RF signal clustersMove probe · click to scanInteractive
Signal feature space
Move to reveal classes · click to probe the spectrum
PSK OFDM Noise UAV
Research portfolio · Hangzhou · 2026
Jinhao WanPhD Researcher · RF & Deep Learning
I build deep-learning systems that classify and identify RF signals.
I am a PhD researcher at Zhejiang University of Technology working on deep learning for radio-frequency signals — automatic modulation classification, RF fingerprinting of unmanned aerial vehicles, and robust recognition under real-world noise and channel conditions.
NOW
Research collaborations
Open to ideas on RF signal identification and UAV detection.
NEXT
Doctoral research
Pursuing a PhD at Zhejiang University of Technology.
Automatic Modulation Classification✦UAV Detection✦RF Fingerprinting✦Signal Identification✦Deep Learning for RF✦Spectrum Awareness✦
✦✦✦✦✦✦
Research atlas
One spectrum,
many signal questions.
Four directions that connect — from identifying the modulation on the wire to finding the drone behind it.
Open the interactive atlas ↗
01
Categorize the spectrum
02RF Signal Identification
Identifying modulation type and signal class directly from raw I/Q samples.
Find the emitter
03UAV Detection
Detecting and recognizing drones from their RF emissions, with per-device fingerprinting.
Learn from waveforms
04Deep Learning for RF
CNN, ResNet and Transformer architectures that turn I/Q into reliable signal decisions.
Hold up under noise
Robust Recognition
Keeping classifiers accurate under low SNR, fading and lab-to-field distribution shift.
Selected work
Work in progress.
Two ongoing manuscripts on RF signal classification and UAV detection — more to follow as the work matures.
Search all publications ↗The library lists each manuscript with status, keywords and one-click BibTeX copying.
Open searchable library ↗Research artifacts
Methods made
tangible.
Open code and captures that turn the research into something reproducible.
View all projects ↗
Explore project ↗Featured method
RFSense
Deep-learning pipelines for automatic modulation classification and RF-based detection and fingerprinting of UAVs.
- Modulations
- 8+
- Architectures
- CNN · ResNet · Transformer
- Framework
- PyTorch
Explore dataset ↗Open dataset
SDRCapture
An in-progress collection of software-defined radio captures across modulation types and UAV emitters for training and evaluation.
- Bands
- 2.4 GHz
- Captures
- Growing
- Status
- In progress
Latest signals
Research
news.
Honest milestones as the work develops.
View timeline ↗Portfolio relaunch — rebuilt this site as an interactive RF research portfolio.
Research focus — centered on deep learning for RF signal categorization and UAV detection.
In preparation — manuscripts on automatic modulation classification and RF fingerprinting underway.
Academic community
Service &
engagement.
Early in the research journey — contributing where I can while the core work develops.
Open science
- Code & capturesReleased as the RF work matures
- ReproducibilitySharing models and SDR pipelines
Peer review
- Open to reviewingRF, signal-processing & ML venues
- Reviews to dateBuilding — happy to be invited
Community
Open sourceMentoringReproducible codeWorkshopsSDR community
Growing into these roles as the research matures.
Have an idea?
Let's explore it
together.
Always glad to talk RF signal identification, UAV detection, collaborations or ideas — reach out anytime.
wanjinhao1@gmail.com ↗