Research

My main areas of research include the performance evaluation, control, and economic analysis of computer networks. In particular, my current research interests include:

Some previous projects include:

Acceleration of (large) DNN Training

Research Purposes:

Our goal is to develop new algorithms and new system architectures to address a variety of cutting-edge research issues in distributed machine learning and federated learning (abbreviated to DML). More specifically, we are interested in the theoretic aspects to cope with acceleration and non-iid data distribution problems, and in the system aspects to cope with the near optimal overlapping between communication and computation. Our on-going studies further incorporate AI job scheduling and foundation model training.

Selected Publications:

  1. Qingyang Duan^, Zeqin Wang^, Yuedong Xu*, Shaoteng Liu, Jun Wu, “Mercury: A Simple Transport Layer Scheduler to Accelerate Distributed DNN Training”, Proc. of IEEE Infocom 2022.
  2. Qingyang Duan^, Chao Peng^, Zeqin Wang^, Yuedong Xu*, Shaoteng Liu, John C.S. Lui, “Accelerating Distributed DNN Training via Transport Layer Scheduling”, IEEE Trans. Parallel and Distributed Systems 2023.
  3. Min Wen^, Chengchang Liu, Yuedong Xu*, “Communication Efficient Distributed Newton Method over Unreliable Networks”, AAAI 2024.

Reinforcement Learning for Networking

Research Purposes:

Our goal is to make use of deep reinforcement learning, multi-armed bandit and general Markov decision approaches for resource allocation in practical networking systems. We also aim to explore the interpretability and robustness of deep reinforcement learning in networking systems.

Selected Publications:

  1. Ying Zheng^, Lixiang Lin^, Tianqi Zhang^, Haoyu Chen^, Qingyang Duan^, Yuedong Xu*, Xin Wang, “Enabling Robust DRL-driven Networking Systems via Teacher-Student Learning”, IEEE JSAC 2022. (extended from IEEE Infocom 2021.)
  2. Xinyu You^, Xuanjie Li^, Yuedong Xu*, Hui Feng, Jin Zhao, Huaicheng Yan, “Toward Packet Routing with Fully-distributed Multi-agent Deep Reinforcement Learning”, IEEE Trans. Systems, Man, and Cybernetics: Systems, 2022. (well cited)
  3. Weijia Chen^, Yuedong Xu* and Xiaofeng Wu, “Deep Reinforcement Learning for Multi-resource Multi-machine Job Scheduling”, Proc. of IEEE ICNP 2017, poster. (an early attempt of using DRL in networking).
  4. Wenqi Pan^, Yuedong Xu*, Shaoteng Liu, “Making TCP BBR Pacing Adaptive with Domain Knowledge Assisted Reinforcement Learning”, IEEE Trans. Network Science and Engineering 2023.
  5. Xutong Liu, Jinhang Zuo, Junkai Wang, Zhiyong Wang, Yuedong Xu*, John C.S. Lui, “Learning Context-Aware Probabilistic Maximum Coverage Bandits: A Variance-Adaptive Approach”, IEEE Infocom 2024.

Blockchain Economics and Data Sharing

Research Purposes:

Our goal is to study the intrinsic economic behaviors of blockchain in theory, and design practical blockchain systems for data sharing as well as distributed machine learning. We have thoroughly analyzed the competition of multiple selfish miners in Bitcoin-like blockchain systems, and have done some preliminary studies in layer-2 payment channel networks. We have developed a novel decentralized query engine on the basis of IPFS and Apache Drill.

Selected Publications:

  1. Qianlan Bai^, Yuedong Xu*, Nianyi Liu^, Xin Wang, “Blockchain Mining with Multiple Selfish Miners”, IEEE Trans. Information Forensics and Security, 2023. (link)
  2. Qianlan Bai^, Yuedong Xu*, Xin Wang, “Understanding the Benefit of Being Patient inPayment Channel Networks”, IEEE Trans. Network Science and Engineering, 2022.
  3. Qianlan Bai^, Xinyan Zhou^, Xing Wang^, Yuedong Xu*, Xin Wang, Qingsheng Kong, “A Deep Dive into Blockchain Selfish Mining”, Proc. of IEEE ICC 2019. (one of the most cited papers in blockchain selfish mining.)
  4. Zhiyi Yao, Bowen Ding, Qianlan Bai, Yuedong Xu*, “Minerva: Decentralized Collaborative Query Processing over InterPlanetary File System”, IEEE Trans. Big Data, 2024. (APNet 2019 Poster)  (Academic Data Sharing and Query Platform based on InterPlanetary File System (IPFS) https://www.datahub.pub)

Bigdata-driven Video Streaming

Selected Publications:

  1. Yuedong Xu, Zhujun Xiao, Hui Feng, Tao Yang, Bo Hu and Yipeng Zhou, “Modeling Buffer Starvations of Video Streaming in Cellular Networks with Large-Scale Measurement of User Behavior”. IEEE Trans. Mobile Computing, 2017. 
  2. Lei Huang^, Bowen Ding^, Aining Wang^, Yuedong Xu*, Yipeng Zhou, Xiang Li, “User Behavior Analysis and Video Popularity Prediction on a Large-Scale VoD System”, ACM Trans. Multimedia Computing, Communications and Applications, 2018.
  3. Chen Zhang^, Yuedong Xu*, Yipeng Zhou and Xiaoming Fu, “On the”Familiar Stranger” Phenomenon in a Large-scale VoD System”, Proc. of NetSciCom 2017.

Quality of Experience for Video Streaming

Selected Publications:

  1. Yuedong Xu, Eitan Altman, Rachid El-Azouzi, Majed Haddad, Salah Elayoubi, Tania Jimenez. “Probabilistic Analysis of Buffer Starvation in Markovian Queues”, Proc. of IEEE Infocom 2012.
  2. Yuedong Xu, Eitan Altman, Rachid El-Azouzi, Majed Haddad, Salah Elayoubi, Tania Jimenez. “Analysis of Buffer Starvation with Application to QoE Optimization of Streaming-like Services”, IEEE Trans. Multimedia, 2014.
  3. Yuedong Xu, Salah Elayoubi, Eitan Altman, Rachid El-Azouzi, “Impact of Flow Dynamics on QoE of Streaming Services in Wireless Networks”, Proc. of IEEE Infocom 2013.
  4. Yuedong Xu, Yipeng Zhou, Dah-ming Chiu, “Analytical QoE Models for Bit-rate Switching in Dynamic Adaptive Streaming Service”, IEEE Trans. Mobile Computing, 2014.
  5. Yuedong Xu, Salah Elayoubi, Eitan Altman, Rachid El-Azouzi, Yinghao Yu,“Flow Level QoE for Video Streaming in Wireless Networks”. IEEE Trans. Mobile Computing, 2016.

Large-scale MIMO System for Future Virtual Reality Transmission

Selected Publications:

  1. Zhe Chen^, Xu Zhang^, Sulei Wang^, Yuedong Xu*, Jie Xiong, Xin Wang, “Enabling Practical Large-Scale MIMO in WLANs With Hybrid Beamforming”, IEEE/ACM Trans. Networking, 2021
  2. Zhe Chen, Xu Zhang, Yuedong Xu*, Jie Xiong, Xin Wang. “BUSH: Empowering Large-Scale MU-MIMO in WLANs With Hybrid Beamforming,” Proc. of IEEE INFOCOM 2017.
  3. Zhe Chen^, Xu Zhang^, Yuedong Xu*, Jie Xiong, Yu Zhu and Xin Wang. “MuVi: Multi-View Video Aware Transmission Over MIMO Wireless Systems,” IEEE Trans. Multimedia, 2017.

Content Dissemination in Peer-to-peer and Opportunistic Networks

 

Selected Publications:

  1. Eitan Altman, Philippe Nain, Adam Shwartz and Yuedong Xu*, “Predicting the Impact of Measures against P2P Networks on the Transient Behavior”, Proc. of IEEE Infocom 2011.
  2. Eitan Altman, Philippe Nain, Adam Shwartz and Yuedong Xu*, “Predicting the Impact of Measures Against P2P Networks: Transient Behavior and Phase Transition”, IEEE/ACM Trans. Networking, 2013.
  3. Eitan Altman, Rachid El-azouzi, Daniel Sadoc Menasche, Yuedong Xu, “Forever Young: Aging Control for Smartphones in Hybrid Networks”, ACM Sigmetrics Performance Evaluation Review, 2010. (The first paper on Age of Information to the best of our knowledge)
  4. Eitan Altman, Rachid El-azouzi, Daniel Sadoc Menasche, Yuedong Xu, “Forever Young: Aging Control for Smartphones in Hybrid Networks”, Proc. of ACM Mobihoc 2019 (Alphabetical ranking).
    Remark: It is mistakenly believed that the idea of Age of Information was first proposed in an Infocom’12 paper.

WiFi Localization and Mobile Sensing

Publications:

  1. Zhihao Gu^, Taiwei He^, Junwei Yin^, Yuedong Xu*, Jun Wu, “TyrLoc: A Low-cost Multi-technology MIMO Localization System with A Single RF Chain”, Proc. of ACM Mobisys 2021.
  2. Zhe Chen^, Zhongmin Li^, Guorong Zhu^, Xu Zhang^, Yuedong Xu*, Jie Xiong, Xin Wang, “AWL: Turning Spatial Aliasing From Foe to Friend for Decimeter-Level WiFi Localization”, Proc. of ACM CoNEXT 2017.
  3. Zhe Chen^, Guorong Zhu^, Sulei Wang^, Yuedong Xu*, Jie Xiong, Xin Wang, Jin Zhao, Jun Luo, “M^3: Multipath Assisted Wi-Fi Localization with a Single Access Point”, IEEE Trans. Mobile Computing, 2019.
  4. Sulei Wang^, Zhe Chen^, Yuedong Xu*, Qiben Yan, Chongbin Xu, Xin Wang, “On User Selective Eavesdropping Attacks in MU-MIMO: CSI Forgery and Countermeasure”, Proc. of IEEE Infocom 2019.
  5. Zhihui Gao^, Yunfan Gao^, Sulei Wang^, Dan Li, Yuedong Xu*, “CRISLoc: Reconstructable CSI Fingerprintingfor Indoor Smartphone Localization”, IEEE Internet of Things Journal, 2020.

Network Economics: Kelly Mechanism and Pricing

Publications:

  1. Yuedong Xu, Zhujun Xiao, Tianyu Ni, Xin Wang, Eitan Altman, “On The Robustness of Price-anticipating Kelly Mechanism”, IEEE/ACM Trans. Networking. 2019.
  2. Yuedong Xu, John C.S. Lui, and Dahming Chiu, “On Oligopoly Spectrum Allocation Game in Cognitive Radio Networks with Capacity Constraints”, Elsevier Computer Networks, 2010.
  3. Eitan Altman, Arnaud Legout, Yuedong Xu, “Network Non-neutrality Debate: An Economic Analysis”, Proc. of  IFIP Networking 2011.
  4. Eitan Altman, Manjesh Hanawal, Jose Rojas, Sulan Wong, Yuedong Xu, “Network Neutrality and Quality of Service”,  Proc. of GameNets 2011.
  5. Yuedong Xu, John C.S. Lui, “Multi-dimensional Network Security Game”, Technical Report, 2014. (link)
  6. Yuedong Xu, Yifan Mao, Yifan Zhou, Xu Chen, “Can Early Joining Participants Contribute More?– Timeliness Sensitive Incentivization for Crowdsensing”, Technical Report. (link)