Naval Postgraduate School
Calhoun, Institutional Archive of the Naval Postgraduate SchoolNot a member yet
71232 research outputs found
Sort by
EXPERIMENTAL TESTING OF SPIRAL AND WAFFLE PROTOTYPES OF 3D-PRINTED ARTIFICIAL MUSCLES
No full textArtificial muscles are essential for emerging technologies such as robotic limbs, exoskeletons, and underwater unmanned vehicles (UUVs), which require lightweight, flexible, and efficient actuators. Traditional systems like electric motors and hydraulics often cannot meet these demands. This study aimed to develop a low-cost, 3D-printed artificial muscle suitable for practical use. Earlier work established the foundation with COMSOL simulations, initial 3D prints, and basic clearing and wiring procedures. However, the “Potato Spiral” design showed limitations, including low survival rates during clearing, sparking, and lack of visible actuation. A new design, the “Belgian Waffle,” was introduced to overcome these challenges. This thesis experimentally evaluated both designs, focusing on improving clearing survival, identifying the cause of sparking, and testing actuation. A refined clearing method using NaOH-assisted sonication significantly improved survival rates. Capacitance and breakdown voltage tests showed that NaOH absorption was not responsible for sparking. Although no strong contractive force was measured, the Belgian Waffle displayed force behavior similar to biological muscles. Overall, the Belgian Waffle architecture demonstrated better performance and reliability, making it the recommended architecture for the next phase of the project.Distribution Statement A. Approved for public release: Distribution is unlimited.Lieutenant, Taiwan Navyoffice of naval researc
THE GEOPOLITICAL ALIGNMENT OF AZERBAIJAN: A CASE STUDY IN SMALL STATE THEORY
No full textAs a small, former-Soviet state sitting at the fault line of Eurasian politics, Azerbaijan represents a unique case study into how small states behave and how they shape their interests. This thesis discusses Azerbaijan’s geopolitical orientation since the fall of the Soviet Union and aims to answer how economic, political, and military factors have shaped its cooperation with external powers. It identifies three major factors that have impacted Azerbaijan’s geopolitical orientation: oil wealth which has given Azerbaijan the relative economic independence to seek multiple avenues for greater global integration, a growing Russian withdrawal from Azerbaijan's political and military domains, and a shift in perceptions of external powers’ ability to help shape regional politics. This thesis utilizes process tracing to show the development of these trends over time, and provides implications for small state behavior, as well as for other former-Soviet states. Analysis of Azerbaijan’s geopolitical alignment suggests that small states can employ strategies which play to their unique advantages, diversify their partnerships to lessen the influence of a regional or global hegemon, and capitalize on relationships that support their political objectives, if and when external conditions allow a shift in the regional balance of power away from the existing hegemon.Distribution Statement A. Approved for public release: Distribution is unlimited.Lieutenant, United States Nav
ANALYSIS OF EGYPTIAN MINISTRY OF DEFENSE E-WASTE PROCESSES AND MANAGEMENT
No full textThis thesis presents an in-depth analysis of the E-waste management process used by the Egyptian Ministry of Defense (MOD), highlighting opportunities to improve efficiency and economic outcomes. Currently, E-waste is stored in centralized warehouses and sold every three months to private sector buyers, a process that neglects the substantial potential value within discarded electronics. By adopting urban mining—extracting valuable metals such as gold, silver, copper, and palladium from electronic waste—MOD could recover resources internally and enhance sustainability. Many private-sector buyers lack a proper recycling infrastructure, often leading to environmentally harmful disposal practices and the loss of critical materials. Establishing internal recycling facilities would allow MOD to control the full process, increase revenue, and minimize environmental risks associated with improper E-waste handling. Recovered materials could directly support Egypt’s defense industrial base, reducing reliance on imported resources and strengthening national supply chain resilience. This study includes a comparative cost-effectiveness analysis as well as detailed process mapping of both the current and proposed E-waste management systems. The findings show that, although internal recycling requires a higher initial investment, it ultimately delivers greater financial returns, stronger regulatory compliance, improved material flow, and better alignment with national sustainability.Distribution Statement A. Approved for public release: Distribution is unlimited.Colonel, Egypt ArmyCaptain, Egypt Arm
EVALUATING SECURE COMMUNICATION PROTOCOLS FOR RADIO FREQUENCY CHALLENGED TACTICAL ENVIRONMENTS
No full textThis research compares the performance of Messaging Layer Security (MLS) with existing secure communication under emulated tactical communication constraints. Whole-stack testing is critical to accelerating the integration of commercial technologies into military systems. Unfortunately, current validation methods and simulation models often fail to accurately reflect the challenging radio frequency (RF) environments encountered in military operations. To address this gap, this study uses physical hardware and an emulated wireless channel with standardized test configurations to measure message delivery rates over degraded RF links. For each trial, key establishment was performed in advance using MLS or Transport Layer Security (TLS). Message delivery rates were measured while utilizing the Transmission Control Protocol (TCP) and NACK-Oriented Reliable Multicast (NORM) under increasing levels of multipath fading. MLS over TCP achieved 2.17 times greater goodput than Transport Layer Security (TLS) over TCP while offering enhanced security guarantees when sending 1000 byte sized message between two nodes. Moreover, when paired with NORM, MLS outperformed TLS message delivery by up to 99% under heavily degraded RF channel conditions. These findings support the suitability of MLS for mission-critical tactical communications and provide a basis for its consideration in future military communication systems.Distribution Statement A. Approved for public release: Distribution is unlimited.Lieutenant Commander, United States Nav
FIELD OBSERVATIONS OF SEA-SWELL NONLINEARITY ON ROCKY SHORES
No full textThe nonlinear evolution of sea-swell waves across high-friction rocky shores is hypothesized to differ from frictionless theories for sandy coasts. Analyses at Pebble Beach, CA, show that cross-shore energy dissipation is four times greater over rocky seabeds than sandy ones. To study bottom friction’s effect on nonlinear wave transformation, pressure data were collected from June to November 2023, using RBR sensors, and processed into sea surface elevation and higher-order wave statistics. This study examines nonlinear wave behavior at Pebble Beach, using hourly averages of skewness (Sk) and asymmetry (As), compared with data from Duck, NC, which features a mild slope, sandy beach. Contrary to expectations that mild slopes produce more substantial nonlinearities, Pebble Beach showed consistently higher Sk and As. Bispectral analysis confirmed strong nonlinear interactions, especially during narrowband storms, however, both narrow- and broadband events generated significant nonlinearity. A dissipation model applied to offshore data underestimated observed Sk by 7.5%, suggesting nonlinear energy transfers do play a role on rocky shores. Strong bottom friction results in frequency-based dissipation, reducing low-frequency energy and amplifying higher harmonics. These results suggest bottom dissipation, slope, and spectral bandwidth play key roles in nonlinear wave transformation in rocky environments, with important implications for coastal wave modeling and naval operations.Distribution Statement A. Approved for public release: Distribution is unlimited.Ensign, United States Nav
FEDERATED LEARNING IN CYBERSECURITY: PERFORMANCE, ROBUSTNESS, AND ADVERSARIAL THREATS
No full textModern military operations require resilient, privacy-preserving data processing across distributed systems, where centralized learning often fails to meet security and operational demands. Federated learning (FL) provides a decentralized solution by enabling model training without transferring raw data. This thesis evaluates FL's performance, robustness, and vulnerability to adversarial attacks in network traffic classification at the tactical edge. Using supervised models—random forest, Extreme Gradient Boosting, and convolutional neural networks (CNNs)—this study compares FL to centralized learning using Internet of Things (IoT)-23 and Naval Postgraduate School (NPS) Enterprise Research Network (ERN) datasets. Results show FL matches centralized performance in binary classification but struggles with multiclass tasks, particularly under class imbalance. Label flipping attacks preserve overall accuracy while degrading minority-class performance. A novel backdoor method is introduced, embedding feature-level triggers without changing labels, reducing macro and weighted F1 scores by up to 10%. This work contributes a federated simulation framework, model comparison, and threat evaluation. Findings reveal the need for improved metrics, class balancing, and integrated defense mechanisms to enhance FL resilience. These insights inform future deployment of FL in military and coalition domains where data sensitivity, resource constraints, and operational security remain paramount.Distribution Statement A. Approved for public release: Distribution is unlimited.Outstanding ThesisLieutenant, United States Nav
Faces of NPS: Jae Jun Kim, PhD
No full textFaces of NPS features interviews spotlighting the students, faculty, staff and alumni of our Nation's premier defense education and research institution
APPLICATION OF TROPICAL CONVOLUTIONAL LAYERS IN NEURAL NETWORKS
No full textNeural networks are increasingly deployed in Department of Defense applications such as threat detection, signal processing, and autonomous systems. However, these models are vulnerable to adversarial attacks, where small, often imperceptible perturbations cause misclassifications by the model. This vulnerability can undermine the reliability of systems that depend on neural networks, such as radar or surveillance platforms. This research explores whether replacing the final rectified linear unit (ReLU) based output layer of a convolutional neural network (CNN) with a tropical embedding layer improves adversarial robustness without sacrificing clean accuracy or computational efficiency. Two models with identical architectures but different output layers (ReLU and tropical) are trained on spectrograms generated from wireless signal recordings. Robustness is evaluated using multiple white-box and blackbox attacks under varying perturbation magnitudes. The results show that the tropical model performs similarly with clean data but achieves 10–20% higher robust accuracy under stronger attacks, particularly in the ∞ norm. This supports the claim that tropical geometry can improve the resilience of neural networks while maintaining computational efficiency. Future work can explore the integration of tropical networks in real-time systems, expansion to more complex signal domains, and the broader use of tropical geometry in network architectures.Distribution Statement A. Approved for public release: Distribution is unlimited.Captain, United States Marine Corp
EARLY ATTRITION WITHIN THE FIRST SIX MONTHS IN THE GERMAN MILITARY: IMPLICATIONS FOR RECRUITMENT AND PERSONNEL RETENTION
Get PDFSince the suspension of conscription in 2011 and the transition to a volunteer army, the German Armed Forces (Bundeswehr) has faced significant recruitment challenges in a fiercely competitive applicant market. A particularly problematic issue is the high dropout rate within the first six months, which now reaches up to 30%. This thesis investigates the reasons for Early Attrition of service and analyzes the extent to which the commuting distance between the place of residence and the workplace plays a role. By enriching existing personnel data with geographic information and incorporating actual measured distances via an application programming interface, the hypothesis that long travel distances are a crucial factor for early termination is tested.Methodologically, this study combines descriptive statistics with heatmaps and density plots to visualize the geographical data, as well as regression analyses to identify correlations between commuting distances and the dropout rate. Additionally, a Difference-in-Differences analysis is conducted to assess the effects of a new Withdrawal Policy introduced in 2024, which had the aim to decrease the Early Attrition Rate.The results indicate that commuting distance has a significant, non-linear impact on early termination and that the new policy is only effective in specific areas. Recommendations for the future include deepening of analyses using advanced methods such as survival analysis and improvement of data collection.Distribution Statement A. Approved for public release: Distribution is unlimited.Fregattenkapitän, German Nav