United States

Spectrum AI Agent for Monitoring a Radio Dynamic Zone
Silvia Ionescu (sionescu@mitre.org)
Curtis Watson (cmwatson@mitre.org)

The increasing reliance on wireless communications highlights the critical need for innovative spectrum sharing strategies to maximize the utilization of finite resources. Dynamic spectrum environments require continuous monitoring and agile decision-making, yet the volume and complexity of radio frequency (RF) data can overwhelm human operator and hinder timely, accurate responses. There is a pressing need for tools that reduce cognitive overload while enabling robust, real-time spectrum coexistence in diverse, evolving operational contexts. 
The NSF National Radio Dynamic Zones (NRDZ) Spectrum Applications with Real-world Knowledge through Integrated Experiments (SPARKIE) project advances spectrum management through long-term dynamic spectrum sharing experimentation and the integration of cutting-edge artificial intelligence (AI) technologies. SPARKIE introduces a spectrum-aware AI agent that leveraged machine learning (ML) to continuously analyze RF data for dynamic, efficient monitoring and decision-making. By deploying sensing capabilities at edge sensor nodes equipped with high-performance GPUs, such as DeepWave AIR-T devices that combine software defined radio functionality with GPU acceleration, the project strives to enable real-time sensing, monitoring, and decision support. These edge-based tools are further augmented with large language model (LLM)-based agents that provide contextual reasoning and decision-making, spectrum monitoring, and evaluation of experimental outcomes. 
In this presentation, we will present results from the NRDZ AI spectrum agent’s evaluation of the November 2025 Hat Creek Radio Observatory spectrum coexistence experiments. Our results highlight the AI agent’s ability to assess experiment outcomes and deliver refinement suggestions that accelerate and improve decision-making. 

Approved for Public Release; Distribution Unlimited Public Release Case Number 25-02739-2. © 2025 The MITRE Corporation. ALL RIGHTS RESERVED


NAML 2026

Spectrum AI Agent for Monitoring a Radio Dynamic Zone

spectrum sharing

ai agent

machine learning

Naval Information Warfare Center Pacific is partnering with AFCEA San Diego chapter to host the workshop.

The NAML workshop is an opportunity for researchers from Navy research and development centers and our partner organizations to exchange technical ideas and to come together as a team.

The goals of NAML are:

To spread awareness of current machine learning research relevant to Naval applications

To connect machine learning researchers with experts in military requirements

To build and strengthen collaborations within the defense research community


<figure class="video_container">
  <iframe src="https://s3.amazonaws.com/pf-user-files-01/u-59356/uploads/2022-03-22/hk13mxo/NAML_ReceptionWithtitle.mp4" width=826  height=526  frameborder="0" allowfullscreen="true"> </iframe>
</figure>

Please log in to explore this event. You need to log in with the email address you registered with. Access credentials have been sent to your email. 

Please be sure to check your spam and other email folders if you do not see an email confirmation right away.

Please log in to explore this event.

Registration for this workshop has now closed. We appreciate your interest and hope to welcome you at a future event. If you’d like to hear about similar upcoming events, please check our website or join our mailing list.

Registration for this workshop has now closed.

NAML 2022

Naval Information Warfare Center Pacific is partnering with AFCEA San Diego chapter to host the workshop. The NAML workshop is an opportunity for researchers from Navy research and development centers and our partner organizations to exchange technical ideas and to come together as a team.

The goals of NAML are:

To spread awareness of current machine learning research relevant to Naval applications

To connect machine learning researchers with experts in military requirements

To build and strengthen collaborations within the defense research community



technical paper

### Welcome to the 10th annual workshop on Naval Applications of Machine Learning!

NAML is the premier annual event showcasing current machine learning research relevant to Naval applications. We welcome participation from the defense community, industry, and academia. The NAML workshop is an opportunity for researchers from Navy research and development centers and our partner organizations to exchange technical ideas and to come together as a team. 
Naval Information Warfare Center Pacific is partnering with AFCEA San Diego chapter to host the workshop.
#### On-Demand Access

* This is an on-demand site and all the recorded content from the workshop is available here. 
* Access to the recordings is exclusively available to registered participants.
* Each registrant has received an email notification, with instructions on how to access the content.

[![](https://assets.underline.io/markdown_image/1/image/60cd12aa376c191d3467ea2ee60944ea.png)](https://naml2026.org/)

The goals of NAML are:

* To spread awareness of current machine learning research relevant to Naval applications
* To connect machine learning researchers with experts in military requirements
* To build and strengthen collaborations within the defense research community

#### *"What we do as a Navy is a human business. Don’t pass up an opportunity to come together as a team. Don’t pass up leadership opportunities, and don’t outsource them to a PowerPoint. If you have an opportunity to come together as a team, to increase belonging, do it."*
Admiral John M. Richardson, 31st Chief of Naval Operations
<br><br>

This is an on-demand site. All recorded content from NAML 2026 is available here and access to it is exclusively available to registered participants.

**Please log in using the email address you registered with.** If you have any issues logging in, please contact us at <naml2026@underline.io>.

Let’s celebrate the past decade by investing in the next. 🚀


This is an on-demand site. All recorded content from NAML 2026 is available here and access to it is exclusively available to registered participants.

**Explore the latest advancements in machine learning tailored to Naval applications 🔎**

With on-demand access to NAML 2026, you can watch all lectures & browse poster presentations at your convenience, while also engaging directly with presenters via the Underline platform - ask questions, exchange ideas, and stay connected.

**If you still haven’t registered, you can do so now.**
👇
**Take advantage of a special one-time offer: full access for just $99, with complimentary registration for students, government, and military employees.**

**Get access here: https://naml2026-ondemand.paperform.co/**

Please check your spam folder, or contact our support team at **<naml2026@underline.io>** if you registered, but still haven't received an access email.

Let’s celebrate the past decade by investing in the next. 🚀

Please register!

The 10th Annual Workshop on Naval Applications of Machine Learning (NAML 2026) was held in San Diego, California, March 2-5, 2026. NAML is a forum for interdisciplinary experts from government, industry, and academia to discuss results of ongoing research on theory and applications of machine learning to address some of the most challenging problems faced by our naval forces. 



Hierarchical Spiking Density: An Event-Based Implementation of HDBSCAN using Neuromorphic Framework

Abstract
The rapid growth of edge-based sensors requires unsupervised learning algorithms capable of processing asynchronous, high-velocity data streams under strict power constraints. This is particularly critical for Event-Based Sensor (EBS) object tracking applications, where immediate, localized cluster analysis is necessary to identify and track objects from high-speed, noisy sensor data. While density-based clustering, specifically Density-Based Spatial Clustering of Applications with Noise (DBSCAN), is a standard for identifying clusters of arbitrary shapes, it struggles with data containing varying density gradients and requires sensitive hyperparameter tuning. The hierarchical extension, H-DBSCAN, mitigates these issues by constructing a hierarchy of cluster stability, but its traditional recursive implementation is computationally prohibitive for real-time, resource-constrained neuromorphic hardware.
This research proposes a novel Spiking Neural Network (SNN) architecture that approximates the HDBSCAN algorithm, simulated within a specific Neuromorphic Computing Framework. This SNN-based H-DBSCAN is designed to efficiently perform the clustering required for EBS object tracking. By triggering a wave-front propagation of spikes, the network inherently computes a Minimum Spanning Tree (MST) equivalent through a "first-to-fire" mechanism. A multi-tiered topology integrates spikes over varying temporal windows, effectively constructing a tree diagram of cluster persistence to extract stable clusters without prior knowledge of the data's scale.
The proposed architecture will be validated through edge and hardware-in-the-loop implementation targeting traditional CUDA based devices and by FPGAs. The SNN design will be synthesized using a known framework neuro-processor backend, configured for a small neuron array with sparse connectivity. Results will include the comparison between the SNN and CUDA based implementation of the network based on accuracy, speed, and power consumption of each platform.



Hierarchical Spiking Density: An Event-Based Implementation of HDBSCAN using Neuromorphic Framework

Naval radar systems operating near the 3.55–3.7 GHz Citizens Broadband Radio Service (CBRS) band face an increasingly complex RF environment, sharing spectrum with dense commercial networks and non-cooperative emissions. In coastal and port regions, these sources introduce clutter-like artifacts and reduce receiver sensitivity, particularly when atmospheric ducting extends the reach of land-based LTE/5G networks. Because this interference can erode radar performance long before operators notice visible degradation, there is a critical need for methods that not only detect interference but also quantify its impact on mission capabilities.

To address this, we present a two-stage machine learning framework designed to enhance both situational awareness and detection resilience in congested spectrum. First, we introduce RFI-Net, a physics-guided U-Net architecture that analyzes time–frequency data to characterize the spectral environment. Unlike simple energy detectors, RFI-Net learns the distinct signatures of radar waveforms, 5G OFDM blocks, narrowband tones, and impulsive bursts. A custom physics-guided loss function enables the model to separate interference from the underlying noise floor, providing operators with a real-time spectral occupancy map. This allows for the direct quantification of spectral interference map, enabling proactive adaptation of radar parameters if needed. 

Secondly, we developed a Convolutional Recurrent Neural Network (CRNN) detector capable of identifying radar pulses in extreme low-SINR regimes that can be used for CBRS Environmental Sensing Capability (ESC) sensors. While current FCC regulations for CBRS-ESC sensors require detection down to approximately -89 dBm/MHz our CRNN model achieves >99% detection probability at -10 dB SINR. This is an improvement of nearly 30 dB over standard regulatory baselines. By leveraging both local spectral features and temporal dependencies, the model maintains high recall even when radar pulses are submerged beneath structured OFDM signals or bursty wideband noise. 

We validated this framework on a comprehensive dataset comprising NTIA-specified radar waveforms mixed with realistic LTE/5G traffic and heterogeneous interference sources (tones, harmonics, and nonstationary bursts). The resulting system is highly efficient: the detector operates on single spectrogram snapshots with sub-4 millisecond inference latency. While the proposed approaches offer a radar detection capability for coastal monitoring and ESC applications, similar architectures provide detection of 5G waveform detection at naval radar receivers as well that can potentially provide enhanced RFI mitigation and situational awareness in shared spectrum environments.


Deep Learning for Resilient Spectrum Sharing: Physics-Guided Interference Mapping and Low-SINR Radar Detection in the CBRS Band

AI-Enhanced Undersea Signal Processing for Long-Range Vessel Classification

The ability to detect and classify surface vessels at long ranges—beyond visual and radar limits—is critical for distributed maritime operations and decision superiority. Acoustic sensing platforms offer unique advantages for this mission, yet classification remains challenging due to highly non-stationary signals, pervasive environmental clutter, and limited labeled datasets. The Office of Naval Research (ONR) has identified this capability as a research priority. Avathon AI proposes an innovative approach that fuses physics-based modeling with advanced machine learning to deliver robust performance in complex acoustic environments.
Our framework integrates multiple technical components:
• Physics + AI Hybrid Models to improve acoustic imaging and classification beyond traditional statistical methods.
• Mass Feature Construction (MFC) across MFCC, spectral, wavelet, and time-domain features to identify predictive patterns at scale.
• Transformer-Based Architectures, including Masked Spectrogram Modeling (MSM-MAE), for unsupervised labeling and resilient embeddings in cluttered environments.
• Data Augmentation leveraging large public acoustic datasets to bootstrap models for naval-specific signatures.
This approach enables rapid adaptation to new vessel classes with minimal labeled data, improves classification accuracy under distortion and interference, and supports operational decision-making at the tactical edge. During the proposed 20-minute session, Avathon will demonstrate feature construction workflows, transformer-based classification on representative datasets, and discuss transition pathways for ONR and NIWC Pacific programs focused on ISR, undersea warfare, and multi-domain awareness.
Avathon AI brings over a decade of experience delivering AI/ML solutions for government and industry, including recent acoustic projects with NSWC. Our proven track record in predictive maintenance, ISR fusion, and autonomy platforms positions us as a strong partner for advancing naval AI capabilities.

AI-Enhanced Undersea Signal Processing for Long-Range Vessel Classification

The maritime industry relies heavily on the Automatic Identification System (AIS) to enhance navigational safety and ensure effective vessel traffic management.[1] AIS is a tracking system that enables ships to communicate their position, speed, heading, and other vital data with one another and with shore-based authorities.[1] However, the efficacy of AIS can be severely compromised when anomalies in the data occur.[2] These anomalies can arise from various sources, including equipment malfunctions, data entry errors, environmental interferences, and intentional malicious actions such as data spoofing.[2]

Such anomalies pose significant risks to maritime safety, security, and operational efficiency, leading to potential accidents, navigational errors, and even legal issues for vessel operators.[2] For instance, an undetected AIS anomaly may result in a vessel being unaware of a nearby ship, increasing the risk of collision. Moreover, anomalies may mask illegal activities such as unauthorized fishing, piracy, or human trafficking, challenging law enforcement and regulatory agencies tasked with maintaining maritime order. The failure to detect these anomalies in real-time can have dire consequences for both vessels and marine ecosystems, highlighting the critical need for robust anomaly detection solutions.[2]

Despite the importance of effective AIS monitoring, traditional methods of data analysis struggle to keep pace with the sheer volume and velocity of maritime traffic data generated daily.[1] As the number of vessels in commercial shipping increases, so does the complexity of data interpretation.[1] [2] To address these significant challenges, there is a growing trend toward integrating Artificial Intelligence (AI) and Machine Learning (ML) technologies into maritime safety frameworks.[2] AI and ML possess the capacity to analyze large volumes of AIS data in real-time, identifying irregular patterns and potential anomalies that may escape human attention.[2] For example, supervised learning models can be trained on historical AIS data to recognize the typical behavior of vessels in various maritime environments, allowing them to flag deviations as potential anomalies.
Moreover, AI methods can integrate various sources of information, such as weather data, historical traffic patterns, and even satellite imagery, to provide a contextual understanding of vessel behaviors, enriching the analysis and improving detection capabilities.[2] 

References: 

[1] Yang, D., Wu, L., Wang, S., Jia, H., & Li, K. X. (2019). How big data enriches maritime research–a critical review of Automatic Identification System (AIS) data applications. Transport reviews, 39(6), 755-773.

[2] Lee, E., Mokashi, A. J., Moon, S. Y., & Kim, G. (2019). The maturity of automatic identification systems (AIS) and its implications for innovation. Journal of Marine Science and Engineering, 7(9), 287.

Anomalies in Automatic Identification System (AIS) Maritime Traffic Data

By implementing an engineering-grade Naval GenAI framework constructed from extensive historical fleet and vessel maintenance datasets—including structured maintenance logs, free-text technician notes, condition-based monitoring outputs, and other unstructured data sources—naval personnel can leverage advanced machine learning and natural language processing pipelines to accurately determine inspection requirements, fault signatures, and component-level repair procedures. The system integrates supervised learning models, anomaly-detection algorithms, and transformer-based language models to correlate past failure modes with current system states, enabling the automated identification of probable root causes and the precise parts, tools, and technical steps required for corrective maintenance.
The GenAI models generate near-real-time, data-driven diagnostic assessments that replicate and enhance traditional engineering analyses by producing recommended inspection sequences, stepwise repair instructions, and decision-support evaluations for component replacement, refurbishment, or overhaul. These recommendations are optimized using reliability-centered maintenance (RCM) principles, failure mode and effects analysis (FMEA) data, and historical mean-time-to-repair (MTTR) and mean-time-between-failure (MTBF) patterns to ensure engineering validity and operational relevance.
Developing a Navy-wide GenAI system for fleet and vessel maintenance offers engineering advantages far beyond conventional schedule-based preventive maintenance models. Instead of relying exclusively on fixed intervals or human judgment, the framework enables condition-informed, data-adaptive maintenance scheduling that reduces unnecessary work, minimizes downtime, and increases operational availability. The integration of historical work packages, parts-usage records, and prior diagnostic outcomes allows sailors, maintainers, and technical support engineers to derive actionable insights from decades of maintenance activity. This supports faster troubleshooting, automated diagnostic triage, higher accuracy in fault localization, and more consistent adherence to technical standards and repair specifications.
Ultimately, the engineering-focused GenAI framework enhances fleet sustainment by enabling predictive maintenance workflows, improving technical decision-making, and providing maintainers with reliable, AI-augmented guidance to optimize repair processes and lifecycle performance across naval platforms.


From Data to Diagnostics: How GenAI Revolutionizes Naval Maintenance

Purpose
The Navy’s introduction of Unmanned Surface Vessels (USVs) into distributed maritime operations requires new approaches to maintenance, readiness, and lifecycle management. Sustainment models designed for crewed platforms depend on manual inspection and reactive maintenance actions, which are not suitable for autonomous or semi-autonomous systems expected to operate persistently, at distance, and in degraded communications environments. This effort explores the application of artificial intelligence (AI) and data-driven methods to enable predictive, condition-based, and adaptive sustainment for the PMS 420 sUSV fleet.
Methodology
The architecture is designed to support integration of multiple data sources, including sensor telemetry, maintenance and configuration records, and operational context, within a unified data environment. 
A key capability currently demonstrated is the use of a generative AI (GenAI) interface that enables users to interact with the system in natural language rather than traditional query syntax. This interface leverages retrieval-augmented generation (RAG) and adaptive agent workflows to allow maintainers, engineers, and analysts to obtain contextual information, system insights, and maintenance recommendations through conversational interaction.
The architecture is designed for secure operation within DoD environments and aligns with DISA STIG, NIST SP 800-53, and IL4/IL5 compliance requirements. Model performance, data provenance, and decision traceability are continuously monitored. The framework supports future integration with Navy logistics systems and Digital Twin environments to enable predictive simulation and fleet-wide readiness assessments.
Results
Initial demonstrations focus on the GenAI interface and its ability to simplify human–machine interaction for maintenance and sustainment tasks. The system has shown that natural-language engagement can reduce diagnostic and troubleshooting time by making technical data more accessible to non-specialist users. Future work will focus on connecting live and historical data sources to further enhance predictive and condition-based capabilities.
Impact
This effort advances the concept of sustainment autonomy by establishing a scalable and secure framework that can evolve as data sources and operational requirements mature. The GenAI-driven interface represents a practical step toward enabling consistent, data-informed decision support across distributed maritime environments. The work contributes to PMS 420’s and the Navy’s broader objective of maintaining readiness through data-centric processes, intelligent automation, and continuous learning systems. By leveraging AI to augment, not replace, human expertise, the approach enhances the speed, transparency, and resilience of sustainment operations in support of future distributed fleet concepts.


From Data to Decision: An AI-Powered Sustainment Platform Enabling Predictive Readiness Across the USV Fleet

Modern destroyers generate far more engineering data than humans can interpret in real time. Legacy maintenance models using planned and casualty maintenance supported by tools like ICAS (Integrated Condition Assessment System) struggle to recognize anomalies in the data streams or turn this deluge of data into actionable recommendations that integrate into sailor workflows.

This talk introduces ERMv4, the Navy's Enterprise Remote Monitoring system which turns roughly 7,000 hull, mechanical, and electrical sensor readings per second into recommendations that feed the ship’s existing maintenance planning system. This shift has moved condition-based maintenance (CBM) from concept to operational tool on USS Fitzgerald (DDG-62).

We focus on three themes that connect ML research with naval requirements and fielded impact: 

ML at the tactical edge. 

ERMv4 enables the MLOps data flow for models to learn from noisy, heterogeneous sensor data across pumps, motors, power distribution, and auxiliary systems. Delivering useful signals from tactical platforms demands time-series preprocessing, anomaly and degradation modeling when failure labels are rare, and handling non-stationary behavior after maintenance or configuration changes. We also cover edge-compute realities: packaging models for afloat deployment, operating under SWaP and bandwidth limits, handling DDIL connectivity, and safely updating models on ships underway for months within policy and accreditation constraints. 

Human-in-the-loop maintenance. 

ERMv4 is built so the computer does not “boss” sailors around. Recommendations are ranked, explained, and integrated into existing maintenance systems in ways that leverage sailor expertise and reduce alert fatigue. UI and workflow choices make it clear why a recommendation is being made, how confident the system is, and what evidence backs it up. We share early lessons on trust: what ship’s force accepts, what they push back on, and how the system must evolve so sailors feel supported, not treated like warehouse workers executing AI-generated task lists. 

Fleet-level implications. 

ERMv4 is one of the first AI programs-of-record continuously monitoring ship health and offers a template for scaling CBM across classes and platforms. We discuss how consistent data models, shared anomaly patterns, and common integration points with Navy logistics and maintenance systems can support Distributed Maritime Operations and higher availability, and how “assured autonomy” for CBM—demonstrating model behavior, reliability, and limits—can build durable trust in recommendations. 

Attendees will leave with a grounded case study of naval ML running on the deck plates, plus practical lessons for building and deploying ML systems that sailors actually use in the fight.

Machine Learning at Sea with Speed: ERMv4 and Condition-Based Maintenance for Mission-Ready Destroyers

Traditional analyses of high-speed craft (HSC) hydrodynamics depend on experimental, semi-empirical, and numerical approaches that can be computationally expensive and poorly suited for applications. This work presents a Neural Ordinary Differential Equation (Neural ODE) framework for learning the continuous-time dynamics directly from experimental data. Rather than discretizing the equations of motion, the model infers a smooth latent flow that maps sensor inputs to the vessel’s dynamic response. Trained with a noise-injection scheme to enhance robustness under perturbations, the Neural ODE achieves high accuracy in one-step-ahead prediction across calm-water and regular-wave conditions.
Additionally, this work will be extended and compared with discrete-time sequence models such as (TCN, LSTM). These results highlight the potential of continuous-time neural models for real-time hydrodynamic simulation and autonomous-vessel prediction within the broader context of computing with physical systems.

Neural Ordinary Differential Equation (Neural ODE) framework for learning the continuous-time dynamics from experimental data

Machine-learning enabled surrogates are increasingly used to emulate complex high-energy-density physics (HEDP) phenomena, yet the scarcity of high-fidelity, multi-material shock-propagation datasets remains a major barrier to advancing predictive modeling and inverse design capabilities for defense-relevant applications. The Perturbed Layered Interface (PLI) partition of the High-Explosives and Affected Targets (HEAT) dataset addresses this gap by providing time-resolved simulations of explosive-driven shocks interacting with heterogeneous materials. Within the PLI dataset, a fixed material suite subjected to varied initial interface perturbations produces a wide range of plastic deformation outcomes, enabling systematic study of geometry-dependent shock dynamics.

Using this dataset, we develop an inverse-design framework that reconstructs probable initial configurations given only the final, post-shock material morphologies. Our approach leverages machine-learning models trained on PLI simulations to infer a distribution of initial conditions consistent with an observed final deformation. These inferred distributions are subsequently validated through forward simulations that propagate the inferred initial states to late-time outcomes, demonstrating strong agreement with target morphologies. This bidirectional modeling capability provides a physics-informed pathway for linking final observable signatures to their causal shock-driven origins.

Our results highlight the utility of high-fidelity simulation datasets for learning complex multi-material shock behavior and demonstrate an end-to-end inverse design protocol relevant for verification of explosive-driven systems, assessment of material responses under extreme loading, and the development of ML-based decision tools for naval and defense applications.


Inverse Design of Shock-Driven Material Deformation Using Machine Learning and the PLI Dataset

A single agent represents a single system capable of ingesting local data, indexing, cataloging information, performing knowledge pattern discovery, and separating patterns and
anomalies from data. Multiple agents work collaboratively in a peer-to-peer network. Each agent has a peer list. Such multiple agents’ collaboration can be modeled as cooperative games. Each
agent optimizes its own objective locally. We show that each agent self-organizes or converges to its best value and the whole agent network achieves the best social welfare based on both the quantum adiabatic evolution transformation (QAET), and quantum intelligence game (QIG) or the QAET-QIG framework. 

The contributions of this paper are summarized as follows:
1) We show that a multi-agent quantum system converges to an equilibrium state which has a high degree of purity and coherence via a QAET-QIG evolution in a super-positioned torus. The convergence results in an equilibrium and measurable property or quantum theoretic value
(QTV) of the system that is optimized. We demonstrate the QAET-QIG framework and show two theorems, one algorithm used to score new data for quantum values, and compare them with the traditional QAET.
2) We apply the QAET-QIG framework to the kill web concept that potentially improve the traditional kill chain process such as find, fix, track, target, engage, and assess
(F2T2EA). The improvement is measured in the values of powerful global optimization, distributed lethality, and load balancing. 

We show a use case of the QAET-QIG framework to a potential application of mixed sensors, platforms, weapons, and effects.

Disclaimer: The views and conclusions contained in this document are those of the authors and should not be interpreted as others or representing the official policies, either expressed or implied of the U.S. Government.

Premium content

Downloads

Next from NAML 2026

Hierarchical Spiking Density: An Event-Based Implementation of HDBSCAN using Neuromorphic Framework

Similar lecture

Machine Learning and Radio Frequency Spectrum Management