Naval Postgraduate School
Calhoun, Institutional Archive of the Naval Postgraduate SchoolNot a member yet
71232 research outputs found
Sort by
Faces of NPS: Jason Jones
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: Tristan Volpe, 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
Faces of NPS: Jeff Appleget, 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
EXTREME LEARNING MACHINES FOR WEATHER MODELING
Get PDFThe growing demand for Unmanned Aerial Systems (UAS) with extended range and endurance has heightened interest in long-endurance flight capabilities. Trajectory optimization using sparse wind data and linear interpolation has demonstrated that wind-aware path planning can significantly enhance UAS performance. However, the computational cost of traditional methods limits real-time application. This study investigated the use of Extreme Learning Machines (ELMs)—a class of fast-training feedforward neural networks with strong generalization ability—to model wind forecast data continuously, replacing conventional interpolation. Several ELM variants were evaluated, including the standard ELM, Incremental ELM (I-ELM), Enhanced Incremental ELM (EI-ELM), and Pruning ELM (P-ELM), with systematic tuning of training parameters. Results showed that ELMs can accurately approximate wind data using far fewer neurons than data points, with training times in seconds and prediction in milliseconds. Initial integration into the trajectory optimization algorithm yielded promising results. Future work should focus on refining this integration and exploring online learning for near-real-time forecast updates.Distribution Statement A. Approved for public release: Distribution is unlimited.Outstanding ThesisEnsign, United States Nav
ELECTRONIC PROTECTION AGAINST AN ENSEMBLE OF TRANSMIT WAVEFORM-SHAPED INTERFERENCE FOR SPACE-TIME ADAPTIVE PROCESSING RADAR
No full textSpace-time adaptive processing (STAP) has many military applications, including in airborne moving target indicator radars. Knowledge-based interference is known to be more effective than traditional spot noise jamming. In this work, we investigate the effect of various types of transmit waveform-shaped interference on STAP radars. We also design an electronic protection technique against transmit waveform-shaped interference for STAP radars. Transmit waveform-shaped interference has the potential to greatly impact the detection performance of moving target indicator radars employing STAP. The research and development of electronic protection techniques to mitigate transmit waveform-shaped interference could have important military applications.Distribution Statement A. Approved for public release: Distribution is unlimited.Captain, United States Marine Corp
SOURCE RANGE ESTIMATION USING FUNCTIONAL VARIATIONAL INFERENCE FOR SINGLE ANTENNA SOFTWARE DEFINED RADIO RECEIVERS.
No full textIt has been previously shown that Deep Neural Networks (DNNs) and Bayesian DNNs outperform the current physics model for radio frequency source range estimation in a standard maritime operational environment with a simple signal antenna Software Defined Radio. This was demonstrated on a synthetic dataset that simulates the radio frequency transmission over Very High Frequency/Ultra High Frequency (VHF/UHF) bands in a manner that exhibits general features of an actual signal waveform. The Bayesian methods employed infer a posterior distribution over the model parameters. However, it can, in many situations, be preferable to learn a distribution over model outputs using methods such as Functional Variational Inference (FVI). This work evaluates FVI for radio frequency source range estimation. Surprisingly, the results of this work show that FVI does not improve upon parameter-based variational inference for this signal intelligence task.Distribution Statement A. Approved for public release: Distribution is unlimited.Lieutenant, United States Nav
RISK AND REACTION: RISK PERCEPTION AND RESPONSE BEHAVIOR IN BOSTON
No full textBoston, Massachusetts has never issued a large-scale mandatory evacuation order, but is facing increased risks of sea level rise, extreme weather events, and flooding, which may necessitate such an order. The lack of evacuation experience, combined with the city’s linguistic and cultural diversity, could present challenges in both the public’s understanding of and compliance with an evacuation order. This thesis synthesizes how Boston can apply findings from literature on risk perception, evacuation behavior and crisis communication to improve its evacuation and emergency public information plans and policies. The research design included a literature review, evaluation of the City of Boston’s Evacuation Support Annex and Emergency Public Information Support Annex, and a focus group with representatives from City of Boston departments. Themes from the literature were used to guide the plan analysis and focus group discussion, integrating both theoretical insights and practitioner perspectives. The analysis found that the evaluated City of Boston plans lack strategies to address emotional responses, cultural influences, and trusted community networks that shape public risk perception and evacuation decisions. This thesis recommends actionable strategies for more effective emergency communication in Boston, including increased pre-incident community engagement, culturally tailored messaging, and the integration of community voices into public messaging.Distribution Statement A. Approved for public release: Distribution is unlimited.Civilian, City of Boston Office of Emergency Managemen
Defense Innovation Leader Doug Beck to Speak at NPS’ Annual Acquisition Symposium and Innovation Summit
No full textARP Celebrates the 250th Issue of the “Need to Know” Newsletter
No full textAn Acquisition Research Program Blog entry on this dat
Faces of NPS: Capt. Christian Thiessen, USMC
Get PDFFaces of NPS features interviews spotlighting the students, faculty, staff and alumni of our Nation's premier defense education and research institution