Cognitive radio · SNN · FPGA/EDA
Research workspace
A Ph.D.-stage hardware workspace focused on spectrum-aware ML, spiking neural network-based detectors, and FPGA/EDA flows—publishing in IEEE TCCN, ICC, and GLOBECOM with NSF and ONR support. Collaborations welcome where goals and timelines are explicit.
Lab overview
Work here sits at the intersection of communications-oriented ML, FPGA prototyping, and silicon-minded discipline: bio-inspired spectral-temporal models for cognitive radio (IEEE TCCN 2024), spiking neural network-based white-space detectors (IEEE ICC 2024, GLOBECOM 2024), federated learning for opportunistic spectrum access (IEEE ICC 2024), and optoacoustic communications demodulation (IEEE ICC 2024)—NSF and ONR funded. Versioned RTL, scripted experiments, and reproducible measurement baselines are non-negotiable.
Industry rotations at Meta Reality Labs (LCoS laser-controller RTL/HLS for AR/VR micro-display), Intel (Xeon SoC high-speed clock distribution with Synopsys Fusion Compiler and ICC II), and Lasarrus Clinic (WearME FPGA prototype for COPD patients, FDA-compliant documentation) ground abstract research in production-grade discipline—informing how hardware problems are scoped, verified, and handed off.
Mission
Publishable hardware research with clear baselines, honest limitations, and teaching-ready notes.
Focus
Spectrum-aware ML (cognitive radio, SNN detectors, federated learning), neuromorphic accelerators, optoacoustic communications, and EDA-aware FPGA/ASIC design.
Collaboration
Joint work when roles, IP, and timelines are explicit—see Contact to start a conversation.
People
Md Mehedi Hassan Galib
Primary researcher · Ph.D. Candidate, UMBC
Ph.D. candidate in Computer Engineering at UMBC (2021–2025, CGPA 3.97); B.Sc. EEE at Islamic University of Technology (2012, CGPA 3.98); M.Sc. Electronics & Radio Engineering at Kyung Hee University (2015, CGPA 4.1/4.3). Research spans bio-inspired spectral-temporal modeling (IEEE TCCN 2024), SNN-based and federated white-space detectors (IEEE ICC 2024, GLOBECOM 2024), optoacoustic demodulation (IEEE ICC 2024), and radiation-hardened SRAM/flip-flop tape-outs (IEEE Trans. Nuclear Science 2015). Industry: Meta Reality Labs, Intel, Lasarrus Clinic.
Mohamed F. Younis
Professor · UMBC CSEE
Faculty advisor; wireless sensor networks, communications, and resilient systems—guiding spectrum-aware ML, SNN-based detection, federated learning, and direction-of-arrival research threads in the group.
Fow-Sen Choa
Professor · UMBC CSEE
Co-investigator on optoacoustic communications research; expertise in photonics, lasers, and semiconductor devices—co-author on the SNN-based demodulation scheme for optoacoustic communications accepted at IEEE ICC 2024.
Muntasir Mahmud
Graduate researcher · UMBC
Co-author on SNN-based demodulation for optoacoustic communications (IEEE ICC 2024) and other ML-for-networking studies; contributes to modeling, evaluation, and reproducible experiment design.
Tasnim Nishat Islam
Ph.D. student · UMBC
Co-author on spectrum and direction-of-arrival work; focuses on signal processing and learning-based wireless sensing with clear measurement baselines.
Brian W. Stevens
Research collaborator
Co-author on the spectral-temporal model for opportunistic spectrum access in cognitive radio networks (IEEE TCCN 2024); contributes to resilient wireless systems and experimental validation.
Sultan Ahmed
Research collaborator
Co-author on cognitive radio and AI-enabled networks research; contributed to the lightweight SNN-based detector for interweave cognitive radios accepted at IEEE ICC 2024 (Cognitive Radio and AI-Enabled Networks track).
Collaboration & visitors
Aligned collaboration (faculty / lab)
Short proposals for joint spectrum-sensing, SNN-on-FPGA, federated learning, or EDA-adjacent studies—with clear deliverables (code, measurement set, or manuscript plan).
- 1–2 page technical outline and timeline
- Agreement on authorship and data handling up front
Course-linked project (UMBC or partner program)
Scoped independent study or thesis work where prerequisites match FPGA, ML, or architecture coursework (e.g., CMPE 611, CMPE 640, CMPE 691, CMSC 678).
- Faculty sponsor at your institution
- Weekly checkpoints and a mid-term design review
