SoC Theses

Permanent URI for this collection


Recent Submissions

Now showing 1 - 5 of 13
  • Item
    Information-theoretic secret sharing: Fundamental limits and coding schemes via deep learning
    (Wichita State University, 2023-12) Rana, Vidhi; Chou, Rémi
    This dissertation aims to study and design coding schemes for information-theoretic security. We focus on two models: the secret sharing model and the Gaussian wiretap channel model. The main contribution of this dissertation is to take practical constraints into account. We consider a rate- limited public communication channel to account for bandwidth constraints and finite block length for practical applications requiring short packet length or low latency.
  • Item
    Computer vision techniques for natural hazard detection
    (Wichita State University, 2023-12) Haridasan, Smitha; Rattani, Ajita
    The escalating frequency and magnitude of wildfires, exacerbated by climate change, present a formidable environmental challenge with devastating consequences. This dissertation addresses the imperative need for early wildfire detection using multi-spectral deep learning models and explores the intersection of social media data analytics for enhanced disaster response. The first chapter introduces a novel multi-spectral deep learning model, leveraging diverse spectral information for improved accuracy in forest fire detection. Utilizing a heterogeneous dataset, the proposed model outperforms single-spectrum models by 1.9% and 14.8% in test and challenge sets, respectively, showcasing its efficacy in challenging environments. Recognizing the pivotal role of social media in disaster reporting, the second chapter delves into the analysis of multi-modal Twitter datasets from natural disasters. We present a fusion-based decision-making technique that surpasses baseline models, achieving a 6.98% improvement in informative tweet classification and an 11.2% enhancement in humanitarian categorization. The third chapter underscores the impact of wildfires on ecosystems and human communities, emphasizing the significance of deep learning in real-time monitoring. Deploying lightweight deep-learning architectures on drones and satellites, our survey elucidates the transformative potential of these technologies in early wildfire detection, offering precise, real-time information crucial for decision-making and resource allocation. In response to computational challenges in real-time applications, the fourth chapter proposes a lightweight model for wildfire detection on unmanned aerial vehicles (UAVs). Evaluating three neural architecture search (NAS) methods, MNAS, RNAS, and BONAS, we identify efficient techniques for optimizing model size, speed, and accuracy. The proposed RNAS model achieves 84.38% accuracy with a remarkable compression ratio of 2.47 × 10−6 compared to the ResNet50 model. In conclusion, our comprehensive review and proposed lightweight model contribute to advancing the state-of-the-art in wildfire detection. The synthesis of multi-spectral deep learning, social media analytics, and lightweight UAV models offers a holistic approach to mitigating the ecological, social, and economic impacts of wildfires.
  • Item
    Harnessing unlabeled data for improving generalization of deep learning methods
    (Wichita State University, 2023-07) Shanmugasundaram, Deepika; Rattani, Ajita
    Recent advancements in Deep Learning, Artificial Intelligence, and Computer Vision have reached a critical stage, enabling researchers to explore the automatic extraction of individual demographic traits, known as soft-biometrics. This research aims to leverage unlabeled data in predicting soft-biometric traits, such as gender and age, using deep learning models. The objective is to develop a model that can accurately classify these traits by utilizing semi-supervised methods that rely on a limited amount of labeled data and a vast amount of unlabeled data. While unlabeled data may initially seem devoid of crucial information, this thesis explores how it can be effectively used to enhance classification accuracy, especially in scenarios where labeled data is scarce. This study evaluated the accuracy of different image classification models on the Celeb-A and NIR-VIS datasets using co-training, mix-up procedure, knowledge distillation, and blind distillation techniques. The results showed that incorporating these methods led to improvements in accuracy across both datasets and various attributes such as gender classification and smiling classification. Exploring the combined use of different techniques and investigating their synergistic effects could lead to further accuracy improvements. Evaluating the models on larger and more diverse datasets, analyzing their generalization capabilities, optimizing hyperparameters and architectures, and applying the techniques to other computer vision tasks were also identified as areas for future research.
  • Item
    Distributed misbehavior detection in UAV flocks
    (Wichita State University, 2023-07) Aguida, Mohamed Anis; Monroy, Sergio A.Salinas
    Unmanned aerial vehicles have become increasingly popular in many applications such as remote surveillance, reconnaissance, and precision agriculture. Often multiple UAVs form a swarm and perform their operations in a distributed, coordinated fashion. A rogue UAV in the flock can negatively disrupt the expected behavior and may jeopardize the objective of the mission, leading other UAVs to make incorrect decisions or even crash. This work introduces GRIFFIN, a distributed and lightweight misbehavior detection framework for UAV flocks. GRIFFIN relies on readily available packet metadata (e.g., GPS coordinates) and signal characteristics (e.g., RSSI measurements) and detects malicious UAVs by employing a “majority voting” protocol. We show that GRIFFIN requires only three honest nodes for correct operations. We implement and evaluate GRIFFIN on (a) a realistic UAV simulator (ArduSim) and (b) a Raspberry Pi+Navio-based drone testbed. We find that GRIFFIN outputs 100% successful detection with zero false negatives as long as less than half of the UAVs in the flock are not compromised. Our implementation on the real UAV testbed shows that runtime overhead of GRIFFIN is minimal (i.e., it requires less than 1 MB of memory and consumes less than 1% of CPU) and computes operation within 2:5 ms.
  • Item
    Private neural network auctions for additive manufacturing
    (Wichita State University, 2023-05) Shukla, Amey; Monroy, Sergio A.Salinas
    Additive Manufacturing is changing the way we construct, deliver, and consume objects by allowing customers to quickly build custom objects on-demand and in locations near to them. Additive manufacturing operators need to optimally control the manufacturing process, monitor tasks in real-time and set the prices for their built objects. Moreover, operators need to protect private data related to the prices paid by customers and their purchased objects. Existing works focus on either optimal control, real-time monitoring, price setting, or privacy, missing the advantages of jointly addressing them. Existing works also require vast computational resources to accomplish only one of these tasks. To address these issues, we develop a differentially-private distributed neural network auction that optimally allocates manufacturing resources and sets prices in a way that maximizes the profit of the manufacturer. The auction protects the privacy of the customers’ bids. Moreover, to reduce the computing time, we design a parallel computation algorithm for the neural network that is executed by a cluster of edge computing devices. We evaluate the proposed neural network through extensive simulations. We observe that it can jointly perform the operators’ tasks while maintaining the privacy of the customers. Our simulations also show that the parallel algorithm reduces the execution time.