biometric, behavior and emotional states with cameras, microphones and other sensors.

      The first set of techniques will be motivated using augmented reality applications. Augmented reality inserts virtual elements into live views of real environments. The human’s position and look direction must be very precisely tracked in real-time in order for insertions to appear stably in the correct location. This precision must be maintained across multiple humans and as they move around complex environments.  In addition, the effects must be rendered realistically, so they appear to be part of the environment and reflect local conditions, including obscuration by terrain and dynamic objects.

      Second we will present techniques to recognize facial expressions, gaze behaviors, gestures, postures, speech and paralinguistics in real-time and use that to assess cognitive states. These measurements will be used to drive multiple applications: assessing driver state, human computer communication and training based on interaction with avatars in simulation environments. We will describe the techniques developed and results, comparable to or better than the state of the art, obtained for each of the behavioral cues, as well as identify avenues for further research.

      Finally, we will provide a brief overview of the Center for Vision Technologies and SRI International.

      About Rakesh “Teddy” Kumar
      Kumar, Ph.D., is Vice President, Information and Computing Sciences and Director of the Center for Vision Technologies at SRI International. In this role, he is responsible for leading research and development of innovative end-to-end vision solutions from image capture to situational understanding that translate into real-world applications such as robotics, intelligence extraction and human computer interaction. He has received the Outstanding Achievement in Technology Development award from his alma mater, University of Massachusetts Amherst, the Sarnoff Presidents Award, and Sarnoff Technical Achievement awards for his work in registration of multi-sensor, multi-dimensional medical images and alignment of video to three-dimensional scene models. The paper “Stable Vision-Aided Navigation for Large-Area Augmented Reality” co-authored by him received the best paper award in the IEEE Virtual Reality 2011 conference. The paper “Augmented Reality Binoculars” co-authored by him received the best paper award in the IEEE International Symposium on Mixed and Augmented Reality (ISMAR) 2013 conference. Kumar has over 50 patents, and was a principal founder for several spin-off companies from SRI/ Sarnoff Corporation, including VideoBrush, LifeClips, SSG, HI-Lytes, Area-17 etc.

    • Biomedical and Chemical Engineering Posters

      A new screening method to quantitatively evaluate antibiotic penetration through bacterial membranes
      Presented by Zhaowei Jiang
      Advisor: Dacheng Ren
      Related research area(s): Health and Well-being
      Poster number BMCE 1-01

      Antibiotic resistance has been on the rise and poses major threats to public health. Bacteria develop multidrug resistance due to various mechanisms such as the limited antibiotic permeability of double-layered Gram-negative cell membranes. However, a robust and fast screening method to determine penetration of candidate agents through bacterial membranes is still missing, which hinders the discovery of new antibiotics. We recently developed a high throughput method for quantifying membrane penetration by optimizing treatment conditions to lyse the inner and outer membranes separately. In this proof-of-concept study, we evaluated the penetration of ciprofloxacin into Escherichia coli MG1665. With optimized experimental conditions, we found that outer membrane is the main barrier to ciprofloxacin penetration. This result can guide the future studies to identify novel antimicrobials and inhibitors of efflux pumps.

      Antibiotics Transport through Outer Membrane Porins of Gram-Negative Bacteria
      Presented by Meishan Wu
      Advisor: Shikha Nangia
      Related research area(s): Health and Well-being
      Poster number BMCE 1-02

      For more than a century, antibiotics were used to treat and prevent bacterial prevention. The beneficial impact of antibiotics during war is proven throughout the world, leading to a sharp increase in production of antibiotics. However, the effect of antibiotics is overly-exaggerated, resulting in a misuse and overuse of antibiotics overtime. A direct consequence is antibiotic resistance. Antibiotic resistance is a term applies when the targeted bacteria become resistant to the antibiotics, lower the effectiveness of antibiotic significantly and making the bacterial infection become harder to treat. Our research addresses to this concern by developing a computational model to explore the penetrating mechanism of drug molecules across the cell wall of gram-negative bacterium, P. aeruginosa. Seven carbapenems were tested and the free energy profiles of each carbapenem transport through the cell wall of P. aeruginosa were obtained to visualize the internal interactions as the molecule travels through the outer membrane porin of the bacterium. The computational platform can be used for any small molecules and provide insight on the molecular-level understanding of the molecules uptake, hence give statistical support on developing a new class of antimicrobial agents in the future.

      Development of a Double Network Hydrogel and the Effects of Network Variables on Mechanical Performance
      Presented by Alexander Jannini
      Advisor: Julie Hasenwinkel
      Related research area(s): Health and Well-being
      Poster number BMCE 1-03

      Double network (DN) synthesis is a technique that is used to strengthen hydrogels by combining a tough, brittle network with a weak, ductile one. This combination results in moduli orders of magnitude above either of the individual networks. An area not yet investigated is the effect between the weight percent of polymer and crosslinking ratio on mechanical properties. To study this, we developed a double network consisting of hyaluronic acid (HA) crosslinked with poly (ethylene glycol) diacrylate (PEGDA) as the brittle network and poly (dimethyl acrylamide) (DMAM) has the ductile network. Initial results show that our DN gels have an average tensile modulus of around 437 kPa, while DMAM gels have a modulus of 23 kPa, and HA/PEGDA gels are too weak to test. The maximum stress shows similar results (387 kPa and 24 kPa, respectively). A factorial analysis will be conducted to investigate the interplay of the two variables.

      Eradicating Bacterial Biofilm Cells by Wireless Electrostimulation
      Presented by Hao Wang
      Advisor: Dacheng Ren
      Related research area(s): Health and Well-being
      Poster number BMCE 1-04

      Bacteria such as Pseudomonas aeruginosa and Staphylococcus aureus can form biofilms on medical implants and cause serious infections that are incurable by conventional antibiotics due to high-level tolerance of biofilm cells to antimicrobials. In this study, we demonstrate that biofilm cells can be effectively eradicated by using electromagnetic coupling to deliver direct current (DC) (50 µA/cm2) wirelessly from a remote power source; and these conditions were found safe to a model epithelial cell line. Based on these results, we developed a prototype to demonstrate the feasibility of assembling such control system in a compacted device. With the capability to kill bacteria without using a directly connected power source, this platform has potential applications in engineering better implantable medical devices and providing therapies to control device-associated infections.

      Generation of 3-Dimensional Cardiac Microchambers for Embryotoxicity Drug Screening
      Presented by Plansky Hoang
      Advisor: Zhen Ma
      Related research area(s): Health and Well-being
      Poster number BMCE 1-05

      Pregnant women can face difficult decisions about taking medications during their pregnancy. Certain drugs taken during pregnancy have been attributed to affect organ development. However, despite well-known teratogens, many drugs are still not well understood regarding their effects on human embryogenesis, nor is there a well-established human embryotoxicity drug screening platform available to test medications that can complicate pregnancy. In addition, federal regulations heavily restrict research and clinical testing on pregnant women. As a result, women often must abstain from medications used to treat chronic conditions. This dilemma not only affects the health of the mother, but also the developing baby. We have combined biomaterials-based cell patterning and human induced pluripotent stem cell technology to generate in vitro 3D cardiac microchambers that represent the early developing heart. We have assessed the effects of a series of drugs across the entire pregnancy category system on the formation of the cardiac microchambers. We show that high concentrations of category D or X drugs significantly disrupts the microchamber formation and the characteristic contractile physiology. This illustrates the potential of this organoid model for the screening and safe prescription of medications during pregnancy.

      Increasing light capture in silicon solar cells with encpsulants incorporating air prisms to reduce metallic contact losses
      Presented by Fu-Hao Chen
      Advisor: Ian Hosein
      Related research area(s): Energy Sources, Conversion, and Conservation
      Poster number BMCE 1-06

      The energy loss arising from metallic contacts on a silicon solar cells accounts for up to 10%. Based on the global solar power capacity of 232 GW, the loss is 23 GW, equivalent to 23 nuclear plants. To recycle the stray light, we proposed an air-prism array to guide the stray light by total internal reflection. This low-cost, easy-integrated encapsulant design not just covers wide angles of incident light, up to ±40 degree, but also yields 6% external quantum efficiency improvement. Thus, the recycling efficiency is up to 60%. Furthermore, the short-circuit current improves 8%, implying that the solar cell efficiency jumps from 10% to 10.8% if we assume the solar cell is 10% efficient, which reduces the cost significantly.

      Kinetic Insights of Acetone Hydrogenation over Supported Platinum Catalyst
      Presented by Xin Gao
      Advisor: Jesse Bond
      Related research area(s): Energy Sources, Conversion, and Conservation
      Poster number BMCE 1-07

      One of the most significant and fundamental strategies of converting biomass is oxygen removal because of the originally large amount of oxygen contents in biomass. In this regard, the elementary process of the catalytic reduction of oxygen-involved chemicals is urgently required to be understood, for example, which is the hydrogenation of carbonyl groups (C=O) to their corresponding hydroxyl groups (CH-OH) over metal catalysts. Pt has been listed as the highest activity in the vapor phase ketone hydrogenation with comparing Ru and Pd. However, the detailed kinetic and thermodynamic behaviors of the Pt-catalyzed ketone hydrogenation is still unclear. To fully investigate the mechanism, we did this kinetic study of acetone hydrogenation over silica supported Pt catalyst, Pt/SiO2.

      Self-Written, Scalable Microstructure Arrays for Advanced Battery Applications
      Presented by Shreyas Pathreeker
      Advisor: Ian D. Hosein
      Related research area(s): Energy Sources, Conversion, and Conservation
      Poster number BMCE 1-08

      Rechargeable batteries undergo multiple cycles of charge and discharge over the duration of their use, and while doing so, must maintain their capacity in a stable manner. The industry standard for Li-ion anodes currently is Carbon, which possesses a low theoretical capacity to retain Lithium during cycling. On the other hand, Silicon is known to possess the highest capacity for Lithium, but suffers from enormous, undesirable volume expansion due to mechanical instabilities. For this reason, Silicon continues to be a focus of Li-ion battery research. Although other chemical approaches to incorporate Silicon into Li-ion batteries have been reported in the literature, we have successfully employed a unique, highly tunable photo-polymerization route to synthesize spatially distinct micro-structures containing Silicon nanoparticles in a Carbon matrix. We demonstrate control over the size and shape of the structures by varying processing parameters like light intensity, curing time, and nanoparticle loading. The Silicon nanoparticles provide high capacity while the high aspect ratio micro-structures can accommodate volume expansion. This translates into a longer battery life and cycling stability. Lastly, we use blue light as the radiation source for polymerization, which is much safer than UV light, requires almost no safety measures and can be scaled up relatively easily given the preexisting nature of radiation curing infrastructure in the industry.

      Sensitizing Bacterial Cells to Antibiotics through Dynamic Topography-triggered Biofilm Detachment
      Presented by Sang Won Lee
      Advisor: Dr Dacheng Ren
      Related research area(s): Health and Well-being
      Poster number BMCE 1-09

      Bacterial biofilms are multicellular structures with bacterial cells attached to a surface and embedded in an extracellular matrix. Due to high-level resistance to antimicrobial agents, the need to develop better bacterial biofilm control strategies is acute. Recently, we engineered surfaces with dynamic topography using a shape memory polymer (SMP) and demonstrated up to 99.9% detachment of biofilm after triggered shape recovery. In this study, Pseudomonas aeruginosa biofilm cells released by shape recovery were treated by four different antibiotics concurrently including tobramycin, ofloxacin, tetracycline, and minocycline, which showed 2-3 logs of increase in antibiotic susceptibility compared to the original biofilm cells. Using a reporter strain, we found the expression of rrnB, encoding ribosomal RNA gene, increased by 4 times upon shape recovery. Collectively, these results indicate that active changes in the topography of substratum material can have profound impact on the physiology of biofilm cells, which can help the design of better biomaterials for infection control.

      Solid Polymer Electrolyte Networks for Lithium Ion Conduction
      Presented by Francielli Genier
      Advisor: Ian Hosein
      Related research area(s): Energy Sources, Conversion, and Conservation
      Poster number BMCE 1-10

      The development of high performance batteries is a central to meeting global energy demands, while also storing energy in a clean, affordable, and safe manner. Towards this end, an area of intense research focus has been the replacement of the liquid or gel electrolyte with a solid polymer material, namely a solid polymer electrolyte (SPE), because of its potential to double the energy density and further address concerns over the volatility and flammability of currently used organic solvents. Of the numerous polymer systems that have been studied, SPEs produced through the crosslinking are quite attractive owing to the ability to enhance conductivity, while also retaining mechanical stability. We have study the synthesis, properties, and ion conductivity of a solid polymer electrolyte produced from polytetrahydrofuran (PTHF) photo-crosslinked with an epoxy, via an active monomer mechanism that facilitates the reaction of the native hydroxyl and epoxide end-groups. Crosslinked samples were loaded with different quantities of lithium tetrafluoroborate (LiBF4) and evaluated by electrochemical spectroscopy impedance (EIS) to determine their ionic conductivity, reaching competitive conductivities at temperatures typical for battery operation. Thermal analysis confirms the amorphous structure and the high thermal stability. Mechanical analysis shows the materials possess suitable stiffness for applications. The results demonstrate a new synthetic route to crosslinked networks.

      Solubility Analysis for γ-Valerolactone and Pentenoic acid
      Presented by Zijian Wang
      Advisor: Jesse Bond
      Related research area(s): Energy Sources, Conversion, and Conservation
      Poster number BMCE 1-11

      Lignocellulosic biomass is an alternative source of industrial carbon. In general, biomass upgrading is predicated upon intermediate formation of functional platform molecules, which can be converted as desired into a range of commodity products. γ-valerolactone (GVL) is an example of one such platform as it provides a spectrum of interesting opportunities for the production of gasoline, jet, and diesel fuels; polymer precursors; and commodity and specialty chemicals. Of interest here is the acid-catalyzed ring opening of GVL to form pentenoic acid (PEA) isomers, which are a family of bifunctional alkene-acids that are of potential interest in the production of nylon.GVL ring opening is kinetically facile, and it occurs readily in the gas phase over solid acids; however, thermodynamic limitations and secondary reactions make it generally difficult to recover pentenoic acids in good yield. Although GVL ring opening is entropically favorable, it is substantially endothermic, such that PEA yields are equilibrium limited below ≈ 20% at low temperatures. Raising operating temperatures will theoretically address this issue; unfortunately, increasing temperatures to the point where ring opening becomes thermodynamically favorable also makes undesired secondary reactions kinetically accessible. Thus, we propose that homogeneous mineral acids can be used in biphasic (aqueous-organic) reactors to navigate thermodynamic and kinetic constraints and deliver high PEA yields.

      Superhydrophobic Microporous Substrates via Photocuring: Coupling Optical Pattern Formation to Phase Separation for Process-Tunable Pore Architectures
      Presented by Saeid Biria
      Advisor: Ian Hosein
      Related research area(s): Other
      Poster number BMCE 1-12

      We studied a new approach to synthesizing superhydrophobic microporous surfaces via photocuring, as a critical aspect of material surface design for numerous applications including tissue engineering, regenerative medicine, and self-healing. Though other approaches have been achieved, all suffer from their inherent trade-off between precise control over structure and scalability. We found that our approach can be leveraged to tune pore size and spacing, to provide a range of porous surface structures, and is attractive for the rational control of surface structure of relevance to anti-wetting properties and surface-related applications.

      Alignment of Human Cardiomyocytes on Nano-Wrinkled Surface
      Presented by Sarah Moore
      Advisor: Zhen Ma
      Related research area(s): Health and Well-being
      Poster number BMCE 2-01

      Cardiomyocytes construct the cardiac muscle for blood pumping functions. Cardiomyocyte alignment is critical to maintain the physiological function of cardiac muscle tissues and need to be in a sustainable environment. Disorganized alignment of these cells will lead to heart failure. To better display in vivo structures, researchers have been searching for new biomaterials and fabrication tools to help align cardiomyocytes in a variety of culture systems. In our project, we mimicked the in vivo environment by using shape memory polymers (SMPs). If a strain is applied to SMPs, the polymer will revert back to its original shape when heated to its glass transition temperature. To induce nano-wrinkles, polyelectrolyte multilayers (PEMs) are coated onto the surface of a strained SMP. The cardiomyocytes derived from human induced pluripotent stem cells (hiPSCs) were seeded onto the topographic wrinkled surface of the SMPs, and the alignment of the cardiomyocytes and their sarcomeres were subsequently studied.

      Calcium Sulfate Addition to Two-Solution Bone Cements:Influence of Concentration on Cement Viscosity and Mechanical Properties
      Presented by Brittany Reed
      Advisor: Julie Hasenwinkel
      Related research area(s): Health and Well-being
      Poster number BMCE 2-02

      Two-solution bone cements (TSBC) have been developed as an alternative to powder-liquid cements on the market. TSBC produce a high residual monomer content upon setting. Unpolymerized monomers in clinical use have been linked to a wide range of adverse health effects. To increase biocompatibility of the cement, calcium sulfate was added to TSBC as the substance is resorbable and Ca2+ ions may stimulate osteoblasts to encourage bony ingrowth. As infections may occur following orthopaedic surgery, the possibility for cements to serve as delivery vehicles for antibiotics has been researched. Cements such as those containing calcium phosphate have been deemed inappropriate for drug delivery as their setting mechanism involves an acid-base reaction, in which drug molecules would be decomposed. The setting mechanism for calcium sulfate involves rehydration and recrystallization, indicating that water soluble drugs may be incorporated into the cement.

      Enzymatically Triggered Shape Memory Polymers
      Presented by Shelby Buffington
      Advisor: James Henderson
      Related research area(s): Other
      Poster number BMCE 2-03

      Shape memory polymers (SMPs) are a special class of materials that change shape in response to an external stimulus. SMPs that respond various stimuli have been previously designed, but the most common triggering mechanism is heat.

      1. We have developed a new SMP system that recovers in response to enzymatic degradation. Enzyme response of materials were made by dual spinning poly(µ-caprolatone) (PCL), an enzymatically degradable material
      2. with Pellethane TM, a non-degradable elastomer to create a fiber composite.
      3. In this system, PCL acts as a shape fixer, to learn the temporary shape, while Pellethane acts as the memory component and provides the force to return to the original shape.

      To program fiber composites are first heated above the melting temperature of PCL to program a strain, stretched to 100% and then cooled to -20°C to allow the PCL to crystallize. Bulk degradation experiments were performed on stretched samples. Briefly, samples were incubated in a phosphate buffer solution at 37°C with varying concentrations of lipase. One sample was removed every day and analyzed for shape recovery. As the PCL degrades, the Pellethane is free to contract back to the original shape. The strain in the materials was measured by measuring the length of the sample in a picture taken before and after programming and after degradation. Mass loss and dynamic scanning calorimetry (DSC) were used to track the degradation of PCL. Fiber composites of different ratios have been tested.

      Expanding, Enzymatic Responsive Shape-Memory Polymer
      Presented by Justine Paul
      Advisor: James Henderson
      Related research area(s): Other
      Poster number BMCE 2-04

      Stimuli responsive biomaterials have been developed to assay or control biological systems, but the potential of these materials has been largely untapped. Here we present the first expanding enzyme-responsive SMP platform and study the relationship between composition and function, including the relationships between material chemistry, cleavable ester bond density, and the sensitivity of the material to soluble enzyme. The SMP is fabricated by dual electrospinning commercialized Pellethane with a synthesized thermoplastic polyurethane (TPU) poly(d,l-lactide-co-caprolactone) (PLCL). A PLCL12k is first synthesized by ring-opening polymerization and acts as the soft segment. PLCL12k is then used for the TPU reaction with hexamethylene diisocyanate acting as the linking agent and hard segment. Samples are then electrospun and heated to a temperature above the transition temperature of the TPU. Upon heating, the strain programed into the fibers during electrospinning is recovered, compressing the Pellethane elastomer fibers. This allows for the SMP to be enzymatically triggered for expansion. Two forms of enzyme responsive shape-memory triggering will be studied in the future and will not be covered by this poster: bulk triggering by administration of soluble enzyme (Amano lipase PS) and localized triggering by cell produced (HepG2) enzyme. This enzymatic responsive shape memory polymer will provide a new approach to controlling the interaction of smart materials and cells.

      Identification of transcription factors binding to the conserved PD motif in the osm-9 TRPV channel gene promoter in Caenorhabditis elegans
      Presented by Emily Pujadas
      Advisor: Sarah Hall
      Related research area(s): Health and Well-being
      Poster number BMCE 2-05

      Environmental stress early in development can program changes in gene expression and physiology in animals. In the nematode C. elegans, exposure to stressful environmental conditions during early development (i.e. low food availability, high temperatures and overcrowding) causes the animals to enter the dauer diapause stage, which arrests development until environmental conditions improve. We have demonstrated that postdauer (PD) adults retain a cellular memory of their developmental history which manifests as changes in gene expression and altered behavior. Research from our lab has shown that the osm-9 TRPV channel gene is down-regulated in the ADL neurons of PD animals, and consequently, their ability to respond to the ascr#3 pheromone component is lost. The down-regulation of osm-9 in PD animals is mediated via chromatin remodeling, endogenous RNAi, and TGF-β signaling pathways. A cis-acting promoter region (PD motif) was identified in the upstream regulatory sequences of 977 genes and discovered to be required for osm-9 down regulation in PD animals. The TGF-β transcription factor DAF-3 Co-SMAD negatively regulates osm-9 expression by binding a region of the PD motif named the DAF-3 binding site (DBS). Using a yeast one-hybrid experimental design, we have shown that the DAF-8 R-SMAD transcription factor also binds to the DBS in vitro. Our findings suggest a new role for DAF-8 in the developmental programming of gene expression in postdauer adults.

      In vivo, Noncontact, Real-time, Optical and Spectroscopic Assessment of the Immediate Local Physiological Response to Spinal Cord Injury in a Rat Model
      Presented by Kyle Bishop
      Advisor: Julie Hasenwinkel
      Related research area(s): Health and Well-being
      Poster number BMCE 2-06

      We report a small study to test a methodology for real-time probing of chemical and physical changes in spinal cords in the immediate aftermath of a localized contusion injury. Raman spectroscopy, optical profilometry and scanning NIR autofluorescence images were obtained simultaneously in vivo on spinal cords that had been surgically exposed between T9 and T10. A total of six rats were studied in two n=3 groups of injured and control. A single 830 nm laser of 100 um round spot size was either spatially scanned across the cord or held at a specified location relative to the injury to improve signal to noise in the Raman spectra. Analysis of the Raman spectra suggest that the tissues were equally hypoxic for both the control and injured animals. On the other hand, only injured cords display Raman features possibly indicating that extensive, localized protein phosphorylation occurs in minutes following spinal cord trauma.

      Manufacturing Mesenchymal Stem Cell Tissue Ring from Induced Pluripotent Stem Cells
      Presented by Tackla Winston
      Advisor: Dr: Zhen Ma
      Related research area(s): Health and Well-being
      Poster number BMCE 2-07

      Mesenchymal stem cells (MSCs) have the potential to be used for applications in cardiac, neural, joint and bone repair because of its immunosuppressive properties and ability to differentiate into different lineage tissues. MSCs are usually harvested from bone marrow or adipose tissues in a highly invasive and painful procedure. We used SB431542, glycogen synthase kinase 3 (Gsk3) and bFGF with Essential 6 to promote the formation of intermediate neural crest cells that were later differentiated in a serum free culture into hiPSC-MSCs. Immunofluorescence and flow cytometry was used to characterize the cells for the expression of common MSC surface protein markers CD73, CD90, CD105, CD144 and CD146. hiPSC-MSCs derived by this method offers a scalable and ethical source for cell therapy and tissue engineering applications. The cells were used to develop a physiologically relevant, robust 3D tissue ring in a pluronic-coated reusable PDMS mold. Collagen a major component of the extracellular matrix, was used as a scaffold. The thickness and inner diameter of the ring was measured over 14 days after which immunohistochemistry of the tissue rings for CD73, CD90 and H&E was conducted.

      Photothermally Activated Shape Memory Polymers
      Presented by Jose Waimin
      Advisor: James Henderson
      Related research area(s): Other
      Poster number BMCE 2-08

      My research focuses on developing photothermally activated Shape Memory Polymers (SMP’s) by utilizing the novel approach of functionalizing the polymers with silver nanoparticles. Due to the plasmonic resonance of silver nanoparticles, heat is produced when the material comes in contact with light, which causes the SMP’s to return to their original shape. Furthermore, a multilayer coating on the material creates wrinkles on the surface as the polymer shrinks. There are many wrinkling techniques being developed, however, this technique has the potential of enabling localized wrinkling on shape memory materials by utilizing the photoactivation method.

      Reducing biofilm formation of Chlamydomonas reinhardtii by controlling surface topography
      Presented by Sweta Roy
      Advisor: Dacheng Ren
      Related research area(s): Energy Sources, Conversion, and Conservation
      Poster number BMCE 2-09

      Algae biofouling on the surfaces of photobioreactors adversely affect cell growth and thus the productivity. To overcome this challenge, we investigated the effects of microstructured topographical patterns on the attachment of algal cells and the following biofilm formation using Chlamydomonas reinhardtii as a model species. By tuning the size of protruding hexagonal patterns and distance between adjacent patterns on poly (dimethylsiloxane) (PDMS), micron scale surface topography was found to have profound impact on algal biofilm formation. For example, 10 µm tall hexagonal patterns with side length of 20 µm and inter-pattern distance of 20 µm reduced algal biofilms by 83.85 ± 4.7% compared to smooth PDMS surfaces. Further development to transfer such topography to bioreactor surfaces may help improve the production of algal biofuel.

    • Civil and Environmental Engineering Posters

      A Project-Level Infrastructure Management Framework For Sustainable Roadways
      Presented by Song He
      Advisor: Baris Salman
      Related research area(s): Sustainable Natural and Built Systems
      Poster number CIE 1-01

      Currently, maintenance, repair, and rehabilitation (MRR) projects for roadway infrastructure are mostly undertaken by traditional techniques, resulting in high overall life cycle impacts. Although non-traditional MRR techniques including accelerated methods can reduce the overall life cycle impacts, there is a lack of frameworks that can facilitate the project-level decision-making and justify the use of non-traditional techniques. The goal of this research is to develop a project-level roadway infrastructure management framework to consider multiple factors in decision-making and to analyze the life cycle economic, social, and environmental impacts of project alternatives involving traditional and non-traditional MRR techniques.

      The proposed framework features decision flowcharts and a multi-criteria decision-making model using analytical hierarchy process and analytical network process to shortlist alternatives that meet the project requirements to support preliminary decision-making. Then, life cycle assessment (LCA) and life cycle cost analysis (LCCA) is performed through the LCA-LCCA model to quantify life cycle economic, social, and environmental impacts of candidate project alternatives following the triple bottom line of sustainability. The LCA-LCCA model is also capable of performing what-if analysis by adjusting variables so that public agencies can apply their own data and make decisions based on their sustainability goals, objectives, and performance measures.

      Analysis and Behavior of Segmented Energy Absorbing Steel Plate Shear Walls
      Presented by Nafiseh Shahbazi Majd
      Advisor: Eric Lui
      Related research area(s): Sustainable Natural and Built Systems
      Poster number CIE 1-02

      Conventional steel plate shear walls (SPSW) are often designed to behave elastically under normal lateral load condition. However, they are expected to yield and absorb energy through inelastic hysteresis under extreme load condition. Depending on the width-to-thickness ratio and boundary conditions of these walls, SPSW may also experience inelastic out-of-plane buckling. Therefore, maintaining structural integrity and preventing building collapse are dependent upon their post-buckling behavior. In this study, a new type of SPSW referred to as the Segmented Energy Absorbing Steel Plate Shear Wall (SEA-SPSW) is proposed and its behavior is investigated. The proposed SPSW is segmented into geometric shapes. These segments, with predetermined gaps along their edges, are connected together by reinforcing steel strips using bolts with oversized or slotted bolt holes. The steel plates and strips not only provide the necessary stiffness to the structure under normal load condition, together they will also act as friction damper to absorb energy under extreme load condition. If the gap length, size, and the amount of overlap between the plate segments and reinforcing strips are properly set, no yielding or buckling of the plate segments will occur. These geometric parameters are optimized by finite element analysis. A parametric study is then carried out to demonstrate the effectiveness of the proposed SEA-SPSW when compared with conventional shear wall structural systems.

      Application of Wavelet Transform to Endurance Time Analysis Method for Performance-based Seismic Design
      Presented by Mohammadhossein Mamaghani
      Advisor: Eric Lui
      Related research area(s): Other
      Poster number CIE 1-03

      Earthquake engineering has made important advances over the last century. What started as a push to save lives from future seismic events has developed into an effort not merely to ensure life safety, but limit structural damage and minimize repair time to predetermined acceptable levels agreed upon by both owners and designers. Performance-based seismic design (PBSD) is a design paradigm by which a structure is designed to satisfy multiple levels of performance expectations for various degrees of potential hazards. There are various methods that can be used to assess earthquake effects on structures. Those that give accurate results for a large sample of earthquakes are usually computational intensive. The endurance time analysis (ETA) method is one that offers high accuracy with relatively low computational effort. This research uses wavelet transform in conjunction with ETA to generate artificial earthquakes with increasing intensities that allow engineers to access structural performance with relative ease and accuracy.

      Electrical Analog Simulation of Water Seepage through Soil Media
      Presented by Matthew Franceschini
      Advisor: Dawit Negussey
      Related research area(s): Other
      Poster number CIE 1-04

      Predicting the flow of subsurface water is an important practical problem that geotechnical engineers and hydrogeologists encounter. Flow nets, 2D graphical representations of subsurface flow, are widely used in practice. Flow of water and electricity depend on potential difference and are analogous. Thus, seepage flow through soils due to head difference can be emulated by current flow through conducting paper, teledeltos, due to voltage difference. The voltage difference is subdivided using a potentiometer to trace lines of equal potential. Potentiometers (short name: “pots”) are similar to resistors, but use three terminals instead of two: two ends, and a wiper to vary the output. Pots are classified by the number of possible turns on the wiper dial. Pots that have higher number of turns provide higher resolution of output. The 10-turn pot used in the experiment allowed for detection of voltage at 1% of the full scale output. The 10 turn pot, conducting paper, 9V battery and digital multi-meter with test probes were used to produce equipotential lines for steady state flow around a sheet pile cutoff wall. A physical model of the sheet pile cutoff through clean sand was also set up to produce matching flow lines. The combination of the equipotential and flow lines represent solutions for an elliptic partial differential equation that is widely known as the Laplace Equation.

      Investigation of Contact Pressure Distribution at EPS Geofoam Interfaces Using Tactile Pressure Sensors
      Presented by Chen Liu
      Advisor: Dawit Negussey
      Related research area(s): Sustainable Natural and Built Systems
      Poster number CIE 1-05

      EPS geofoam blocks are made of pre-expanded resin beads that form fused cellular micro-structures. The strength and deformation properties of geofoam blocks are determined by unconfined compression of small test samples between rigid loading plates. Applied loads are presumed to be supported uniformly over the entire mating end areas. Predictions of field performance on the basis of such laboratory tests widely over-estimate actual post-construction settlements and exaggerate predictions of long-term creep deformations. This investigation examined the development of contact pressures at a large number of discrete points at low and large strain levels for different densities of geofoam. Development of pressure patterns for fine and coarse interface material textures as well as for molding skin and hot wire cut geofoam surfaces were examined. The research findings provide a better understanding of EPS geofoam behavior for improvement of design methods and performance prediction of critical infrastructures.

      Non Destructive Testing for EPS Geofam Quality Assurance
      Presented by Engda Temesgen
      Advisor: Dawit Negussey
      Related research area(s): Sustainable Natural and Built Systems
      Poster number CIE 1-06

      EPS geofoam blocks are used as light weight material for infrastructure construction and are produced in different density grades. Strength and deformation properties of geofoam blocks depend on production quality and increase with density. The price of EPS blocks increase with density and volume is the common basis of payment for acquisition. Producers aim to deliver as low density grade as acceptable to maximize profits. An effective quality assurance program must be in place to ensure installation of the specified geofoam grade. The ideal method of quality assurance involves destructive testing in a laboratory. The method is rarely used in practice due to the associated material waste and delay in turnover of test result. Ultrasound velocities through EPS blocks were measured to estimate Young’s moduli for different densities produced at four molding plants. Moduli values derived from non-destructive testing (NDT) compared favorably with previously reported results. For the same density, blocks that contained recycled material had high impedance and lower velocities than blocks made of virgin material. The results indicate NDT is a practical method for onsite quality assurance, evaluation of moduli, and to identify substandard blocks that have recycled content.

      Using Instrumentation to Detect Sub-surface Movements and Groundwater Level Fluctuations
      Presented by Larissa Takou-Ayaoh
      Advisor: Dawit Negussey
      Related research area(s): Other
      Poster number CIE 1-07

      In Onondaga County, the 1993 Tully Valley landslide caused severe damage to homes and facilities. Water level fluctuation is believed to be one of the catalysts of this landslide. Hence, monitoring groundwater level and ground movement is important. A slope indicator (inclinometer) and water logger were used to mimic ground movement and water level fluctuations respectively in the laboratory. Inclinometers can be placed either vertically or horizontally in the field to measure the magnitude, rate, and direction of soil movement. Inclinometers are used in embankments and earth dams which may experience settlement of soil layers and slope failures. Some vertical inclinometers have openings to allow groundwater entry and simultaneously measure water level. Water loggers are placed within vertically placed inclinometers to obtain a continuous water level detection. The data obtained from these instruments are beneficial to prevent and evaluate failures under dynamic field conditions. In addition, the data collected can help predict future ground movements using numerical analysis and modeling.

      Combining Suspect Screening and Fluorescence Analysis to Characterize Organic Micropollutants in Onondaga Lake, NY
      Presented by Shiru Wang
      Advisor: Teng Zeng
      Related research area(s): Sustainable Natural and Built Systems
      Poster number CIE 2-01

      Over the past decade, water quality monitoring in the U.S. has gradually shifted towards the analysis of an increasing number of polar and persistent compounds, or the so-called organic micropollutants (OMPs). While OMPs (e.g., pharmaceuticals) are typically present at low levels in aquatic ecosystems, their occurrence has raised considerable concerns among the scientific community due to potential risks to aquatic life and human health. Onondaga Lake is an urban lake located in Syracuse and has long been stressed by pollution from point sources such as wastewater discharge and diffuse inputs from urban and agricultural runoff. Notably, 20-25% of the lake inflow is contributed by a municipal wastewater treatment plant serving the Syracuse metropolitan area. However, the occurrence and fate of OMPs in the Onondaga Lake system remains largely uncharacterized. Without this knowledge, assessing the potential ecotoxicological risk associated with OMPs or developing mitigation strategies is hindered. Our main objective for this research is to combine liquid chromatography-high resolution mass spectrometry and excitation-emission matrix fluorescence spectroscopy to assess the spatiotemporal occurrence of OMPs in water samples collected from Onondaga Lake and its four major tributaries. Such information can serve as a basis for developing an adaptive, comprehensive monitoring strategy for OMPs at larger scales.

      Rising Heatwave Trends:  A Case Study in Ten Communities across the USA
      Presented by Javad Shafiei Shiva
      Advisor: David Chandler
      Related research area(s): Health and Well-being; Energy Sources, Conversion, and Conservation; Sustainable Natural and Built Systems
      Poster number CIE 2-02

      Heatwaves are an important type of extreme climate events and result in more than 130 deaths per year across the US. Heatwaves have been described by several attributes, combinations of which constitute various event typologies. We studied long-term heatwaves in ten cities during 1950-2016 to better understand how these attributes determine variability in local heatwaves and how climate change is affecting heat-waves across the USA. Our results indicated that at least five harmful attributes of heatwave have increased simultaneously in Miami, New York, Phoenix, and Portland. In addition, we found the largest change in heatwave season length, frequency, and timing occurred in Miami from the 1950s to 2010s.  Meanwhile, despite the significantly high values of heatwave attributes in warm climates, Bismarck, ND and Syracuse, NY have the greatest mean heatwave intensity during 1950s to 2010s. Similar results across much of the domain are presented to clarify the many differences in quantitative heatwave attributes and variance in approaches across climates. This work explores the nexus of quantitative description and social construction of heatwaves through the lens of the various regional metrics to describe heatwaves. Ultimately, this understanding will lead to assessment of various strategies to help communities understand and prepare for heat resilience based on local heat-waves components.

      Analyzing the Toxicity of Cationic Polyacrylamide vs. Cationic Starch on Aquatic Life
      Presented by Katie Duggan
      Advisor: Shobha Bhatia
      Related research area(s): Health and Well-being; Sustainable Natural and Built Systems
      Poster number CIE 2-03

      My research project is focused on the geotextile tube dewatering industry. Geotextile tubes were used in the Onondaga Lake cleanup process, and are used extensively for dewatering applications. In this process, contaminated sediments from the bottom of the lake are dredged and pumped into the tubes. This is then treated with a polymer for flocculation (i.e. bonding sediments together to form larger particles) which ultimately helps retain the contaminated sediments within the tube while releasing the water. Unfortunately under situations where too much polymer is added to the slurry, the polymer flocculant remains in the water that drains from the tubes. I am currently experimenting with both cationic starch (C. Starch) and cationic polyacrylamide (CPAM) to compare the toxicity of each of them on zebrafish embryos (model organism). C. Starch is a natural-based polymer flocculant that is a potential alternative to CPAM, as it been seen to have similar dewatering capabilities. Switching to a less toxic flocculant would result in less toxic water being released from the geotextile tubes and into the surrounding bodies of water.

    • Electrical Engineering and Computer Science: Computer/Information Science Posters

      A Data-Driven Study on the Correlation Between Network Structure and Small-World Phenomenon
      Presented by Pegah Hozhabrierdi
      Advisor: Reza Zafarani
      Related research area(s): Other
      Poster number EECS 2-01

      This study tackles a multi-disciplinary problem of improving the average shortest-path length in a network. A network is described by its members (nodes) and the connections among them which can lead to the formation of groups (communities). Understanding the characteristics of the shortest- paths among these nodes is the key to solving problems in information diffusion theory, power grids, gene regulatory networks and social networks. This concept in social network was first popularized by Milgram in the 60’s. The result of his studies is known as the small-world phenomenon which states every individual in the world is accessible from any other individual by at most six hops on average. Most often, our knowledge of the network structure is limited to the position of its nodes and the possible communities. The question of how these communities and their members affect the small-world phenomenon is our main challenge. The study is divided into two levels; microscopic level (member-based) and macroscopic level (community-based). Real-world social networks of different sizes (from tens of connections to more than a million) are gathered and tested on. The results show that members alone do not affect the average shortest-path (unless they are bridges) while community structure plays an important role.

      Certified Security By Design (CSBD)  Applied to Patrol Base Operations
      Presented by Lorii Pickering
      Advisor: Shiu-Kai Chin
      Related research area(s): Security
      Poster number EECS 2-02

      We all want our systems to be secure and do only what they are designed to do. CSBD is an approach to designing secure systems that satisfy the requirements of TRUSTWORTHINESS outlined in NIST 800-160 Systems Security Engineering Framework. CSBD accomplishes this with an access-control logic (ACL) that reasons about authentication and authorization. Trustworthiness is assured because CSBD employs computer-aided reasoning tools such as the higher order logic (HOL) Interactive Theorem Prover to verify the ACL. To date, CSBD focuses on automated systems. We also want to focus CSBD on non-automated military systems because secure systems are critical to national security, the safety of military personnel, and mission success. We begin with the patrol base operations described in the Ranger Manual.  We describe these as a hierarchy of secure state machines. We identify the needed authentication and authorization and verify everything with ACL in HOL. CSBD also elucidates ways to improve these operations and design future accountability systems for military operations. From designing home-emergency plans for our families to disaster emergency plans for our community to special forces operations for our nation, we all want SECURE SYSTEMS that do what they are designed to do. This research is important because it confirms that CSBD spans the range of automated to non-automated, human-centered systems. Security is important, and we should apply security to all systems we design.

      Create Trusted Path on Untrusted Mobile OS
      Presented by Kailiang Ying
      Advisor: Wenliang Du
      Related research area(s): Security
      Poster number EECS 2-03

      As more and more people are using mobile devices for daily tasks (like authentication, mobile payment), the security of mobile systems becomes very essential. A compromised device can lead to compromise of all sensitive data and operations. To counter such a threat, many processors provide a Trust Execution Environment (TEE), which is isolated from the environment (called normal world) where normal apps run, so even if the normal-world OS is compromised, the data and code inside TEE are still protected. TEE has been used widely, but mainly by apps developed by vendors, while normal apps are unable to benefit from TEE. Also, the way to integrate TEE functionalities with apps is vendor specific and ad hoc. To enable a wide adoption of TEE among apps, providing a seamless and device-neutral integration is essential. In this project, we tackle this integration problem. We address two important aspects: app’s interaction with the user interface (UI) inside TEE, and TEE-assisted interaction between app and server. Our work focuses on changing the OS, so that the way apps interact with UI and server remains the same as before, even though the underlying interaction is quite different due to the usage of TEE. The key idea is to hide the TEE-specific interaction logic inside the OS, making the interaction transparent to the app.

      LSTM Based Soil Moisture Prediction
      Presented by Shivanand Venkanna Sheshappanavar
      Advisor: Chilukuri K. Mohan
      Related research area(s): Intelligence Systems
      Poster number EECS 2-04

      Soil moisture has a colossal impact on several hydrological processes including infiltration, evapotranspiration and subsurface flow. Accurate prediction of soil moisture allows the quantification of drought conditions and the prediction of flash floods caused by precipitation runoff. Economic consequences of accurate soil moisture prediction are significant, assisting improvements in crop productivity and agricultural management practices, and permitting precise control over the root zone environment,  leads to healthier crops and higher yields. In addition to improving weather forecasts monitoring, soil moisture provides us with a better understanding of how water, energy, and carbon are exchanged between land and air. Neural networks such as MLP have been useful in prediction of streamflow based on snow accumulation. More recently, deep networks have been used for prediction, providing greater flexibility in mapping diverse, complex functions. Researchers have recently used Deep Belief Networks for feature learning or extraction. Recurrent Neural Nets with LSTM rank among the state-of-the-art networks for predicting future values of a time series, with potential application to hydrology. Deep Feed-forward Neural Networks have also been used on datasets obtained using Visible Infrared Imaging Radiometer Suite from cropland China. We applied LSTM models to predict soil moisture content, using datasets collected from ground through SCAN and SNOTEL networks.

      LSTM Based Time-Series Link Prediction
      Presented by Zeinab Saghati Jalali
      Advisor: Chilukuri K. Mohan
      Related research area(s): Intelligence Systems
      Poster number EECS 2-05

      The problem of predicting future links is an important task in several research areas such as data mining, social networks, network science, and biology. This problem is relevant to many applications in bioinformatics, e-commerce, social networks and interactions of individuals over the internet. Traditional link prediction methods are often based on modeling and analyzing static snapshots of the network and fail to predict future links in real world networks that are dynamic. In this work, we propose a new time series link prediction method using recurrent networks based on Long Short-Term Memory (LSTM) units. LSTMs have been widely used in different areas and were able to achieve satisfactory results compared to other methods in different areas such as language modeling, translation, and acoustic modeling of speech. To the best of our knowledge, the RNN approach has not been previously applied to address link prediction using LSTM. In the proposed approach, input data passes through the LSTM units, in a chain consisting a series of snapshots at time 1 to time T-1, with information about the presence of various links; predictions are then made about the existence of links at time T. Our experiments show better results compared to other methods (such as dynamic methods that use baseline methods like common neighborhood, Jaccard-coefficient, and Adamic-Adar) in terms of different accuracy measures.

      Predicting TV Audience Viewership during Commercial Breaks through the Application of Neural Networks
      Presented by Sushanth Suresh
      Advisor: Chilukuri K. Mohan
      Related research area(s): Other
      Poster number EECS 2-06

      TV ad viewership determines the price advertisers pay for their ads, therefore, we work intently to examine audience viewership decline during commercial breaks predicated on moment-to-moment data tracking with a goal to predict audience’s mechanical ad avoidance behaviour and identify key attributes leading to ad viewership decline.

      We investigated three problem statements to accomplish our objective

      1. Falloff based on ads vs one Tv show
      2. Fall off based on relevant advertisement features like ads age, product age.
      3. Fall off based on one add vs Tv shows and network.

      Different supervised and unsupervised neural network models such as Backpropagation, Support Vector Machine and Linear Vector Quantization were implemented to predict the drop in the viewership and find out which model best suits our data.


      1. The accuracies suggested that our data best fitted with the support vector machine model and using the weights we created a relative importance table which gave light as in Duration of the advertisement is the most important feature used to predict the drop-in viewership.
      2. Our efforts to train our models using the ad features and different Tv shows resulted in discovering the data was unpredictable.

      Future Work:

      1. The effect of the number of advertisements in a number of advertisement towards the drop in the audience.
      2. The analysis where the genre of the advertisement and the program are similar.

      Protecting Sensitive Data in Android SQLite Databases in a Compromised Execution Environment
      Presented by Francis Akowuah
      Advisor: Wenliang Du
      Related research area(s): Security
      Poster number EECS 2-07

      SQLite is a free in-process library that implements SQL database engine. It has been widely deployed in many systems. Given the wide use of smartphones in our daily lives, SQLite database in a mobile app may contain information such as social security number, credit card number, health information, private email conversation, and in some cases passwords. SQLite has no built-in security like SQL Server and Oracle. It relies on its environment, for example, the operating system, to provide security for the database content. Security provided by Android for SQLite has been shown to be inadequate. Proposed solutions by the research community do not take into consideration when the OS becomes compromised. Therefore, we propose a hardware isolation solution that protects sensitive database content in a compromised environment by leveraging the security guarantees of the ARM TrustZone. Mainly, we encrypt and encode the sensitive data in the trusted execution environment (secure world) where a malicious app cannot access. We modify classes in the Android Framework such that the developer can express what constitutes sensitive data to the operating system.  We also design and develop a TrustZone-base UI such that data entry is also done in isolation from the normal OS. Evaluation results indicate our solution is practical. This solution enables developers to ensure user sensitive data is protected even if the database file is relocated.

      Selective Parts-Based Tracking Through Occlusions
      Presented by Maria Cornacchia
      Advisor: Senem Velipasalar
      Related research area(s): Other
      Poster number EECS 2-08

      Visual tracking is a difficult task due to numerous scale, occlusion, motion blur, and other deformation changes throughout a video sequence. While correlation filter trackers have recently shown promise, it still remains a challenge to account for the numerous different changes of an object during tracking. In this paper, we propose a selective parts-based approach, using correlation filters, that makes choices based on a consensus of the parts and global tracking to track through occlusions. In contrast to existing part-based methods, the proposed method does not dilute accurate tracking by averaging results over multiple parts at every frame. Instead, we only make location corrections when a part diverges and rely on these corrections to maintain an accurate appearance model. The proposed approach was evaluated for scenarios obtained from two different challenging benchmark datasets. Our approach has resulted in better overall precision and success rates compared to recent parts-based approaches, and has performed better especially in occlusion scenarios.

      Simultaneous Exploration and Harvesting in Multi-Robot Foraging
      Presented by Zilong Jiao
      Advisor: Jae Oh
      Related research area(s): Unmanned Systems; Intelligent Systems
      Poster number EECS 2-09

      We study the multi-robot foraging task in unknown environment with risks. The multi-robot foraging task is for deploying a group of robots to explore unknown environment and transport discovered targets back to their home base. In the task, multi-robot exploration is essential, and it has many applications in practice, including planetary exploration, cleaning, harvesting and environmental data collecting. In our work, the frontier-based exploration algorithm is integrated with the auction-based task allocation method, so that a team of homogeneous robots can simultaneously explore the unknown environment and collect discovered targets. The experiment results demonstrate that our proposed algorithm can effectively balance the tasks of environment exploration and target collection.

      T-SGX: Eradicting Controlled-Channel Attacks against Enclave programs
      Presented by Kai Li
      Advisor: Yuzhe Tang
      Related research area(s): Security
      Poster number EECS 2-10

      To protect your private data from leaking to the hacker in the cloud storage. Previous works such as TSGX tried to defend the control-channel attacks for Intel Enclave program by leveraging the TSX to do atom transaction commit. However, it protects the privacy of enclave program with the cost of data scalability. We mitigate such limitation by dynamic calculating the transaction size at runtime, we can find the tradeoff point of TSX, that at this point, the data security and data scalability are both can be guaranteed.

      Measuring and Improving the Core Resilience of Networks
      Presented by Ricky Laishram. Advisor: Sucheta Soundarajan
      Related research area(s): Other
      Poster number EECS 2-12

      It is often valuable to understand the resilience of the k-cores of a network to attacks and dropped edges (i.e., damaged communications links). We provide a formal definition of a network’s core resilience, and examine the problem of characterizing core resilience in terms of the network’s structural features: in particular, which structural properties cause a network to have high or low core resilience? To measure this, we introduce two novel node properties, Core Strength and Core Influence, which measure the resilience of individual nodes’ core numbers and their influence on other nodes’ core numbers. Using these properties, we propose the Maximize Resilience of k- Core (MRKC) algorithm to add edges to improve the core resilience of a network. We find that on average, for edge deletion attacks, MRKC improves the resilience of a network by 11.1% over the original network, as compared to the best baseline method, which improves the resilience of a network by only 2%. For node deletion attacks, MRKC improves the core resilience of the original network by 19.7% on average, while the best baseline improves it by only 3%.

    • Electrical Engineering and Computer Science: Electrical/Computer Engineering Posters

      Presented by Swatantra Kafle
      Advisor: Pramod Varshney
      Related research area(s): Other
      Poster number EECS 1-01

      In this paper, we consider the problem of sparse signal reconstruction with 1-bit compressed measurements using GAMP algorithm. This work proposes reconstruction method for a general setting, i.e., when the compressed measurements are corrupted with additive Gaussian noise before quantization and with the bit-flip error after quantization. This work is first of its kind which tries to analyze the effect of pre-quantization and post-quantization noises in the signal reconstruction from the 1-bit quantized measurement. We can always improve the performance of signal reconstruction by making extra assumptions on sparse signals or taking side information into account. We believe that the reconstruction performance provided by this algorithm can be used as a benchmark performance for every effort in improving reconstruction performance from noisy 1-bit compressed measurements.

      Presented by Yu Zheng
      Advisor: Senem Velipasalar
      Related research area(s): Unmanned Systems; Intelligent Systems; Security
      Poster number EECS 1-02

      Target re-identification across non-overlapping camera views is a challenging task due to variations in target appearance, illumination, viewpoint and intrinsic parameters of cameras. Brightness transfer function (BTF) was introduced for inter-camera color calibration, and to improve the performance of target re-identification methods. There have been several works based on BTFs, more specifically using weighted BTFs (WBTF), cumulative BTF (CBTF) and mean BTF (MBTF). In this work, we present a novel method to model the appearance variation across different camera views. We propose building a codebook of BTFs composed of the most representative BTFs for a camera pair. We also propose an ordering and trimming criteria to avoid using all possible combinations of codewords for different color channels. In addition, to obtain a better appearance model, we present a different way to segment a target from the background. Evaluations on VIPeR, CUHK01 and CAVIAR4REID datasets show that the proposed method outperforms other approaches focusing on BTFs, including WBTF, CBTF and MBTF. As proven by the results, the proposed method provides an improved brightness transfer across different camera views, and any target ReID approach incorporating color/brightness histograms can benefit from it.

      A Deep Reinforcement Learning-Based Framework for Cloud Resource Allocation
      Presented by Ning Liu
      Advisor: Yanzhi Wang
      Related research area(s): Intelligent Systems
      Poster number EECS 1-03

      Automatic decision-making approaches, such as reinforcement learning (RL), have been applied to (partially) solve the resource allocation problem adaptively in the cloud computing system. However, a complete cloud resource allocation framework exhibits high dimensions in state and action spaces, which prohibit the usefulness of traditional RL techniques.

      In addition, high power consumption has become one of the critical concerns in design and control of cloud computing systems, which degrades system reliability and increases cooling cost. An effective dynamic power management (DPM) policy should minimize power consumption while maintaining performance degradation within an acceptable level.

      Thus, a joint virtual machine (VM) resource allocation and power management framework is critical to the overall cloud computing system. Moreover, novel solution framework is necessary to address the even higher dimensions in state and action spaces.

      A Deep Reinforcement Learning-Based Framework for Content Caching
      Presented by Chen Zhong
      Advisor: Cenk Gursoy
      Related research area(s): Intelligent Systems
      Poster number EECS 1-04

      Content caching at the edge nodes is a promising technique to reduce the data traffic in next-generation wireless networks. Inspired by the success of Deep Reinforcement Learning(DRL) in solving complicated control problems, this work presents a DRL-based framework with Wolpertinger architecture for content caching at the base station. The proposed framework is aimed at maximizing the long-term cache hit rate, and it requires no knowledge of the content popularity distribution. To evaluate the proposed framework, we compare the performance with other caching algorithms, including Least Recently Used(LRU), Least Frequently Used (LFU), and First-In First-Out(FIFO) caching strategies. Meanwhile, since the Wolpertinger architecture can effectively limit the action space size, we also compare the performance with Deep Q-Network to identify the impact of dropping a portion of the actions. Our results show that the proposed framework can achieve improved short-term cache hit rate and improved and stable long-term cache hit rate in comparison with LRU, LFU, and FIFO schemes. Additionally, the performance is shown to be competitive in comparison to Deep Q-learning, while the proposed framework can provide significant savings in runtime.

      An Approach to Classification of Power System Dynamic Events and False Data Injection Using PMU Data
      Presented by Rui Ma
      Advisor: Sara Eftekharnejad
      Related research area(s): Security
      Poster number EECS 1-05

      Identification of the real-time transient events is essential for power system protection and control. Phasor measurement units (PMUs) provide synchronized voltage phasors, current phasors, and frequency measurements at a high resolution. A targeted false data injection attack on PMUs can prompt operators to take wrong actions that eventually jeopardize power system reliability.  In this work, a multivariate-based model is developed to not only verify the correctness of the PMU data but also classify dynamic events by considering the attributes of each PMU data and their relationship. In addition, a method, inspired by text mining is proposed to help efficiently identify each transient event.

      Behavioral Hypothesis Testing
      Presented by Baocheng Geng
      Advisor: Pramod Varshney
      Related research area(s): Intelligent Systems
      Poster number EECS 1-06

      Traditional sensor networks have found applications in lots of areas. These sensor networks only contain machines to collect information to a fusion center. However, many mission-oriented control system requires humans as an essential part of the decision-making process. Unlike traditional sensors which can be programmed to do whatever we want, when analyzing human’s decision making we need to consider the cognitive limitation, uncertainty and unpredictability of human beings. In this work, we model how human make decisions in a hypothesis testing scenario as an individual, as well as in population level.

      Convolutional Neural Network Based Multi-Scale 3D Object Detection
      Presented by Burak Kakillioglu
      Advisor: Senem Velipasalar
      Related research area(s): Intelligent Systems
      Poster number EECS 1-07

      The advent of LiDAR sensors and increasing availability of 3D depth sensors such as Microsoft Kinect sensors induced research using 3D data for many application areas including virtual reality, autonomous navigation and surveillance. Moreover, smart phone companies are expeditiously moving towards implementing 3D sensors in their devices. As new and reliable techniques have become available, the demand for 3D data processing has increased in many industries, such as transportation, avionics, defense and entertainment. Current state-of-the-art approaches for 2D image classification and detection, which are based on convolutional neural networks (CNN), have achieved very high accuracy rates on various applications surpassing human performance. Although there have been successful attempts at object classification and detection in 3D volumetric domain, the performance of the state-of-the art approaches are still not as improved as that of their 2D counterparts. In this work we created a 3D object detector which detects and localizes objects in 3D point clouds of scene captures which are possibly acquired by LiDAR sensors or depth cameras. We first encode the input scene cloud into 3D occupancy grid. Then we extract multi-scale 3D features of the scene using CNNs and apply sliding anchors on these features where these anchors act as a classifier for every region on each feature scale. Our algorithm gives promising preliminary detection results on SUN-RGBD dataset.

      Distributed Self Localization and tracking with an unknown number of targets
      Presented by Pranay Sharma
      Advisor: Pramod Varshney
      Related research area(s): Unmanned Systems; Intelligent Systems
      Poster number EECS 1-08

      In this work, we propose an algorithm for sequentially localizing multiple mobile agents, while simultaneously using these agents to track an unknown number of mobile targets. We work under the additional constraint of measurement origin uncertainty, i.e., it is unknown which target a measurement originated from, or if it was caused by a false alarm. For better scalability along with distributed implementation, we use particle-based belief propagation on appropriately designed factor graphs. Since each agent node needs to estimate its own state, as well as the states of all targets, we use suitable “gossip” schemes for communicating relevant information among agents. To the best of our knowledge, this is the first work to consider simultaneous localization and tracking, in a distributed setting, with measurement origin uncertainty.

      Beyond the obvious application in surveillance, this problem also has potential application in autonomous driving (in which case other manual driven vehicles and pedestrians become targets), indoor localization, deploying unmanned first-responder agents in hazardous environments, biomedical analytics (for e.g., tracking cells in multidimensional time-lapse fluorescence microscopy), and crowd counting.

      Downlink Analysis in Unmanned Aerial Vehicle (UAV) Assisted Cellular Networks with Clustered Users
      Presented by Esma Turgut
      Advisor:Cenk Gursoy
      Related research area(s): Unmanned Systems
      Poster number EECS 1-09

      The use of unmanned aerial vehicles (UAVs) operating as aerial base stations (BSs) has emerged as a promising solution especially in scenarios requiring rapid deployments (e.g., in the cases of crowded hotspots, sporting events, natural disasters) to assist the ground BSs. An analytical framework is provided to analyze the signal-to-interference-plus-noise ratio (SINR) coverage probability of UAV assisted cellular networks with clustered user equipments (UEs). Locations of UAVs and ground BSs are modeled as Poison point processes (PPPs), and UEs are assumed to be distributed according to a Poisson cluster process (PCP) around the projections of UAVs on the ground. The complementary cumulative distribution function (CCDF) and probability density function (PDF) of path losses for both UAV and ground BS tiers are derived. Association probabilities with each tier are obtained. SINR coverage probability is derived for the entire network using tools from stochastic geometry. Area spectral efficiency (ASE) of the entire network is determined, and SINR coverage probability expression for a more general model is presented by considering that UAVs are located at different heights. We have shown that UAV height and path-loss exponents play important roles on the coverage performance. Coverage probability can be improved with smaller number of UAVs, while better area spectral efficiency is achieved by employing more UAVs and having UEs more compactly clustered around the UAVs.

      Fast and Energy-Aware Resource Provisioning and Task Scheduling for Cloud Systems
      Presented by Hongjia Li
      Advisor: Yanzhi Wang
      Related research area(s): Other
      Poster number EECS 1-10

      Cloud computing has become an attractive computing paradigm in recent years to offer on demand computing resources for users worldwide. Through Virtual Machine (VM) technologies, the cloud service providers (CSPs) can provide users the infrastructure, platform, and software with a quite low cost. In this paper, we propose a fast and energy-aware resource provisioning and task scheduling algorithm to achieve low energy cost with reduced computational complexity for CSPs.

      In our iterative algorithm, we divide the provisioning and scheduling to multiple steps which can effectively reduce the complexity and minimize the run time while achieving a reasonable energy cost. Experimental results demonstrate that compared to the baseline algorithm, the proposed algorithm can achieve up to 79.94% runtime improvement with an acceptable energy cost increase.

      Fusion of Correlated Decisions Using Regular Vine Copulas
      Presented by Shan Zhang
      Advisor: Pramod Varshney
      Related research area(s): Other
      Poster number EECS 1-11

      In this paper, we propose a regular vine copula based methodology for the fusion of correlated decisions. Regular vine copula is an extremely flexible and powerful graphical model to characterize complex dependence among multiple modalities. It can express a multivariate copula by using a cascade of bivariate copulas, the so-called pair copulas. Assuming that local detectors are single threshold binary quantizers and taking complex dependence among sensor decisions into account, we design an optimal fusion rule using a regular vine copula under the Neyman-Pearson framework. In order to reduce the computational

      complexity resulting from the complex dependence, we propose an efficient and computationally light regular vine copula based optimal fusion algorithm. Numerical experiments are conducted to demonstrate the effectiveness of our approach.

      Hardware Acceleration of Bayesian Neural Networks
      Presented by Ruizhe Cai
      Advisor: Yanzhi Wang
      Related research area(s): Intelligent Systems
      Poster number EECS 1-12

      Bayesian Neural Networks (BNNs) have been proposed to address the problem of model uncertainty in training and inference. By introducing weights associated with conditioned probability distributions, BNNs are capable of resolving the overfitting issue commonly seen in conventional neural networks and allow for small-data training, through the variational inference process. Frequent usage of Gaussian random variables in this process requires a properly optimized Gaussian Random Number Generator (GRNG). The high hardware cost of conventional GRNG makes the hardware implementation of BNNs challenging.In this paper, we propose VIBNN, an FPGA-based hardware accelerator design for variational inference on BNNs. We explore the design space for massive amount of Gaussian variable sampling tasks in BNNs. Specifically, we introduce two high performance Gaussian (pseudo) random number generators:

      1. the RAM-based Linear Feedback Gaussian Random Number Generator (RLF-GRNG),which is inspired by the properties of binomial distribution and linear feedback logics
      2. the Bayesian Neural Network-oriented Wallace Gaussian Random Number Generator.

      To achieve highscalability and efficient memory access, we propose a deep pipelined accelerator architecture with fast execution and good hardware utilization. Experimental results demonstrate that the proposed VIBNN implementations on an FPGA can achieve throughput of 321,543.4 Images/s and energy efficiency upto 52,694.8 Images/J while maintaining good accuracy.

      Inducing Cascading Failures in Transportation Networks
      Presented by Griffin Kearney
      Advisor: Makan Fardad
      Related research area(s): Unmanned Systems,Intelligent Systems,Security
      Poster number EECS 1-13

      In this work we examine the effect of malicious attacks in disrupting optimal routing algorithms for transportation networks. Highway networks, disaster evacuation plans, water supply networks, and (the routing of data packets in) computer networks can all be described with transportation network models. We focus on modeling traffic networks using the cell transmission model, which is a spatiotemporal discretization of kinematic wave equations. Here, vehicles are modeled as masses and roads as cells, and traffic flow is subject to conservation of mass and capacity constraints. At time zero a resource-constrained malicious agent reduces the capacities of cells so as to maximize the amount of time mass spends in the network. For the resulting set of capacities the network router then solves a linear program to determine the flow configuration that minimizes the amount of time mass spends in the network. This two-player problem can be written as a max-min that can be transformed to an equivalent bilinear maximization problem.  Optimization problems with bilinear objectives are non-convex, and known to be NP-hard in general. Linearization techniques are applied to the formulation to find solutions. These techniques scale gracefully in network size but may converge to globally non-optimal values. Analyzing fundamental examples shows that attackers with relatively small resource budgets can cause widespread failure in a traffic network.

      Integrating ARM TrustZone with Android to Protect VoIP Calls
      Presented by Amit Ahlawat
      Advisor: Wenliang Du
      Related research area(s): Security
      Poster number EECS 1-14

      An important trend in the use of mobile devices to make phone calls is the use of Voice-over-IP (VoIP) apps, such as Facebook Messenger, Google Hangouts, Signal etc.  When using a VoIP app, users place implicit trust on the VoIP protocol, VoIP infrastructure and the mobile device. A compromise in any one of them can lead to privacy loss. In this work, we focus on the trust placed on a mobile device. Focusing on the Android OS, the absolute trust placed by users on the device is misplaced, as data shows that the number of vulnerabilities discovered in Android has increased from year to year. With such weak trusted computing base (TCB), mobile devices cannot provide the privacy guarantee needed by users during a VoIP call, as a compromised mobile OS can examine the audio content of a VoIP call.

      This work intends to design a solution for secure VoIP call on Android using a technology called ARM TrustZone. The design leverages the partitioning of device resources allowed by TrustZone and applies it to secure audio peripherals on a device while a VoIP call is being made. The goal is that during a VoIP call, a compromised Android OS should be unable to access the audio from the mic and the speaker. The design aims to provide TrustZone support transparently to VoIP app developers s.t. with minimal configuration, they should be able to convert an existing VoIP app to a TrustZone-secured version.

      Making Computers Emotion Aware
      Presented by Danushka Bandara
      Advisor: Senem Velipasalar
      Related research area(s): Intelligent Systems
      Poster number EECS 1-15

      The human brain is the genesis of our emotions. Our work uses the brain as an objective indicator of emotion. Emotion in the brain is associated with the Limbic system and the Prefrontal Cortex. We focus on the Blood Flow Activity in the Pre Frontal Cortex to infer a person’s emotional state (Positivity or Negativity of emotion in particular: AKA Valence). The contributions of this work includes improving on current state of the art in Valence classification as well as proposing a novel approach to capturing the spatial nature of fNIRS data in the classification of fNIRS data in general. Such improvement in emotion classification can impact fields such as robotics, psychology, education, autonomous vehicles, interactive media and assistive technologies where an objective awareness of human emotion can provide the computer the ability to better adapt to the user.

      Microscopic Origin of the Chiroptical Response of Plasmonic Media
      Presented by Matthew Davis. Advisor: Jay Lee
      Related research area(s): Health and Well-being,Other
      Poster number EECS 1-17

      Chiral plasmonic systems are compelling components in a range of impactful technologies such as actively controlled ultrafast CP modulators, high efficiency visible wavelength holograms, and enhanced circular dichroism (CD) spectroscopy. Chiroptical techniques such as CD spectroscopy can be used to identify structural and handedness information of a chiral medium, which is of great importance in the study of pharmaceuticals, physiology, and the origins of life itself. CD spectroscopy plays an important role in both the identification of enantiomorphic compounds and conformational analysis, however CD responses in natural molecules is weak. The promise of enhanced CD spectroscopy has attracted an intense research effort in the field of chiral plasmonics. Providing orders of magnitude signal enhancement, plasmonic systems seem poised to continue making significant contributions to chiroptical measurements. Understanding the chiroptical properties of plasmonic structures is, therefore, vital. In this work, a Lorentzian coupled-oscillator model is introduced to facilitate a comprehensive study of the chiroptical properties of plasmonic media to aid in the study and identification of CD responses. The GPBK model is shown to unite several previously reported chiroptical phenomena into a unified theoretical framework, illuminating the origins of parasitic non-CD response types in chiral structures and providing guidance to the thoughtful design of chiroptical measurements.

      Presented by Prashant Khanduri
      Advisor: Pramod Varshney
      Related research area(s): Energy Sources, Conversion, and Conservation,Intelligent Systems
      Poster number EECS 1-18

      In this work, we present an efficient methodology to design precoders for distributed detection of unknown high dimensional signals. We consider a wireless sensor network, where several distributed sensors collaborate to perform binary hypothesis testing based on observations of an unknown high dimensional signal corrupted by noise. The sensors collect data over both temporal and spatial domains. Due to network resource constraints, each sensor performs a linear compression (through precoding) of the observed high dimensional signal at each time instant and forwards the compressed signal to the fusion center (FC). The FC then employs the generalized likelihood ratio test (GLRT) to make a decision on the presence or absence of the signal. We propose online linear precoding/compression strategies for such sensors that collect data over spatio-temporal domain, so that the detection performance at the FC is maximized under certain network resource constraints. Through the measure of non-centrality parameter and receiver operating characteristics (ROC), we show that our proposed precoder design achieves very good detection performance.

      Power Control and Mode Selection for VBR Video Streaming in D2D Networks
      Presented by Chuang Ye
      Advisor: Mustafa Gursoy
      Related research area(s): Other
      Poster number EECS 1-19

      In this work, we investigate the problem of power control for streaming variable-bit-rate (VBR) videos in a device-to-device (D2D) wireless network. A VBR video traffic model that considers video frame sizes and playout buffers at the mobile users is adopted. A setup with one pair of D2D users (DUs) and one cellular user (CU) is considered and three modes, namely cellular mode, dedicated mode and reuse mode, are employed. Mode selection for the data delivery is determined and the transmit powers of the base station (BS) and device transmitter are optimized with the goal of maximizing the overall transmission rate while VBR video data can be delivered to the CU and DU without causing playout buffer underflows or overflows. A low-complexity algorithm is proposed. Through simulations with VBR video traces over fading channels, we demonstrate that video delivery with mode selection and power control achieves a better performance than just using a single mode throughout the transmission.

      Robust Decentralized Learning with Unreliable Agents
      Presented by Qunwei Li
      Advisor: Pramod Varshney
      Related research area(s): Intelligent Systems
      Poster number EECS 1-20

      Many machine learning problems can be formulated as consensus optimization problems which can be solved efficiently via a cooperative multi-agent system.  However, the agents in the system can be unreliable due to a variety of reasons: noise, faults and attacks. Thus, providing erroneous data leads the problem solving process in a wrong direction, and degrades the performance of distributed machine learning algorithms. This paper considers the problem of decentralized learning in the presence of unreliable agents.  First, we rigorously analyze the effect of erroneous updates (in ADMM-based learning iterations) on the convergence behavior in solving optimization problems in the multi-agent system. We show that the algorithm converges to a neighborhood of the optimal solution and characterize the neighborhood size analytically. Next, we provide guidelines for multi-agent system design to achieve a faster problem-solving capability. We also provide necessary conditions on the falsified updates for exact convergence to the optimal solution. Finally, to mitigate the influence of unreliable agents, we propose a robust variant of ADMM and show its  resilience to unreliable agents.

      SC-DCNN: Highly-Scalable Deep Convolutional Neural Network using Stochastic Computing
      Presented by Ao Ren
      Advisor: Yanzhi Wang
      Related research area(s): Intelligent Systems
      Poster number EECS 1-21

      With recent advancing of wearable devices and Internet of Things (IoTs), it becomes very attractive to implement the deep convolutional neural networks (DCNNs) onto embedded and portable systems. Stochastic Computing (SC), which uses a bit-stream to represent a probability number within [-1, 1] by counting the number of ones in the bit-stream, has high potential for implementing DCNNs with high scalability and ultra-low hardware footprint. Since multiplications and additions can be calculated using AND gates and multiplexers in SC, significant reductions in power (energy) and hardware footprint can be achieved compared to the conventional binary arithmetic implementations. We present the first comprehensive design and optimization framework of SC-based DCNNs (SC-DCNNs), using a bottom-up approach. We first present the optimal designs of function blocks that perform the basic operations, i.e., inner product, pooling, and activation function, in DCNN. Then we propose the optimal design of four types of combinations of basic function blocks, named feature extraction blocks, which are in charge of extracting features from input feature maps. Besides, weight storage methods are proposed and investigated to reduce the area and power (energy) consumption for storing weights. Finally, the whole SC-DCNN implementation is optimized, with feature extraction blocks carefully selected, to minimize area and power (energy) consumption while maintaining a high network accuracy level.

      Statistical Reinforcement Learning Based Joint Antenna Selection and User Scheduling for Single-Cell Massive MIMO Systems
      Presented by Mangqing Guo
      Advisor: M. Cenk Gursoy
      Related research area(s): Other
      Poster number EECS 1-22

      A statistical reinforcement learning based joint antenna selection and user scheduling method to improve the Energy Efficiency (EE) of single-cell Massive MIMO system is investigated. The uplink and downlink transmissions are considered together. We also consider a limitation on the number of Radio Frequency chains in Massive MIMO systems. The original energy-efficiency-maximizing problem is solved via a two-step approach: Determine the optimal subset of antennas at the Base Station (BS) first, and then obtain the optimal subset of users with the subset of antennas selected before. We note that EE initially increases with the number of antennas at the BS, and then decreases, and random antenna selection is already very close to the optimum antenna selection. So we could determine the optimum number of antennas that should be selected at the BS via a typical bisection algorithm. Then the original problem can be transformed into a combinatorial optimization problem. Finally, we use statistical reinforcement learning methods to solve this combinatorial optimization problem.

      Throughput Analysis with Content Caching and RRH Association in Cloud-Radio Access Networks
      Presented by Yang Yang
      Advisor: Mustafa Gursoy
      Related research area(s): Energy Sources, Conversion, and Conservation
      Poster number EECS 1-23

      Cloud-Radio Access Network (C-RAN) is a centralized, cloud computing-based architecture for radio access networks that supports 2G, 3G, 4G and future wireless communication standards. However, due to the introduction of caches at the base station and cloud center, how to cache contents and associate users with the Remote Radio Heads (RRHs) will have significant

      influence on the network. In our recent work, local content caching algorithms and RRH association are investigated under C-RAN constraints. In particular, we first propose a pre-ideal user association algorithm to demonstrate the association scenario in the desired best performance case. Then, a content caching method is presented to explain how the contents should be stored. Effective capacity is used as the metric in performance comparison, which is also the objective we want to maximize in practical RRH association scenarios.

      Towards Ultra-High Performance and Energy Efficiency of Deep Learning Systems: An Algorithm-Hardware Co-Optimization Framework
      Presented by Caiwen Ding
      Advisor: Yanzhi Wang
      Related research area(s): Intelligent Systems,Other
      Poster number EECS 1-24

      Hardware accelerations of deep learning systems have been extensively investigated in industry and academia. The aim of this paper is to achieve ultra-high energy efficiency and performance for hardware implementations of deep neural networks (DNNs). An algorithm-hardware co-optimization framework is developed, which is applicable to different DNN types, sizes, and application scenarios. The algorithm part adopts the general block-circulant matrices to achieve a fine-grained tradeoff of accuracy and compression ratio. It applies to both fully-connected and convolutional layers and contains a mathematically rigorous proof of the effectiveness of the method. The proposed algorithm reduces computational complexity per layer from O($n^2$) to O($n\log n$) and storage complexity from O($n^2$) to O($n$), both for training and inference. The hardware part consists of highly efficient \emph{Field Programmable Gate Array} (FPGA)-based implementations using effective reconfiguration, batch processing, deep pipelining, resource re-using, and hierarchical control. Experimental results demonstrate that the proposed framework achieves at least 152X speedup and 71X energy efficiency gain compared with IBM TrueNorth processor under the same test accuracy. It achieves at least 31X energy efficiency gain compared with the reference FPGA-based work.

      Cache-miss Oblivious Data Shuffling on Harware Enclaves
      Presented by Ju Chen
      Advisor: Yanzhi Tang
      Related research area(s): Security
      Poster number EECS 1-25

      Memory-access based side-channel attacks are real and serious security problems. By exploiting the vulnerabilities, attackers can extract sensitive information such as secret keys and users’ profiles from the programs run on a shared execution platform. Recently discovered attacks such as Meltdown and Spectra even allowing a program to read arbitrary information in memory by combining the vulnerabilities with hardware design flaws. This work takes the initiative to mitigate memory-access based side-channel attacks. Our approach is to enforce the obliviousness of cache-misses emitted during a program execution against sensitive data. The novelty is to combine the detection of the cache-misses with the externally oblivious algorithms. The evaluation results show that our approach outperforms the related works in the performance and the scalability, without losing security.

      Uplink Coverage in Heterogeneous mmWave Cellular Networks with User-Centric Small Cell Deployments
      Presented by Xueyuan Wang
      Advisor: Mustafa Gursoy
      Related research area(s): Other
      Poster number EECS 1-26

      Demand for mobile data has been growing rapidly in recent years resulting in a global bandwidth shortage for wireless service providers. For future cellular networks including 5G wireless systems, two key techniques for capacity improvement will be network densification and the use of higher frequencies such as in millimeter wave (mmWave) bands. Yet as another trend, heterogeneous cellular wireless networks are being developed to support higher data rates to satisfy the increasing user demand for broadband wireless services, by supporting the coexistence of denser but lower power small-cell base stations (BSs) with the conventional high-power and low density large-cell BSs. Motivated by the fact that mmWave communications and user-centric deployments have been attracting growing attention as significant components in next-generation wireless networks, our work focus on K-tier heterogeneous uplink mmWave cellular networks with UE-centric small cell deployments. Our work gives insights for setting up next generation wireless networks, and how to deploy base stations to obtain high communication quality.

    • Mechanical and Aerospace Engineering Posters

      Antiplane Shear of Cylinders and Layered Systems: Cohesive Fracture and Instability
      Presented by Yueming Song
      Advisor: Alan Levy
      Poster number MAE1-01

      Mode-III (antiplane shear) fracture is one of the three fundamental fracture modes that can happen by itself or couple with other modes. This research exmaines the mechanics of mode-III defect initiation and quasi-static grwoth by analyzing cylinder and layered systems. The analysis is based on exact solution which combines linear elasticity formulation of deformation and nonlinear cohesive zone models governing interfacial bahavior. For a particular geometry and defect configuration, these solutions are shown to lead to a pair of interfacial integral equations which capture the entire defect evolution process from incipient growth through complete failure. General features of the solutions to antiplane shear cohesive fracture in both geometries indicate that quasi-static defect initiation and propagation occur under increasing load. For small values of characteristic force length, brittle behavior occurs that is readily identifiable with the growth of a sharp crack. At larger values of force length, ductile response occurs which is more characteristic of a linear spring interface. Both behaviors ultimately give rise to abrupt failure of the interface. Results for the stiff, strong interface under a small applied load show consistency with static fracture mechanics solutions in the literature. Other problems, such as interface stability, multilayer and multidefect system behavior, mixed mode fracture and frictional effects can also be addressed.

      Automatic Generation of Near-Body Structured Grids
      Presented by Yuyan Hao
      Advisor: John Dannenhoffer
      Poster number MAE1-02

      Overset grid is used in simulating complex fluid flow problems. It decomposes a complex geometry into near-body and far-field grids. Hyperbolic Grid Generation (HYPGEN) and Grids about Airfoil and other shapes by the use of Poisson Equations (GRAPE) are the two-main near-body grid generation techniques, which have their own strengths and weaknesses. My research is to generate a new scheme which combines these two methods exploits their advantages and reduces the number of user-inputs. In HYPGEN, a mesh is generated by propagating in the normal direction from a known level of points to a new level. There are a bunch of mechanisms to make HYPGEN works in different cases. However, it will become unstable in the far field. On the other hand, GRAPE requires a user defined outer boundary and generate the entire mesh simultaneously by iteration. The scheme works well but require much more time to solve. In my combined scheme, the mesh is generated by HYPGEN at first with some “bad” points. Then, we find out those bad points by checking the length and skewness of the grids and cut those parts off. After that, we patch those parts by GRAPE method with boundary conditions given from the rest of HYPGEN. The combined scheme generates grids fast, get a useful grid without overlapping, and user parameters about process in not needed.

      Development of a smartphone-based app for determination of formaldehyde concentration indoors
      Presented by Zhenlei Liu
      Advisor: Jianshun Zhang
      Poster number MAE1-03

      Formaldehyde concentrations in residences and offices are typically in the range from a few ppb to about 100 ppb. Measuring formaldehyde concentration at these low levels is challenging and no low-cost badge sensors currently exist. The objective of this work/task is to experimentally test and evaluate the formaldehyde badges to be developed in the proposed NSF research project for the detection and quantification of the formaldehyde under well-controlled temperature, relative humidity and airflow conditions that are similar to that experienced in a typical indoor environment. The goal of the present project will be accomplished via the following objectives:

      1. Record images of badges exposed to different concentrations of formaldehyde;
      2. Use these images to quantify color change and create a calibration curve for the smartphone app;
      3. Evaluate the performance of badge from two groups.

      Generalized Fillets/Chamfers via B-spline Surfaces
      Presented by Zachary Eager
      Advisor: John Dannenhoffer
      Poster number MAE1-04

      A common desire in solid modeling is the ability to create a transition (blend) between two surfaces or adjacent faces of a solid. Currently existing approaches include fillets (or rounds) and chamfers. The issue with fillets and chamfers is that they are, by definition or implementation, highly restrictive. In the case of fillets, the radius is fixed and thus is the location where the fillet meets the original surface. Likewise, chamfers also have a fixed “radius”. We wish to create a method using open-source tools that allows the creation of transition surfaces where the curvature of the resulting surface is an input parameter. Furthermore, we wish to provide the user with the ability to choose where the transition surface begins/ends on the original surfaces. A comprehensive method that achieves this would greatly increase the flexibility of any such feature in a CAD system, furthermore allowing a designer to more freely express their creativity while greatly reducing the amount of time required to do so.

      Impact of Gaseous Contamination and High Humidity on the Reliable Operation of Information Technology Equipment in Data Center
      Presented by Rui Zhang
      Advisor: Jianshun Zhang
      Poster number MAE1-05

      Data centers have the highest energy usage intensity among all building types and consume increasingly more energy due to the significant increase in the number of data centers worldwide. In an effort to reduce energy consumption, an increasing number of data centers have adopted air-side economizers to enable free-cooling. However, gaseous and particulate contaminants can enter data center when using air-side economizers, which may cause environment-related IT and Datacom equipment failures. Due to lack of data on the severity of the corrosion effects under realistic concentration levels and thermal environmental condition, gaseous contamination limits for the reliable operation of electronic equipment cannot be specified presently in terms of the concentration of gaseous contaminants in the air. ITEs can operate in a wide range of thermal environmental conditions including relative humidity levels up to 80% according to the current standards (ASHRAE, 2012c). There is clearly a need to determine the allowable gaseous concentration limits for the data center environment, especially under higher relative humidity conditions. According to review the literature, it need to summarize the gaseous concentration range and define the “worst realistic ” pollutant level in data center. The synergistic effects for NO2, SO2, O3, Cl2 and H2S is another challenge in order to get pollutant limits in data center.

      Intrusion Detection for Cyber-Physical Attacks  in CyberManufacturing System (CMS)
      Presented by Mingtao Wu
      Advisor: Young Moon
      Poster number MAE1-06

      Security is a problem needs to be addressed for the future manufacturing visions such as CyberManufacturing System, Industry 4.0 and Cloud Manufacturing. In those visions, the integration of network and computational resources introduces threats such as Stuxnet, that can intrude via cyber and cause physical damage. We call it cyber-physical attacks. In manufacturing, unplanned downtime can cost as much as $20,000 potential profit loss per minute and $2 million for a single incident. My doctoral work is the first step to address the cyber-physical attack problem: detect the intrusion of cyber-physical attacks in CyberManufacturing System. The objective is to detect the cyber-physical intrusion in the early stage of the manufacturing flow with low false alarm rate. This work can contribute to manufacturing enterprises that highly integrated with the Internet,  and lack of ability to identify cyber-physical attacks; It can also provide security guidelines for the development of future manufacturing visions. The methodology incorporates techniques including unsupervised machine learning, alert correlation, physical data feature extraction in manufacturing processes, the taxonomy of cross-domain and cyber-physical attack, network and host-based intrusion detection systems, and quality control engineering. This research develops along with a cyber-physical testbed simulates additive manufacturing, subtractive manufacturing, industrial robot and other representative processes.

      PIV Results of Delta Wing at Low/High Reynolds Number
      Presented by Han Tu
      Advisor: Melissa Green
      Poster number MAE1-07

      Triangular delta wing planforms have found important applications in current and proposed unmanned combat air vehicles (UCAVs). A proper understanding of fluid dynamics around UCAVs would provide critical insight for effective flow-control of combat tactical maneuvers. Force measurement and time-averaged three-dimensional flow visualization of the low Reynolds number baseline cases have been carried out on a steady delta wing. Flow visualization, achieved by stereoscopic particle image velocimetry, can reconstruct the flow field around delta wing and helps us to discover the relation of flow structure and force production. Low Reynolds number data is compared with high Reynolds number data from collaborators.  Current result shows an interesting range of angle of attack where the flow transitions to fully stalled, and as a result, the delta wing loses lift. Further detailed analysis will be applied to have further understanding of this transition.

      Research on High-Density Server packaging improving fan efficiency
      Presented by Tong Lin
      Advisor: Thong Dang
      Poster number MAE 1-08

      The demand of computational power keeps increase. As a result tightly packed electronic components in High-Density Servers (F.1) are common. In this case the available flow areas for air cooling are reduced (i.e. higher flow resistance) while airflow rate demands go up, and hence fans are required to generate more pressure rise and higher flow rates. More importantly, the flow-interaction between the fan and the surrounding components has to be taken into account. Therefore, choosing proper fans is the key to optimize efficiency. In current practice, computer server thermal designers pick fans to match with the flow resistance by assuming that the performance of the fan is not affected the flow-interaction between the fan and the surrounding components , i.e. they assume that the manufacturer fan performance curves obtained from fan alone testing are correct. However, it is often the case that the selected fans will not provide the correct pressure rise and flow rate, and time-consuming design iterations are required to satisfy the thermal constraints. Moreover, the fans will not operate at the optimum point, resulting in high energy consumption and high noise level. The proper fans have to operate efficiently and quietly in the presence of highly distorted flows in both radial and circumferential directions. The goal of this research is to better understand the flow interaction.

      Sustainability Benefits Analysis of CyberManufacturing Systems
      Presented by Zhengyi Song
      Advisor: Young Moon
      Poster number MAE 1-09

      CyberManufacturing System (CMS) is emerging as a new promising manufacturing paradigm as well as an integrated managerial approach, aiming to the shaping of an on-demand, data-driven, highly-collaborative, knowledge-intensive and sustainability-oriented manufacturing platform. The recent developments in the Internet of Things, Cloud Computing, Service-Oriented Technologies, etc., all contribute to the development of CMS. Under the umbrella of CMS, each manufacturer will be able to package their resources/capability and know-hows into services, and allowing them to be conveniently shared through pay-per-use pricing strategy. In industrial production field, sustainable manufacturing has been a widely-recognized strategy for handling with the environmental degradation and natural resource depletion issues. CMS owns advanced sustainability-bearing features (e.g., resource sharing, servitization and self-manage capabilities) and, therefore, has great potentials in address sustainability issues. This research presents a comprehensive framework of CMS paradigm and performance analysis of CMS from the perspective of sustainability. A CMS architecture is proposed to elaborate the constitution of CMS, and CMS functions are accordingly identified. Finally, a case study is used to illustrate to (i) show how an initial manufacturing request is processed and solved by production services and CMS functions and to (ii) evaluate the sustainability advancements of the proposed framework.

      Topology optimization of turning vanes using potential flow analysis
      Presented by Jack Rossetti
      Advisor: John F. Dannenhoffer III
      Poster number MAE 1-10

      In current engineering fluid flow systems, space is restricted and turning a fluid while trying to minimize pressure loss and maintain flow uniformity can be a challenge. Attempts to solve this problem have mainly been through the use of a trial and error methodology, both experimentally and numerically, which invokes some experiential knowledge or intuition. This design involves finding the shape of the vanes as well as their number and distribution. Topology optimization (TO) presents a general design optimization approach that can produce non-conventional designs. Current designs obtained using TO are dependent on initial conditions. Many cases need to be tested to ensure that a true optimum has been found. Unfortunately, many fluid flow calculations can be prohibitively expensive when using a high-fidelity model. The work presented here investigates using potential flow analysis as the low-fidelity model for initial topology optimization of the fluid turning devices. The source panel method is used to model the flow boundaries and point vortices are used to represent the turning vanes. Each vortex is placed in an optimal location that maximizes flow uniformity at the exit of the domain. An initial design topology will then be defined using the vortex locations. The goal is to use this initial topology in a high-fidelity viscous flow model for shape optimization, to obtain the optimal design and layout of turning vanes in moderate to high Reynolds number flow.

      Wrinkle Patterns on Torus
      Presented by Xiaoxiao Zhang
      Advisor: Teng Zhang
      Poster number MAE1-11

      We investigate the wrinkle patterns in a triple-layered torus, where the thin outer layer is under expansion to drive the formation and the evolution of wrinkles and the inner core has a tunable modulus to adjust the confinement of global expansion of the torus. We show from large-scale finite element simulations that hexagonal patterns will form at strong confinement (i.e. a stiff core) and stripe wrinkles will develop at weak confinement (i.e. a soft core). Hexagons and stripes can co-exist to form hybrid patterns at an intermediate confinement. As the outer layer further expands, stripe and hexagon patterns will evolve into zigzag and segment labyrinth, respectively. In addition, we observe stripe wrinkles tend to initiate from the inner surface of the torus while hexagonal wrinkles start from the outer surface. We further quantitatively analyze the topological defect distribution for a representative hexagonal pattern.

      Stability of Healthy and Diseased Arterial Tissues
      Presented by Xinyu Zhang
      Advisor: Alan Levy
      Poster number MAE1-12

      Aneurysm is the leading cause of death in the United States. Aneurysm can be viewed as initiating from a nonlinear localized instability of the arterial wall which occurs when local imperfections in vessel geometry and tissue properties grow abnormally due to the presence of pathologies in the tissue constituents. Stability of healthy and diseased arterial tissue, subject to initial geometrical and/or material imperfections, is investigated based on the long wavelength approximation. The study employs existing constitutive models for elastin, collagen and vascular smooth muscle which comprise the medial layer of large elastic (conducting) arteries. A composite constitutive model, is presented based on the concept of the musculoelastic fascicle which is taken to be the essential building block of medial arterial tissue. Nonlinear equations governing the evolution of the principal stretch imperfection growth quantities are obtained and solved numerically. Results reveal a complexity of behaviors depending on the constitutive relation, the kind of imperfection (e.g., geometrical or material) and the nominal loads. Because the character of incipient imperfection growth in diseased arteries is a precursor to aneurysm growth and development, some comments are provided as to how such a process might evolve.

    • Research Pitch Presentations

      Rapid, scalable production of anti-wetting coatings
      Presented by Saeid Biria. Advisor: Ian Hosein

      Extensive effort has focused on developing methods to fabricate these surfaces, particularly with tailored porosity. Such porous structures provide the necessary microscopic surface porosity and roughness to induce hydrophobicity, which can be enhanced through additional coating or surface functionalization. While significant progress has been achieved with such methods, all are challenged by the inherent trade-off between precise control over structure and scalability. Hence, such a synthetic approach is highly desirable to tune the structure and functionality for large-scale applications. Here, I present a new approach to synthesize microporous surfaces through the combination of photopolymerization- induced phase separation (PIPS) and light pattern formation in photopolymer−solvent mixtures. A congruently arranged microporous structure is attained. The microporous surface structure can be varied by changing the irradiating light profile via photomask design. All surfaces become superhydrophobic when spray-coated with a thin layer of polytetrafluoroethylene nanoparticles. Herein, it was found that my work led to scalable, precise, periodic microporous patterned surfaces by combining PIPS with the transmission of an incoherent light source under nonlinear optical conditions. This approach can fabricate microporous as a critical aspect of material surface design for applications as functional surfaces, water collection, antifouling, self-healing, and regenerative medicine.

      Numerical Analysis of The Imperfection-Sensitivity of Thin Shells
      Presented by Junbo Chen. Advisor: Teng Zhang

      High-performance aerospace shell structures are inherently thin walled because of weight and performance considerations. The buckling loads of these thin shells are especially sensitive to geometric, loading and material imperfections such that the experimentally measured buckling loads are substantially lower than the corresponding simplified analytical predictions of perfect shells. Thus, design factors determined from the lower bounds of experimental data, such as those recommended by NASA, are currently used to knockdown the un-conservative analytical predictions. However, these experiment-based design factors are typically over conservative and lacking information to quantify robustness, resulting in overweight structures with unknown reliability.

      The demand for analysis-based design factors has driven recent research interests in quantifying the robustness of thin shells in terms of the energy barrier between buckled and unbuckled states. A bottle neck of fully harnessing the potential of this idea is the lack of numerical method to compute the energy barrier in continuum bodies. The Nudged Elastic Band (NEB) method is often used in the study of atomic defect motions or chemical reactions to find the saddle points and minimum energy paths between two stable states. However, whether it can be applied to continuum bodies remains unknown. Our future work will include integrating the NEB method and the triangular lattice model for thin shells. We hope our work will lead to a new analysis-based shell buckling design factor and therefore save the weight, cycle time and cost in aerospace structural designs.

      Cache-Miss Obliviousness Data Shuffling on Hardware Enclaves
      Presented by Ju Chen. Advisor: Yuzhe Tang

      Description coming soon

      A generalized model for predicting the chiral response of plasmonic media
      Presented by Matthew Davis. Advisor: Jay Lee

      Description coming soon

      A Project-level Infrastructure Management Framework for Sustainable Roadways
      Presented by Song He. Advisor: Baris Salman

      Currently, maintenance, repair, and rehabilitation (MRR) projects for roadway infrastructure are mostly undertaken by traditional techniques, resulting in high overall life cycle impacts. Although non-traditional MRR techniques including accelerated methods can reduce the overall life cycle impacts, there is a lack of frameworks that can facilitate the project-level decision-making and justify the use of non-traditional techniques.

      The goal of this research is to develop a project-level roadway infrastructure management framework to consider multiple factors in decision-making and to analyze the life cycle economic, social, and environmental impacts of project alternatives involving traditional and non-traditional MRR techniques.

      The proposed framework features decision flowcharts and a multi-criteria decision-making model using analytical hierarchy process and analytical network process to shortlist alternatives that meet the project requirements to support preliminary decision-making. Then, life cycle assessment (LCA) and life cycle cost analysis (LCCA) is performed through the LCA-LCCA model to quantify life cycle economic, social, and environmental impacts of candidate project alternatives following the triple bottom line of sustainability. The LCA-LCCA model is also capable of performing what-if analysis by adjusting variables so that public agencies can apply their own data and make decisions based on their sustainability goals, objectives, and performance measures.

      Can we ensure prescription safety for expecting mothers
      Presented by Plansky Hoang. Advisor: Zhen Ma

      Description coming soon

      How Network Structure Manipulates Shortest Paths
      Presented by Pegah Hozhabrierdi. Advisor: Reza Zafarani

      Our objective is to:

      1. study how network structure in community and node levels affect the average shortest path
      2. formulate the findings to improve the network connectivity by introducing minimum

      connections. We used real-world networks with over a million connections to study the effective network structure measures.

      Then, we used these measures as the parameters in an optimization problem to improve the average shortest path with minimum connections added. This idea can be used in variety of fields that deal with networks and their connectivity. In social networks, they can be used in fake news/rumor propagation studies and recommendation systems. In power grids, it can help to design robust grids in times of power outage and in gene regulatory networks, it will be useful in determining a topology for a given transcriptional-regulatory network. The future direction of this study is to introduce an algorithm package that gives the optimal network design based on user’s need.

      Identification of Cascading Failures in Infrastructure Networks
      Presented by Griffin Michael Kearney. Advisor: Markan Fardad

      Description coming soon

      How can we use computers to discover new antibiotics
      Presented by Huilin Ma. Advisor: Shikha Nangia

      Antibiotics was one of the greatest discoveries of the last century, which is being used widely to treat and prevent bacterial infections. However, the overuse of antibiotics has led to a big problem called antibiotic resistance. The World Health Organization (WHO) describes antibiotics resistance as a serious threat to anyone, of any age, in any country. However, the misuse of antibiotics is still accelerating the process to make many common infections harder to treat and lead to longer hospital stays, higher medical costs and increased mortality. Development of antibiotics is not profitable due to its short-term use and high cost of trials. We have developed a computation platform that can predict the energy barrier of small molecules transport through bacteria outer membrane. This provides us the information of the key features that the molecules need to kill the bacteria. We have used it to test the transport the carbepenem, a class of β-lactam antibiotics used to treat P.aeruginosa, a multi-drug resistant pathogen. It perfectly reflects the free energy profiles of transport. These important information would help people design new antibiotics against P.aeruginosa. Our platform can be easily extended to most bacteria and small molecules that needed to test. A website server will be launched later, by then, people from all of the world can utilize this tool to speed up the discovery of new antibiotics and lower the total cost significantly.

      Analysis and Behavior of Segmented Energy Absorbing Steel Plate Shear Walls (SEA-SPSW)
      Presented by Nafiseh Shahbazi Majd. Advisor: Eric M. Lui

      Description coming soon

      Occurrence, Fate and Composition of N-Nitrosamines in Wastewater
      Presented by Changcheng Pu. Advisor: Teng Zeng

      Description coming soon

      Topology Optimization for Aerodynamic Applications
      Presented by Jack Rossetti. Advisor: John Dannenhoffer

      In current engineering fluid flow systems, space is restricted and turning a fluid while trying to minimize pressure loss and maintain flow uniformity can be a challenge. Attempts to solve this problem have mainly been through the use of a trial and error methodology, both experimentally and numerically, which invokes some experiential knowledge or intuition. This design involves finding the shape of the vanes as well as their number and distribution.

      Topology optimization (TO) presents a general design optimization approach that can produce non-conventional designs. Current designs obtained using TO are dependent on initial conditions. Many cases need to be tested to ensure that a true optimum has been found. Unfortunately, many fluid flow calculations can be prohibitively expensive when using a high-fidelity model.

      The work presented here investigates using potential flow analysis as the low-fidelity model for initial topology optimization of the fluid turning devices. The source panel method is used to model the flow boundaries and point vortices are used to represent the turning vanes. Each vortex is placed in an optimal location that maximizes flow uniformity at the exit of the domain. An initial design topology will then be defined using the vortex locations. The goal is to use this initial topology in a high-fidelity viscous flow model for shape optimization, to obtain the optimal design and layout of turning vanes in moderate to high Reynolds number flow.

      A scalable 3D printed model for bone tissue engineering
      Presented by Stephen Sawyer. Advisor: Pranav Soman

      Bone is the second most transplanted tissue in the world, used in approximately 2.2 million bone graft procedures annually.  In the United States alone, more than 500,000 bone grafts have been performed, with a forecasted economic impact of $6.6 billion by 2020. Unfortunately, due to the complexities associated with creating a bone tissue substitute which include a hierarchical inorganic-organic composite containing embedded blood vessels, current tissue engineering based solutions have not generated outcomes that are significantly improved over modern orthopedic reconstruction procedures.

      In this project, we design and develop a bone tissue model using commercial 3D printers. Our group has used a MakerBot 3D printer to print hollow channels within a hydrogel biomaterial laden with bone-forming cells. This process enhances the mass transport of nutrients to encapsulated cells and allows deposition of bone mineral around printed channels. The scalability of this method can be potentially used to generate human-scale bone tissue substitutes using bone marrow-derived stem cells. Additionally, this model can be modified to contain different cellular, structural, and vascular components in order to identify and study niche bone environments in normal bone as well as pathologies such as cancer and osteoporosis. This work was highlighted on a PBS program, ‘SciTech Now’ and on ‘HealthLink in Air’, an award-winning talk show.

      Distributed Self-localization and Tracking with an unknown number of targets
      Presented by Pranay Sharma. Advisor: Pramod Varshney

      Description coming soon

      Rising Heatwave Trends:  A Case Study in Ten Communities across the USA
      Presented by Javad Shafiei Shiva. Advisor: David G Chandler

      Heatwaves are an important type of extreme climate events and result in more than 130 deaths per year across the US. Heatwaves have been described by several attributes, combinations of which constitute various event typologies. We studied long-term heatwaves in ten cities during 1950-2016 to better understand how these attributes determine variability in local heatwaves and how climate change is affecting heat waves across the USA.

      Our results indicated that at least five harmful attributes of heatwave have increased simultaneously in Dallas, Miami, New York, Phoenix, and Portland. In addition, we found the largest change in heatwave season length, frequency, and timing occurred in Miami from the 1950s to 2010s.  Meanwhile, despite the significantly high values of heatwave attributes in warm climates, Bismarck, ND and Syracuse, NY have the greatest mean heatwave intensity during 1950s to 2010s. Similar results across much of the study sites domain are presented to clarify the many differences in quantitative heatwave attributes and variance in approaches across climates.

      This work explores the nexus of quantitative description and social construction of heatwaves through the lens of the various regional metrics to describe heatwaves. Ultimately, this understanding will lead to assessment of various strategies to help communities understand and prepare for heat resilience based on local heat waves components.

      Is the future of CyberManufacturing System at risk
      Presented by Mingtao Wu. Advisor: Young Moon

      Description coming soon

      Real-time Obstacle Free Trajectory Planning for Next Generation Autonomous UAV
      Presented by Hang Yin. Advisor: Utpal Roy

      Autonomous unmanned aerial vehicles (UAVs) are unmanned aircrafts that are flown autonomously, without human control. Research on UAV has become an important because of their significant contribution to both military and civil applications, such as infrastructure inspection, field surveying, scientific data collection and security monitoring. Furthermore, they are suited for the delivery of goods, mapping and precision agriculture. One of the current research topics is essentially an optimization problem that involves multiple areas of flight dynamics, control, navigation, and information science. A key point in the future advancement of autonomous UAVs is the development of obstacle avoidance technology. However, the obstacle avoidance function of Sense and Avoid capability is relatively in infancy. Although obstacle avoidance has advanced in the past years, it must be improved if autonomous UAV are to become a widespread reality and used in challenging environments.

      To improve autonomous control level of the next generation UAV, there is a demand of developing a novel method which could handle environment uncertainty, static, dynamic, solid and soft obstacles while generating an optimized trajectory. The ongoing research will yield a new method which can be implemented in an onboard portable computer and run in real-time. The proposed method consists of two primary objectives: 1) uncertainty management; 2) real-time/near real-time obstacle free trajectory generation.

    • Winners and Poster Listing
      Winners of the Poster and Research Pitch Competitions
      • Biomedical and Chemical Engineering
        • Development of a “2-D” Test System for Visualizing Fretting Corrosion: A Study of the Fretting Corrosion Behavior of CoCrMo Alloy, presented by Dongkai Zhu. Advisor: Jeremy Gilbert
        • Femtosecond Laser Processing of Gelatin Methacrylate Hydrogel, presented by Zheng Xiong. Advisor: Pranav Soman
        • Molecular Transport through Blood-Brain Barrier Pores, presented by Flaviyan Jerome Irudayanathan. Advisor: Shikha Nangia
        • Modeling Diversity in Structures of Bacterial Outer Membrane Lipids, presented by Huilin Ma. Advisor: Shikha Nangia
        • Controlling Streptococcus mutans and Staphylococcus aureus Biofilms with Direct Current and Chlorhexidine, presented by Hao Wang. Advisor: Dacheng Ren
        • Fabrication of Cell-laden Hydrogel Microspheres for Bone Regeneration, presented by Sanika Suvarnapathaki. Advisor: Pranav Soman
        • Ketone Oxidative Cleavage Over Vanadium Oxides Supported on Metal Oxides, presented by Ran Zhu. Advisor: Jesse Quentin Bond
        • Perfusion directed 3D bone mineral formation, presented by Kairui Zhang. Advisor: Pranav Soman
        • Thermodynamic and kinetic analysis of y-valerolactone ring opening in multiphase reactors, presented by Xinlei Huang. Advisor: Jesse Quentin Bond
        • Continuous synthesis of biodiesel fuel under sub/supercritical conditions, presented by Jiuxu Liu. Advisor: Lawrence Tavlarides
        • Electrospun Polyurethane Blends Exhibiting Shape Memory and Self-Healing Properties, presented by Wenbin Kuang. Advisor: Patrick Mather
        • A Novel Approach to Isolation and Characterization of Viable but Nonculturable Bacterial Cells, presented by Nicholas Kelley. Advisor: Dacheng Ren
        • Nanocomposite Architectures for Batteries via a Novel Photopolymerization Route, presented by Shreyas Pathreeker. Advisor:  Ian Hosein
        • Effect of palmitoylation in membrane proteins at the blood-brain barrier interface, presented by Nandhini Rajagopal. Advisor: Shikha Nangia
        • Removal of Radioactive Iodine and Tritiated Water from Spent Nuclear Fuel Reprocessing Off-gas Streams by Silver Containing Adsorbents, presented by Yue Nan. Advisor: Lawrence Tavlarides
        • Sensitizing Bacterial Cells to Antibiotics Through Dynamic Topography-Triggered Biofilm Detachment, presented by Sang Lee. Advisor: Dacheng Ren
        • Microcontact Printing on Shape Memory Polymers for Cell Culture Applications, presented by Fred Donelson. Advisor: James Henderson
      • Civil and Environmental Engineering
        • Suspect Screening of Emerging Contaminants in Onondaga Lake and Its Tributaries Using Orbitrap High-Resolution Mass Spectrometry, presented by Shiru Wang. Advisor: Teng Zeng
        • Exploring the Fate of N-nitrosamines in Municipal Wastewater Treatment Plants, presented by Changcheng Pu. Advisor: Teng Zeng
        • A Cost Effective Method to Retrofit Steel Girders, presented by Omar Youssef El Masri. Advisor: Eric Lui
        • A Survey and Decision Support System for Innovative Maintenance, Repair, and Reconstruction Techniques for Asphalt Roadways, presented by Song He. Advisor: Sam Salem
        • Application of an Integrated Biogeochemical Model to Different Timber-harvesting Techniques, presented by Mahnaz Valipour. Advisor: Charles Driscoll, Chris Johnson
        • Effect of Degraded Fluid Viscous Damper in Seismic Response of Structures, presented by Kamiar Kalbasi-Anaraki. Advisor: Hossein Ataei
      • Electrical Engineering and Computer Science
        • Classification of Affect using Deep Learning on Brain Blood Flow Data, presented by Danushka Bandara. Advisor: Senem Velipasalar
        • Intercell Interference-Aware Scheduling for Delay Sensitive Applications in C-RAN, presented by Yi Li . Advisor: Senem Velipasalar
        • Solar Generation Forecasting Based on Different Data Sizes, presented by Guangyuan Shi. Advisor: Sara Efteharnejad
        • Will A Group Split? Measure the Divisibility and Uniformity, presented by Shuchang Liu.
        • Capacity Characterization for State-Dependent Gaussian Channel with a Helper, presented by Yunhao Sun. Advisor: Yingbin Liang
        • A Low-Computation-Complexity, Energy-Efficient, and High-Performance Linear Program Solver Using Memristor Crossbars, presented by Ruizhe Cai. Advisor: Yanzhi Wang
        • Dual Polarized Directivity Enhanced Active Metameaterial Antenna, presented by Dongyin Ren. Advisor: Jun Choi
        • On Classification of Environmental Acoustic Data Using Crowds, presented by Shan Zhang. Advisor: Pramod Varshney
        • Milimeter Wave Communications over Heterogeneous Cellular Networks, presented by Esma Turgut. Advisor: Mustafa Cenk Gursoy
        • Selective Parts-Based Tracking Using Local and Global Correlation Filtering, presented by Maria Lynn Scalzo. Advisor: Senem Velipasalar
        • Predicting the Starting Distance of the Far Field, presented by Mohammad Najib Abdallah. Advisor: Tapan Kumar Sarkar
        • Optimal Resource Allocation for Full-Duplex Wireless Video Transmission under Delay Constraints, presented by Chuang Ye. Advisor: Senem Velipasalar, Mustafa Cenk Gursoy
        • Mobile Neurosurgery Imaging and Surgical Navigation System, presented by Yu Zheng. Advisor: Senem Velipasalar
        • A Graph Theory-Based Model for Power Grid Security Assessment, presented by Mirjavad Hashemi Gavgani. Advisor: Sara Eftekharnejad
        • Heat Leakage Detection from Thermal Images for Autonomous Aerial Building Inspection, presented by Burak Kakillioglu. Advisor: Senem Velipasalar
        • Prediction of Biological Functions by Histone Modification Patterns Profiling, presented by Yiou Xiao. Advisor: Kishan Mehrotra, Chilukuri Mohan
        • Fast and Energy-Aware Resource Provisioning and Task Scheduling for Cloud Systems, presented by Hongjia Li. Advisor: Yanzhi Wang
        • Critical PMU identification and Optimal Redundant PMU placement, presented by Rui Ma. Advisor: Sara Eftekharnejad
        • Influence of the Probe when Computing Far Field from Near Field Measurements, presented by Heng Chen. Advisor: Tapan Kumar Sarkar
        • SC-DNN: Highly-Scalable Deep Convolutional Neural Network using Stochastic Computing, presented by Ao Ren. Advisor: Yanzhi Wang
        • Improving Emotion Classification of Functional Near-Infrared Spectroscopy Data, presented by Natalie Sommer. Advisor: Senem Velipasalar
        • Reconfigurable Thermoelectric Generators for Vehicle Radiators Energy Harvesting, presented by Caiwen Ding. Advisor: Yanzhi Wang
        • An Area and Energy Efficient Design of Domain-Wall Memory-Based Deep Convolutional Neural Networks using Stochastic Computing, presented by Xiaolong Ma. Advisor: Yanzhi Wang
        • Predicted Max Degree Sampling: Sampling in Directed Networks to Maximize Node Coverage, presented by Ricky Laishram. Advisor: Sucheta Soundarajan
        • Max-Node Sampling: an Expansion Densification Algorithm for Data Collection, presented by Katchaguy Areekijseree. Advisor: Sucheta Soundarajan
        • Emotional Patterns in Social Media: From Users to Communities, presented by Shengmin Jin. Advisor: Reza Zafarani
        • Seeing Red: Locating People of Interest in Networks, presented by Pivithuru Wijegunawardana. Advisor: Sucheta Soundarajan
        • A Hierarchical Framework of Cloud Resource Allocation and Power Management Using Deep Reinforcement Learning, presented by Ning Liu. Advisor: Yanzhi Wang
        • LPAD – Authenticated Data Storage with Minimal Trust, presented by Ju Chen. Advisor: Yuzhe Tang
        • Robust Footstep Counting and Traveled Distance Calculation by Mobile Phones Incorporating Camera Geometry, presented by Yantao Lu. Advisor: Senem Velipasalar
        • Deep Reinforcement Learning Content Caching for Base station, presented by Chen Zhong. Advisors: Mustafa Cenk Gursoy, Senem Velipasalar
        • A Unified Diversity Measure for Distributed Inference, presented by Prashant Khanduri. Advisor: Pramod Varshney
        • Inferring Communication Network Topology via Transfer Entropy, presented by Pranay Sharma. Advisor: Pramod Varshney
        • Fast Anomaly Detection using a Forest for High-dimensional Streaming Data, presented by Zhiruo Zhao. Advisors: Kishan Mehrotra, Chilukuri Mohan
        • Crawling Community Structure in Dynamic Online Social Networks, presented by Humphrey Mensah. Advisor: Sucheta Soundarajan, Chilukuri Mohan
        • Metaprogramming to Defeat Slide-Channel Attacks, presented by Scott Constable. Advisor: Steve Chapin
        • Layered Decoding and Secrecy Over Degraded Broadcast Channel, presented by Shaofeng Zou. Advisor: Yingbin Liang
        • Development of Dual-Band/duel-Polarized Metematerial Antenna in an Electrically-Small Form Factor for Efficient Wireless Data and Power Transfer, presented by Komlan Payne. Advisor: Jun Choi
        • Linear Programming Approach to the Optimal Design of Cascading Failures, presented by Griffin Kearney. Advisor: Makan Fardad
        • Influential Node Detection in Implicit Social Networks using Gaussian Copula Models, presented by Qunwei Li. Advisor: Pramod Varshney
        • Catch Me If You Can!, presented by Mahmuda Rahman. Advisor: Jae Oh
        • Evolving the Electric Utility Distribution Substation into a Microgrid, presented by William Maxwell. Advisor: Tomislav Bujanovic
        • PNAS: Privacy-preserving Network Analysis based on Intel SGX, presented by Chenghong Wang. Advisor: Sucheta Soundarajan
        • Your Smartphone Security is at Risk!, presented by Diksha Shukla. Advisor: Vir Phoha
        • A password free world!, presented by Rajesh Kumar. Advisor: Vir Phoha
        • Bayesian Compressive Detection with Laplacian prior, presented by Swatantra Kafle. Advisor: Pramod Varshney
      • Mechanical and Aerospace Engineering
        • Integrated Guidance and Control through Given Waypoints for Unmanned Aerial Vehicles with Four Control Inputs, presented by Reza Hamrah. Advisor: Amit Sanyal
        • A PLM-Based Data Analytics Approach for Improving Product Development Lead Time in an Engineer-to-order Manufacturing Firm, presented by Kai Sun. Advisor: Utpal Roy
        • Intrusion Detection in Cybermanufacturing System, presented by Mingtao Wu. Advisor: Young Moon
        • Near Field Propagation Pathways and Acoustic Events of a Two Stream Supersonic Rectangular Jet, presented by Pinqing Kan. Advisor: Jacques Lewalle
        • Fan System Modeling via Body Force Approach, presented by Yinbo Mao. Advisor: Thong Dang
        • Operation Study and Performance Analysis of CyberManufacturing Systems, presented by Zhengyi Song. Advisor: Young Moon
        • Automatic Generation of Near-Body Structured Grids, presented by Yuyan Hao. Advisor: John Francis Dannenhoffer
        • Efficient Personal Cooling and Heating Terminal Design, presented by Meng Kong. Advisor: Ezzat Khalifa
        • An Ontology based Information Framework for Smart Manufacturing Systems Interoperability, presented by Heng Zhang. Advisor: Utpal Roy
        • A Smart-Product Lifecycle Management (sPLM) Architecture for Unmanned Aircraft Systems (UAS), presented by Yunpeng Li. Advisor: Utpal Roy
        • Modified Artificial Potential Field for UAV Formation Generation and Changing, presented by Hang Yin. Advisor: Utpal Roy
        • Research on High-Density Server packaging improving fan efficiency, presented by Tong Lin. Advisor: Thong Dang
        • Design & Characterization of Biomimetic Tribologically Enhanced Hydrogels, presented by Allen Osaheni. Advisor: Michelle Blum
        • Unmanned Air Systems Performance Benchmarking, presented by Joe Weiner. Advisor: Mark Glauser
        • Biaxial Stability of Medial Arterial Tissues, presented by Xinyu Zhang. Advisor: Alan Levy
        • An Exact Analysis of Mode III Cohesive Fracture, presented by Yueming Song. Advisor: Alan Levy
        • Applying Topology Optimization to Fluid Dynamics Problems, presented by Jack Rossetti. Advisor: John Francis Dannenhoffer, Melissa Green
        • Latent Heat Storage Device, presented by Riley Gourde. Advisor: Ezzat Khalifa
        • CFD Analysis of Distributed Propulsion Systems for Vertical Takeoff and Landing, presented by Andrew Welles. Advisor: Thong Dang
        • An sPLM Supported UAVs System with Distributed Flight Control, presented by Heng Zhang. Advisor: Utpal Roy
      • Research Pitches
        • Northeastern US – Synergy of Climate and Land, presenter: Rouzbeh Berton (CIE). Advisor: Charles Driscoll
        • Trajectory Generation on SE(3) for Underactuated Vehicles, presenter: Mani Dhullipalla (MAE). Advisor: Amit Sanyal
        • Reconfigurable thermoelectric generators for vehicle energy harvesting, presenter: Caiwen Ding (EECS). Advisor: Yanzhi wang
        • Succinic Acid Hydrodeoxygenation Over Noble Metal Catalysts, presenter: Joshua Gopeesingh (BMCE). Advisor: Jesse Bond
        • Can we find a cure for Alzheimer’s Disease?!, presenter: Flavian Jerome Irudayanathan (BMCE). Advisor: Shikha Nangia
        • Emotional Patterns in social media: from users to communities, presenter: Shengmin Jin (EECS). Advisor: Reza Zafarani
        • Linear programming approach to the optimal design of cascading failures, presenter: Griffin Kearney (EECS). Advisor: Makan Fardad
        • A password free world: your behavioral traits are your signature, presenter: Rajesh Kumar (EECS). Advisor: Vir Phoha
        • A Smart-Product Lifecycle Management (sPLM) Architecture for Unmanned Aircraft Systems (UAS), presenter: Yunpeng Li (MAE). Advisor: Utpal Roy
        • A Single-Surgery Approach For Accelerated Healing of Bone Defects Using a Radial Defect Model in Rabbits, presenter: Michelle Pede (BMCE). Advisor:  James Henderson
        • Oxygen transport membrane for oxy-fuel combustion and carbon capture purpose, presenter: Matt Rushby (MAE). Advisor: Jeongmin Ahn
        • Inferring communication network topology via transfer entropy, presenter: Pranay Sharma (EECS). Advisor: Pramod Varshney
        • Application of an Integrated Biogeochemical Model, PnET-BGC to Different Timber-harvesting Techniques, presenter: Mahnaz Valipour (CIE). Advisor: Charles Driscoll
        • Seeing red: Locating people of interest in networks, presenter: Pivithuru Wijegunawardana (EECS). Advisor: Sucheta Soundarajan
        • Prediction of biological functions by histone modification patterns profiling, presenter: Yiou Xiao (EECS), Advisor: Kishan Mehrotra

    Graduate Student Funding

  • Graduate Student Funding
  • PhD students in the College of Engineering and Computer Sciences may be supported through assistantships or fellowships. Graduate assistants (teaching or research) assist in teaching courses or conduct research in support of their thesis or dissertation. The links below provide further information about each type of support.

    Financial support for MS students is given on a competitive basis and is typically in the form of a tuition discount.