Research

最新研究方向(2024~): 融合生成式AI與智能感測技術的永續6G智慧網路

隨著 6G 技術的快速發展，永續運算、先進感測技術和人工智慧驅動的智慧系統正逐漸與 6G 融合。本研究致力構建支援自主移動系統的 6G 架構，涵蓋無人機網路及無線基礎設施，以提升系統的適應性與智慧化程度。

本實驗室的研究重點包括：

轉向語意和目標導向的通訊模式: 處理具情境相關性和目標驅動的數據，以提升網路效率。
結合生成式 AI 技術: 增強數據篩選與優化。
實現資源受限環境下的即時數據處理: 確保系統穩定性與快速回應。

我們的研究預期能顯著提升自主移動系統的連接穩定性與數據處理效率，降低 6G 網路的能耗和成本，推動 AI 技術在通訊領域的應用，構建永續智慧的 6G 生態系統，為智慧城市和智慧交通提供關鍵技術支援。

B5G/6G 三維行動無線網路 (B5G/6G 3D Mobile Wireless Networks)

Autonomous Mobile Relay for Satellite-Aerial-Terrestrial Networks

Optimization and Learning in UAV-assisted Cellular Networks

Low-latency Communications with Aerial-Terrestrial Edge Intelligence

Research Summary

Utilizing unmanned aerial vehicles (UAVs) as flying base stations (BSs) has been regarded as a promising approach to assist existing terrestrial cellular networks. Besides, the maneuverability of UAVs offers great potential for network performance improvement by dynamically adjusting the positions of UAVs to fit the dynamic network environment. It is expected that the future cellular networks will consist of macro-cells and many kinds of small-size aerial/territorial BSs.

To support real-time control of unmanned vehicles in aerial-terrestrial networks, it is essential to design low-latency wireless communications. In 5G/6G networks, ultra-reliable low-latency communication (URLLC) is one of the targeting service scenarios. Mobile edge computing and caching provides a promising enabler for URLLC by moving the network functions and resources to proximate edge nodes. Also, integrating AI into the edge (i.e., edge-AI) has been attracting growing attention because of potential applications in smart cities. This is different from traditional centralized learning in which users have to transmit the private raw data to the cloud server.

In ICAN lab, we exploit network optimization and learning theory to obtain comprehensive design principles for intelligent edge assisted low-latency networks. These tasks include the worst-case delay performance evaluation, efficient radio resource utilization, and autonomous network management.

AI-enabled Autonomous UAV-assisted mmWave Wireless Networks

Publication/Patent

Y. J. Chen, W. Chen, and M. L. Ku, "Trajectory Design and Link Selection in UAV-assisted Hybrid Satellite-terrestrial Network," IEEE Communications Letters, 2022.
Y. J. Chen, K. M. Liao, M. L. Ku, F. P. Tso, and G. Y. Chen, "Multi-Agent Reinforcement Learning Based 3D Trajectory Design in Aerial-Terrestrial Wireless Caching Networks," IEEE Transactions on Vehicular Technology, 2021.
Y. J. Chen and D. Y. Huang, "Joint Trajectory Design and BS Association for Cellular-Connected UAV: An Imitation Augmented Deep Reinforcement Learning Approach," IEEE Internet of Things Journal, 2021.
Y. J. Chen, K. M. Liao, and Y. F. Chen, "End-to-End Delay Analysis in Aerial-Terrestrial Heterogeneous Networks," IEEE Transactions on Vehicular Technology, 2021.
K. M. Liao, G. Y. Chen, Y. J. Chen and Y. F. Chen, "End-to-End Delay Analysis in mmWave UAV-assisted Wireless Caching Networks", IEEE Wireless Communications and Networking Conference (WCNC), 2020.
Y. J. Chen, D. Y. Huang, "Trajectory Optimization for Cellular-Enabled UAV With Connectivity Outage Constraint," IEEE Access, 2020.
Y. J. Chen, K. M. Liao, M. L. Ku and F. P. Tso, "Mobility-aware Probabilistic Caching in UAV-assisted Wireless D2D Networks," IEEE Global Communication Conference (GLOBECOM), 2019.
G. J. Nunns, Y. J. Chen, D. K. Chang, K. M. Liao, F. P. Tso and L. Cui, "Autonomous Flying WiFi Access Point," IEEE Symposium on Computers and Communications (ISCC), 2019.
Y. J. Chen, W. Y. Cheng and L. C. Wang, "Learning-assisted Beam Search for Indoor mmWave Networks," IEEE Wireless Communications and Networking Conference Workshops (WCNC Workshop), 2018.

自主無人載具之智慧感測與控制技術 (Intelligent Sensing and Control of Autonomous Unmanned Vehicles)

Autonomous Multi-UAV System for RF Source Localization

Device-free Human Detection and Passive Localization via Wi-Fi Signal

Trajectory Control of Multiple UAVs with Connectivity Guarantees

Research Summary

In this research, we are interested in the possibilities created by UAV based wireless sensing. For example, imagine UAVs arriving behind thick concrete walls. They have no prior knowledge of the area behind these walls. But they were able to sense the invisible area behind the wall. We envision a new era of the smart city enabled by intelligent UAVs with high-accuracy target tracking and activity recognition functionalities. In the long-term goals, we hope to design efficient and reliable flying robots for mission-critical applications. Through the human-robot collaboration, we can achieve the vision of smart disaster relief and smart security. Moreover, to support safety-critical control, it requires further investigation on trajectory design considering an underlying cellular network and its impact on outage performance. It is important to maintain reliable wireless connectivity between UAVs and ground base station (GBS). Also, since carrying a sensor or package can significantly reduce the battery life, the flight efficiency of UAVs requires careful consideration.

In ICAN lab, we have been working on the problem of localization and tracking of RF sources by UAV swarms since 2018. Multi-UAVs are designed to locate and track the RF source by using machine learning (e.g., deep reinforcement learning). Also, we develop UAV communication network testbed for performance evaluation and demonstration. Our designed fully programmable UAV system is based on open architecture and open source. The developed system integrates wireless sensing, flight control, and wireless communication technology. It is expected to become the foundation of autonomous UAV research and achieve the important milestone of AI democratization in the age of drone.

OpenNCU: An Open-source Testbed for Research in Networking and Communications of Autonomous UAVs

Publication/Patent

Y. J. Chen, K. M. Liao, M. L. Ku, F. P. Tso, and G. Y. Chen, "Multi-Agent Reinforcement Learning Based 3D Trajectory Design in Aerial-Terrestrial Wireless Caching Networks," IEEE Transactions on Vehicular Technology, 2021.
Y. J. Chen and D. Y. Huang, "Joint Trajectory Design and BS Association for Cellular-Connected UAV: An Imitation Augmented Deep Reinforcement Learning Approach," IEEE Internet of Things Journal, 2021.
Y. J. Chen, D. K. Chang, and C. Zhang, "Autonomous Tracking Using a Swarm of UAVs: A Constrained Multi-agent Reinforcement Learning Approach," IEEE Transactions on Vehicular Technology, 2020.
Y. J. Chen, D. Y. Huang, "Trajectory Optimization for Cellular-Enabled UAV With Connectivity Outage Constraint," IEEE Access, 2020.
G. J. Nunns, Y. J. Chen, D. K. Chang, K. M. Liao, F. P. Tso and L. Cui, "Autonomous Flying WiFi Access Point," IEEE Symposium on Computers and Communications (ISCC), 2019.

區塊鏈與物聯網資安技術 (Blockchain and IoT Security)

Privacy Protection for Internet of Autonomous Vehicles

Attack and Defense Technology Based on Machine Learning

Energy-efficient Blockchain for IoT Data Integrity

Research Summary

With the recent advances of artificial intelligence (AI), sensing technologies, and wireless communication/networking, AI-enabled Internet of things (IoT) applications are becoming ever more pervasive in everyday life. For example, autonomous unmanned aerial vehicles (UAVs) or Internet of drones (IoD) play a vital role to support various applications including medical, industrial, agricultural, and public safety. To realize real-time perception and autonomous control, computing and communications in AI-enabled IoT can be more complex and heterogeneous than before. Especially for mission-critical applications, such as autonomous driving, a reliable and trustworthy framework for AI-enabled IoT is in an urgent need.

Due to the spatial diversity (e.g., high mobility of nodes) and temporal features (e.g., unreliable link connectivity), security and privacy in IoT are challenging problems. Also, data collected from sensors for AI-based approaches, including deep neural networks, offer great commercial potentials but also pose new challenges on privacy protection. Since IoT devices with limited computing power cannot perform complex encryption and decryption of large datasets, traditional security solution cannot directly be applied.

In ICAN lab, we focus on the two security issues of UAV systems. First, we investigate the AI-based attack in UAV networks to prevent private information from being eavesdropped. Also, we develop security technology which can detect abnormal UAV behaviors such as unauthorized flight and message replay.

Secure UAV Design with Physical Layer Key Generation

Publication/Patent

Y. J. Chen, X. C. Chen, and M. Pan, "Defense against Machine Learning Based Attacks in Multi-UAV Networks: A Network Coding Based Approach," IEEE Transactions on Network Science and Engineering, 2022.
Y. J. Chen, L. C. Wang, and H. Wang, "Stochastic Blockchain for IoT Data Integrity," IEEE Transactions on Network Science and Engineering, 2020.
X. C. Chen and Y. J. Chen, "A Machine Learning Based Attack in UAV Communication Networks," IEEE Vehicular Technology Conference (VTC Fall), 2019.
Y. J. Chen and L. C. Wang, "Privacy Protection for Internet of Drones: A Network Coding Approach," IEEE Internet of Things Journal, 2019.
O. I. Abdullaziz, L. C. Wang, and Y. J. Chen, "HiAuth: Hidden Authentication for Protecting Software Defined Networks," IEEE Transactions on Network and Service Management, 2019.
Y. J. Chen and L. C. Wang, "An Overflow Problem in Network Coding for Secure Cloud Storage," IEEE Transactions on Parallel and Distributed Systems, 2019.
Y. J. Chen, L. C. Wang and C. H. Liao, "Eavesdropping Prevention for Network Coding Encrypted Cloud Storage Systems," IEEE Transactions on Parallel and Distributed Systems, 2016.

Report abuse