Nunan Research Day: Cyber, Wireless and Big Data

April 15, 2015

On April 6th Syracuse University's College of Engineering and Computer Science celebrated Nunan Research & Lecture Day. The day featured a record number of research posters being presented by graduate student candidates from each of the College’s four departments. ``The significant increase in the number of posters, from about 30 in 2008 to 130 in 2015, clearly points to the extensive growth in our Ph.D. programs within the College and is a testimony to our outstanding faculty, the numbers of which have expanded by more than 30 new faculty hires in this time period," said Mark Glauser, Associate Dean for Research and Graduate Programs for the College of Engineering and Computer Science. The theme of this year’s Nunan Day was Cyber, Wireless and Big Data. A number of faculty working in these research areas presented short talks about the importance and impact of this theme. This year, the College partnered with the Center for Advanced Technology in Computer Applications and Software Engineering (CASE) to present this event. Additionally, the event featured a talk by H. Vincent Poor, Dean of the School of Engineering and Applied Science at Princeton University. His talk was entitled, “Fundamental Limits on Information Security and Privacy.” Nunan Research Day was established in 2006 with a generous gift from the estate of Syracuse University alumnus James D. Nunan and his wife, Marge.

The College presented a number of best poster awards to graduate students. Here are the winners (the abstracts are listed below the winner list):

Overall Winner

Leading Edge Embedded Fan Airfoil Concept – New Powered High Lift Technology Nhan Phan, 4th year Ph.D. Advisor: Thong Dang Department of Mechanical and Aerospace Engineering

Department of Biomedical and Chemical Engineering

Computational Rheology of Surfactant Micelle and Micelle- Nanoparticle Solutions: A Molecular Dynamics Study Abhinanden Sambasivam, Ph.D. Dr. Subas Dhakal Advisor: Radhakrishna Sureshkumar

Department of Civil and Environmental Engineering

Sustainable and Innovative Design Methods for Geotextile Tube Dewatering Mahmoud Khachan, Ph.D. Candidate Shobha Bhatia and Zeru Kiffle Advisor: Shobha Bhatia

Department of Electrical Engineering and Computer Science

Guided-Wave Nanophotonic Devices based on Networked Plasmonic Waveguides Ashish S Chanana, 2nd Year Masters Matthew Davis, Amit Agrawal, Jay Kyoon Lee Advisor: Jay Kyoon Lee Secure Distributed Inference in the Presence of Eavesdroppers Sid Nadendla, 6th year Ph.D. Advisor: Pramod K. Varshney Define-Use Vulnerabilities in Android Yousra Aafer, 4th year Ph.D. Xiao Zhang, Zhongwen Zhang, Nan Zhang Advisor: Wenliang Du

Department of Mechanical and Aerospace Engineering

Design & Characterization of Low Friction Zwitterionic Hydrogel for use as Articular Cartilage Replacement Allen Osaheni, 2nd year Ph.D. Patrick T. Mather, Rebecca A. Bader Advisor: Michelle M. Blum Laser Ignition Studies of Methane and Biogas Nathan Peters, 2nd year Ph.D. Henry Morrow Advisor: Benjamin Akih-Kumgeh

Practical Application Winner

Understanding and Mitigating Security Hazards in Android Applications Uninstallation Xiao Zhang, 5th Year Ph.D. Zhenshen Qiu, Kailiang Ying, Yousra Aafer Advisor: Wenliang Du

Read the Abstracts of the Winners

Overall Winner

Leading Edge Embedded Fan Airfoil Concept – New Powered High Lift Technology Nhan Phan, 4th year Ph.D. Advisor: Thong Dang Department of Mechanical and Aerospace Engineering The LEEF airfoil concept is a powered high-lift airfoil concept capable of generating very high lift coefficient at high angle-of attack (AoA). This technology is developed for Extremely Short Take-Off and Landing (ES­TOL) aircraft applications. Unlike existing powered high lift systems, the LEEF airfoil uses a local high-pressure air source from cross-flow fans, does not require ducting, and is able to be deployed using distributed electric power systems for general aviation aircrafts. In addition to lift augmentation, the LEEF airfoil can provide additional thrust during take­off and landing operation to supplement the primary cruise propulsion system. Two-dimensional and three-dimensional CFD simulations of orig­inal NACA 63-3-418 airfoil/wing and LEEF airfoil/wing were carried out to evaluate the advantages of and the cost associated with implementing the LEEF concept. An experimental study of the LEEF concept was also performed to validate the concept. Finally, the LEEF technology is applied for a Remote Control model and DARPA’s Aerial Reconfigurable Embed­ded System. Tiltrotor aircrafts have wing-mounted ducted rotors that can be tilted to enable VTOL capability while providing high-speed cruising flight. One challenge is the potential loss of aircraft control during the transition period between horizontal flight and hovering period. The LEEF technology presents a potential solution for this problem. 65

Department of Biomedical and Chemical Engineering

Computational Rheology of Surfactant Micelle and Micelle- Nanoparticle Solutions: A Molecular Dynamics Study Abhinanden Sambasivam, Ph.D. Dr. Subas Dhakal Advisor: Radhakrishna Sureshkumar Coarse grained molecular dynamics (MD) simulations [1-2] are used to study the equilibrium structure, shear-induced structure changes and rheology of cationic cylindrical micelle solutions. Systems that consist of up to a million coarse grained particles are simulated in the presence of explicit solvent and salt interactions. Addition of nanoparticles (NPs) to cationic wormlike micelles results in the formation of electrostati­cally stabilized micelle-NP junctions. The molecular mechanism and the kinetics of such junction formation will be elucidated. The effect of salt concentration and NP volume fraction on the structure and rheology of micelle-NP networks will be discussed. At sufficiently high salt concen­trations, it is observed that branching in micelles results in a decrease in zero-shear viscosity of the solution. Non-equilibrium MD simulations are performed for micelle and micelle-NP solutions, showing shear thickening due to shear-induced structure (SIS) formation and shear thinning behav­ior depending on the salt and surfactant concentration. A shear-induced isotropic to nematic transition was observed for Weissenberg number Wi O(1), where Wi is defined as the product of shear rate and the intrinsic structure relaxation time of the system. For Wi > 1, micelles align in the flow direction. Mean square displacement of NPs in such solutions sug­gests that particles could undergo transient trapping due to entanglement constraints arising from the micelle structure.
  1. A.V. Sangwai & R. Sureshkumar, Langmuir, 27, 6628 (2011)
  2. A.V. Sangwai & R. Sureshkumar, Langmuir, 28, 1127 (2012)

Department of Civil and Environmental Engineering

Sustainable and Innovative Design Methods for Geotextile Tube Dewatering Mahmoud Khachan, Ph.D. Candidate Shobha Bhatia and Zeru Kiffle Advisor: Shobha Bhatia Several industries have adopted geotextile tubes as a cost effective and successful dewatering technology over the past three decades. The dewatered materials include dredged sediments, industrial wastes, coal combustion products, and municipal wastes. In conjunction with the success and efficacy gained by using geotextile tubes, there were increased environmental concerns associated with the use of synthetic chemical flocculants to enhance to dewatering rate, in addition to con­cerns about the strength and stability of the dewatered materials. This research aims to address these concerns by introducing sustainable and practical alternatives to current dewatering practices. Ecofriendly floc­culants (starch-based) that are made at Syracuse University Geotechnical Laboratory as an alternative to the widely used synthetic flocculants are introduced in this research. The effect of green flocculants on dewatering rate and on retained sediments properties was evaluated using pressure filtration and centrifuge tests. Tests results showed that green flocculants effects are very similar to those of synthetic flocculants. In some cases, where organic sediments were used, the green flocculants were more effective than the synthetics. Furthermore, two new methods for measur­ing the concentration of the residual flocculants in the effluent that seeps through geotextile tubes were introduced. The two methods are Stream­ing Current Method, and Kaolin Settling Rate Method. Additionally, a study about possible methods to improve the stability of the dewatered sediments was conducted. Randomly distributed discrete fibers were found to be effective in increasing the shear strength of the dewatered sediments by about 80%. Finally, two-dimensional mathematical model that incorporates geotextile tube dewatering physical processes, such as sedimentation and filtration, is introduced in this research.

Department of Electrical Engineering and Computer Science

Guided-Wave Nanophotonic Devices based on Networked Plasmonic Waveguides Ashish S Chanana, 2nd Year Masters Matthew Davis, Amit Agrawal, Jay Kyoon Lee Advisor: Jay Kyoon Lee The prospect of controlling the interaction of light with matter at na­noscale has been widely studied in recent years, and entails character­izing optical and optoelectronic devices at resolution higher than the diffraction limit. One technique that allows localization of light to sub-wavelength dimensions is through the use to surface-plasmon-polaritons (SPPs) wherein the interaction of light with free-electrons on a metal surface can lead to a bound surface electromagnetic-field that is confined to deep-subwavelength dimensions. Studies based on SPPs merged with the field of nanotechnology have resulted in novel imaging technologies, nonlinear and quantum-optic devices and the ability to design materials with unusual electromagnetic properties with potential applications rang­ing from enhancing the efficiency of photovoltaic devices to detection of bio-molecules at ultra-small concentrations. Here we report the design of nanophotonic devices based on SPP wave­guide structures that would act as a true counterpart to today’s electronic devices, providing orders of increase in data-speeds while maintaining nanoscale dimensions. The devices are based on metal-dielectric-metal (MDM) waveguide structures composed of Ag/SiO2/Ag heterostructure that utilizes interference effect within multiple intersecting plasmonic-waveguides. The specific devices we have been exploring include guided-wave devices such as L and T-bends, 4-way-splitters and 2x2-net­worked structures, wherein by altering the device geometry one can tune its operating frequency, and by changing the angle of incidence one can switch these devices between ON/OFF states. We plan to fabricate and experimentally characterize these devices for applications in color rout­ing, as directional-filters and optical-switches. We will discuss preliminary design rules and constraints based on results obtained from the finite-difference-time-domain simulations. Secure Distributed Inference in the Presence of Eavesdroppers Sid Nadendla, 6th year Ph.D. Advisor: Pramod K. Varshney The distributed detection framework comprises of a group of distributed sensing units which acquire observations about a phenomenon of inter­est (PoI) and send processed data to a central node called the fusion cen­ter (FC) where a global inference is made about the presence or absence of the PoI. This framework has a wide range of applications in domains such as communication systems, stock-market forecasting, anomaly detection, psychology, human decision-making and medical instrumenta­tion. The sensitivity of these application-domains to information-leakage drives us to investigate security in the framework of distributed detection networks. In this work, we analyze and design a distributed detection network in the presence of an eavesdropper. We consider the problem of designing binary quantizers at the sensors for a distributed detection network under a secrecy constraint imposed on the eavesdropper. The performance metric chosen is the Kullback- Leibler Divergence at both the fusion center (FC) and the eavesdropper (Eve). First, we consider the problem of secure distributed detection in the presence of identical sensors and channels. We prove that the optimal quantizer can be implemented as a likelihood ratio test, whose threshold depends on the specified secrecy constraint on the Eve. We present an algorithm to find the optimal threshold in the case of Addi­tive White Gaussian Noise (AWGN) observation models at the sensors. In the numerical results, we discuss the tradeoff between the distributed detection performance and the secrecy constraint on the eavesdropper. We show how the system behavior varies as a function of the secrecy constraint imposed on Eve. Finally, we also investigate the problem of designing the binary quantizers for a distributed detection network with non-identical sensors and channels. We propose a greedy algorithm by decomposing the problem into N sequential problems, where each individual problem has the same structure as the scenario with identi­cal sensors and channels. Optimum binary quantizers are obtained and numerical results are presented for illustration. Define-Use Vulnerabilities in Android Yousra Aafer, 4th year Ph.D. Xiao Zhang, Zhongwen Zhang, Nan Zhang Advisor: Wenliang Du In this poster, we present a new class of Android vulnerabilities. The vulnerability is present when a privileged app uses resources on the device that might not be present without a strong identification mechanism. This creates attack opportunities for malware writers to squat these resources and thus achieve various damages. We develop a detection mechanism for this type of vulnerabilities that looks for our predefined rules in system apps within a specified Android image statically. Using our tool, we conduct a large scale investigation on tens of collected system images and apps from various vendors such as Google, LG, Moto, Samsung and HTC and show that this new vulnerability is actually prevailing in several of them. We also verify some of the identified vulnerabilities to demonstrate the imposed risks in real world scenarios. We were able to achieve various damages ranging from denial of service, intent redirection, and squatting system apps. Finally, we propose a lightweight protection mechanism to defend against the define-use vulnerability.

Department of Mechanical and Aerospace Engineering

Design & Characterization of Low Friction Zwitterionic Hydrogel for use as Articular Cartilage Replacement Allen Osaheni, 2nd year Ph.D. Patrick T. Mather, Rebecca A. Bader Advisor: Michelle M. Blum Treatment of focal chondral defects is a widespread and intractable problem; especially with people living longer, or accumulating damage from progressive debilitating diseases such as osteoarthritis. In response, use of hydrogels has been investigated to repair this necrotic tissue in an inexpensive, non-invasive manner. These investigations routinely focus on mimicking the biomechanics of the natural tissue or the fluid pres­surization mechanism. However, these approaches neglect the surface boundary lubrication mechanism, making these potential load-bearing substitutes incompatible with natural cartilage, resulting in construct wear, failure, and healthy cartilage damage. Therefore, there is a need to develop a material with combined mechanical and tribological properties comparable to articular cartilage. Poly (vinyl alcohol) (PVA) hydrogels are an attractive option due to their inherent biomimetic properties. How­ever, their use for weight bearing applications is limited due to inferior tribological properties. This presentation outlines the effectiveness of two fabrication approaches to enhance the tribological properties of PVA hydrogels. The first approach consists of blending ratios of PVA with zwitterionic polymer, 2-(Methacryloyloxy) ethyl]dimethyl-(3-sulfopropyl) ammonium hydroxide) (MEDSAH) to form a hydrogel blend. In the second approach poly(MEDSAH) is functionalized to the surface of PVA hydrogels, resulting in the formation of a zwitterionic brush layer. The structure-property relationships of the hydrogels were investigated by evaluating the chemical composition (ATR), physical properties (water content, contact angle), elastic compressive modulus and coefficient of friction. Preliminary results suggest the functionalization of these zwitter­ionic polymers result in as much as a 40% reduction in average coeffi­cient of friction. Laser Ignition Studies of Methane and Biogas Nathan Peters, 2nd year Ph.D. Henry Morrow Advisor: Benjamin Akih-Kumgeh Laser-induced ignition is a promising technology for combustion initia­tion in gas turbines and internal combustion engines. There is renewed interest in this technology in recent years due to its ability to ignite lean mixtures which are desirable for cleaner combustion. Apart from fossil fuels, such as methane and propane, biofuels are increasingly considered for applications in combustion systems in an attempt to improve energy sustainability. One of these bio-derived fuels is biogas. Unlike natural gas that consists mostly of methane, biogas contains a high percentage of carbon dioxide in addition to methane and other trace gases. The high carbon dioxide content can hinder flame propagation which in some cases may lead to quenching of the flame kernel near the ignition source. This clearly has further implications in the laser energy requirement for biogas ignition, which warrants investigation. In this work we study laser-induced ignition of natural gas and biogas. A Nd:YAG laser is used to induce breakdown and ignition of fuel/air mixtures in a cylindrical stainless steel vessel, equipped with 6 optical accesses. Plasma formation, flame initiation, quenching, and successful flame propagation are captured using a high speed Schlieren imaging system and laser interferometry. These imaging techniques allow for a detailed investigation of differences between the fuels as well as fully ignited and quenched cases during the early stages of flame formation. Minimum pulse and minimum ignition energies are determined for a wide range of equivalence ratios, also permitting the determination of flammability limits.

Practical Application Winner

Understanding and Mitigating Security Hazards in Android Applications Uninstallation Xiao Zhang, 5th Year Ph.D. Zhenshen Qiu, Kailiang Ying, Yousra Aafer Advisor: Wenliang Du Uninstalling an application from mobile devices is among the most com­mon user practices in smartphone’s daily usage. It may sound trival, but the entire process involves each system component coordinating to re­move data belonging to the application being uninstalled. With its univer­sality and complexity, little has been done, however, to understand the security risks of the application’s uninstallation process. In this project, we report the first security analysis of Android’s data clean up mecha­nism after application is removed, which reveals the pervasiveness of subtle yet significant security flaws in them, leading to multiple data residue instances. Each instance is connected back to one specific system service’s data mishandling, and it can be further utilized to launch DoS, privilege escalation and data hijacking attacks. To mitigate this threat, we provide guidelines that helps Android framework developers correctly clean up data from each system service after application is removed. We show that, by following our guideline