Course: Advanced Network Computing

From MobiNetS
Revision as of 10:29, 22 September 2021 by Shiqi (talk | contribs)
Jump to: navigation, search

Introduction

This course will introduce recent advances in network computing. Specifically, it contains four sections: SDN/NFV, Edge Computing, IoT, and Network Big Data. In SDN/NFV, we mainly talk about two topics: Consistent update, traffic engineering and VNF embedding. In Edge Computing, we mainly talk about the topics of deployment and task offloading. In Internet-of-Things, we mainly talk about the topics of low-power protocols, long-range IoT and IoT applications/systems. In Network Big Data, we mainly talk about the topics of wireless sensing and various applications based on smartphone big data.

Anc bg img.png

Reference books:

  • Software Defined Networks: A Comprehensive Approach, 2014
  • Network Function Virtualization: Ken Gray, Thomas D. Nadeau, 2016
  • Big Data and The Internet of Things, 2015
  • Edge Computing: From Hype to Reality, 2019

Requirements (2021)

Final score = 40% x class presentation+ 60% x exam.

Class presentation

The class score comes from the class presentations.

  • Each student is required to participate in a presentation (15 min) of a research paper selected from the top conferences/journals in network computing.
  • Each presentation allows up to four presenters.
  • The fourth presenter will have 10 points fewer than the first three members.
  • You may also choose to serve as a session chair so that you do not have to present papers. The session chair will share the same score with the highest score of the session papers.

Paper selection

Conference/journal rankings:

  • Class A: SigComm/INFOCOM/MobiCom/ToN/TMC/TPDS/JSAC/TC/NSDI/
  • Class B: MobiHoc/MobiSys/SenSys/ICDCS/IPSN/ICNP/IWQoS/other IEEEACM Transactions
  • Class C: SECON/MASS/DCOSS/Posters from Class-A conferences

If you have no idea after going through the above list, you may select papers from the list.

Time slots

We have in total 32 time slots for presentations. Please send a mail to "anc_course@mobinets.org" and apply for a specific slot. The format of the mail is shown as below. 

Title: 时间序列申请 (Time slot application)
Mail: 
论文 (Paper info)：XXXXX, Conference/Journal name, Year
申请序列 (Preferred slots)：14,21,1
演讲成员 (Members)：
刘华强 (Name1)；202101233210
鹿晗 (Name2)； 201901233211
赵本山 (Name3)； 202001233212

If you are applying for a session chair role, please follow the template below. 

Title: Session chair application
Mail: 
Session ID: 5, 3, 1
StuInfo: 刘华强，202101233211

Please refer to the following time table before applying for a specific time slot. Green blocks are available slots.

Time\Date 10/11 10/18 10/25 11/01
19:30-20:00 1 2 9 10 23 24 37 38
20:05-20:35 3 4 11 12 25 26 39 40
20:40-21:10 5 6 13 14 27 28 41 42
21:15-21:45 7 8 15 16 29 30 43 44
Time\Date 10/20 10/27
14:30-15:15 17 18 19 31 32 33
15:20-16:05 20 21 22 34 35 36


Presentations:

  • Session 1 10/11 19:30-20:35(1-4) (Chair: 王翰生)
    1. 刘婧媛，张鑫 Analyzing User-Level Privacy Attack Against Federated Learning，JSAC'20
    2. 王朴真，蒋国庆 Interpreting Deep Learning-Based Networking Systems，SigComm‘20
    3. 徐昊，江威峰 Meaningful Availability NSDI'20
    4. 郑凯文,徐显阳,丛荣 L2D2: low latency distributed downlink for LEO satellites，SigComm‘21
  • Session 2 10/11 20:40-21:10(5-8) (Chair: 张煦)
  1. 张林元齐，陈嘉俊 When to Hedge in Interactive Services ，NSDI 2021
  2. 蒋欣辰 Adaptive Clustering-based Malicious Traffic Classification at the Network Edge, INFOCOM'21
  3. 于子青，彭鑫才 Learning-Based Mobile Edge Computing Resource Management to Support Public Blockchain Networks TMC'21
  4. 刘祥坤,彭志昊 EagleEye: Wearable Camera-based Person Identification in Crowded Urban Spaces （MobiCom）2021
  • Session 3 10/18 19:30-20:35(9-12) (Chair: 杨轶)
  1. 黄宇胜，刘嘉贺 GCN-GAN: A Non-linear Temporal Link Prediction Model for Weighted Dynamic Networks INFOCOM'19
  2. 张钦栋，徐盛 Edge Computing Assisted Adaptive Mobile Vedio Streaming，IEEE Transactions on Mobile Computing'21
  3. 孙凡，陈柯宇 Distributed Threshold-based Offloading for Large-Scale Mobile Cloud Computing，Infocom 2021
  4. 张瀚文,张铭露，翟圣杰 Beyond Inferring Class Representatives: User-Level Privacy Leakage From Federated Learning, INFOCOM’19
  • Session 4 10/18 20:40-21:10(13-16) (Chair: 周在磊)
  1. 彭翔宇，陈睿 5G Wireless Communications with High Mobility，JSAC'2020
  2. 刘佳欣，张宁，邓力玮 Predictability and Prediction of Human Mobility Based on Application-Collected Location Data, TMC, 2021
  3. 张欣然，陈维龙，王鹏 Optimizing Federated Learning on Non-IID Data with Reinforcement Learning, INFOCOM'20
  4. 苑辉辉,曹武豪 EdgeSharing: Edge Assisted Real-time Localization and Object Sharing in Urban Streets,INFOCOM'2021
  • Session 5 (17-19) (Chair: XXX)
  1. 葛孟雨 A Federated Learning Approach for Smart Health Sensing with Class Imbalance in Resource Constrained Edge Computing, INFOCOM'21
  2. 陈一，卢庚源 Learning-Driven Decentralized Machine Learning in Resource-Constrained Wireless Edge Computing, IEEE INFOCOM, 2021
  3. 李欣悦，戴子玟 WiForce: Wireless Sensing and Localization of Contact Forces on a Space Continuum
  • Session 6 (20-22) (Chair: 周子涵)
  1. 邓紫菱,周楠馨,刘文君 Server-driven video streaming for deep learning inference, SigComm'20
  2. 李思怡，武宁宁 Network Planning with Deep Reinforcement Learning, SIGCOMM, 2021
  3. 冯丹阳，罗涵 Reducto: On camera filtering for resource-efficient real-time video analytics, SIGCOMM 2020
  • Session 7 (23-26) (Chair: XXX)
  1. 邵贤栋，彭宇博 WiTAG: Seamless WiFi Backscatter Communication， Sigcomm2020
  2. 常智珩，韩孟群 (How Much) Does a Private WAN Improve Cloud Performance?，infocom'2020
  3. 刘靖，陈玄 NitroSketch: Robust and General Sketch-based Monitoring in Software Switches，INFOCOM'19
  4. 李思思，苑舒雨 Fire in Your Hands: Understanding Thermal Behavior of Smartphones, MobiCom'19
  • Session 8 (27-30) (Chair: XXX)
  1. 周子涵，张棋 HeadFi: Bringing Intelligence to All Headphones (MobiCom2021)
  2. 王亚鑫，赵昊，张鹏宇 Hop-by-Hop Multipath Routing:Choosing the Right Nexthop Set，INFOCOM
  3. 李仲意，施柯 Understanding Operational 5G: A First Measurement Study on Its Coverage, Performance and Energy Consumption，SigComm '20
  4. 刘卓柳，王海燕 Robust Service Mapping in Multi-Tenant Clouds， INFOCOM21
  • Session 9 (31-33) (Chair: 张馨月)
  1. 张廷意，李林峰 Neural Tensor Completion for Accurate Network Monitoring，INFOCOM'20
  2. 魏凡越，孙翔宇 Swift: Delay is Simple and Effective for Congestion Control in the Datacenter，SigComm ' 20
  3. 张馨月,杨瑞,林贵鑫 Sdn candidate selection in hybrid ip/sdn networks for single link failure protection，ToN '20
  • Session 10 (34-36) (Chair: XXX)
  1. 王瀚生 Ownership: A Distributed Futures System for Fine-Grained Tasks，NSDI'21
  2. 程柯菡,张煦,张宇轩 Large-scale Vehicle Trajectory Reconstruction with Camera Sensing Network, Mobicom, 2021
  3. 吴同宇，罗凯文 Layer Aware Microservice Placement and Request Scheduling at the Edge，infocomm，2021
  • Session 11 (37-40) (Chair: XXX)
  1. 孙震，巩嘉诚 PDVocal: Towards Privacy-preserving Parkinson’s Disease Detection using Non-speech Body Sounds，MobiCom，19
  2. 卫文杰,罗通,汪玉洁 Sensing and Communication Integrated System for Autonomous Driving Vehicles
  3. 柯德劲，Your Home is Insecure: Practical Attacks on Wireless Home Alarm Systems，INFOCOM'21
  4. 李洪平，赵卓逸 PTAuth: Temporal Memory Safety via Robust Points-to Authentication NSDI’21
  • Session 12 (41-44) (Chair: XXX)
  1. 张春潮，Interference Recycling: Exploiting Interfering Signals to Enhance Data Transmission，INFOCOM‘19
  2. 倪孝泽，宋书玮 Understanding Ethereum via Graph Analysis, INFCOM, 2018
  3. 冯佳铭，严然 Keep Others From Peeking At Your Mobile Device Screen!，MobiCom，2019

Exam

The exam will contain two types of questions: Concept understanding (7 questions) and problem-solving (2 questions). Don't worry if you do not major in computer networks because all questions will be selected from class tests (answers will be discussed in class).

Bonus system

You have chance to get bonus points before the presentations and the final exam. Here are the ways how you get bonus:

  • Select papers from top venues.

We encourage all students to select papers from the Class-A conferences and journals.

  1. A Class-A paper gets 4 bonus points for presentation.
  2. A Class-B paper gets 2 bonus points for presentation.
  • Select earlier time slots

As most of you have no experience of public presentations, I believe pioneers will take more pressure. As a result, the earlier you do the presentation, more bonus you get. Specifically,

  1. time slots 1-6 get 3 bonus points
  2. time slots 9-14 get 2 bonus points
  3. time slots 7,8, 17-22 get 1 bonus points
  • Raise questions to presenters

To encourage the discussion of the presentation, each raised question will be given a 2 points bonus. Each student can gain 6 bonus (three questions) points at most.

Slides

The slides will be sent to all students in class. Note: the questions and answers will not be shared.

Reference books:

  • Software Defined Networks: A Comprehensive Approach, 2014
  • Network Function Virtualization: Ken Gray, Thomas D. Nadeau, 2016
  • Big Data and The Internet of Things, 2015
  • Edge Computing: From Hype to Reality, 2020

Referenced links:

Retrieved from "http://mobinets.cn/site/index.php?title=Course:Advanced_Network_Computing&oldid=2065"