Ryan M. Dreifuerst

Ryan M. Dreifuerst, Ph.D.

Lead, Computational Signals and Spectrum Modeling at MITRE · Ph.D., NC State

I research AI-enhanced PHY algorithms, foundation models, digital twins, mission-critical wireless systems, and all things computationally heavy in communications modeling.

CV Google Scholar GitHub Email

Published inIEEE Transactions on Wireless Communications · IEEE Transactions on Signal Processing · IEEE Communications Magazine · ICASSP · MILCOM

Ink drawing of antenna beam-pattern lobes fanning out on warm paper

Featured work

Learning the parts of a wireless system that fixed designs get wrong.

IEEE TWC 2025IEEE TMLCN 2024

Neural codebooks for extreme MIMO

Fixed DFT codebooks don't scale to extreme arrays.→Learned, site-specific hierarchical codebooks→less beam-training overhead, with beamforming gain kept.

PaperarXivCode
ICASSP 2021with Meta Connectivity

ML for coverage & capacity optimization

Tuning tilt and power across a cellular network is a huge search.→Bayesian optimization and RL→better coverage–capacity trade-offs, found automatically.

PaperarXivCode
IEEE TSP 2022

SignalNet: low-resolution sinusoid estimation

One- and few-bit ADCs save power but wreck classic estimators.→A neural decomposition network→sinusoids recovered from heavily quantized samples.

PaperarXivCode
IEEE AP-S 2025MILCOM 2025

Propagation modeling & digital twins

Accurate links need accurate terrain.→Terrain and surface processing for cluttered propagation models→inputs for threat-aware aerial network control.

PaperarXivCode

Publications

Papers, newest first.

2025

  1. Neural codebook design for MIMO network beam management

    RM Dreifuerst, RW Heath

    JournalIEEE Transactions on Wireless Communications 24(5)PDFBibTeX

  2. Threat-aware Aerial Network Control

    T Cuvelier, RM Dreifuerst, N Pang, C Minor, C Jenney, S Vishwanath, et al.

    ConferenceMILCOM 2025PDFBibTeX

  3. Terrain and Surface Processing for Cluttered Propagation Modeling

    RM Dreifuerst, WJ Durkin, WF Young, JR Burdin, MJ McAdams, et al.

    ConferenceIEEE AP-S/URSI 2025PDFBibTeX

2024

  1. Hierarchical ML codebook design for extreme MIMO beam management

    RM Dreifuerst, RW Heath

    JournalIEEE Transactions on Machine Learning in Communications and NetworkingPDFBibTeX

  2. Context-aware codebook design for 6G extreme MIMO systems

    RM Dreifuerst, I Kilinc, RW Heath

    ConferenceIEEE SPAWC 2024PDFBibTeX

  3. Beam training in mmWave vehicular systems: Machine learning for decoupling beam selection

    I Kilinc, RM Dreifuerst, J Kim, RW Heath

    ConferenceIEEE BlackSeaCom 2024PDFBibTeX

2023

  1. Massive MIMO in 5G: How beamforming, codebooks, and feedback enable larger arrays

    RM Dreifuerst, RW Heath

    MagazineIEEE Communications Magazine 61(12)PDFBibTeX

  2. Machine learning codebook design for initial access and CSI type-II feedback in sub-6-GHz 5G NR

    RM Dreifuerst, RW Heath

    JournalIEEE Transactions on Wireless Communications 23(6)PDFBibTeX

  3. Intent-aware radio resource scheduling in a RAN slicing scenario using reinforcement learning

    CV Nahum, VHL Lopes, RM Dreifuerst, P Batista, I Correa, KV Cardoso, et al.

    JournalIEEE Transactions on Wireless Communications 23(3)PDFBibTeX

  4. CSI type-II codebook of codebooks

    RM Dreifuerst, RW Heath

    ConferenceIEEE SPAWC 2023PDFBibTeX

2022

  1. SignalNet: A low resolution sinusoid decomposition and estimation network

    RM Dreifuerst, RW Heath

    JournalIEEE Transactions on Signal Processing 70PDFBibTeX

  2. Deep reinforcement learning-based scheduling for multiband massive MIMO

    VHL Lopes, CV Nahum, RM Dreifuerst, et al.

    JournalIEEE Access 10PDFBibTeX

  3. Massive MIMO Beam Management in Sub-6 GHz 5G NR

    RM Dreifuerst, RW Heath Jr, A Yazdan

    PreprintarXiv 2204.06064PDFBibTeX

2021

  1. Optimizing coverage and capacity in cellular networks using machine learning

    RM Dreifuerst, S Daulton, Y Qian, P Varkey, M Balandat, S Kasturia, et al.

    ConferenceICASSP 2021PDFBibTeX

  2. Load balancing and handover optimization in multi-band networks using deep reinforcement learning

    M Gupta, RM Dreifuerst, A Yazdan, PH Huang, S Kasturia, JG Andrews

    ConferenceIEEE GLOBECOM 2021PDFBibTeX

2020

  1. Deep learning-based carrier frequency offset estimation with one-bit ADCs

    RM Dreifuerst, RW Heath, MN Kulkarni, J Charlie

    ConferenceIEEE SPAWC 2020PDFBibTeX

  2. End-to-End Radio Fingerprinting with Neural Networks

    RM Dreifuerst, A Graff, S Kumar, C Unger, D Bray

    PreprintarXiv 2010.05169PDFBibTeX

Full list on Google Scholar →

News

Recently.

  1. Jun 2026

    Promoted to Lead for Computational Signals and Spectrum Modeling at MITRE.

  2. 2025

    Papers at MILCOM 2025 and IEEE AP-S 2025.

  3. 2025

    Neural codebook design published in IEEE TWC.

  4. 2024

    Joined MITRE as a researcher.

  5. Mar 20, 2024

    Defended Ph.D. at NC State, advised by Prof. Robert W. Heath Jr.

  6. Dec 2021

    M.S., UT Austin.

About

Signal processing meets machine learning, for real networks.

I lead Computational Signals and Spectrum Modeling at MITRE, working at the intersection of signal processing and machine learning for wireless networks — from learned beam management for massive and extreme MIMO to propagation modeling and network control for aerial and mission-critical systems.

I completed my Ph.D. in Electrical Engineering at NC State in 2024 with Prof. Robert W. Heath Jr., after starting my graduate work with him at UT Austin. My thesis developed machine-learning based codebook design for beam management.

Ph.D. EE
NC State '24
M.S. EE
UT Austin '21
B.S., dual degree
EE, Milwaukee School of Engineering & Electrical and Communications Engineering, TH Lübeck '19

Contact

Always glad to talk about AI for wireless, beam management, and digital twins.

Email Google Scholar GitHub LinkedIn