Computer Engineering Commons^™

Machine Learning (Ml) - Assisted Tools For Enhancing Security And Privacy Of Edge Devices, Santosh Kumar Nukavarapu

Theses and Dissertations

The rapid growth of edge-based IoT devices, their use cases, and autonomous communication has created new challenges with privacy and security. Side-channel attacks are one of the examples of security and privacy vulnerabilities that can cause inference at Internet-Service Provider (ISP) and local Wi-Fi networks. Such an attack would leak user’s sensitive information such as home occupancy, medical activity, and daily routines. Another example is that these devices have weak authentication and low encryption standards, making them an easy target for malware-based attacks such as denial of service or launching other network attacks using these infected devices. This thesis dissertation …

Learning Robot Motion From Creative Human Demonstration, Charles C. Dietzel

Theses and Dissertations

This thesis presents a learning from demonstration framework that enables a robot to learn and perform creative motions from human demonstrations in real-time. In order to satisfy all of the functional requirements for the framework, the developed technique is comprised of two modular components, which integrate together to provide the desired functionality. The first component, called Dancing from Demonstration (DfD), is a kinesthetic learning from demonstration technique. This technique is capable of playing back newly learned motions in real-time, as well as combining multiple learned motions together in a configurable way, either to reduce trajectory error or to generate entirely …

Optimized Cancer Detection On Various Magnified Histopathological Colon Imagesbased On Dwt Features And Fcm Clustering, Tina Babu, Tripty Singh, Deepa Gupta, Shahin Hameed

Turkish Journal of Electrical Engineering and Computer Sciences

Due to the morphological characteristics and other biological aspects in histopathological images, the computerized diagnosis of colon cancer in histopathology images has gained popularity. The images acquired using the histopathology microscope may differ for greater visibility by magnifications. This causes a change in morphological traits leading to intra and inter-observer variability. An automatic colon cancer diagnosis system for various magnification is therefore crucial. This work proposes a magnification independent segmentation approach based on the connected component area and double density dual tree DWT (discrete wavelet transform) coefficients are derived from the segmented region. The derived features are reduced further shortened …

Cnn Based Sensor Fusion Method For Real-Time Autonomous Robotics Systems, Berat Yildiz, Aki̇f Durdu, Ahmet Kayabaşi, Mehmet Duramaz

Turkish Journal of Electrical Engineering and Computer Sciences

Autonomous robotic systems (ARS) serve in many areas of daily life. The sensors have critical importance for these systems. The sensor data obtained from the environment should be as accurate and reliable as possible and correctly interpreted by the autonomous robot. Since sensors have advantages and disadvantages over each other they should be used together to reduce errors. In this study, Convolutional Neural Network (CNN) based sensor fusion was applied to ARS to contribute the autonomous driving. In a real-time application, a camera and LIDAR sensor were tested with these networks. The novelty of this work is that the uniquely …

Fft Enabled Ecc For Wsn Nodes Without Hardware Multiplier Support, Utku Gülen, Selçuk Baktir

Turkish Journal of Electrical Engineering and Computer Sciences

ECC is a popular cryptographic algorithm for key distribution in wireless sensor networks where power efficiency is desirable. A power efficient implementation of ECC without using hardware multiplier support was proposed earlier for wireless sensor nodes. The proposed implementation utilized the number theoretic transform to carry operands to the frequency domain, and conducted Montgomery multiplication, in addition to other finite field operations, in that domain. With this work, we perform in the frequency domain only polynomial multiplication and use the fast Fourier transform to carry operands between the time and frequency domains. Our ECC implementation over $GF((2^{13}-1)^{13})$ on the MSP430 …

Using Vertical Areas In Finite Set Model Predictive Control Of A Three-Level Inverter Aimed At Computation Reduction, Alireza Jaafari, Alireza Davari, Cristian Garcia, Jose Rodriguez

Turkish Journal of Electrical Engineering and Computer Sciences

In power electronics applications, finite set model predictive control (FS-MPC) has proven to be a viable strategy. However, due to the high processing power required, using this technology in multilevel converters is difficult. This strategy, which is based on predicting the behavior of the system for all conceivable states, has an issue with a numerous of possible switching states. A recent and useful strategy for dealing with the problem is the limiting of calculations based on triangle regions. Despite its success, this method has several limitations, including the computation required to locate the right triangle and the boundary modes. In …

Application Of Long Short-Term Memory (Lstm) Neural Network Based On Deeplearning For Electricity Energy Consumption Forecasting, Mehmet Bi̇lgi̇li̇, Ni̇yazi̇ Arslan, Ali̇i̇hsan Şekerteki̇n, Abdulkadi̇r Yaşar

Turkish Journal of Electrical Engineering and Computer Sciences

Electricity is the most substantial energy form that significantly affects the development of modern life, work efficiency, quality of life, production, and competitiveness of the society in the ever-growing global world. In this respect, forecasting accurate electricity energy consumption (EEC) is fairly essential for any country?s energy consumption planning and management regarding its growth. In this study, four time-series methods; long short-term memory (LSTM) neural network, adaptive neuro-fuzzy inference system (ANFIS) with subtractive clustering (SC), ANFIS with fuzzy cmeans (FCM), and ANFIS with grid partition (GP) were implemented for the short-term one-day ahead EEC prediction. Root mean square error (RMSE), …

Deep Learning-Aided Automated Personal Data Discovery And Profiling, Apdullah Yayik, Vedat Aybar, Hasan Hüseyi̇n Apik, Sevcan İçöz, Beki̇r Bakar, Tunga Güngör

Turkish Journal of Electrical Engineering and Computer Sciences

In Turkey, Turkish Personal Data Protection Rule (PDPR) No. 6698, in force since 2016, provides protection to citizens for the legal existence of their personal data. Although the law provides excellent guidance, companies currently face challenges in complying with its regulations in terms of storing, sharing, or monitoring personal data. Since any specially designed software with wide industrial usage is not on the market, almost all of the companies have no other choice but to take expensive and error-prone operations manually to ensure their compliance. In this paper, we present an automated solution to facilitate and accelerate PDPR compliance. In …

Evaluating The English-Turkish Parallel Treebank For Machine Translation, Onur Görgün, Olcay Taner Yildiz

Turkish Journal of Electrical Engineering and Computer Sciences

This study extends our initial efforts in building an English-Turkish parallel treebank corpus for statistical machine translation tasks. We manually generated parallel trees for about 17K sentences selected from the Penn Treebank corpus. English sentences vary in length: 15 to 50 tokens including punctuation. We constrained the translation of trees by (i) reordering of leaf nodes based on suffixation rules in Turkish, and (ii) gloss replacement. We aim to mimic human annotator?s behavior in real translation task. In order to fill the morphological and syntactic gap between languages, we do morphological annotation and disambiguation. We also apply our heuristics by …

A Novel Energy Consumption Model For Autonomous Mobile Robot, Gürkan Gürgöze, İbrahi̇m Türkoğlu

Turkish Journal of Electrical Engineering and Computer Sciences

In this study, a novel predictive energy consumption model has been developed to facilitate the development of tasks based on efficient energy consumption strategies in mobile robot systems. For the proposed energy consumption model, an advanced mathematical system model that takes into account all parameters during the motion of the mobile robot is created. The parameters of inclination, load, dynamic friction, wheel slip and speed-torque saturation limit, which are often neglected in existing models, are especially used in our model. Thus, the effects of unexpected disruptors on energy consumption in the real world environment are also taken into account. As …

Automatically Classifying Familiar Web Users From Eye-Tracking Data:A Machine Learning Approach, Meli̇h Öder, Şükrü Eraslan, Yeli̇z Yesi̇lada

Turkish Journal of Electrical Engineering and Computer Sciences

Eye-tracking studies typically collect enormous amount of data encoding rich information about user behaviours and characteristics on the web. Eye-tracking data has been proved to be useful for usability and accessibility testing and for developing adaptive systems. The main objective of our work is to mine eye-tracking data with machine learning algorithms to automatically detect users' characteristics. In this paper, we focus on exploring different machine learning algorithms to automatically classify whether users are familiar or not with a web page. We present our work with an eye-tracking data of 81 participants on six web pages. Our results show that …

Shape Investigations Of Structures Formed By The Self-Assembly Of Aromaticamino Acids Using The Density-Based Spatial Clustering Of Applications With Noise Algorithm, Mehmet Gökhan Habi̇boğlu, Helen W. Hernandez, Şahi̇n Uyaver

Turkish Journal of Electrical Engineering and Computer Sciences

Tyrosine, tryptophan, and phenylalanine are important aromatic amino acids for human health. If they are not properly metabolized, severe rare mental or metabolic diseases can emerge, many of which are not researched enough due to economic priorities. In our previous simulations, all three of these amino acids are discovered to be self-organizing and to have complex aggregations at different temperatures. Two of these essential stable formations are observed during our simulations: tubular-like and spherical-like structures. In this study, we develop and implement a clustering analyzing algorithm using density-based spatial clustering of applications with noise (DBSCAN) to measure the shapes of …

Stability Regions In Time Delayed Two-Area Lfc System Enhanced By Evs, Ausnain Naveed, Şahi̇n Sönmez, Saffet Ayasun

Turkish Journal of Electrical Engineering and Computer Sciences

With the extensive usage of open communication networks, time delays have become a great concern in load frequency control (LFC) systems since such inevitable large delays weaken the controller performance and even may lead to instabilities. Electric vehicles (EVs) have a potential tool in the frequency regulation. The integration of a large number of EVs via an aggregator amplifies the adverse effects of time delays on the stability and controller design of LFC systems. This paper investigates the impacts of the EVs aggregator with communication time delay on the stability. Primarily, a graphical method characterizing stability boundary locus is implemented. …

An Active Contour Model Using Matched Filter And Hessian Matrix For Retinalvessels Segmentation, Mahtab Shabani, Hossein Pourghassem

Turkish Journal of Electrical Engineering and Computer Sciences

Medical image analysis, especially of the retina, plays an important role in diagnostic decision support tools. The properties of retinal blood vessels are used for disease diagnoses such as diabetes, glaucoma, and hypertension. There are some challenges in the utilization of retinal blood vessel patterns such as low contrast and intensity inhomogeneities. Thus, an automatic algorithm for vessel extraction is required. Active contour is a strong method for edge extraction. However, it cannot extract thin vessels and ridges very well. In this research, we propose an improved active contour method that uses discrete wavelet transform for energy minimization to solve …

Stressed Or Just Running? Differentiation Of Mental Stress And Physical Activityby Using Machine Learning, Yekta Sai̇d Can

Turkish Journal of Electrical Engineering and Computer Sciences

Recently, modern people have excessive stress in their daily lives. With the advances in physiological sensors and wearable technology, people?s physiological status can be tracked, and stress levels can be recognized for providing beneficial services. Smartwatches and smartbands constitute the majority of wearable devices. Although they have an excellent potential for physiological stress recognition, some crucial issues need to be addressed, such as the resemblance of physiological reaction to stress and physical activity, artifacts caused by movements and low data quality. This paper focused on examining and differentiating physiological responses to both stressors and physical activity. Physiological data are collected …

Spectrum Sensing With Energy Detection In Multiple Alternating Time Slots, Călin Vlădeanu, Alexandru Marţian, Dimitrie C. Popescu

Electrical & Computer Engineering Faculty Publications

Energy detection (ED) represents a low complexity approach used by secondary users (SU) to sense spectrum occupancy by primary users (PU) in cognitive radio (CR) systems. In this paper, we present a new algorithm that senses the spectrum occupancy by performing ED in K consecutive sensing time slots starting from the current slot and continuing by alternating before and after the current slot. We consider a PU traffic model specified in terms of an average duty cycle value, and derive analytical expressions for the false alarm probability (FAP) and correct detection probability (CDP) for any value of K . Our …

Bfv-Based Homomorphic Encryption For Privacy-Preserving Cnn Models, Febrianti Wibawa, Ferhat Ozgur Catak, Salih Sarp, Murat Kuzlu

Engineering Technology Faculty Publications

Medical data is frequently quite sensitive in terms of data privacy and security. Federated learning has been used to increase the privacy and security of medical data, which is a sort of machine learning technique. The training data is disseminated across numerous machines in federated learning, and the learning process is collaborative. There are numerous privacy attacks on deep learning (DL) models that attackers can use to obtain sensitive information. As a result, the DL model should be safeguarded from adversarial attacks, particularly in medical data applications. Homomorphic encryption-based model security from the adversarial collaborator is one of the answers …

Collective Action On Behalf Of Women: Testing The Conceptual Distinction Between Traditional Collective Action And Small Acts In College Women, Anca M. Miron, Thomas C. Ball, Nyla R. Branscombe, Monica Fieck, Cristinel Ababei, Serena Raymer, Baylee Tkaczuk, Megan M. Meives

Electrical and Computer Engineering Faculty Research and Publications

The current study examines the nature of actions that U.S. college women (N = 267) engage in to promote, protect, or enhance the welfare of other women. The study had two goals: 1) to distinguish between traditional forms of action (traditional collective action) and more informal, interpersonal, forms of action (small acts) among college women; and 2) to test whether the classic antecedents of collective action (gender identity, feminist identity, women’s activist identity, efficacy, appraisals of gender inequality, and injustice standards) are differentially predictive of these two types of participation. A confirmatory factor analysis provided strong support for these two …

A Comparison Of Two Generalizations To The Linear Sampling Method For Inverse Scattering, Yeasmin Sultana, James E. Richie

Electrical and Computer Engineering Faculty Research and Publications

The linear sampling method (LSM) is a very popular method for determining the boundary of an object from the scattered field. However, there are instances where LSM provides the convex hull of the boundary rather than the true boundary. There are two common generalizations to LSM: the Generalized Linear Sampling Method (GLSM) and the Multipoles-based Linear Sampling Method (MLSM). In this paper, the ability of GLSM and MLSM to overcome some of the deficiencies of LSM are investigated. It is found that GLSM may be ideal for imaging thin features of scatterers and that MLSM can provide an improvement over …

Artificial Intelligence (Ai): The New Look Of Customer Service In A Cybersecurity World, Sharon L. Burton

Publications

Cybersecurity leaders are not adequately developed to guide the re-engineering of quality customer service (QCS) workflows, designed with automation and AI, that interrelate with people through customers’ perceptions. Realizing re-engineering processes should be a team effort with well-versed leadership and stakeholders guiding the successful design through a follow-up process. Leaders must shape compelling and straightforward needs to learn and teach employees and chat boxes indispensable customer service skills demonstrating patience, self-discipline, flexibility, and resourcefulness in communication with irritated customers or difficult circumstances. Whether the analysis, design, development, and implementation struggles are vacuums in cybersecurity knowledge, skill, and abilities or a …

An Analysis Of Compressive Convolutional Autoencoders For Image Archiving In Medical Informatics, Charles Warren

Dissertations, Master's Theses and Master's Reports

Within a given enterprise network, an array of data types needs to be communicated. These network transmissions consist of images, videos, text, and binaries that have unique requirements of bandwidth and computational overhead to transmit. With respect to medical informatics, these include a multitude of varying subjects, standards, and modalities which are communicated to and from imaging equipment, clinicians, and medical archives. To reduce the required bandwidth to transmit, or provide adequate storage capacity for archival purposes, the data may be compressed in such a way that reduces the size of the image when it is transferred or stored. The …

Development And Implementation Of Robot Operating Systems For Undergraduates, Chelsey Spitzner

Dissertations, Master's Theses and Master's Reports

The purpose of this project was to create an undergraduate junior lab to teach students about Robotic Operating System (ROS). The labs were designed to highlight the usefulness of ROS and the process used. Designing algorithms, how to send/receive messages, and the hierarchy of how nodes work with each other are emphasized. Taking packages that are open-source then modifying them is also emphasized. This is done so that students can transfer their knowledge from this course to other robot operating systems.

Off Road Autonomous Vehicle Modeling And Repeatability Using Real World Telemetry Via Simulation, Matthew Paul Spencer

Dissertations, Master's Theses and Master's Reports

One approach to autonomous control of high mobility ground vehicle platforms operating on challenging terrain is with the use of predictive simulation. Using a simulated or virtual world, an autonomous system can optimize use of its control systems by predicting interaction between the vehicle and ground as well as the vehicle actuator state. Such a simulation allows the platform to assess multiple possible scenarios before attempting to execute a path. Physically realistic simulations covering all of these domains are currently computationally expensive, and are unable to provide fast execution times when assessing each individual scenario due to the use of …

Collective Hydrodynamics Of Robotic Fish, Rohit S. Pandhare

Dissertations, Master's Theses and Master's Reports

Many animals in nature travel in groups either for protection, survival, or endurance. Among these, fish do so under the burden of hydrodynamic loads, which incites questions as to the significance of the multi-body fluid-mediated interactions that facilitate collective swimming. We study such interactions in the idealized setting of a rotational array of robotic fish whose tails undergo a prescribed flapping motion, but whose swimming speed is determined as a natural result of the hydrodynamic effects. Specifically, we examine how the measured collective speed of the swimmers varies with the imposed frequency and amplitude of their tail flapping, and with …

Adapting Deep Learning For Underwater Acoustic Communication Channel Modeling, Li Wei

Dissertations, Master's Theses and Master's Reports

The recent emerging applications of novel underwater systems lead to increasing demand for underwater acoustic (UWA) communication and networking techniques. However, due to the challenging UWA channel characteristics, conventional wireless techniques are rarely applicable to UWA communication and networking. The cognitive and software-defined communication and networking are considered promising architecture of a novel UWA system design. As an essential component of a cognitive communication system, the modeling and prediction of the UWA channel impulse response (CIR) with deep generative models are studied in this work.

Firstly, an underwater acoustic communication and networking testbed is developed for conducting various simulations and …