Advanced Intelligent Systems Laboratory
Prof. 양현종
Research Communication Systems, distributed learning systems
WEBSITE https://aislab.snu.ac.kr/
Doc poster.pdf
About the Laboratory & Research Area
Advanced Intelligent Systems Lab (AISLab)
Pioneering Real-World Intelligence through the Convergence of 6G, AI, and Robotics
The Advanced Intelligent Systems Lab (AISLab) at Seoul National University, Department of Electrical and Computer Engineering, is dedicated to the research and development of next-generation intelligent communication systems. Under the leadership of Prof. Hyun Jong Yang, we bridge the gap between rigorous theoretical foundations and practical real-world impact.
Core Research Areas:
AI-Native RAN and Wireless Foundation Models: We design AI-native RAN intelligence that translates high-level human intent into real-time network actions. Utilizing NVIDIA Sionna and Aerial, we build proactive, GPU-accelerated control loops for the 6G PHY/MAC layers.
Distributed Learning and Privacy-Preserving Computation: We develop distributed machine learning frameworks optimized for resource-constrained wireless edge devices. Our research focuses on channel-aware task offloading and privacy-first model updates to meet strict QoS targets.
Connected Collaborative Robotics: We implement fleet-level navigation systems where autonomous robots coordinate via seamless 6G connectivity. By integrating Edge LMMs (Large Multimodal Models), we enable holistic path planning and congestion management for large-scale multi-robot systems.
At AISLab, students master the essential essence of top-tier research through our specialized in-house curriculum, covering everything from fundamental optimization and communication theories to cutting-edge AI architectures. We pursue research that transcends academic boundaries, aiming to build the intelligent infrastructure that defines the future of connected society.
Pioneering Real-World Intelligence through the Convergence of 6G, AI, and Robotics
The Advanced Intelligent Systems Lab (AISLab) at Seoul National University, Department of Electrical and Computer Engineering, is dedicated to the research and development of next-generation intelligent communication systems. Under the leadership of Prof. Hyun Jong Yang, we bridge the gap between rigorous theoretical foundations and practical real-world impact.
Core Research Areas:
AI-Native RAN and Wireless Foundation Models: We design AI-native RAN intelligence that translates high-level human intent into real-time network actions. Utilizing NVIDIA Sionna and Aerial, we build proactive, GPU-accelerated control loops for the 6G PHY/MAC layers.
Distributed Learning and Privacy-Preserving Computation: We develop distributed machine learning frameworks optimized for resource-constrained wireless edge devices. Our research focuses on channel-aware task offloading and privacy-first model updates to meet strict QoS targets.
Connected Collaborative Robotics: We implement fleet-level navigation systems where autonomous robots coordinate via seamless 6G connectivity. By integrating Edge LMMs (Large Multimodal Models), we enable holistic path planning and congestion management for large-scale multi-robot systems.
At AISLab, students master the essential essence of top-tier research through our specialized in-house curriculum, covering everything from fundamental optimization and communication theories to cutting-edge AI architectures. We pursue research that transcends academic boundaries, aiming to build the intelligent infrastructure that defines the future of connected society.
Research Interests & Projects
AI-Native RAN and Wireless Foundation Models
Research on intelligent Radio Access Network architectures that interpret high-level human intent and translate it into real-time network policies. This includes developing Wireless Foundation Models that fuse digital twin data with field measurements for proactive channel prediction.
Distributed Learning and Wireless Edge Intelligence
Development of efficient distributed machine learning algorithms for resource-constrained wireless devices. Our focus includes privacy-preserving computation and channel-aware task offloading frameworks that adapt to time-varying wireless conditions.
Connected Collaborative Robotics
Establishing seamless coordination systems for multi-robot fleets via 6G connectivity. We leverage Edge LMMs (Large Multimodal Models) to analyze fleet-wide telemetry and provide holistic guidance for navigation and congestion management.
Integrated Sensing and Communications (ISAC)
Studies on integrating high-precision environmental sensing capabilities with wireless communication functions into a single system. This is a core technology for environmental perception and autonomous systems in the 6G era.
Research on intelligent Radio Access Network architectures that interpret high-level human intent and translate it into real-time network policies. This includes developing Wireless Foundation Models that fuse digital twin data with field measurements for proactive channel prediction.
Distributed Learning and Wireless Edge Intelligence
Development of efficient distributed machine learning algorithms for resource-constrained wireless devices. Our focus includes privacy-preserving computation and channel-aware task offloading frameworks that adapt to time-varying wireless conditions.
Connected Collaborative Robotics
Establishing seamless coordination systems for multi-robot fleets via 6G connectivity. We leverage Edge LMMs (Large Multimodal Models) to analyze fleet-wide telemetry and provide holistic guidance for navigation and congestion management.
Integrated Sensing and Communications (ISAC)
Studies on integrating high-precision environmental sensing capabilities with wireless communication functions into a single system. This is a core technology for environmental perception and autonomous systems in the 6G era.
Journals & Patents
Selected Publications (by Research Area)
1. AI-Native RAN and Wireless Foundation Models
Research focuses on designing AI-native communication architectures that translate human intent into network policies or utilize Large Language Models (LLMs) within communication layers for autonomous optimization.
Selected Papers:
* H. J. Yang, H. Kim, H. Noh, S. Kim, and B. Shim, "Large Language and Multimodal Models for Task-Oriented Autonomous Communications: Opportunities and Challenges," accepted, IEEE Vehicular Technology Magazine (VTM), 2026.
* S. Ryu and H. J. Yang, "Standards-Compliant DM-RS Allocation via Temporal Channel Prediction for Massive MIMO Systems," accepted, IEEE Transactions on Vehicular Technology (TVT), 2026.
* S. Park, H. Noh, and H. J. Yang, "Robust Transmission of Punctured Text With Large Language Model-Based Recovery," IEEE Transactions on Vehicular Technology (TVT), vol. 75, no. 1, 2026.
2. Distributed Learning and Wireless Edge Intelligence
Focuses on algorithms for training models in communication-constrained edge environments while maintaining data privacy. Emphasis is placed on federated learning and framework designs that maximize computation offloading and communication efficiency.
Selected Papers:
* J. Jang, H. Lyu, D. J. Love, and H. J. Yang, "Fed-ZOE: Communication-Efficient Over-the-Air Federated Learning via Zeroth-Order Estimation," under review, IEEE Journal on Selected Areas in Communications (JSAC), 2026.
* M. Kim, J. Jang, Y. Choi, and H. J. Yang, "Distributed Task Offloading and Resource Allocation for Latency Minimization in Mobile Edge Computing Networks," IEEE Transactions on Mobile Computing (TMC), vol. 23, no. 12, 2024.
* S. Ryu, J. Jang, and H. J. Yang, "Noise Variance Optimization in Differential Privacy: A Game-Theoretic Approach Through Per-Instance Differential Privacy," IEEE Access, vol. 12, 2024.
3. Connected Collaborative Robotics
Involves communication-control co-design for organic cooperation among multiple robots over 6G networks. It utilizes edge intelligence to optimize fleet navigation paths and integrate intelligent sensing for large-scale multi-robot systems.
Selected Papers:
* H. J. Yang, H. Lee, K. Shim, J. Kwak et al., "Advancing Multi-Robot Networks via MLLM-Driven Sensing, Communication, and Computation: A Comprehensive Survey," under review, IEEE Communications Surveys and Tutorials (COMST), 2025.
* H. Lyu, H. Noh, H. J. Yang, and K. Chowdhury, "Secure Multi-Hop Relaying in Large-Scale Space-Air-Ground-Sea Integrated Networks," under review, IEEE Transactions on Wireless Communications (TWC), 2025.
* J. Jang, H. Lyu, H. J. Yang, M. Oh, and J. Lee, "Deep Learning-Based Autonomous Scanning Electron Microscope," Proc. IEEE/RSJ Int. Conf. on Intelligent Robots and Systems (IROS), 2020.
Summary of Research Content
AI-RAN: We develop technologies to autonomously manage complex 6G networks using LLM/LMMs and dramatically improve the accuracy of signal recovery and channel prediction through generative AI at the physical layer (PHY).
Distributed Learning: To overcome the power and bandwidth limitations of wireless terminals, we design high-efficiency distributed learning frameworks that utilize zeroth-order optimization or exchange only model parameters.
Connected Robot: Leveraging the ultra-low latency characteristics of 6G, we build collaborative navigation systems where robot fleets sense the environment in real-time and move along optimal, congestion-free paths guided by edge intelligence.
1. AI-Native RAN and Wireless Foundation Models
Research focuses on designing AI-native communication architectures that translate human intent into network policies or utilize Large Language Models (LLMs) within communication layers for autonomous optimization.
Selected Papers:
* H. J. Yang, H. Kim, H. Noh, S. Kim, and B. Shim, "Large Language and Multimodal Models for Task-Oriented Autonomous Communications: Opportunities and Challenges," accepted, IEEE Vehicular Technology Magazine (VTM), 2026.
* S. Ryu and H. J. Yang, "Standards-Compliant DM-RS Allocation via Temporal Channel Prediction for Massive MIMO Systems," accepted, IEEE Transactions on Vehicular Technology (TVT), 2026.
* S. Park, H. Noh, and H. J. Yang, "Robust Transmission of Punctured Text With Large Language Model-Based Recovery," IEEE Transactions on Vehicular Technology (TVT), vol. 75, no. 1, 2026.
2. Distributed Learning and Wireless Edge Intelligence
Focuses on algorithms for training models in communication-constrained edge environments while maintaining data privacy. Emphasis is placed on federated learning and framework designs that maximize computation offloading and communication efficiency.
Selected Papers:
* J. Jang, H. Lyu, D. J. Love, and H. J. Yang, "Fed-ZOE: Communication-Efficient Over-the-Air Federated Learning via Zeroth-Order Estimation," under review, IEEE Journal on Selected Areas in Communications (JSAC), 2026.
* M. Kim, J. Jang, Y. Choi, and H. J. Yang, "Distributed Task Offloading and Resource Allocation for Latency Minimization in Mobile Edge Computing Networks," IEEE Transactions on Mobile Computing (TMC), vol. 23, no. 12, 2024.
* S. Ryu, J. Jang, and H. J. Yang, "Noise Variance Optimization in Differential Privacy: A Game-Theoretic Approach Through Per-Instance Differential Privacy," IEEE Access, vol. 12, 2024.
3. Connected Collaborative Robotics
Involves communication-control co-design for organic cooperation among multiple robots over 6G networks. It utilizes edge intelligence to optimize fleet navigation paths and integrate intelligent sensing for large-scale multi-robot systems.
Selected Papers:
* H. J. Yang, H. Lee, K. Shim, J. Kwak et al., "Advancing Multi-Robot Networks via MLLM-Driven Sensing, Communication, and Computation: A Comprehensive Survey," under review, IEEE Communications Surveys and Tutorials (COMST), 2025.
* H. Lyu, H. Noh, H. J. Yang, and K. Chowdhury, "Secure Multi-Hop Relaying in Large-Scale Space-Air-Ground-Sea Integrated Networks," under review, IEEE Transactions on Wireless Communications (TWC), 2025.
* J. Jang, H. Lyu, H. J. Yang, M. Oh, and J. Lee, "Deep Learning-Based Autonomous Scanning Electron Microscope," Proc. IEEE/RSJ Int. Conf. on Intelligent Robots and Systems (IROS), 2020.
Summary of Research Content
AI-RAN: We develop technologies to autonomously manage complex 6G networks using LLM/LMMs and dramatically improve the accuracy of signal recovery and channel prediction through generative AI at the physical layer (PHY).
Distributed Learning: To overcome the power and bandwidth limitations of wireless terminals, we design high-efficiency distributed learning frameworks that utilize zeroth-order optimization or exchange only model parameters.
Connected Robot: Leveraging the ultra-low latency characteristics of 6G, we build collaborative navigation systems where robot fleets sense the environment in real-time and move along optimal, congestion-free paths guided by edge intelligence.
