Naval Postgraduate School
Calhoun, Institutional Archive of the Naval Postgraduate SchoolNot a member yet
71232 research outputs found
Sort by
Data Governance in Support of Joint C2
Get PDFNPS NRP Executive SummaryThe study analyzed how to apply United States Marine Corps (USMC) data governance advances to Joint All-Domain Command and Control (JADC2) to improve the quality of decisions and speed up data-driven command and control (C2). The long-term objective is to improve processes for leveraging data and machine learning to achieve decision advantage. Specific study questions included the following: 1. How can recent USMC advances in data governance be applied to advance Project OVERMATCH and JADC2? 2. How can formalized data governance processes and automation improve the quality of decisions and/or speed up data-driven C2? 3. What new data sources hold the most value for improving C2 and decision advantage? This study is part of the Navy/USMC contribution to JADC2 and has potential to impact capability development and integration efforts to related programs of record. Leveraging faculty expertise and consulting with subject matter experts, students carried out literature surveys and case studies to find answers to the study questions and assess the merits of new approaches addressed by the study. The study produced 13 data and workflow models spanning multiple aspects of data governance for joint command and control, and three MS theses on related subjects are in progress. Recommendations include developing a common data model to enable full data interoperability for joint C2 to span all services and all domains of warfare, integrating this data model into systems and communications links, exploring distributing decision-making authority at lower levels to reduce delays due to multistep human communication, and exploring feasibility of further reducing such delays via synchronous distributed virtual meetings when sufficient communication bandwidth is available.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)HQMC Information (DCI
Building Supervised ML Models with Geospatial & Time Series Analyses to Predict Adversary’s Action
No full textCurrently, the intelligence community leverages heuristics and historic information to inform analysis of data and provide intelligence recommendations to commanders. This research was requested directly by PACFLEET to provide a reliable predictive model for the intelligence community to leverage data more fully in a quantitative and analytical manner and provide probabilities with confidence. To develop predictive models, we propose to build machine learning models based on the classified data provided by COMPACFLT staff, collected from 2016 to 2021 related to a specific maritime event. Using such information, first, we can apply classification models, such as random forests model, to see which factors strongly correlate to each event. Second, we can conduct time series analyses to see any patterns of events in terms of time horizon such as seasonal patterns. Third, we will conduct survival analyses, such as Cox regression model and random forest survival analysis model, to predict how long will take between events with probability. Finally, combining all such supervised models, we can build ensemble models to predict an event to happen with its probability from observed factors, and we can predict its location and time.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)Naval Research Program and N2/N6 - Information Warfar
Military Operations Research Society (MORS) Oral History Project Interview of Dr. Rafael E. Matos
No full textInterviewer: Dr. Bob SheldonDr. Rafael Matos was President of MORS from 2014 to 2015. In 2017, Rafael was elected a Fellow of the Society (FS). Dr. Matos was the requirements generation analyst for the Office of the Chief of Naval Operations Assessments Division at the Pentagon. In that assignment, Rafael became familiar with other people’s studies and identified limitations in the analysis, as well as recognized terrific advances in analytical skills. The initial interview was conducted on June 20, 2016, at Quantico, Virginia; a follow-up interview was conducted on February 15, 2025
Improving High-Frequency Voice Communications
No full textThe goal of this project was to investigate methods to improve high-frequency (HF) voice communications such that operators can understand each other without having to repeat voice messages. Improved HF voice quality can save precious time that could significantly improve mission success. HF voice can also be a viable complement when other wireless voice systems are compromised. HF voice quality has always been inferior compared to many modern wireless and cellular telephone systems. It is time to replace decades old single-sideband amplitude modulation (SSB-AM) with modern modulation techniques such as digital frequency modulation (FM), high-order digital modulation, and multiple input multiple output (MIMO) methods. Along with advanced digital signal processing and error correction, the goal is to show that much improved HF voice communications are possible.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)Naval Information Warfighting Development CenterNaval Research Progra
Navy Security Force (NSF) Community Workforce Analysis
No full textNPS NRP Executive SummaryThe Navy Security Forces (NSF) is the primary personnel group who defends our Navy base facilities on a day-to-day basis. Initial discussions with our topic sponsor suggested that lower levels of compensation in comparison to civilian employment opportunities may be causing manpower issues with recruiting and retaining NSF personnel. Furthermore, the sponsor indicated that it would be useful to track some general workforce trends to see how the composition of the NSF workforce has changed over time. To address these concerns, this project examines workforce trends for civilian police officers and enlisted Master at Arms (MA) in the NSF. Data is compiled through Defense Manpower Data Center (DMDC) records from 2017-2024 for police officers and 2001-2024 for MA personnel. The research identifies several patterns in the NSF police force, including high turnover rates with new hires comprising up to 22% of the force annually. Research also shows that comparing 2024 to 2017, the NSF police force is 15% less experienced, 7% younger, and its share of police officers with prior military service is 26% less. In terms of pay, law enforcement officers nationally were found to earn approximately 33% more than NSF officers. This gap is even more profound in states where NSF personnel are predominantly stationed. Officers in California, Florida, and Hawaii earn 76%, 58%, and 55% more respectively than NSF officers. In addition, the study also tracks changes within the MA force. Significant findings in the MA community include a force that is pursuing college more often, the proportion of females increasing from less than 15% to almost 25%, and the average age dropping about five years when comparing pre and post 9/11 data. Our main recommendation is that a substantial pay increase (around 33%) is needed for the NSF to ensure a viable workforce going forward.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/nrpN4 - Material Readiness & Logistic
The Navy Security Force (NSF) Community Workforce Analysis
Get PDFThis report examines civilian police officers and enlisted Master at Arms (MA) in the Navy Security Forces (NSF). Data is compiled through Defense Manpower Data Center (DMDC) records from 2017-2024 for police officers and 2001-2024 for MA personnel. The research identifies several patterns in the NSF police force, including high turnover rates with new hires comprising up to 22% of the force annually. Research also shows that comparing 2024 to 2017, the NSF police force is 15% less experienced, 7% younger, and its share of police officers with prior military service is 26% less. Most notably, the study exposes a substantial pay disparity between NSF police officers and other police officers. Law enforcement officers nationally earn approximately 33% more than NSF officers. This gap is even more profound in states where NSF personnel are predominantly stationed. Officers in California, Florida, and Hawaii earn 76%, 58%, and 55% more respectively than NSF officers. This disparity is particularly problematic considering NSF officers are commonly stationed in expensive coastal areas in these states. The report also tracks changes within the MA force, growing by 418% between 2001 and 2005 before reaching a nearly constant size. Significant findings in the MA community include a force that is pursuing college more often, the proportion of females increasing from less than 15% to almost 25%, and the average age dropping about five years when comparing pre and post 9/11 data.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)N4 - Fleet Readiness & Logistic
Tactical Mission Commander Drills Implementation
No full textDuring Naval Special Warfare’s (NSW) Junior Officer Training Course (JOTC) Team Leaders are limited in the number of repetitions of mission rehearsals to hone their tactical decision-making skills during Ground Force Commander (GFC) Drills. Structured as table-top exercises, GFC Drills are used during and after JOTC to create environments for Team Leaders to make decisions in a tactical environment employing various assets, systems, and capabilities. Following the drill, there is an after-action review (AAR) process, but it is not sufficiently structured, so participants are receiving feedback that is non-standardized. This research improved GFC Drills through the analysis and refinement of requirements drafted to field a virtual training system. Our research team surveyed and assessed existing modeling and simulation (M&S) tools and reported on their ability to enhance the realism of GFC Drills. To identify the right tools, we studied the GFC training and readiness (T&R) objectives. This task analysis enabled our research team to map training requirements to the existing capabilities of constructive simulation systems and mixed reality (xR) displays. Our research was conducted in close coordination with NSW students from the Defense Analysis (DA) Department and there was an opportunity to integrate this work into the curriculum through guided discussions and technology demonstrations during DA4500, Special Topics in Strategic Analysis. Since this work was executed as a thesis or capstone project, we delivered a final report and presentation detailing our analysis and findings.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)Naval Special Warfare Group ONENaval Research Progra
Advanced Analytics for Emerging Maritime Threat Activity Detection
No full textCorrectly identifying potentially adversary vessels or groups of vessels in the open ocean or in dense shipping lanes is a challenge. Difficultly arises as adversary vessels perform repetitive behaviors to condition observers and establish observer bias and can reduce signatures to disrupt observability. Big data have been collected for observing maritime traffic and activities, especially activities of adversary fleets and proxy vessels (coast guard, fishing, civilian, and merchant ships). Our objective focused upon developing advanced detection algorithms to discriminate the indicators of malicious, deceptive, and adversarial behaviors to warn fleet warning officers and focus intelligence, surveillance, and reconnaissance assets. Using Variational Autoencoder, which is a class of self-learning or unsupervised learning algorithms that can learn patterns and reconstruct time series using models represented as a series of transformers, we were able to show an improved methodology to greatly filter and identify tracks and events that bear further investigation in areas of concern. Power Spectral Density heatmaps were also developed to localize anomalies to space and frequency by performing unsupervised learning. Future efforts could include incorporating distributed acoustic sensors as a new source with other sensors can potentially validate and corroborate events for object detection and classification.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)Naval Research ProgramNaval Information Warfare Center Pacifi