Welcome to Ziyao Zhang's Home Page      

alt text 

Ziyao (Spike) Zhang (Chinese: 张子尧)
PhD Candidate
Communication and Signal Processing Group
Electrical and Electronic Engineering Department
Imperial College London
Exhibition Road
London SW7 2BT
United Kingdom

Email: ziyao.zhang15@imperial.ac.uk
Homepage: http://www.commsp.ee.ic.ac.uk/~zz315/

My name is Ziyao (feel free to call me Spike), I am currently a PhD student with the Communication and Signal Processing Group in the Department of Electrical and Electronics Engineering at Imperial College London. I started my PhD program under the supervision of Prof. Kin K. Leung in December 2016. Since then, I have been mainly working in collobration with Dr. Liang Ma (IBM T.J. Watson Research Center) on the International Technology Alliance in Distributed Analytics and Information Sciences (DAIS-ITA) project. My research interests include performance analysis of the Software-Defined Networking (SDN) paradigm, trade-offs among different network resources, structures of complex networks, and the control & management of SDN networks using deep reinforcement learning.
Before starting the PhD program, I obtained my Master of Science (MSc) degree in Communication and Signal Processing from Imperial College London, also under the supervision of Prof. Kin K. Leung, in September 2016. My MSc degree thesis was on caching algorithms in Content Delivery Networks (CDNs).
Prior to joining Imperial College London, I completed my Bachelor of Engineering (BEng) degree in Electronic and Information Engineering at Northwestern Polytechnical University (NPU), graduating from the first Internation Class. My final year project for BEng degree focused on network throughput analysis in 5G mmWave system. During my undergraduate program, I studied at the University of Strathclyde for one year between 2013 and 2014 as a visiting student.

Research Interests

  • Distributed Software-defined Networking (SDN)

  • Distributed Optimization Techniques

  • Structures and Properties of Complex Networks

  • Reinforcement Learning (RL) Techniques in Optimization

  • SDN Network Control and Management using Deep RL

Education

  • PhD candidate in Electrical Engineering, Imperial College London, London, United Kingdom     – December 2016 - present

    • Topic: Performance Quantification and Controller Synchronization Policy Design of Distributed SDN Architecture (Part of the DAIS-ITA project )

    • Supervisor: Prof. Kin K. Leung

    • Funded By: DAIS-ITA project

    • Collaborating with Dr. Liang Ma at IBM T. J. Watson Research Center

  • MSc in Communications and Signal Processing, Imperial College London, London, United Kingdom     – October 2015 - September 2016

    • Thesis Topic: Caching Algorithms and their Performance Analysis for Content Delivery Networks

    • Thesis Advisor: Prof. Kin K. Leung

  • BEng in Electronics and Information Engineering, Northwestern Polytechnical University, Xi'an, P.R.China     – September 2011 - June 2015

    • Thesis Topic: Wireless Resources Allocation Algorithms for Contention-based and Concurrent Transmission in 5G mmWave Network

    • Thesis Advisor: Dr. Zhongjinag Yan

    • Thesis Grade: 93/100; Degree Grade: 91.26/100 (Class Top 1)

  • Non-Degree Exchange Student, University of Strathclyde, Glasgow, United Kingdom     – September 2013 - June 2014

    • Overall Grade: 88/100

  • Work Experience

  • Postgraduate Researcher - Imperial College London     – December 2016 - present
    • Project: DAIS-ITA

    • London, United Kingdom

  • Research Intern - IBM T.J. Watson Research Center     – June - present 2019

    • Project: Distributed, Robust, and Scalable SDN Controller Synchronization Policy Design

    • Mentor: Dr. Liang Ma

    • Manager: Dr. Mudhakar Srivatsa

    • Yorktown Heights, NY, United States

  • Research Intern - IBM T.J. Watson Research Center     – July - November 2018

    • Project: SDN Controller Synchronization Policy Design Using Deep Reinforcement Learning

    • Mentor: Dr. Liang Ma

    • Manager: Dr. Mudhakar Srivatsa

    • Yorktown Heights, NY, United States

  • Teaching Assistant - Imperial College London     – November 2017 - January 2018
    • Course: Introduction to Signals and Communications

    • Task: Organize MATLAB tutorial manuals and instructions

    • London, United Kingdom

  • MSc Lab Demonstrator - Imperial College London     – October 2017 - January 2018

    • Course: Multi-Input Multi-Output (MIMO) Communication System

    • Task: Demonstrate lab experiments; Q & A and logbook marking

    • London, United Kingdom

  • Visiting Researcher - IBM T.J. Watson Research Center     – June - July 2017

    • Project: DAIS-ITA

    • Host: Dr. Liang Ma

    • Yorktown Heights, NY, United States

  • Publications and Presentations

  • Ziyao Zhang, Liang Ma, Konstantinos Poularakis, Kin K. Leung, Jeremy Tucker, Ananthram Swami, “MACS: Deep Reinforcement Learning based SDN Controller Synchronization Policy Design”, 2019 IEEE 27th Internation Conference on Network Protocols, Chicago, United States (Available online on arXiv)

  • (BEST PAPER AWARD) Ziyao Zhang, Liang Ma, Konstantinos Poularakis, Kin K. Leung, Lingfei Wu, “DQ Scheduler: Deep Reinforcement Learning Based Controller Synchronization in Distributed SDN”, 2019 IEEE 53rd Internation Conference on Communications, Shanghai, China (Available online on arXiv)

  • Ziyao Zhang, Liang Ma, Kin K. Leung, Leandros Tassiulas, Jeremy Tucker, “Q-placement:Reinforcement-Learning-Based Service Placement in Software-Defined Networks”, 2018 IEEE 38th Internation Conference on Distributed Computing Systems, Vienna, Austria (Available online on IEEE Xplore)

  • Ziyao Zhang, Liang Ma, Kin K. Leung, Franck Le, Sastry Kompella, Leandros Tassiulas, “How Advantageous Is It? An Analytical Study of Controller-Assisted Path Construction in Distributed SDN”, IEEE/ACM Transactions on Networking (Available online on IEEE Xplore)

  • Ziyao Zhang, Liang Ma, Kin K. Leung, Franck Le, “More Is Not Always Better: An Analytical Study of Controller Synchronizations in Distributed SDN”, submitted to IEEE Journal on Selected Areas in Communications, under review

  • Ziyao Zhang, Liang Ma, Konstantinos Poularakis, Kin K. Leung, Jeremy Tucker, Ananthram Swami,“DRL based SDN Controller Synchronization Policy Design for joint Communication and Computation Resource Optimisations ”, Proceedings of the DAIS-ITA Annual Fall Meeting 2019, Weybridge, United Kingdom (Available online on Dais-ITA Science Library)

  • Ziyao Zhang, Qiaofeng Qin, Liang Ma, Konstantinos Poularakis, Franck Le, Kin K. Leung, Sastry Kompella, Leandros Tassiulas,“Routing Performance in Distributed SDN under Synchronization Constraint”, Proceedings of the DAIS-ITA Annual Fall Meeting 2018, New York, United States (Available online on Dais-ITA Science Library)

  • Ziyao Zhang, Liang Ma, Kin K. Leung, “On Quantifying Performance Enhancemnet of Distributed SDN Architecture”, Proceedings of the DAIS-ITA Annual Fall Meeting 2017, London, United Kingdom (Available online on Dais-ITA Science Library) (presentation) (poster)

  • Liang Ma, Ziyao Zhang, Bongjun Ko, Mudhakar Srivatsa, Kin K. Leung, “Resource management in distributed SDN using reinforcement learning”, SPIE 2018, Orlando, United States (Available online on SPIE)