AI-enabled Integrated Sensing and Communication (ISAC)

Our lab explores AI-enabled integrated approaches that unify physical-layer sensing, high-speed communication, and distributed computation. By developing novel PHY/MAC algorithms and scalable protocols, we aim to achieve systems that can dynamically adapt to their environment, optimize resource utilization, and enable seamless interaction among connected intelligent devices.

Multiple Active STAR-RIS-Assisted Secure Integrated Sensing and Communication via Cooperative Beamforming, Preprint
DCFNet: Doppler Correction Filter Network for Integrated Sensing and Communication in Multi-User MIMO-OFDM Systems, Preprint
ICI-Robust Transceiver Design for Integration of MIMO-OFDM Radar and MU-MIMO Communication, TVT'23

Large Language Model (LLM)-based Wireless Orchestrator

Our lab envisions a Large Language Model (LLM)-based Wireless Orchestrator that serves as an intelligent controller for next-generation networks. Leveraging the reasoning and adaptation capabilities of LLMs, the orchestrator can interpret network states, predict resource demands, and autonomously generate control strategies across multiple layers of the wireless stack.

Large Multimodal Models-Empowered Task-Oriented Autonomous Communications: Design Methodology and Implementation Challenges, VTM'25
Robust Transmission of Punctured Text with Large Language Model-based Recovery" TVT'25
Adaptive Resource Allocation Optimization Using Large Language Models in Dynamic Wireless Environments, TVT'25

AI-aided Wireless Communication

Our research focuses on making artificial intelligence a central component in the design, optimization, and management of next-generation wireless networks. By leveraging machine learning and large-scale AI models, we aim to build systems that can sense their environment, anticipate traffic and resource demands, and adapt communication strategies in real time. Through this approach, our lab is working to create wireless technologies that are not only more efficient, but also more resilient and scalable across all layers of the wireless stack.

Deep Reinforcement Learning-based Resource Allocation and Mode Selection for Semantic Communication, WiOpt'24
Joint Optimization on Uplink OFDMA and MU-MIMO for IEEE 802.11ax: Deep Hierarchical Reinforcement Learning Approach, CL'24

Deep learning-based Sensing

Deep Learning-based Sensing is reshaping the way wireless systems interact with the world. Instead of relying on handcrafted features or rigid signal models, deep neural networks can automatically discover patterns hidden in complex wireless data. This capability enables wireless signals to serve as powerful sensors, capturing information about human behavior, device activity, and environmental changes. By fusing communication with perception, deep learning-based sensing opens new opportunities for building adaptive, intelligent, and human-aware wireless technologies.

Driver Activity Recognition with Input-Size-Agnostic Vision Transformer and Domain Fusion for IR-UWB, Preprint

Google Sites

Report abuse