Engineering Commons^™

Design And Implementation Of Uvm-Based Verification Framework For Deep Learning Accelerators, Randa Ahmed Hussein Aboudeif

Theses and Dissertations

Recent advancements in deep learning (DL) have made hardware accelerators, known as deep learning accelerators (DLAs), a preferred solution for numerous high-performance computing (HPC) applications, including speech recognition, computer vision, and image classification. DLAs are composed of hundreds of parallel processing engines to speed up computations and can gain access to pre-trained networks from the cloud or through on-chip memory to implement the DNN inference process. DLA verification is becoming an important and challenging phase. The verification process is required to handle the complex DLA design. Moreover, the reliability of DLAs is critical for assessment as they are involved in …

Efficient Connectivity Management And Path Planning For Iot And Uav Networks, Amirahmad Chapnevis

Theses and Dissertations

This dissertation explores how to better manage resources in mobile networks, especially for enhancing the performance of Unmanned Aerial Vehicles (UAV)-supported IoT networks. We explored ways to set up a flexible communication architecture that can handle large IoT deployments by making good use of mobile core network resources like bearers and data paths. We developed strategies that meet the needs of IoT networks and enhance network performance. We also developed and tested a system that combines traffic from several mobile devices that use the same user identity and network resources within the core mobile network. We used everyday smartphones, SIM …

Exploring End-User Environments For The Control And Programming Of Collaborative Robots, Luiz Felipe Fronchetti Dias

Theses and Dissertations

To collaborate with the ongoing development of robotics, this thesis highlights three research contributions to collaborative robot programming. The first study evaluates block-based programming as an alternative for two-armed robots. A commercial solution is put in contrast with a block-based programming language. Both programming solutions are evaluated by 52 participants in an experiment involving a pick-and-place task. This study brings insights into human-robot collaboration, including robot positioning and interaction challenges. The second study discusses using mixed-reality devices as a potential workaround to the manual positioning of industrial and collaborative robots. Five different control interfaces implemented in mixed reality were used …

Qasm-To-Hls: A Framework For Accelerating Quantum Circuit Emulation On High-Performance Reconfigurable Computers, Anshul Maurya

Theses and Dissertations

High-performance reconfigurable computers (HPRCs) make use of Field-Programmable Gate Arrays (FPGAs) for efficient emulation of quantum algorithms. Generally, algorithm-specific architectures are implemented on the FPGAs and there is very little flexibility. Moreover, mapping a quantum algorithm onto its equivalent FPGA emulation architecture is challenging. In this work, we present an automation framework for converting quantum circuits to their equivalent FPGA emulation architectures. The framework processes quantum circuits represented in Quantum Assembly Language (QASM) and derives high-level descriptions of the hardware emulation architectures for High-Level Synthesis (HLS) on HPRCs. The framework generates the code for a heterogeneous architecture consisting of a …

Wifi Sensing At The Edge Towards Scalable On-Device Wireless Sensing Systems, Steven M. Hernandez

Theses and Dissertations

WiFi sensing offers a powerful method for tracking physical activities using the radio-frequency signals already found throughout our homes and offices. This novel sensing modality offers continuous and non-intrusive activity tracking since sensing can be performed (i) without requiring wearable sensors, (ii) outside the line-of-sight, and even (iii) through the wall. Furthermore, WiFi has become a ubiquitous technology in our computers, our smartphones, and even in low-cost Internet of Things devices. In this work, we consider how the ubiquity of these low-cost WiFi devices offer an unparalleled opportunity for improving the scalability of wireless sensing systems. Thus far, WiFi sensing …

Improving The Performance, Energy Efficiency And Security Of Gpus, Xin Wang

Theses and Dissertations

The work in this dissertation achieves to enhance the performance, energy-efficiency, and security of the GPUs. We noticed that, as the demand of hardware resources keeps rising in GPUs, the energy consumption becomes unaffordable and places barriers for further performance boost. To resolve this issue, we have proposed several novel GPU micro-architectures that are able to assist the GPUs to execute in an energy-efficient manner. They also provide the potential for further performance enhancement in GPUs. Firstly, we proposed a GPU register packing scheme that stores multiple narrow-width operands to a single register to save register file resources. The unoccupied …

Analyzing Microarchitectural Residue In Various Privilege Strata To Identify Computing Tasks, Tor J. Langehaug

Theses and Dissertations

Modern multi-tasking computer systems run numerous applications simultaneously. These applications must share hardware resources including the Central Processing Unit (CPU) and memory while maximizing each application’s performance. Tasks executing in this shared environment leave residue which should not reveal information. This dissertation applies machine learning and statistical analysis to evaluate task residue as footprints which can be correlated to identify tasks. The concept of privilege strata, drawn from an analogy with physical geology, organizes the investigation into the User, Operating System, and Hardware privilege strata. In the User Stratum, an adversary perspective is taken to build an interrogator program that …

Infiniband Network Monitoring: Challenges And Possibilities, Kyle D. Hintze

Theses and Dissertations

Within the realm of High Performance Computing, the InfiniBand Architecture is among the leading interconnects used today. Capable of providing high bandwidth and low latency, InfiniBand is finding applications outside the High Performance Computing domain. One of these is critical infrastructure, encompassing almost all essential sectors as the work force becomes more connected. InfiniBand is not immune to security risks, as prior research has shown that common traffic analyzing tools cannot effectively monitor InfiniBand traffic transmitted between hosts, due to the kernel bypass nature of the IBA in conjunction with Remote Direct Memory Access operations. If Remote Direct Memory Access …

A General Framework For Characterizing And Evaluating Attacker Models For Cps Security Assessment, Christopher S. Deloglos, Christopher Deloglos

Theses and Dissertations

Characterizing the attacker’s perspective is essential to assessing the security posture and resilience of cyber-physical systems. The attacker’s perspective is most often achieved by cyber-security experts (e.g., red teams) who critically challenge and analyze the system from an adversarial stance. Unfortunately, the knowledge and experience of cyber-security experts can be inconsistent leading to situations where there are gaps in the security assessment of a given system. Structured security review processes (such as TAM, Mission Aware, STPA-SEC, and STPA-SafeSec) attempt to standardize the review processes to impart consistency across an organization or application domain. However, with most security review processes, the …

Dbknot: A Transparent And Seamless, Pluggable Tamper Evident Database, Islam Khalil

Theses and Dissertations

Database integrity is crucial to organizations that rely on databases of important data. They suffer from the vulnerability to internal fraud. Database tampering by internal malicious employees with high technical authorization to their infrastructure or even compromised by externals is one of the important attack vectors.

This thesis addresses such challenge in a class of problems where data is appended only and is immutable. Examples of operations where data does not change is a) financial institutions (banks, accounting systems, stock market, etc., b) registries and notary systems where important data is kept but is never subject to change, and c) …

A Study Of Execution Performance For Rust-Based Object Vs Data Oriented Architectures, Joseph A. Vagedes

Theses and Dissertations

To investigate the Data-Oriented Design (DOD) paradigm, in particular, an architecture built off its principles: the Entity-Component-System (ECS). ECS is commonly used by video game engines due to its ability to store data in a way that is optimal for the cache to access. Additionally, the structure of this paradigm produces a code-base that is simple to parallelize as the workload can be distributed across a thread-pool based on the data used with little to no need for data safety measures such as mutexes and locks. A final benefit, although not easily measured, is that the DOD paradigm produces a …

Systematic Model-Based Design Assurance And Property-Based Fault Injection For Safety Critical Digital Systems, Athira Varma Jayakumar

Theses and Dissertations

With advances in sensing, wireless communications, computing, control, and automation technologies, we are witnessing the rapid uptake of Cyber-Physical Systems across many applications including connected vehicles, healthcare, energy, manufacturing, smart homes etc. Many of these applications are safety-critical in nature and they depend on the correct and safe execution of software and hardware that are intrinsically subject to faults. These faults can be design faults (Software Faults, Specification faults, etc.) or physically occurring faults (hardware failures, Single-event-upsets, etc.). Both types of faults must be addressed during the design and development of these critical systems. Several safety-critical industries have widely adopted …

A Quantitative Analysis Of The Fusion Of 3-D Scanning Lidar Systems And 2-D Imaging Systems, Michael F. Milton Jr.

Theses and Dissertations

This research will demonstrate the feasibility of fusing the superior spatial resolution of a 2-D imaging system with the precise range to target information of a 3-D imaging system to create a LIDAR imaging system that can accurately find what and where a target is. The 3-D imaging system will use a scanning method as opposed to a flash method that has been used in similar research. The goal of this research is to improve performance of scanning LIDAR so it has better spatial resolution. The research in this thesis proves that incorporating 2-D imaging data into 3-D scanning LIDAR …

A Multi-Vehicle Cooperative Localization Approach For An Autonomy Framework, Edwin A. Mora

Theses and Dissertations

Offensive techniques produced by technological advancement present opportunities for adversaries to threaten the operational advantages of our joint and allied forces. Combating these new methodologies requires continuous and rapid development towards our own set of \game-changing" technologies. Through focused development of unmanned systems and autonomy, the Air Force can strive to maintain its technological superiority. Furthermore, creating a robust framework capable of testing and evaluating the principles that define autonomy allows for the exploration of future capabilities. This research presents development towards a hybrid reactive/deliberative architecture that will allow for the testing of the principles of task, cognitive, and peer …

A Framework For Cyber Vulnerability Assessments Of Infiniband Networks, Daryl W. Schmitt

Theses and Dissertations

InfiniBand is a popular Input/Output interconnect technology used in High Performance Computing clusters. It is employed in over a quarter of the world’s 500 fastest computer systems. Although it was created to provide extremely low network latency with a high Quality of Service, the cybersecurity aspects of InfiniBand have yet to be thoroughly investigated. The InfiniBand Architecture was designed as a data center technology, logically separated from the Internet, so defensive mechanisms such as packet encryption were not implemented. Cyber communities do not appear to have taken an interest in InfiniBand, but that is likely to change as attackers branch …

Unresolved Object Detection Using Synthetic Data Generation And Artificial Neural Networks, Yong U. Sinn

Theses and Dissertations

This research presents and solves constrained real-world problems of using synthetic data to train artificial neural networks (ANNs) to detect unresolved moving objects in wide field of view (WFOV) electro-optical/infrared (EO/IR) satellite motion imagery. Objectives include demonstrating the use of the Air Force Institute of Technology (AFIT) Sensor and Scene Emulation Tool (ASSET) as an effective tool for generating EO/IR motion imagery representative of real WFOV sensors and describing the ANN architectures, training, and testing results obtained. Deep learning using a 3-D convolutional neural network (3D ConvNet), long short term memory (LSTM) network, and U-Net are used to solve the …

A Stochastic Game Theoretical Model For Cyber Security, Michael T. Larkin

Theses and Dissertations

The resiliency of systems integrated through cyber networks is of utmost importance due to the reliance on these systems for critical services such as industrial control systems, nuclear production, and military weapons systems. Current research in cyber resiliency remains largely limited to methodologies utilizing a singular technique that is predominantly theoretical with limited examples given. This research uses notional data in presenting a novel approach to cyber system analysis and network resource allocation by leveraging multiple techniques including game theory, stochastic processes, and mathematical programming. An operational network security problem consisting of 20 tactical normal form games provides an assessment …

Energy Efficient Spintronic Device For Neuromorphic Computation, Md Ali Azam

Theses and Dissertations

Future computing will require significant development in new computing device paradigms. This is motivated by CMOS devices reaching their technological limits, the need for non-Von Neumann architectures as well as the energy constraints of wearable technologies and embedded processors. The first device proposal, an energy-efficient voltage-controlled domain wall device for implementing an artificial neuron and synapse is analyzed using micromagnetic modeling. By controlling the domain wall motion utilizing spin transfer or spin orbit torques in association with voltage generated strain control of perpendicular magnetic anisotropy in the presence of Dzyaloshinskii-Moriya interaction (DMI), different positions of the domain wall are realized …

Cyber Data Anomaly Detection Using Autoencoder Neural Networks, Spencer A. Butt

Theses and Dissertations

The Department of Defense requires a secure presence in the cyber domain to successfully execute its stated mission of deterring war and protecting the security of the United States. With potentially millions of logged network events occurring on defended networks daily, a limited staff of cyber analysts require the capability to identify novel network actions for security adjudication. The detection methodology proposed uses an autoencoder neural network optimized via design of experiments for the identification of anomalous network events. Once trained, each logged network event is analyzed by the neural network and assigned an outlier score. The network events with …

Digital Forensics Event Graph Reconstruction, Daniel J. Schelkoph

Theses and Dissertations

Ontological data representation and data normalization can provide a structured way to correlate digital artifacts. This can reduce the amount of data that a forensics examiner needs to process in order to understand the sequence of events that happened on the system. However, ontology processing suffers from large disk consumption and a high computational cost. This paper presents Property Graph Event Reconstruction (PGER), a novel data normalization and event correlation system that leverages a native graph database to improve the speed of queries common in ontological data. PGER reduces the processing time of event correlation grammars and maintains accuracy over …

An Analysis Of Multi-Domain Command And Control And The Development Of Software Solutions Through Devops Toolsets And Practices, Mason R. Bruza

Theses and Dissertations

Multi-Domain Command and Control (MDC2) is the exercise of command and control over forces in multiple operational domains (namely air, land, sea, space, and cyberspace) in order to produce synergistic effects in the battlespace, and enhancing this capability has become a major focus area for the United States Air Force (USAF). In order to meet demands for MDC2 software, solutions need to be acquired and/or developed in a timely manner, information technology infrastructure needs to be adaptable to new software requirements, and user feedback needs to drive iterative updates to fielded software. In commercial organizations, agile software development methodologies and …

Sentiment Analysis Of Twitter Data, Evan L. Munson

Theses and Dissertations

The rapid expansion and acceptance of social media has opened doors into users’ opinions and perceptions that were never as accessible as they are with today's prevalence of mobile technology. Harvested data, analyzed for opinions and sentiment can provide powerful insight into a population. This research utilizes Twitter data due to its widespread global use, in order to examine the sentiment associated with tweets. An approach utilizing Twitter #hashtags and Latent Dirichlet Allocation topic modeling were utilized to differentiate between tweet topics. A lexicographical dictionary was then utilized to classify sentiment. This method provides a framework for an analyst to …

Toward Biologically-Inspired Self-Healing, Resilient Architectures For Digital Instrumentation And Control Systems And Embedded Devices, Shawkat Sabah Khairullah

Theses and Dissertations

Digital Instrumentation and Control (I&C) systems in safety-related applications of next generation industrial automation systems require high levels of resilience against different fault classes. One of the more essential concepts for achieving this goal is the notion of resilient and survivable digital I&C systems. In recent years, self-healing concepts based on biological physiology have received attention for the design of robust digital systems. However, many of these approaches have not been architected from the outset with safety in mind, nor have they been targeted for the automation community where a significant need exists. This dissertation presents a new self-healing digital …

Improving The Performance And Energy Efficiency Of Emerging Memory Systems, Yuhua Guo

Theses and Dissertations

Modern main memory is primarily built using dynamic random access memory (DRAM) chips. As DRAM chip scales to higher density, there are mainly three problems that impede DRAM scalability and performance improvement. First, DRAM refresh overhead grows from negligible to severe, which limits DRAM scalability and causes performance degradation. Second, although memory capacity has increased dramatically in past decade, memory bandwidth has not kept pace with CPU performance scaling, which has led to the memory wall problem. Third, DRAM dissipates considerable power and has been reported to account for as much as 40% of the total system energy and this …

A Framework For Understanding, Prioritizing, And Applying Systems Security Engineering Processes, Activities, And Tasks, Stephen Khou

Theses and Dissertations

Current systems security practices lack an effective approach to prioritize and tailor systems security efforts to develop and field secure systems in challenging operational environments, which results in business and mission stakeholders becoming more susceptible to an array of disruptive events. This work informs Systems Engineers on recent developments in the field of system security engineering and provides a framework for more fully understanding the application of Systems Security Engineering (SSE) processes, activities, and tasks as described in the recently released National Institute of Standards and Technology (NIST) Special Publication 800-160. This SSE framework uniquely offers a repeatable and tailorable …

Development Of The Architecture Cost Effectiveness Framework And Application To Open Systems Architectures, Donald A. Barrett

Theses and Dissertations

The Modular Open System Approach (MOSA) is an initiative to, among other things, reduce cost and schedule for acquisitions programs. While programs have experienced savings using MOSA, the majority of programs have not, in part due to a lack of a logical method for evaluating architecture alternatives. This research develops the Architecture Cost Effectiveness Framework (ACEF), which combines multi-attribute utility analysis with cost effectiveness analysis throughout the program lifecycle. Step 1 is the establishment of a business strategy that includes an Attribute Hierarchy of selected operational requirements. The business strategy also defines Lifecycle Utility Reference Profiles (LURPs) that document changing …

Improving The Performance And Time-Predictability Of Gpus, Yijie Huangfu

Theses and Dissertations

Graphic Processing Units (GPUs) are originally mainly designed to accelerate graphic applications. Now the capability of GPUs to accelerate applications that can be parallelized into a massive number of threads makes GPUs the ideal accelerator for boosting the performance of such kind of general-purpose applications. Meanwhile it is also very promising to apply GPUs to embedded and real-time applications as well, where high throughput and intensive computation are also needed.

However, due to the different architecture and programming model of GPUs, how to fully utilize the advanced architectural features of GPUs to boost the performance and how to analyze the …

Towards Design And Analysis For High-Performance And Reliable Ssds, Qianbin Xia

Theses and Dissertations

NAND Flash-based Solid State Disks have many attractive technical merits, such as low power consumption, light weight, shock resistance, sustainability of hotter operation regimes, and extraordinarily high performance for random read access, which makes SSDs immensely popular and be widely employed in different types of environments including portable devices, personal computers, large data centers, and distributed data systems.

However, current SSDs still suffer from several critical inherent limitations, such as the inability of in-place-update, asymmetric read and write performance, slow garbage collection processes, limited endurance, and degraded write performance with the adoption of MLC and TLC techniques. To alleviate these …

Mitigating Interference During Virtual Machine Live Migration Through Storage Offloading, Morgan S. Stuart

Theses and Dissertations

Today's cloud landscape has evolved computing infrastructure into a dynamic, high utilization, service-oriented paradigm. This shift has enabled the commoditization of large-scale storage and distributed computation, allowing engineers to tackle previously untenable problems without large upfront investment. A key enabler of flexibility in the cloud is the ability to transfer running virtual machines across subnets or even datacenters using live migration. However, live migration can be a costly process, one that has the potential to interfere with other applications not involved with the migration. This work investigates storage interference through experimentation with real-world systems and well-established benchmarks. In order to …

A Reused Distance Based Analysis And Optimization For Gpu Cache, Dongwei Wang

Theses and Dissertations

As a throughput-oriented device, Graphics Processing Unit(GPU) has already integrated with cache, which is similar to CPU cores. However, the applications in GPGPU computing exhibit distinct memory access patterns. Normally, the cache, in GPU cores, suffers from threads contention and resources over-utilization, whereas few detailed works excavate the root of this phenomenon. In this work, we adequately analyze the memory accesses from twenty benchmarks based on reuse distance theory and quantify their patterns. Additionally, we discuss the optimization suggestions, and implement a Bypassing Aware(BA) Cache which could intellectually bypass the thrashing-prone candidates.

BA cache is a cost efficient cache design …