Engineering Commons^™

Fiber-Optic Temperature And Flow Sensory System And Methods, Ming Han, Guigen Liu, Weilin Hou, Qiwen Shen

Department of Electrical and Computer Engineering: Faculty Publications

A fiber optic sensor, a process for utilizing a fiber optic sensor, and a process for fabricating a fiber optic sensor are described, where a double-side-polished silicon pillar is attacked to an optical fiber tip and forms, a Fabry-Perot cavity. In an implementation, a fiber optic sensor in accordance with an examplary embodiment includes an optical fiber configured to be coupled to a light source and a spectrometer; and a single silicon layer or multiple silicon layers disposed on an end face of the optical fiber, where each of the silicon layer(s) defines a Fabry-Perot interferometer, and where the sensor …

Generating Energy Data For Machine Learning With Recurrent Generative Adversarial Networks, Mohammad Navid Fekri, Ananda M. Ghosh, Katarina Grolinger

Electrical and Computer Engineering Publications

The smart grid employs computing and communication technologies to embed intelligence into the power grid and, consequently, make the grid more efficient. Machine learning (ML) has been applied for tasks that are important for smart grid operation including energy consumption and generation forecasting, anomaly detection, and state estimation. These ML solutions commonly require sufficient historical data; however, this data is often not readily available because of reasons such as data collection costs and concerns regarding security and privacy. This paper introduces a recurrent generative adversarial network (R-GAN) for generating realistic energy consumption data by learning from real data. Generativea adversarial …

Interpretable Deep Neural Network For Cancer Survival Analysis By Integrating Genomic And Clinical Data, Jie Hao, Youngsoon Kim, Tejaswini Mallavarapu, Jung Hun Oh, Mingon Kang

Computer Science Faculty Research

Background: Understanding the complex biological mechanisms of cancer patient survival using genomic and clinical data is vital, not only to develop new treatments for patients, but also to improve survival prediction. However, highly nonlinear and high-dimension, low-sample size (HDLSS) data cause computational challenges to applying conventional survival analysis. Results: We propose a novel biologically interpretable pathway-based sparse deep neural network, named Cox-PASNet, which integrates high-dimensional gene expression data and clinical data on a simple neural network architecture for survival analysis. Cox-PASNet is biologically interpretable where nodes in the neural network correspond to biological genes and pathways, while capturing the nonlinear …

Technology Safety Audit In Computer Laboratories Using Iso/Iec 17799 : 2005 (Case Study: Ftk Uin Sunan Ampel Surabaya), Muqoffi Khosyatulloh, Nurul Fariidhotun Nisaa, Ilham, M. Kom

Library Philosophy and Practice (e-journal)

Management audit is very important for assessment of their information technology management to gain efficient and effective business running process. Information technology security as an effort of internal controlling for risk and threat security minimization, is mainly considered due to all learning and lecturing administration activities use information technology. Allso, the implementation of a number of computer labs to facilitate learning processes and access to information in order to support the lectures and personal development of students To find out how secure technology information is, it is then requiring an audit to make sure everything run based on procedure. Management …

Point Cloud Processing With Neural Networks, Stephanie Miller, Jiahao Li

All Computer Science and Engineering Research

In this project, we explore new techniques and architectures for applying deep neural networks when the input is point cloud data. We first consider applying convolutions on regular pixel and voxel grids, using polynomials of point coordinates and Fourier transforms to get a rich feature representation for all points mapped to the same pixel or voxel. We also apply these ideas to generalize the recently proposed "interpolated convolution", by learning continuous-space kernels as a combination of polynomial and Fourier basis kernels. Experiments on the ModelNet40 dataset demonstrate that our methods have superior performance over the baselines in 3D object recognition.

Localized Dielectric Loss Heating In Dielectrophoresis Devices, Tae Joon Kwak, Imtiaz Hossen, Rashid Bashir, Woo-Jin Chang, Chung-Hoon Lee

Electrical and Computer Engineering Faculty Research and Publications

Temperature increases during dielectrophoresis (DEP) can affect the response of biological entities, and ignoring the effect can result in misleading analysis. The heating mechanism of a DEP device is typically considered to be the result of Joule heating and is overlooked without an appropriate analysis. Our experiment and analysis indicate that the heating mechanism is due to the dielectric loss (Debye relaxation). A temperature increase between interdigitated electrodes (IDEs) has been measured with an integrated micro temperature sensor between IDEs to be as high as 70 °C at 1.5 MHz with a 30 V_pp applied voltage to our ultra-low …

Founding The Domain Of Ai Forensics, Ibrahim Baggili, Vahid Behzadan

Electrical & Computer Engineering and Computer Science Faculty Publications

With the widespread integration of AI in everyday and critical technologies, it seems inevitable to witness increasing instances of failure in AI systems. In such cases, there arises a need for technical investigations that produce legally acceptable and scientifically indisputable findings and conclusions on the causes of such failures. Inspired by the domain of cyber forensics, this paper introduces the need for the establishment of AI Forensics as a new discipline under AI safety. Furthermore, we propose a taxonomy of the subfields under this discipline, and present a discussion on the foundational challenges that lay ahead of this new research …

Using Machine Learning Classification Methods To Detect The Presence Of Heart Disease, Nestor Pereira

Dissertations

Cardiovascular disease (CVD) is the most common cause of death in Ireland, and probably, worldwide. According to the Health Service Executive (HSE) cardiovascular disease accounting for 36% of all deaths, and one important fact, 22% of premature deaths (under age 65) are from CVD.

Using data from the Heart Disease UCI Data Set (UCI Machine Learning), we use machine learning techniques to detect the presence or absence of heart disease in the patient according to 14 features provide for this dataset. The different results are compared based on accuracy performance, confusion matrix and area under the Receiver Operating Characteristics (ROC) …

Factor Analysis Of Mixed Data (Famd) And Multiple Linear Regression In R, Nestor Pereira

Dissertations

In the previous projects, it has been worked to statistically analysis of the factors to impact the score of the subjects of Mathematics and Portuguese for several groups of the student from secondary school from Portugal.

In this project will be interested in finding a model, hypothetically multiple linear regression, to predict the final score, dependent variable G3, of the student according to some features divide into two groups. One group, analyses the features or predictors which impact in the final score more related to the performance of the students, means variables like study time or past failures. The second …

Amodal Instance Segmentation And Multi-Object Tracking With Deep Pixel Embedding, Yanfeng Liu

Department of Electrical and Computer Engineering: Dissertations, Theses, and Student Research

This thesis extends upon the representational output of semantic instance segmentation by explicitly including both visible and occluded parts. A fully convolutional network is trained to produce consistent pixel-level embedding across two layers such that, when clustered, the results convey the full spatial extent and depth ordering of each instance. Results demonstrate that the network can accurately estimate complete masks in the presence of occlusion and outperform leading top-down bounding-box approaches.

The model is further extended to produce consistent pixel-level embeddings across two consecutive image frames from a video to simultaneously perform amodal instance segmentation and multi-object tracking. No post-processing …

Hardware Implementation Of Assistive Technology Robot, Joycephine Li

Publications and Research

SuperHERO is an on-going research project in Computer Engineering Technology department which involves upgrading Heathkit Education Robot (HERO) hardware circuits and features by using modern hardware devices and sensors. The current phase of the project will focus on upgrading the motor drive system hardware as well as implementation and testing of features such as mobile robot obstacle detection and other assistive technologies to help people with disabilities. This involves the reattachment of the robot arm after repairing and updating with 3D printing and using modern hardware and software technology. We observed that the robotic arm has rotary and translation movements …

Aluminum/Carbon Composites Materials Fabricated By The Powder Metallurgy Process, Amélie Veillère, Hiroki Kurita, Akira Kawasaki, Yongfeng Lu, Jean-Marc Heintz, Jean-François Silvain

Department of Electrical and Computer Engineering: Faculty Publications

Aluminum matrix composites reinforced with carbon fibers or diamond particles have been fabricated by a powder metallurgy process and characterized for thermal management applications. Al/C composite is a nonreactive system (absence of chemical reaction between the metallic matrix and the ceramic reinforcement) due to the presence of an alumina layer on the surface of the aluminum powder particles. In order to achieve fully dense materials and to enhance the thermo-mechanical properties of the Al/C composite materials, a semi-liquid method has been carried out with the addition of a small amount of Al-Si alloys in the Al matrix. Thermal conductivity and …

Chapman Ambassador Tour Robot, Alexandra Lewandowski, Yanni Parissis, Khiry Carter, Hilary Lee

Student Scholar Symposium Abstracts and Posters

Being a student ambassador improves a student's confidence and leadership skills. With an increasing demand for technology skills, our project will display how the ambassador robot can assist student ambassadors while improving upon their efficiency, by discussing information during college campus tours and familiarizing students with robot applications and their technology. The ambassador robot can support students during tours by answering a question about specific knowledge that may have slipped an ambassador's mind. The robot will also be able to create a group-focused atmosphere that will allow ambassadors to have the opportunity to lean on a dependable teammate for specific …

Advanced Security Analysis For Emergent Software Platforms, Mohannad Alhanahnah

Department of Computer Science and Engineering: Dissertations, Theses, and Student Research

Emergent software ecosystems, boomed by the advent of smartphones and the Internet of Things (IoT) platforms, are perpetually sophisticated, deployed into highly dynamic environments, and facilitating interactions across heterogeneous domains. Accordingly, assessing the security thereof is a pressing need, yet requires high levels of scalability and reliability to handle the dynamism involved in such volatile ecosystems.

This dissertation seeks to enhance conventional security detection methods to cope with the emergent features of contemporary software ecosystems. In particular, it analyzes the security of Android and IoT ecosystems by developing rigorous vulnerability detection methods. A critical aspect of this work is the …

International Space Object Orbit Tracker, Nicole Navarro

Publications and Research

This software engineering project involves development of machine learning algorithms for embedded applications. High speed 32-bit hardware devices such as Raspberry Pi and ARM microcontrollers has become inexpensive and readily available. Machine learning algorithms for applications such as image processing and image recognition are computationally intensive. But with the availability of low cost 32-bit embedded computing devices, it is now feasible to implement then on embedded hardware. This project will explore embedded applications of machine learning algorithms by following a software engineering design and test approach.

Self-Driving Toy Car Using Deep Learning, Fahim Ahmed, Suleyman Turac, Mubtasem Ali

Publications and Research

Our research focuses on building a student affordable platform for scale model self-driving cars. The goal of this project is to explore current developments of Open Source hardware and software to build a low-cost platform consisting of the car chassis/framework, sensors, and software for the autopilot. Our research will allow other students with low budget to enter into the world of Deep Learning, self-driving cars, and autonomous cars racing competitions.

Static Taint Analysis Of Binary Executables Using Architecture-Neutral Intermediate Representation, Elaine Cole

All Computer Science and Engineering Research

Ghidra, National Security Agency’s powerful reverse engineering framework, was recently released open-source in April 2019 and is capable of lifting instructions from a wide variety of processor architectures into its own register transfer language called p-code. In this project, we present a new tool which leverages Ghidra’s specific architecture-neutral intermediate representation to construct a control flow graph modeling all program executions of a given binary and apply static taint analysis. This technique is capable of identifying the information flow of malicious input from untrusted sources that may interact with key sinks or parts of the system without needing access to …

How Degrees Of Freedom Affects Sense Of Agency, Akima Connelly, Jungsu Pak, Tian Lan, Uri Maoz

Student Scholar Symposium Abstracts and Posters

Can the rubber-hand illusion be extended to a moving robotic arm in different degrees of freedom (DOF), inducing sense of ownership & agency over the arm? We hypothesize that DOF closer to what humans possess will result in a stronger sense of ownership and agency.

Low-Cost Near Infrared Diffuse Optical Imaging System, Mohammed Z. Shakil, Chen Xu

Publications and Research

Diffuse Optical Tomography (DOT) and Optical Spectroscopy using near-infrared (NIR) diffused light has demonstrated great potential for the initial diagnosis of tumors and in the assessment of tumor vasculature response to neoadjuvant chemotherapy. The NIR technique utilizes intrinsic hemoglobin contrast, which is directly related to tumor angiogenesis development, a key process required for tumor growth and metastasis. The NIR diffuse tomography holds great promise in distinguishing early-stage invasive breast cancers from benign lesions. This technique also provides insight into tumor metabolism and tumor hypoxia, important indicators of tumor response to various forms of therapy. Currently, the high cost of the …

The Trolley Problem In Virtual Reality, Jungsu Pak, Ariane Guirguis, Nicholas Mirchandani, Scott Cummings, Uri Maoz

Student Scholar Symposium Abstracts and Posters

Would people react to the Trolley problem differently based on the medium? Immersive Virtual Reality Driving Simulator was used to examine participants respond to the trolley problem in a realistic and controlled simulated environment.

Voice Controlled Augmented Reality: A Comparison Of Speech-Recognition Tools For Ar Applications, Juan Estrella

Publications and Research

Augmented Reality (AR) refers to the technologies that enhance the version of the physical environment with computer-generated sensory input such as sound and graphics overlaid on top of the user’s view of the real world. Artificial Intelligence (AI) studies how to make computer programs and machines "smart" and take decisions. Our research project focuses on exploring the Integration of AI in AR applications. Specifically, on using Speech Recognition or Natural Language Processing for controlling virtual AR objects and enhancing the human-computer interaction. It is obvious that integration of AI and AR is of great value. However, for developers, it is …

Second-Order Fault Tolerant Extended Kalman Filter For Discrete Time Nonlinear Systems, Xin Wang, Edwin E. Yaz

Electrical and Computer Engineering Faculty Research and Publications

As missing sensor data may severely degrade the overall system performance and stability, reliable state estimation is of great importance in modern data-intensive control, computing, and power systems applications. Aiming at providing a more robust and resilient state estimation technique, this paper presents a novel second-order fault-tolerant extended Kalman filter estimation framework for discrete-time stochastic nonlinear systems under sensor failures, bounded observer-gain perturbation, extraneous noise, and external disturbances condition. The failure mechanism of multiple sensors is assumed to be independent of each other with various malfunction rates. The proposed approach is a locally unbiased, minimum estimation error covariance based nonlinear …

Comparative Study Of Winding Configurations Of A Five-Phase Flux-Switching Pm Machine, Hao Chen, Xiangdong Liu, Ayman M. El-Refaie, Jing Zhao, Nabeel Demerdash, Jiangbiao He

Electrical and Computer Engineering Faculty Research and Publications

This paper introduces a general method for determination of the most suitable winding configurations for five-phase flux-switching permanent magnet (FSPM) machines, associated with feasible stator/rotor-pole combinations. Consequently, the effect of winding configurations on the performance of a five-phase outer-rotor FSPM machine is thoroughly investigated, including non-overlapping concentrated windings (single-layer, double-layer, and multi-layer) as well as distributed winding. The electromagnetic characteristics in the low-speed region, the flux-weakening capability in the high-speed region, and the fault-tolerant capability under faulty situations are evaluated and compared in detail. This work shows that compared with the conventional single-layer or double-layer concentrated windings, the FSPM machine …

An Approach To Fast Multi-Robot Exploration In Buildings With Inaccessible Spaces, Matt Mcneill, Damian Lyons

Faculty Publications

The rapid exploration of unknown environments is a common application of autonomous multi-robot teams. For some types of exploration missions, a mission designer may possess some rudimentary knowledge about the area to be explored. For example, the dimensions of a building may be known, but not its floor layout or the location of furniture and equipment inside. For this type of mission, the Space- Based Potential Field (SBPF) method is an approach to multirobot exploration which leverages a priori knowledge of area bounds to determine robot motion. Explored areas and obstacles exert a repulsive force, and unexplored areas exert an …

Domain Adaptation In Unmanned Aerial Vehicles Landing Using Reinforcement Learning, Pedro Lucas Franca Albuquerque

Department of Computer Science and Engineering: Dissertations, Theses, and Student Research

Landing an unmanned aerial vehicle (UAV) on a moving platform is a challenging task that often requires exact models of the UAV dynamics, platform characteristics, and environmental conditions. In this thesis, we present and investigate three different machine learning approaches with varying levels of domain knowledge: dynamics randomization, universal policy with system identification, and reinforcement learning with no parameter variation. We first train the policies in simulation, then perform experiments both in simulation, making variations of the system dynamics with wind and friction coefficient, then perform experiments in a real robot system with wind variation. We initially expected that providing …

Formal Modeling And Analysis Of A Family Of Surgical Robots, Niloofar Mansoor

Department of Computer Science and Engineering: Dissertations, Theses, and Student Research

Safety-critical applications often use dependability cases to validate that specified properties are invariant, or to demonstrate a counterexample showing how that property might be violated. However, most dependability cases are written with a single product in mind. At the same time, software product lines (families of related software products) have been studied with the goal of modeling variability and commonality and building family-based techniques for both modeling and analysis. This thesis presents a novel approach for building an end to end dependability case for a software product line, where a property is formally modeled, a counterexample is found and then …

Seer: An Explainable Deep Learning Midi-Based Hybrid Song Recommender System, Khalil Damak, Olfa Nasraoui

Faculty Scholarship

State of the art music recommender systems mainly rely on either matrix factorization-based collaborative filtering approaches or deep learning architectures. Deep learning models usually use metadata for content-based filtering or predict the next user interaction by learning from temporal sequences of user actions. Despite advances in deep learning for song recommendation, none has taken advantage of the sequential nature of songs by learning sequence models that are based on content. Aside from the importance of prediction accuracy, other significant aspects are important, such as explainability and solving the cold start problem. In this work, we propose a hybrid deep learning …

Ldakm-Eiot: Lightweight Device Authentication And Key Management Mechanism For Edge-Based Iot Deployment, Mohammad Wazid, Ashok Kumar Das, Sachin Shetty, Joel J. P. C. Rodrigues, Youngho Park

VMASC Publications

In recent years, edge computing has emerged as a new concept in the computing paradigm that empowers several future technologies, such as 5G, vehicle-to-vehicle communications, and the Internet of Things (IoT), by providing cloud computing facilities, as well as services to the end users. However, open communication among the entities in an edge based IoT environment makes it vulnerable to various potential attacks that are executed by an adversary. Device authentication is one of the prominent techniques in security that permits an IoT device to authenticate mutually with a cloud server with the help of an edge node. If authentication …

Comparison And Design Optimization Of A Five-Phase Flux-Switching Pm Machine For In-Wheel Traction Applications, Hao Chen, Xiangdong Liu, Nabeel Demerdash, Ayman M. El-Refaie, Jing Zhao, Jiangbiao He

Electrical and Computer Engineering Faculty Research and Publications

A comparative study of five-phase outer-rotor flux-switching permanent magnet (FSPM) machines with different topologies for in-wheel traction applications is presented in this paper. Those topologies include double-layer winding, single-layer winding, C-core, and E-core configurations. The electromagnetic performance in the low-speed region, the flux-weakening capability in the high-speed region, and the fault-tolerance capability are all investigated in detail. The results indicate that the E-core FSPM machine has performance advantages. Furthermore, two kinds of E-core FSPM machines with different stator and rotor pole combinations are optimized, respectively. In order to reduce the computational burden during the large-scale optimization process, a mathematical technique …

Pixel-Level Deep Multi-Dimensional Embeddings For Homogeneous Multiple Object Tracking, Mateusz Mittek

Department of Electrical and Computer Engineering: Dissertations, Theses, and Student Research

The goal of Multiple Object Tracking (MOT) is to locate multiple objects and keep track of their individual identities and trajectories given a sequence of (video) frames. A popular approach to MOT is tracking by detection consisting of two processing components: detection (identification of objects of interest in individual frames) and data association (connecting data from multiple frames). This work addresses the detection component by introducing a method based on semantic instance segmentation, i.e., assigning labels to all visible pixels such that they are unique among different instances. Modern tracking methods often built around Convolutional Neural Networks (CNNs) and additional, …