Naval Postgraduate School
Calhoun, Institutional Archive of the Naval Postgraduate SchoolNot a member yet
71232 research outputs found
Sort by
EVOLVING THE RISK MANAGEMENT FRAMEWORK TO ENHANCE DEPARTMENT OF DEFENSE CYBERSECURITY
No full textThe Department of Defense (DoD) relies on its robust communication infrastructure to perform essential functions and enable command and control. Ensuring the security of this infrastructure is critical, which the DoD aims to achieve through the Risk Management Framework (RMF) assessment requirements established by the National Institute of Standards and Technology (NIST). Despite its significance to improving standardization and security across the DoD, the current RMF process contains subjectivity, inefficiency, and a compliance-based approach that fails to keep pace with rapidly evolving technology and threats. This thesis aims to explore potential revisions to the RMF to make it more objective, efficient, and threat-based, thus enhancing its effectiveness. The author conducted a qualitative analysis of the DoD’s implementation of the RMF by conducting interviews with subject matter experts. These findings informed the author’s recommendations to improve the RMF. Stakeholders can use the recommendations to implement targeted improvements to the current RMF, which will provide a more effective framework that will improve the security posture of information technology systems within the DoD.Distribution Statement A. Approved for public release: Distribution is unlimited.Outstanding ThesisCaptain, United States Marine Corp
Faces of NPS: Ross Anthony Eldred
No full textFaces of NPS features interviews spotlighting the students, faculty, staff and alumni of our Nation's premier defense education and research institution
Faces of NPS: Maj. Patrik Liljegard, Swedish Armed Forces & Lt. Cmdr. Maximilian J. Leutermann Sr
Get PDFFaces of NPS features interviews spotlighting the students, faculty, staff and alumni of our Nation's premier defense education and research institution
Faces of NPS: SMSgt. Gessica Lillich, USA
No full textFaces of NPS features interviews spotlighting the students, faculty, staff and alumni of our Nation's premier defense education and research institution
EXPERIMENTAL ACCURACY OF NILM LOAD DISAGGREGATION ON THREE-PHASE MICROGRID TEST BED
No full textNon-intrusive load monitoring (NILM) is a powerful tool that uses machine learning and special sensors to disaggregate large amounts of electrical data from a single central point of a power system. This service provides valuable insight into the system, allowing for anomaly detection and improvements to preventative maintenance. An important aspect of NILM is the accuracy of the sensors collecting the power data. The United States Navy is currently experimenting with the Mobile Power Meter (MPM), a load monitoring sensor that is proprietary to the Army Research Lab (ARL). These sensors are still being tested, optimized, and improved with every iteration. MPMs have not yet been validated for widespread use. In this thesis, we test the accuracy of MPM data collection and load disaggregation when applied to a three-phase microgrid testbed. The MPMs are run through tests with various loads and then meticulously compared to the ground truth data, which is obtained from oscilloscope probes attached at every phase of every load. Various power features, such as the frequency, voltage, and phase currents, are then selected for analysis. Based on percent error calculations and visual analysis, the MPM data is compared against the oscilloscopes for accuracy. Validating the accuracy of the MPMs is a significant step in ensuring NILM is precisely executed, thereby bolstering the energy security of the United States Navy.Distribution Statement A. Approved for public release: Distribution is unlimited.Ensign, United States Nav
SUPERVISED LEARNING FOR DAMAGE IDENTIFICATION IN SIMULATED STRUCTURES WITH FEM AND XFEM
No full textThis thesis investigated the use of machine learning (ML) for localization and assessment of structural damage using synthetically generated data from finite element simulations. Two modeling approaches were employed to capture different representations of damage: traditional Finite Element Method (FEM) was used to simulate localized stiffness degradation in one-dimensional beams, while the Extended Finite Element Method (XFEM) was applied to two-dimensional domains to model explicit surface cracks. For each case, synthetic datasets were generated under varying damage scenarios and loading conditions. Supervised ML models were trained on these datasets to predict damage characteristics, including element-level stiffness reduction (FEM) and crack start/tip location (XFEM). The XFEM model was validated against analytical solutions for stress intensity factors, confirming accuracy. Neural networks were then evaluated using error metrics such as Mean Squared Error (MSE) and Euclidean distance. Results showed strong predictive performance, demonstrating that data-driven methods can effectively identify flaws from simulated structural response data, and that ML-based damage detection systems could generalize across varying representations of structural damage.Distribution Statement A. Approved for public release: Distribution is unlimited.Lieutenant, United States Nav
Military Operations Research Society (MORS) Oral History Project Interview of Dr. Robert S. Sheldon, FS
Get PDFInterviewers: Bill Dunn and Dr. Tim Hope. This MORS interview was conducted on August 2, 2006, at the MORS Office in Alexandria, Virginia, and a follow up interview was conducted on July 3, 2024
ADVERSARIAL EXPERIMENTATION IN ATLATL: LEVERAGING SENSITIVITY ANALYSIS, NEIGHBORHOOD SEARCH HEURISTICS, AND PROBABILISTIC SCENARIO GENERATION TO EXPOSE AI WEAKNESSES
No full textModern military decision aids must remain reliable under adversarial conditions that typically exceed their developer’s testing regimen. This thesis presents a reproducible experimentation framework built atop the Atlatl hex-grid wargame, which probes artificial intelligence (AI) vulnerabilities through probabilistic scenario generation, global sensitivity analysis, and local adversarial search. To test the framework, three reference agents are evaluated on a small scenario: NAMaiV5 and NAMaiV9 (scripted AI) and Pascal (a neural network trained on the test scenario). Latin Hypercube Sampling generates 20,000 diverse scenarios, each evaluated using a score differential between Blue-vs-Red and Red-vs-Red matches, from which Sobol indices isolate influential parameters. A neighborhood search heuristic procedure then degrades model performance by up to 65%, outperforming differential evolution in efficiency while achieving better score differential reduction. Behavioral heatmaps reveal consistent spatial biases, particularly when perturbing terrain near the map center. Results show that the scripted AIs fail most under force imbalance and opponent variation, while the neural network is more sensitive to scenario length and unseen terrain clusters. This testbed provides a scalable and interpretable process and tool for adversarial validation of military AI systems, offering actionable insight into operational robustness.Distribution Statement A. Approved for public release: Distribution is unlimited.Outstanding ThesisLieutenant, United States Nav
AUSTRALIA'S PURSUIT OF A SOVEREIGN NUCLEAR SUBMARINE CAPABILITY: CHALLENGES AND STRATEGIC IMPLICATIONS UNDER PILLAR I OF THE AUKUS AGREEMENT
No full textThis study investigates whether the Commonwealth of Australia can develop, field, and operate a sovereign nuclear-powered submarine fleet under Pillar I of the AUKUS Trilateral Security Agreement in a timely, sustainable, independent manner. While the U.S. and U.K. spent over five decades developing mature nuclear naval capabilities, Australia must absorb complex technical, regulatory, and operational knowledge within a much shorter timeframe. Using qualitative analysis of historical case studies, defense policy reviews, and alliance training pipelines, this study evaluates the feasibility and limitations of Australia’s nuclear ambitions. Key findings reveal that while Australia can leverage allied expertise to build a credible undersea deterrent, it will remain significantly dependent on U.S. and U.K. infrastructure for training, maintenance, and regulatory oversight into the 2040s. This dependency challenges claims of full sovereignty. The study reveals that achieving a sovereign nuclear-powered submarine capability will require long-term investment in domestic education, regulation, and industrial infrastructure. Recommendations include the creation of an Australian nuclear power school, embedded allied instructional support, and phased development of sovereign maintenance and regulatory systems to reduce reliance on foreign partners. These measures are essential if Australia is to meet Indo-Pacific security challenges while preserving national decision-making autonomy.Distribution Statement A. Approved for public release: Distribution is unlimited.Captain, United States Marine Corp
Data Synchronization Service for Improving Decision Advantage in Distributed C2
No full textNPS NRP Executive SummaryThe overall study objective was to gain decision advantage by improving the accuracy and timeliness of the common operational picture to support command and control (C2) for Navy platforms that must operate in denied, disrupted, intermittent, or limited (DDIL) environments. The project analyzed methods for synchronizing distributed data sources to achieve eventual data consistency. The project’s hypothesis was that synchronization methods that have been developed for use in Not-Only Structured Query Language (NOSQL) database systems to support massive concurrency in analytics for Big Data would be useful for Navy C2. Faculty and students searched the literature on efficiently achieving consistency of distributed data and carried out case studies that analyzed the results of the search to assess pros and cons of alternative approaches. We found that existing NOSQL systems did support eventual consistency of distributed data and that CouchDB is the most applicable of the systems analyzed in the case studies.Approved for public release. Distribution is unlimited.This research is supported by funding from the Naval Postgraduate School, Naval Research Program (PE0605853N/2098). https://nps.edu/nrpChief of Naval Operations (CNO)N9 - Warfare System