RF Belief Inference & LOCUS-DT
Posterior RF localization methods that turn sparse AoA/SNR and multipath observations into calibrated spatial beliefs.
About
I am a Ph.D. candidate in Electrical and Computer Engineering at New York University, advised by Prof. Sundeep Rangan at NYU WIRELESS.
My research develops spatially aware, uncertainty-aware wireless intelligence for embodied autonomy and adaptive 6G systems. I build algorithms, wireless digital twins, and physical FR3/mmWave testbeds that turn sparse RF and multimodal observations into calibrated spatial beliefs and closed-loop decisions.
Education
Ph.D. in Electrical and Computer Engineering
2022
Expected 2027
New York University
M.S. in Computer Engineering
2020
2022
New York University
B.E. in Electrical Engineering and Automation
2015
2019
China Agricultural University
Research Interests
Spatially aware RF sensing and wireless localization
Probabilistic uncertainty quantification for 6G positioning
Wireless digital twins and ray tracing
Multimodal scene graphs and foundation-model agents
Wireless robotics, localization, navigation, and closed-loop control
Research Overview
My work starts from a simple problem: future wireless and robotic systems rarely see the world through clean measurements. A receiver may observe only a few multipath components; a robot may have partial visual context; a handset may only measure the bands and antenna modules it chooses to activate. In these settings, a single point estimate is often less useful than a belief over competing spatial hypotheses.
I use this view to connect four threads: posterior RF localization through MC-CLE and LOCUS-DT, wireless digital twins for zero-shot robot navigation and SLAM, UE-centric multi-band adaptation under mobility and blockage, and object-centric spatial memory for wearable and embodied agents. The long-term goal is to make wireless systems not only communicate, but also reason about space, uncertainty, and action.
Research Themes
Belief-Aware RF Sensing
Posterior localization methods that retain multimodal spatial hypotheses for 6G and robotics.
Wireless Digital Twins
Ray-tracing priors for zero-shot indoor navigation, wireless SLAM, and robot policies.
Wireless Robotics Systems
FR3/mmWave RFSoC/Pi-Radio testbeds with TurtleBot4, Jackal UGV, D48 pan-tilt, and linear-track motion.
Closed-Loop UE Adaptation
Multi-cell multi-band handset digital twins for array, band, and rate prediction under mobility.
Object-Centric Spatial Memory
Sparse egocentric sensing and semantic 3D object memories for wearable and embodied agents.
Key Projects
Wireless systems and embodied AI pipelines that move from probabilistic inference to real-world experiments.
Wireless Robotics Platform (FR3 / TurtleBot4)
RFSoC/Pi-Radio FR3/mmWave channel-sounding platform for robotic localization and navigation experiments.
Wireless Indoor Navigation & SLAM
Ray-tracing digital-twin priors and physics-informed RL for zero-shot wireless navigation, localization, and wireless SLAM.
Multi-Band UE Coordination Under Mobility
UE-centric multi-cell multi-band handset digital twins for closed-loop array, band, and rate prediction under mobility.
Multimodal Spatial Reasoning & Object Memory
Lightweight object-centric semantic 3D memory for wearable and embodied spatial intelligence.
Key Publications
- 2026 · IEEE TWC under review Site-Agnostic Posterior Inference for Indoor Localization with Ray-Tracing Wireless Digital Twins
- 2026 · IEEE TSP under review Learning a Measurement-to-Posterior Map for Wireless Localization
- 2026 · IEEE GLOBECOM under review LOCUS-DT: Localization via Observation-Conditioned Uncertainty Scoring with Digital Twins
- 2026 · Asilomar under review Beyond Point Estimates: Likelihood-Based Full-Posterior Wireless Localization
- 2026 · IEEE JSAC under review MCMB-HDT: A Multi-Cell Multi-Band Handset Digital Twin for Learning-Based Closed-Loop Array Activation
- 2026 · IEEE ICC Workshops Transformer-Based Rate Prediction for Multi-Band Cellular Handsets
- 2025 · IEEE OJ-COMS Digital Twin-Enhanced Wireless Indoor Navigation: Achieving Efficient Environment Sensing with Zero-Shot Reinforcement Learning
- 2025 · RLC Reinforcement Learning with Physics-Informed Symbolic Program Priors for Zero-Shot Wireless Indoor Navigation
- 2024 · IEEE ICRA Zero-Shot Wireless Indoor Navigation through Physics-Informed Reinforcement Learning
Our Center: NYU WIRELESS
I am proud to be part of NYU WIRELESS, a leading 6G research center at NYU Tandon and the home base for my work on wireless sensing, localization, digital twins, and robotic measurement systems. The center gives my research a rare mix of theory, simulation, RF hardware, and mobile robotic platforms.
Contact
Email: hl4155@nyu.edu