Computer Engineering Commons^™

Network Health Monitoring System, Zeeshan Siddiqui

Undergraduate Student Research Internships Conference

No abstract provided.

Evaluating Algorithms Used For Fetal Brain Scan Segmentation, Connor Stewart Burgess

Undergraduate Student Research Internships Conference

The goal for this project was to successfully segment a fetal brain scan (fetal scan) using the algorithms provided by the program Slicer3D. To better understand the hurdles that arose when segmenting a fetal scan, we first look at the segmentation of an adult brain scan. This will allow us to see the straightforward nature of a brain segmentation when a high quality, high resolution volume with distinct structures is available. After examining the adult brain scan, attention will be moved to the segmentation of the fetal scan, where we’ll first look at the algorithms used and methods followed. Finally …

Distributed Parallel Processing, Adrian Wu

Undergraduate Student Research Internships Conference

This report summarizes the development of testing new microcontrollers in performing image processing and parallel processing to solve the problem of testing and deploying expensive computer hardware technology in space. The project determined that the I2C communication protocol should be converted to an alternative protocol to maximize data transfer in parallel processing. This report also analyzes the software components and hardware components of the Distributed Parallel Processing project.

This project was to research the alternative protocols for the recently developed project Distributed Parallel Processing with CubeSats. There project was to develop a suitable microcontroller to perform image processing techniques with …

Leveraging Machine Learning Techniques Towards Intelligent Networking Automation, Cesar A. Gomez

Electronic Thesis and Dissertation Repository

In this thesis, we address some of the challenges that the Intelligent Networking Automation (INA) paradigm poses. Our goal is to design schemes leveraging Machine Learning (ML) techniques to cope with situations that involve hard decision-making actions. The proposed solutions are data-driven and consist of an agent that operates at network elements such as routers, switches, or network servers. The data are gathered from realistic scenarios, either actual network deployments or emulated environments. To evaluate the enhancements that the designed schemes provide, we compare our solutions to non-intelligent ones. Additionally, we assess the trade-off between the obtained improvements and the …

Mechanical Design And Development Of A Compliant 5-Dof Manipulator Using Magneto-Rheological Actuators, Sergey Pisetskiy

Electronic Thesis and Dissertation Repository

Compliance in robotic systems became a very important and desirable characteristic in recent years. Existing compliant actuation approaches have either limited performance or significant mechanical and control complexity. Keeping high performance while maintaining the necessary level of compliance at low cost and minimum complexity is a challenging goal that should be achieved to boost the propagation of human-safe robots and systems capable to perform delicate tasks in an unknown environment.

This study presents a novel five degrees-of-freedom compliant manipulator. The compliancy of the manipulator is achieved using antagonistically working pairs of magneto-rheological (MR) clutches in each joint of the robot. …

Visual Cues For Semi-Autonomous Control Of Transradial Prosthetics, Mena S.A. Kamel

Electronic Thesis and Dissertation Repository

Upper-limb prosthetics are typically driven exclusively by biological signals, mainly electromyography (EMG), where electrodes are placed on the residual part of an amputated limb. In this approach, amputees must control each arm joint iteratively, in a proportional manner. Research has shown that sequential control of prosthetics usually imposes a cognitive burden on amputees, leading to high abandonment rates. This thesis presents a control system for upper-limb prosthetics, leveraging a computer vision module capable of simultaneously predicting objects in a scene, their segmentation mask, and a ranked list of the optimal grasping locations. The proposed system shares control with an amputee, …

Deep Learning For High-Impedance Fault Detection And Classification, Khushwant Rai

Electronic Thesis and Dissertation Repository

High-Impedance Faults (HIFs) are a hazard to public safety but are difficult to detect because of their low current amplitude and diverse characteristics. Supervised machine learning techniques have shown great success in HIF detection; however, these approaches rely on resource-intensive signal processing techniques and fail in presence of non-HIF disturbances and even for scenarios not included in training data. This thesis leverages unsupervised learning and proposes a Convolutional Autoencoder framework for HIF Detection (CAE-HIFD). In CAE-HIFD, Convolutional Autoencoder learns only from HIF signals by employing cross-correlation; consequently, eliminating the need for diverse non-HIF scenarios in training. Furthermore, this thesis proposes …

A Black-Box Approach For Containerized Microservice Monitoring In Fog Computing, Shi Chang

Electronic Thesis and Dissertation Repository

The goal of the Internet of Things (IoT) is to convert the physical world into a smart space in which physical objects, called things, are equipped with computing and communication capabilities. Those things can connect with anything, anyone at any time, any space via any network or service. The predominant Internet of Things (IoT) system model today is cloud centric. This model introduces latencies into the application execution, as data travels first upstream for processing and secondly the results, i.e., control commands, travel downstream to the devices. In contrast with the cloud-model, the cloud-fog-based model pushes computing capability to the …

Dynamic Planning Networks, Norman Tasfi, Miriam A M Capretz

Electrical and Computer Engineering Publications

We introduce Dynamic Planning Networks (DPN), a novel architecture for deep reinforcement learning, that combines model-based and model-free aspects for online planning. Our architecture learns to dynamically construct plans using a learned state-transition model by selecting and traversing between simulated states and actions to maximize information before acting. DPN learns to efficiently form plans by expanding a single action conditional state transition at a time instead of exhaustively evaluating each action, reducing the number of state-transitions used during planning. We observe emergent planning patterns in our agent, including classical search methods such as breadth-first and depth-first search. DPN shows improved …

A Technique For Evaluating The Health Status Of A Software Module Using Process Metrics, . Ria

Electronic Thesis and Dissertation Repository

Identifying error-prone files in large software systems has been an area where significant attention has been paid over the years. In this thesis, we propose a process-metrics based method for predicting the health status of a file based on its commit profile in its GitHub repository. Precisely, for each file and each bug fixing commit a file participates, we compute a dependency score of the committed file with its other co-committed files. The calculated score is appropriately decayed if the file does not participate in the new bug-fixing commits in the system. By examining the trend of the dependency score …

Arm-Hand Systems As Hybrid Parallel-Serial Systems: A Novel Inverse Kinematics Solution, Shuwei Qiu, Mehrdad Kermani Ph.D., P.Eng.

Electrical and Computer Engineering Publications

No abstract provided.

Development Of A Wearable Haptic Feedback Device For Upper Limb Prosthetics Through Sensory Substitution, Marco B.S. Gallone

Electronic Thesis and Dissertation Repository

Haptics can enable a direct communication pipeline between the artificial limb and the brain; adding haptic sensory feedback for prosthesis wearers is believed to improve operation without drawing too much of the user's attention. Through neuroplasticity, the brain can become more cognizant of the information delivered through the skin and may eventually interpret it as inherently as other natural senses. In this thesis, a wearable haptic feedback device (WHFD) is developed to communicate prosthesis sensory information. A 14-week, 6-stage, between subjects study was created to investigate the learning trajectory as participants were stimulated with haptic patterns conveying joint proprioception. 37 …

Deep Neural Networks For Human Activity Recognition With Wearable Sensors, Davoud Gholamiangonabadi

Electronic Thesis and Dissertation Repository

Human Activity Recognition (HAR) has been attracting significant research attention because of a wide range of applications from healthcare to security. Recently, deep learning approaches have demonstrated great success in the HAR area. However, these models are often evaluated on the same subjects as those used to train the model; thus, the provided accuracy estimates do not pertain to new subjects. Consequently, this thesis examines the generalization capability of different machine learning architectures using Leave-One-Subject-Out Cross-Validation (LOSOCV) and then proposes a personalized model. The accuracy is improved by considering two feature selection directions, time- and frequency-domain, and by dynamically selecting …

Transfer Learning By Similarity Centred Architecture Evolution For Multiple Residential Load Forecasting, Santiago Gomez-Rosero, Miriam A M Capretz, Syed Mir

Electrical and Computer Engineering Publications

The development from traditional low voltage grids to smart systems has become extensive and adopted worldwide. Expanding the demand response program to cover the residential sector raises a wide range of challenges. Short term load forecasting for residential consumers in a neighbourhood could lead to a better understanding of low voltage consumption behaviour. Nevertheless, users with similar characteristics can present diversity in consumption patterns. Consequently, transfer learning methods have become a useful tool to tackle differences among residential time series. This paper proposes a method combining evolutionary algorithms for neural architecture search with transfer learning to perform short term load …

‘Digits’ App - Smartphone Augmented Reality For Hand Telerehabilitation, Hongdao Dong, Edward Ho, Herbert Shin, Tania Banerjee, Geoffrey Masschelein, Jacob Davidson, Sandrine De Ribaupierre, Roy Eagleson, Caitlin Symonette

Electrical and Computer Engineering Publications

Hand telerehabilitation currently has limitations for accurate and remote assessment of range of motion (ROM) in small finger joints. ‘DIGITS’ application utilises the front smartphone camera to measure finger ROM in a reliable and rapid assessment protocol. Our initial beta-phase testing examined the consistency of our software measurements to in-person goniometry. 6 to 9 degrees of difference existed between the smartphone application recorded data versus the in-person measurements. This range is within acceptable 7 to 9 degree tolerance for interrater goniometry measurements. The effect of environmental factors such as hand distance, lightings and hand orientation was evaluated. The intraclass correlation …

A Systematic Review Of Convolutional Neural Network-Based Structural Condition Assessment Techniques, Sandeep Sony, Kyle Dunphy, Ayan Sadhu, Miriam A M Capretz

Electrical and Computer Engineering Publications

With recent advances in non-contact sensing technology such as cameras, unmanned aerial and ground vehicles, the structural health monitoring (SHM) community has witnessed a prominent growth in deep learning-based condition assessment techniques of structural systems. These deep learning methods rely primarily on convolutional neural networks (CNNs). The CNN networks are trained using a large number of datasets for various types of damage and anomaly detection and post-disaster reconnaissance. The trained networks are then utilized to analyze newer data to detect the type and severity of the damage, enhancing the capabilities of non-contact sensors in developing autonomous SHM systems. In recent …

Deep Neural Network For Load Forecasting Centred On Architecture Evolution, Santiago Gomez-Rosero, Miriam A M Capretz, London Hydro

Electrical and Computer Engineering Publications

Nowadays, electricity demand forecasting is critical for electric utility companies. Accurate residential load forecasting plays an essential role as an individual component for integrated areas such as neighborhood load consumption. Short-term load forecasting can help electric utility companies reduce waste because electric power is expensive to store. This paper proposes a novel method to evolve deep neural networks for time series forecasting applied to residential load forecasting. The approach centres its efforts on the neural network architecture during the evolution. Then, the model weights are adjusted using an evolutionary optimization technique to tune the model performance automatically. Experimental results on …

Predicting Residential Energy Consumption Using Wavelet Decomposition With Deep Neural Network, Dagimawi Eneyew, Miriam A M Capretz, Girma Bitsuamlak, London Hydro

Electrical and Computer Engineering Publications

Electricity consumption is accelerating due to economic and population growth. Hence, energy consumption prediction is becoming vital for overall consumption management and infrastructure planning. Recent advances in smart electric meter technology are making high-resolution energy consumption data available. However, many parameters influencing energy consumption are not typically monitored for residential buildings. Therefore, this study’s main objective is to develop a data-driven energy consumption forecasting model (next-hour consumption) for residential houses solely based on analyzing electricity consumption data. This research proposes a deep neural network architecture that combines stationary wavelet transform features and convolutional neural networks. The proposed approach utilizes automatically …

Inverse Mapping Of Generative Adversarial Networks, Nicky Bayat

Electronic Thesis and Dissertation Repository

Generative adversarial networks (GANs) synthesize realistic samples (image, audio, video, etc.) from a random latent vector. While many studies have explored various training configurations and architectures for GANs, the problem of inverting a generative model to extract latent vectors of given input images/audio has been inadequately investigated. Although there is exactly one generated output per given random vector, the mapping from an image/audio to its recovered latent vector can have more than one solution. We train a deep residual neural network (ResNet18) architecture to recover a latent vector for a given target that can be used to generate a face …

Machine Learning Prediction Of Shear Capacity Of Steel Fiber Reinforced Concrete, Wassim Ben Chaabene

Electronic Thesis and Dissertation Repository

The use of steel fibers for concrete reinforcement has been growing in recent years owing to the improved shear strength and post-cracking toughness imparted by fiber inclusion. Yet, there is still lack of design provisions for steel fiber-reinforced concrete (SFRC) in building codes. This is mainly due to the complex shear transfer mechanism in SFRC. Existing empirical equations for SFRC shear strength have been developed with relatively limited data examples, making their accuracy restricted to specific ranges. To overcome this drawback, the present study suggests novel machine learning models based on artificial neural network (ANN) and genetic programming (GP) to …

A New Approach For Homomorphic Encryption With Secure Function Evaluation On Genomic Data, Mounika Pratapa

Electronic Thesis and Dissertation Repository

Additively homomorphic encryption is a public-key primitive allowing a sum to be computed on encrypted values. Although limited in functionality, additive schemes have been an essential tool in the private function evaluation toolbox for decades. They are typically faster and more straightforward to implement relative to their fully homomorphic counterparts, and more efficient than garbled circuits in certain applications. This thesis presents a novel method for extending the functionality of additively homomorphic encryption to allow the private evaluation of functions of restricted domain. Provided the encrypted sum falls within the restricted domain, the function can be homomorphically evaluated “for free” …

Ontology-Driven Semantic Data Integration In Open Environment, Islam M. Ali

Electronic Thesis and Dissertation Repository

Collaborative intelligence in the context of information management can be defined as "A shared intelligence that results from the collaboration between various information systems". In open environments, these collaborating information systems can be heterogeneous, dynamic and loosely-coupled. Information systems in open environment can also possess a certain degree of autonomy. The integration of data residing in various heterogeneous information systems is essential in order to drive the intelligence efficiently and accurately. Because of the heterogeneous, loosely-coupled, and dynamic nature of open environment, the integration between these information systems in the data level is not efficient. Several approaches and models have …

Terramechanics And Machine Learning For The Characterization Of Terrain, Bryan W. Southwell

Electronic Thesis and Dissertation Repository

An instrumented rover wheel can collect vast amounts of data about a planetary surface. Planetary surfaces are changed by complex geological processes which can be better understood with an abundance of surface data and the use of terramechanics. Identifying terrain parameters such as cohesion and angle of friction hold importance for both the rover driver and the planetary scientist. Knowledge of terrain characteristics can warn of unsafe terrain and flag potential interesting scientific sites. The instrumented wheel in this research utilizes a pressure pad to sense load and sinkage, a string potentiometer to measure slip, and records motor current draw. …

Optimized Machine Learning Models Towards Intelligent Systems, Mohammadnoor Ahmad Mohammad Injadat

Electronic Thesis and Dissertation Repository

The rapid growth of the Internet and related technologies has led to the collection of large amounts of data by individuals, organizations, and society in general [1]. However, this often leads to information overload which occurs when the amount of input (e.g. data) a human is trying to process exceeds their cognitive capacities [2]. Machine learning (ML) has been proposed as one potential methodology capable of extracting useful information from large sets of data [1]. This thesis focuses on two applications. The first is education, namely e-Learning environments. Within this field, this thesis proposes different optimized ML ensemble models to …

Noisy Importance Sampling Actor-Critic: An Off-Policy Actor-Critic With Experience Replay, Miriam A M Capretz, Norman Tasfi

Electrical and Computer Engineering Publications

This paper presents Noisy Importance Sampling Actor-Critic (NISAC), a set of empirically validated modifications to the advantage actor-critic algorithm (A2C), allowing off-policy reinforcement learning and increased performance. NISAC uses additive action space noise, aggressive truncation of importance sample weights, and large batch sizes. We see that additive noise drastically changes how off-sample experience is weighted for policy updates. The modified algorithm achieves an increase in convergence speed and sample efficiency compared to both the on-policy actor-critic A2C and the importance weighted off-policy actor-critic algorithm. In comparison to state-of-the-art (SOTA) methods, such as actor-critic with experience replay (ACER), NISAC nears the …

Edge-Cloud Iot Data Analytics: Intelligence At The Edge With Deep Learning, Ananda Mohon M. Ghosh

Electronic Thesis and Dissertation Repository

Rapid growth in numbers of connected devices, including sensors, mobile, wearable, and other Internet of Things (IoT) devices, is creating an explosion of data that are moving across the network. To carry out machine learning (ML), IoT data are typically transferred to the cloud or another centralized system for storage and processing; however, this causes latencies and increases network traffic. Edge computing has the potential to remedy those issues by moving computation closer to the network edge and data sources. On the other hand, edge computing is limited in terms of computational power and thus is not well suited for …

A Blockchain Approach To Social Responsibility, Augusto Bedin, Wander Queiroz, Miriam A M Capretz, London Hydro

Electrical and Computer Engineering Publications

As blockchain technology matures, more sophisticated solutions arise regarding complex problems. Blockchain continues to spread towards various niches such as government, IoT, energy, and environmental industries. One often overlooked opportunity for blockchain is the social responsibility sector. Presented in this paper is a permissioned blockchain model that enables enterprises to come together and cooperate to optimize their environmental and societal impacts. This is made possible through a private or permissioned blockchain. Permissioned blockchains are blockchain networks where all the participants are known and trust relationships among them can be fostered more smoothly. An example of what a permissioned blockchain would …

A Lightweight Magnetorheological Actuator Using Hybrid Magnetization, Masoud Moghani, Mehrdad Kermani Ph.D., P.Eng.

Electrical and Computer Engineering Publications

Copyright © 2020, IEEE

This paper presents the design and validation of a lightweight Magneto-Rheological (MR) clutch, called Hybrid Magneto-Rheological (HMR) clutch. The clutch utilizes a hybrid magnetization using an electromagnetic coil and a permanent magnet. The electromagnetic coil can adjust the magnetic field
generated by the permanent magnet to a desired value, and fully control the transmitted torque. To achieve the maximum torque to mass ratio, the design of HMR clutch is formulated as a multiobjective optimization problem with three design objectives, namely the transmitted torque, the mass of the clutch, and the
magnetic field strength within the clutch …

Geometric State Observers For Autonomous Navigation Systems, Miaomiao Wang

Electronic Thesis and Dissertation Repository

The development of reliable state estimation algorithms for autonomous navigation systems is of great interest in the control and robotics communities. This thesis studies the state estimation problem for autonomous navigation systems. The first part of this thesis is devoted to the pose estimation on the Special Euclidean group $\SE(3)$. A generic globally exponentially stable hybrid estimation scheme for pose (orientation and position) and velocity-bias estimation on $\SE(3)\times \mathbb{R}^6$ is proposed. Moreover, an explicit hybrid observer, using inertial and landmark position measurements, is provided.

The second part of this thesis is devoted to the problem of simultaneous estimation of the …

A Visual Analytics System For Making Sense Of Real-Time Twitter Streams, Amir Haghighatimaleki

Electronic Thesis and Dissertation Repository

Through social media platforms, massive amounts of data are being produced. Twitter, as one such platform, enables users to post “tweets” on an unprecedented scale. Once analyzed by machine learning (ML) techniques and in aggregate, Twitter data can be an invaluable resource for gaining insight. However, when applied to real-time data streams, due to covariate shifts in the data (i.e., changes in the distributions of the inputs of ML algorithms), existing ML approaches result in different types of biases and provide uncertain outputs. This thesis describes a visual analytics system (i.e., a tool that combines data visualization, human-data interaction, and …