Computer engineering

Mobility monitoring in urban environments can provide valuable insights into pedestrian and vehicle movement – where people want to go, how they get there, and the challenges they face along the way. Today, local governments can automate the acquisition of such data using video surveillance to understand the potential impact of investment and policy decisions. However, public disapproval of computer vision due to privacy concerns opens opportunities for research into alternative tools built with privacy constraints at the core of the design. WiFi sensing emerges as a promising solution. Modern mobile devices ubiquitously support the 802.11 standard and regularly emit WiFi probe requests for network discovery. We can passively monitor this traffic to estimate the levels of congestion in public spaces.
In this dissertation, we address three fundamental research problems pertaining to developing streetscape-scale mobility intelligence: scalable infrastructure for WiFi signal capture, passive device localization, and device re-identification.

Every passenger vehicle must rely on a safe and optimal trajectory to eliminate traffic incidents and congestion as well as to reduce environmental impact, and travel time. Autonomous intersection management systems (AIMS) enable large scale optimization of vehicular trajectories with connected and autonomous vehicles (CAVs). The first contribution of this dissertation is the fastest trajectory planner (FTP) method which is geared for computing the fastest waypoint trajectories via performing graph search over a discretized space-time (ST) graph (Gt), thereby constructing collision-free space-time trajectories with variable vehicular speeds adhering to traffic rules and dynamical constraints of vehicles. The benefits of navigating a connected and autonomous vehicle (CAV) truly capture effective collaboration between every CAV during the trajectory planning step. This requires addressing trajectory planning activity along with vehicular networking in the design phase. For complementing the proposed FTP method in decentralized scenarios, the second contribution of this dissertation is an application layer V2V solution using a coordinator-based distributed trajectory planning method which elects a single leader CAV among all the collaborating CAVs without requiring a centralized infrastructure. The leader vehicular agent calculates and assigns a trajectory for each node CAV over the vehicular network for the collision-free management of an unsignalized road intersection. The proposed FTP method is tested in a simulated road intersection scenario for carrying out trials on scheduling efficiency and algorithm runtime. The resulting trajectories allow high levels of intersection sharing, high evacuation rate, with a low algorithm single-threaded runtime figures even with large scenarios of up to 1200 vehicles, surpassing comparable systems.

This dissertation explores one-class classification (OCC) in the context of big data and fraud detection, addressing challenges posed by imbalanced datasets. A detailed survey of OCC-related literature forms a core part of the study, categorizing works into outlier detection, novelty detection, and deep learning applications. This survey reveals a gap in the application of OCC to the inherent problems of big data, such as class rarity and noisy data. Building upon the foundational insights gained from the comprehensive literature review on OCC, the dissertation progresses to a detailed comparative analysis between OCC and binary classification methods. This comparison is pivotal in understanding their respective strengths and limitations across various applications, emphasizing their roles in addressing imbalanced datasets. The research then specifically evaluates binary and OCC using credit card fraud data. This practical application highlights the nuances and effectiveness of these classification methods in real-world scenarios, offering insights into their performance in detecting fraudulent activities. After the evaluation of binary and OCC using credit card fraud data, the dissertation extends this inquiry with a detailed investigation into the effectiveness of both methodologies in fraud detection. This extended analysis involves utilizing not only the Credit Card Fraud Detection Dataset but also the Medicare Part D dataset. The findings show the comparative performance and suitability of these classification methods in practical fraud detection scenarios. Finally, the dissertation examines the impact of training OCC algorithms on majority versus minority classes, using the two previously mentioned datasets in addition to Medicare Part B and Durable Medical Equipment, Prosthetics, Orthotics and Supplies (DMEPOS) datasets. This exploration offers critical insights into model training strategies and their implications, suggesting that training on the majority class can often lead to more robust classification results. In summary, this dissertation provides a deep understanding of OCC, effectively bridging theoretical concepts with novel applications in big data and fraud detection. It contributes to the field by offering a comprehensive analysis of OCC methodologies, their practical implications, and their effectiveness in addressing class imbalance in big data.

This sturty presents a method for automatically generating and
animating 3-D models of planar linkages. A computer program called
ANIMEC is introduced which serves as a link between two existing
programs KAPCA, and MOVIE. The algorithms of ANIMEC are described in
detail and a program listing is provided.