Ali Cafer Gurbuz, PhD

Research

My research topics are in the intersection of signal processing and machine learning, with applications on radar, remote sensing and communications.

IMPRESS Lab research overview: signal processing and machine learning at the core, connecting radar systems, remote sensing, and communications — enabling next-generation intelligent sensing and communication technologies

The Information Processing and Sensing (IMPRESS) Lab covers basic and applied research ranging from the design, building, and experimentation of radar and sensing systems to information processing and machine learning, with an emphasis on remote sensing, computational imaging, and signal detection, estimation, and classification. My research has been funded by NSF — including the NSF CAREER award — NASA, DoD, NOAA, USDA, and other sponsors.

My current research interests are organized under the four thrusts below. For detailed project descriptions, the full list of current and past projects, and news from my group, please visit the IMPRESS Lab website.

Compressive Sensing Deep Learning for Inverse Problems Computational Imaging Interpretable Learning Architectures Integrated Sensing and Communication (ISAC) UAV-Based Smart Sensing Radar & Array Signal Processing Passive Radar & Signals of Opportunity Software Defined Radios

Research Thrusts

Learning to sense: joint learning of data acquisition and reconstruction
Thrust 01

Learning to Sense — Task-Cognizant, Physics-Aware Learning

Signal reconstruction from sensor measurements is the core of many application areas, including computational imaging, radar, bio-imaging, remote sensing, and communications. Building on compressive sensing and sparse signal reconstruction, I develop learning-based sensing frameworks that integrate data acquisition with reconstruction and inference in learnable network structures. An optimal set of measurements can be learned for a given signal class and task in an adaptive manner, incorporating sensor constraints and physical models — leading to resource-efficient sensing systems in time, power, and space.

Learning based cyber-physical and autonomous systems
Thrust 02

Learning-Based Cyber-Physical & Autonomous Systems

RF sensors are non-contact, remotely operable, effective in the dark, and highly capable of capturing the kinematics of human motion. I develop multi-modal collaborative sensing systems that integrate RF, camera, and LiDAR sensors for human activity recognition, gesture and American Sign Language recognition, and smart environments. This thrust also spans machine learning for autonomy and mobility — object detection and classification for off-road autonomy, subterranean sensing and threat detection, and gas seep detection in sonar imagery — with experimental systems and processing solutions.

Spectrum sharing and joint radar communication
Thrust 03

Spectrum Sharing & Integrated Sensing and Communication

Coexistence of communication, radar, and passive sensing in an increasingly crowded spectrum is a major challenge. My research develops AI-based RF spectrum coexistence between active and passive users, radio frequency interference detection and mitigation for satellite and UAS-based radiometers, and interpretable machine learning for waveform recognition with software-defined radios. I also work on cognitive radar with multifunctional reconfigurable antenna arrays and dual-use radar/communication systems for the next generation of joint spectrum utilization.

Machine learning based remote sensing and precision agriculture
Thrust 04

ML-Based Remote Sensing & Precision Agriculture

With growing global strain on food supplies, optimal crop production matters more than ever. I develop UAS and ground-robotic remote sensing platforms and the signal processing and machine learning to match — collecting visual, multispectral, hyperspectral, LiDAR, and microwave RF observations along with in-situ soil and plant measurements. A particular focus is passive microwave soil moisture sensing from UAS and satellites using GNSS reflectometry, re-utilizing existing navigation and communication transmissions. These capabilities combine toward the smart farms of the future.

Current Funded Research

25+ awarded external grants totaling over $14M. One-line summaries below — full project details on the IMPRESS Lab website.

  • NSF CAREER (PI, 2021–2026) — Learning to Sense: joint learning of task-oriented cognitive sensing with data-driven reconstruction and inference · Award #2047771
  • NSF SWIFT-SAT (PI, 2024–2026) — INTERACT: end-to-end learning-based interference mitigation for radiometers
  • NASA (Co-PI, 2025–2028) — Enhancing SMAP radiometer performance: calibration and RFI detection via deep learning
  • LAS (PI, 2026) — RF-SHIELD: AI-driven RF sensemaking for intelligent spectrum sensing and threat detection
  • NOAA Ocean Exploration (Co-PI) — Machine learning-based automated detection of seafloor gas seeps
  • AFRL (PI, 2023–2025) — Interpretable complex Sinc-Nets for RF waveform detection and classification