LLM × Sensor Fusion
activeExploring large language models as reasoning engines for multi-modal sensor data. Aims to enable natural language queries over continuous physiological and environmental streams.
LLMSensor FusionPythonPyTorchResearch
Projects
Research systems, sensing platforms, and tools built across my PhD and postdoc work.
Active
Completed
mmFER
Privacy-preserving facial expression recognition using millimetre-wave radar. No camera required — captures subtle facial muscle movements from RF reflections.
mmWaveDeep LearningSignal ProcessingPythonPyTorch
mmBP
Contact-free blood pressure monitoring using commodity mmWave radar. Estimates systolic and diastolic BP from pulse wave velocity measured via radar reflections.
mmWaveHealth SensingSignal ProcessingPython
MDLdroid
ChainSGD-Reduce framework for distributed on-device deep learning across a chain of smartphones. Enables collaborative training without a central server.
On-device MLDistributed LearningAndroidPythonJava
MDLdroidLite
Resource-efficient DNN execution on mobile devices via release-and-inhibit control. Dynamically adapts network depth to real-time resource constraints.
Mobile MLResource EfficiencyAndroidPython
AccCall
Real-time phone call quality enhancement using the smartphone's built-in accelerometer. Detects call conditions and adapts audio processing without extra hardware.
Signal ProcessingAndroidAudioAccelerometer