Seminar on Internet Technologies (Summer 2017): Difference between revisions

From NET Wiki
Jump to navigation Jump to search
 
(28 intermediate revisions by 10 users not shown)
Line 32: Line 32:
==Schedule==
==Schedule==
* '''Apr. 20, 16:00ct''': Introduction meeting  
* '''Apr. 20, 16:00ct''': Introduction meeting  
* '''TBA''' : Deadline for registration
* '''Jun. 22, 2017''' : Deadline for registration
* '''TBA''' : Presentations
* '''Jun. 29, 2017''' : Presentations
* '''September. 30, 2017, 23:59''': Deadline for submission of report (should be sent to the topic advisor!)
* '''September. 30, 2017, 23:59''': Deadline for submission of report (should be sent to the topic advisor!)


== Topics ==
== Topics ==
Line 46: Line 45:
|-
|-
|-
|-
| '''Deep into Google Translate'''   
| '''Deep into Google Translate (assigned to Monisha Khurana)'''   
This study is to provide a comprehensive study of one of the Google products - Google translate and aim to understand the technologies behind it.
This study is to provide a comprehensive study of one of the Google products - Google translate and aim to understand the technologies behind it.
| [http://www.net.informatik.uni-goettingen.de/people/Hong_Huang Hong Huang]
| [http://www.net.informatik.uni-goettingen.de/people/Hong_Huang Hong Huang]
Line 56: Line 55:
| [http://science.sciencemag.org/content/350/6264/1073][http://www.sciencedirect.com/science/article/pii/S0378873314000033]
| [http://science.sciencemag.org/content/350/6264/1073][http://www.sciencedirect.com/science/article/pii/S0378873314000033]
|-
|-
|'''An overview on deep learning framework'''
|'''An overview on deep learning framework (assigned to Fangxi Deng)'''
In this work, you will be asked to do a survey on all popular deep learning framework either in academe or industry, like tensorflow, caffe and so on. You shall elaborate their shortcomings and advantages.
In this work, you will be asked to do a survey on all popular deep learning framework either in academe or industry, like tensorflow, caffe and so on. You shall elaborate their shortcomings and advantages.
|[http://www.net.informatik.uni-goettingen.de/people/Hong_Huang Hong Huang]
|[http://www.net.informatik.uni-goettingen.de/people/Hong_Huang Hong Huang]
|[https://deeplearning4j.org/compare-dl4j-torch7-pylearn]
|[https://deeplearning4j.org/compare-dl4j-torch7-pylearn]
|-
|-
| '''Industrie 4.0: Networking prospective and challenges'''   
| '''Industrie 4.0: Networking prospective and challenges (Assigned to Hailiang Li)'''   
Germany is targeting reach Industry 4.0 stage in factories. You should survey all requirements from networking prospective and the main challenges.
Germany is targeting reach Industry 4.0 stage in factories. You should survey all requirements from networking prospective and the main challenges.
'''NOTE:'''This topic could be a good entry for master project and thesis later.  
'''NOTE:'''This topic could be a good entry for master project and thesis later.  
Line 72: Line 71:
|[http://www.cryptovest.co.uk/resources/Bitcoin%20paper%20Original.pdf][https://www.usenix.org/system/files/login/articles/03_meiklejohn-online.pdf]
|[http://www.cryptovest.co.uk/resources/Bitcoin%20paper%20Original.pdf][https://www.usenix.org/system/files/login/articles/03_meiklejohn-online.pdf]
|-
|-
| '''Legacy devices support in SDN controllers'''
| '''Legacy devices support in SDN controllers (Assigned to Ankita Bajpai)'''
'''NOTE:''' This topic could be a good entry for master project and thesis later.  
'''NOTE:''' This topic could be a good entry for master project and thesis later.  
Supporting legacy network is an active research area in SDN. You should survey all techniques used up to date to solve this problem. Details may be provided later.  
Supporting legacy network is an active research area in SDN. You should survey all techniques used up to date to solve this problem. Details may be provided later.  
Line 79: Line 78:
|-
|-
| '''Google QUIC'''
| '''Google QUIC'''
QUIC is an experimental transport layer network protocol designed by Jim Roskind at Google, initially implemented in 2012. Investigate QUIC in detail and conduct some simple experiments to compare QUIC with TCP. The experiments should be designed by the student himself/herself.
QUIC is an experimental transport layer network protocol designed by Jim Roskind at Google, initially implemented in 2012. Investigate QUIC in detail and compare QUIC with TCP and TCP variants.
| [http://www.net.informatik.uni-goettingen.de/people/enhuan_dong Enhuan Dong]
| [http://www.net.informatik.uni-goettingen.de/people/enhuan_dong Enhuan Dong]
|[https://en.wikipedia.org/wiki/QUIC][https://docs.google.com/document/d/1RNHkx_VvKWyWg6Lr8SZ-saqsQx7rFV-ev2jRFUoVD34/edit][https://datatracker.ietf.org/wg/quic/about/][https://github.com/google/proto-quic][https://groups.google.com/a/chromium.org/forum/#!topic/proto-quic/CioG51ecKB8]
|[https://en.wikipedia.org/wiki/QUIC][https://docs.google.com/document/d/1RNHkx_VvKWyWg6Lr8SZ-saqsQx7rFV-ev2jRFUoVD34/edit][https://datatracker.ietf.org/wg/quic/about/][https://github.com/google/proto-quic][https://groups.google.com/a/chromium.org/forum/#!topic/proto-quic/CioG51ecKB8]
|-
|-
| '''Google TCP BBR'''
| '''Google TCP BBR'''
TCP BBR is developed by Google. Investigate BBR in detail and conduct some simple experiments to compare BBR with TCP Cubic.The experiments should be designed by the student himself/herself.
TCP BBR is developed by Google. Investigate BBR in detail and compare TCP BBR with TCP and TCP variants.
| [http://www.net.informatik.uni-goettingen.de/people/enhuan_dong Enhuan Dong]
| [http://www.net.informatik.uni-goettingen.de/people/enhuan_dong Enhuan Dong]
|[http://queue.acm.org/detail.cfm?id=3022184][https://github.com/google/bbr][https://groups.google.com/forum/#!forum/bbr-dev]
|[http://queue.acm.org/detail.cfm?id=3022184][https://github.com/google/bbr][https://groups.google.com/forum/#!forum/bbr-dev]
|-
|-
| '''Commercial usage of Multipath TCP'''
| '''Commercial usage of Multipath TCP (assigned to Mojtaba Shabani)'''
MultiPath TCP (MPTCP) is an emerging extension for TCP and it is under discussion in IETF now. Study  MPTCP protocol including architecture, data transmission, default congestion control, etc. Investigate how MPTCP is used in companies.   
MultiPath TCP (MPTCP) is an emerging extension for TCP and it is under discussion in IETF now. Study  MPTCP protocol including architecture, data transmission, default congestion control, etc. Investigate how MPTCP is used in companies.   
| [https://www.net.informatik.uni-goettingen.de/people/enhuan_dong Enhuan Dong]
| [https://www.net.informatik.uni-goettingen.de/people/enhuan_dong Enhuan Dong]
| [https://tools.ietf.org/html/rfc6824][http://link.springer.com/chapter/10.1007%2F978-3-642-20757-0_35][https://www.usenix.org/conference/nsdi12/technical-sessions/presentation/raiciu][http://dl.acm.org/citation.cfm?id=2342476][http://dl.acm.org/citation.cfm?id=2631977][https://www.usenix.org/legacy/event/nsdi11/tech/full_papers/Wischik.pdf][http://blog.multipath-tcp.org/blog/html/2015/12/25/commercial_usage_of_multipath_tcp.html]
| [https://tools.ietf.org/html/rfc6824][http://link.springer.com/chapter/10.1007%2F978-3-642-20757-0_35][https://www.usenix.org/conference/nsdi12/technical-sessions/presentation/raiciu][http://dl.acm.org/citation.cfm?id=2342476][http://dl.acm.org/citation.cfm?id=2631977][https://www.usenix.org/legacy/event/nsdi11/tech/full_papers/Wischik.pdf][http://blog.multipath-tcp.org/blog/html/2015/12/25/commercial_usage_of_multipath_tcp.html]
|-
|-
| '''Traffic Data Analysis'''
| '''Traffic Data Analysis (assigned to Michael Debono)'''
Great amount of traffic data are generated everyday from private cars, subway, taxi and buses, etc. Traffic data analysis is of great help to understand the patterns of people mobility, transport planning, urban management and policymaking. And it is also an interesting way to learn some basic knowledge about big data and machine learning.
Great amount of traffic data are generated everyday from private cars, subway, taxi and buses, etc. Traffic data analysis is of great help to understand the patterns of people mobility, transport planning, urban management and policymaking. And it is also an interesting way to learn some basic knowledge about big data and machine learning.
| [https://www.net.informatik.uni-goettingen.de/people/shichang_ding Shichang Ding]
| [Shichang Ding --  shichang.ding@informatik.uni-goettingen.de]
| [http://journals.plos.org/plosone/article?id=10.1371/journal.pone.0149222][https://pdfs.semanticscholar.org/7d15/0a9390d569750978d9abcee4524f1974961f.pdf]
| [http://journals.plos.org/plosone/article?id=10.1371/journal.pone.0149222][https://pdfs.semanticscholar.org/7d15/0a9390d569750978d9abcee4524f1974961f.pdf]
|-
|-
| '''Robo advisors and AI'''
| '''Robo advisors and AI'''
A robo-advisor (robo-adviser) is an online wealth management service that provides automated, algorithm-based portfolio management advice without the use of human financial planners. Robo-advisor is one of new examples which show how AI begin to take place of human beings in high-end service like finance, laws, education and even research. Beside gaining basic knowledge about AI, it is also a good way to understand how AI change our future work markets.
A robo-advisor (robo-adviser) is an online wealth management service that provides automated, algorithm-based portfolio management advice without the use of human financial planners. Robo-advisor is one of new examples which show how AI begin to take place of human beings in high-end service like finance, laws, education and even research. Beside gaining basic knowledge about AI, it is also a good way to understand how AI change our future work markets.
| [https://www.net.informatik.uni-goettingen.de/people/shichang_ding Shichang Ding]
| [Shichang Ding --  shichang.ding@informatik.uni-goettingen.de]
| [http://onlinepresent.org/proceedings/vol141_2016/21.pdf]
| [http://onlinepresent.org/proceedings/vol141_2016/21.pdf]
|-
|-
Line 110: Line 109:


|-
|-
| '''Mobile network data for public health - A Survey'''   
| '''Large-Scale Mobile Traffic Analysis - A Survey (Assigned to Mian Athar Naqash)'''   
This study is to provide a comprehensive study of Mobile network data for public health.
This study is to provide a comprehensive study of large-scale mobile traffic analysis. You should survey the related work about this topic and focus on some classical research work. You also need to give your own opinion on the topic.
| [http://www.net.informatik.uni-goettingen.de/people/tao_zhao Tao Zhao]
| [http://www.net.informatik.uni-goettingen.de/people/tao_zhao Tao Zhao]
| [http://journal.frontiersin.org/article/10.3389/fpubh.2015.00189/full]
| [http://perso.citi-lab.fr/mfiore/data/naboulsi_comst15.pdf]


|-
|-
| '''Understanding and modelling individual human mobility'''   
| '''Understanding and modelling individual human mobility'''   
This study is to provide a comprehensive study of understanding and modelling individual human mobility.
This study is to provide a comprehensive study of understanding and modelling individual human mobility. You should survey the related work about this topic and focus on some classical research work. You also need to give your own opinion on the topic.
| [http://www.net.informatik.uni-goettingen.de/people/tao_zhao Tao Zhao]
| [http://www.net.informatik.uni-goettingen.de/people/tao_zhao Tao Zhao]
| Take a look at related papers in well known conferences/workshops/journals, e.g., [http://www.ccsb.dfci.harvard.edu/web/export/sites/default/ccsb/publications/papers/2010/Song--Barabasi_NatPhysics_10.pdf]  
| Take a look at related papers in well known conferences/workshops/journals, e.g., [http://www.ccsb.dfci.harvard.edu/web/export/sites/default/ccsb/publications/papers/2010/Song--Barabasi_NatPhysics_10.pdf]  


|-
|-
| '''Recommendations in Location-based Social Networks - A Survey'''   
| '''Recommendations in Location-based Social Networks - A Survey (Assigned to Hussain Nauman)'''   
This study is to provide a comprehensive study of recommendations in Location-based Social Networks.
This study is to provide a comprehensive study of recommendations in Location-based Social Networks. You should survey the related work about this topic and focus on some classical research work. You also need to give your own opinion on the topic.
| [http://www.net.informatik.uni-goettingen.de/people/tao_zhao Tao Zhao]
| [http://www.net.informatik.uni-goettingen.de/people/tao_zhao Tao Zhao]
| [https://www.microsoft.com/en-us/research/publication/recommendations-in-location-based-social-networks-a-survey/]  
| [https://www.microsoft.com/en-us/research/publication/recommendations-in-location-based-social-networks-a-survey/]  


|-
|-
|'''ICN - Information Centric Networking'''
|'''ICN - Information Centric Networking (Assigned to Wazed Ali)'''  


Content Centric Networking (CCN) is a new ambitious proposal to replace the IP protocol. A better and faster content distribution, improved privacy, integrated cryptography and easy P2P communication are among the key elements of this architecture. On the other hand problems like efficiency and scalability of the name-based routing, support of existing application and new ones and the possibility to actually deploy this technology are still open and actively discussed, making CCN one of the most active research field in networking.  
Content Centric Networking (CCN) is a new ambitious proposal to replace the IP protocol. A better and faster content distribution, improved privacy, integrated cryptography and easy P2P communication are among the key elements of this architecture. On the other hand problems like efficiency and scalability of the name-based routing, support of existing application and new ones and the possibility to actually deploy this technology are still open and actively discussed, making CCN one of the most active research field in networking.  
Line 175: Line 174:
|-
|-
|-
|-
| '''Learning from Imbalanced Data'''
| '''Learning from Imbalanced Data''' (assigned to Christoph Rauterberg)
When building and training classifiers for classification problems, one commonly encountered problem is that of imbalanced data. For instance, in the case of a binary classifier, this means that one class is hugely overrepresented in the data available. Training classifiers for this kind of datasets has been a problem for some time. In this work, your task is to i) precisely introduce the imbalanced data problem, ii) discuss the state of the art of approaches for mitigating this problem (both from the perspective of learning algorithms and data manipulation techniques) and iii) find out what issues still remain open until today. Note that this topic requires a background in data science, and in particular in classification algorithms. Also, this topic requires a comparatively high reading effort.
When building and training classifiers for classification problems, one commonly encountered problem is that of imbalanced data. For instance, in the case of a binary classifier, this means that one class is hugely overrepresented in the data available. Training classifiers for this kind of datasets has been a problem for some time. In this work, your task is to i) precisely introduce the imbalanced data problem, ii) discuss the state of the art of approaches for mitigating this problem (both from the perspective of learning algorithms and data manipulation techniques) and iii) find out what issues still remain open until today. Note that this topic requires a background in data science, and in particular in classification algorithms. Also, this topic requires a comparatively high reading effort.
| [https://www.net.informatik.uni-goettingen.de/people/David_Koll David Koll ]
| [https://www.net.informatik.uni-goettingen.de/people/David_Koll David Koll ]
| [http://ieeexplore.ieee.org/xpl/articleDetails.jsp?arnumber=5128907&tag=1]
| [http://ieeexplore.ieee.org/xpl/articleDetails.jsp?arnumber=5128907&tag=1]
|-
|-
| '''Deep Learning and its (possible) flaws'''
| '''Deep Learning and its (possible) flaws''' (assigned to Sven Voigt)
One recent trend in machine learning is 'deep learning', where neural networks are employed for solving a wide range of problems. One prominent example of such problems is image classification. While neural networks are in fact delivering sometimes great results, they may also have some weak spots. In this work, your task is to i) make yourself familiar with neural networks, ii) discuss the state-of-the-art in image classification, and iii) to investigate some possible flaws in neural networks. Note that for this topic a background in data science, and in particular in classification algorithms, is strongly recommended. Also, this topic requires a comparatively high reading effort.
One recent trend in machine learning is 'deep learning', where neural networks are employed for solving a wide range of problems. One prominent example of such problems is image classification. While neural networks are in fact delivering sometimes great results, they may also have some weak spots. In this work, your task is to i) make yourself familiar with neural networks, ii) discuss the state-of-the-art in image classification, and iii) to investigate some possible flaws in neural networks. Note that for this topic a background in data science, and in particular in classification algorithms, is strongly recommended. Also, this topic requires a comparatively high reading effort.
| [https://www.net.informatik.uni-goettingen.de/people/David_Koll David Koll ]
| [https://www.net.informatik.uni-goettingen.de/people/David_Koll David Koll ]
Line 186: Line 185:
|-
|-
| '''How do self-driving cars work?'''   
| '''How do self-driving cars work?'''   
The topic title is pretty self-explanatory :)
The topic title is pretty self-explanatory :) (however, you need to understand the math behind it to some extent).
| [https://www.net.informatik.uni-goettingen.de/people/David_Koll David Koll ]
| [https://www.net.informatik.uni-goettingen.de/people/David_Koll David Koll ]
| [http://cs.stanford.edu/people/teichman/papers/iv2011.pdf]
| [http://cs.stanford.edu/people/teichman/papers/iv2011.pdf]
Line 196: Line 195:
| [https://pdfs.semanticscholar.org/e682/85544e01b2075d8d5fe65569232a3de840cc.pdf] [http://web.cs.ucla.edu/classes/fall03/cs218/paper/pgmcc.pdf] [http://conferences2.sigcomm.org/acm-icn/2016/proceedings/p11-chen.pdf]
| [https://pdfs.semanticscholar.org/e682/85544e01b2075d8d5fe65569232a3de840cc.pdf] [http://web.cs.ucla.edu/classes/fall03/cs218/paper/pgmcc.pdf] [http://conferences2.sigcomm.org/acm-icn/2016/proceedings/p11-chen.pdf]
|-
|-
| '''Adaptive Video Streaming'''   
| '''Adaptive Video Streaming (assigned to Nikolaj Kopp)'''   
Today's Internet is a heterogeneous networking environment. In such an environment, resources available to multimedia applications vary. To adapt to the changes in network conditions, both networking techniques and application layer techniques have been proposed. The study must give an overview of the different techniques proposed and some real use-case scenarios (ever heard about a company named Netflix??)
Today's Internet is a heterogeneous networking environment. In such an environment, resources available to multimedia applications vary. To adapt to the changes in network conditions, both networking techniques and application layer techniques have been proposed. The study must give an overview of the different techniques proposed and some real use-case scenarios (ever heard about a company named Netflix??)
| [https://www.net.informatik.uni-goettingen.de/people/jacopo_de%20benedetto Jacopo De Benedetto ]
| [https://www.net.informatik.uni-goettingen.de/people/jacopo_de%20benedetto Jacopo De Benedetto ]
Line 206: Line 205:
| [https://www.net.informatik.uni-goettingen.de/people/jacopo_de%20benedetto Jacopo De Benedetto ]
| [https://www.net.informatik.uni-goettingen.de/people/jacopo_de%20benedetto Jacopo De Benedetto ]
| [http://ieeexplore.ieee.org/stamp/stamp.jsp?arnumber=6807945] [https://www.qualcomm.com/invention/research/projects/lte-direct] [https://www.wi-fi.org/discover-wi-fi/wi-fi-aware]
| [http://ieeexplore.ieee.org/stamp/stamp.jsp?arnumber=6807945] [https://www.qualcomm.com/invention/research/projects/lte-direct] [https://www.wi-fi.org/discover-wi-fi/wi-fi-aware]
|-
| '''Microsoft Natick'''
Natick is Microsoft research project to manufacture and operate an underwater datacenter. The goal of this topic is study the impact of underwater datacenters on environment and performance compared to modular datacenters.
| [http://www.net.informatik.uni-goettingen.de/people/abhinandan%20s_prasad Abhinandan S Prasad]
| [http://natick.research.microsoft.com/]
|-
| '''Big Data Optimization Algorithms'''
Big data is a current buzz word in both industry and academia. The goal of this topic is to study atleast two convex optimization based big data optimizations like firts-order, randomization, etc.
| [http://www.net.informatik.uni-goettingen.de/people/abhinandan%20s_prasad Abhinandan S Prasad]
| [https://arxiv.org/pdf/1411.0972.pdf]
|-
| '''Prediction Markets'''
Prediction markets are exchange-traded markets created for the purpose of trading the outcome of events. The market prices indicate the probability of an event. The goal is to study and understand how prediction markets work.
| [http://www.net.informatik.uni-goettingen.de/people/abhinandan%20s_prasad Abhinandan S Prasad]
| [https://en.wikipedia.org/wiki/Prediction_market][http://www.nature.com/news/the-power-of-prediction-markets-1.20820][https://dash.harvard.edu/handle/1/5027266]
|-
| '''Privacy-Aware Platform for Personal Data (Assigned to Alireza) '''
Take a look at the open IoT platforms and provde a summary of the differences, similarities and vision.
| [http://www.net.informatik.uni-goettingen.de/people/mayutan_arumaithurai Mayutan Arumaithurai]
| Take a look at Fiware, Sophia, Crystal and other open platforms
|-
| '''Edge Computing for Internet of Things (Assigned to Andrea Melina) '''
A study of the various edge computing solutions that exist for IoT
| [http://www.net.informatik.uni-goettingen.de/people/mayutan_arumaithurai Mayutan Arumaithurai]
| Take a look at Amazon Lambda, Amazon IoT, Amazon greengrass and solutions by other companies
|-
| '''Brief Introduction to High Performance Computing Acceleration (assigned to DOAN Ho Anh Triet)'''
The aim of this work is to take a look at HPC and study their similarities, differences, pros and cons
| [http://www.net.informatik.uni-goettingen.de/people/mayutan_arumaithurai Mayutan Arumaithurai]
| Take a look at recent papers in well known conferences/workshops.
|-
|-

Latest revision as of 11:23, 26 June 2017

Details

Workload/ECTS Credits: 5 ECTS (BSc/MSc AI); 5 (ITIS)
Module: M.Inf.1124 -or- B.Inf.1207/1208; ITIS Module 3.16: Selected Topics in Internet Technologies
Lecturer: Dr. Hong Huang
Teaching assistant: Tao Zhao
Time: Apr 20, 16:00ct: Introduction Meeting
Place: IFI Building, Room 3.101
UniVZ [1]


Course description

This course covers selected topics on the up-to-date Internet technologies and research. Each student takes a topic, does a presentation and writes a report on it. Besides the introduction meeting, there are no regular meetings, lectures or classes for this course. The purpose of this course is to familiarize the students with new technologies, enable independent study of a specific topic, and train presentation and writing skills.

The informational meeting at the beginning of the course will cover some guidelines on scientific presenting and writing.


Due to topic advisors' workload limitation, we could only provide limited topics, and the topic assignment will be on the basis of first come first serve principle. Please contact the topic advisor directly for the topic availability.

Passing requirements

  • Actively and frequently participate in the project communication with your topic advisor. The topic advisor has the right to decide whether a student is eligible for the final presentation or not according to their communication.
    • This accounts for 20% of your grade.
  • Present the selected topic (20 min. presentation + 10 min. Q&A).
    • This accounts for 40% of your grade.
  • Write a report on the selected topic (12-15 pages) (LaTeX Template:[2]).
    • This accounts for 40% of your grade.
  • Please check the #Schedule and adhere to it.

Schedule

  • Apr. 20, 16:00ct: Introduction meeting
  • Jun. 22, 2017 : Deadline for registration
  • Jun. 29, 2017 : Presentations
  • September. 30, 2017, 23:59: Deadline for submission of report (should be sent to the topic advisor!)

Topics

Topic Topic Advisor Initial Readings
Deep into Google Translate (assigned to Monisha Khurana)

This study is to provide a comprehensive study of one of the Google products - Google translate and aim to understand the technologies behind it.

Hong Huang [3]
Inferring social capital from big data

This study is to discover the state of art of social capital measuring, particularly, from big data perspective.

Hong Huang [4][5]
An overview on deep learning framework (assigned to Fangxi Deng)

In this work, you will be asked to do a survey on all popular deep learning framework either in academe or industry, like tensorflow, caffe and so on. You shall elaborate their shortcomings and advantages.

Hong Huang [6]
Industrie 4.0: Networking prospective and challenges (Assigned to Hailiang Li)

Germany is targeting reach Industry 4.0 stage in factories. You should survey all requirements from networking prospective and the main challenges. NOTE:This topic could be a good entry for master project and thesis later.

Osamah Barakat [7][8][9]
Bitcoin: state of the art and position paper (Assigned to Amine Lasfar)

This study is to provide a comprehensive study of the current situation on Bitcoin. Latest advances in its structure, security and furture.

Osamah Barakat [10][11]
Legacy devices support in SDN controllers (Assigned to Ankita Bajpai)

NOTE: This topic could be a good entry for master project and thesis later. Supporting legacy network is an active research area in SDN. You should survey all techniques used up to date to solve this problem. Details may be provided later.

Osamah Barakat a good start from [12][13]
Google QUIC

QUIC is an experimental transport layer network protocol designed by Jim Roskind at Google, initially implemented in 2012. Investigate QUIC in detail and compare QUIC with TCP and TCP variants.

Enhuan Dong [14][15][16][17][18]
Google TCP BBR

TCP BBR is developed by Google. Investigate BBR in detail and compare TCP BBR with TCP and TCP variants.

Enhuan Dong [19][20][21]
Commercial usage of Multipath TCP (assigned to Mojtaba Shabani)

MultiPath TCP (MPTCP) is an emerging extension for TCP and it is under discussion in IETF now. Study MPTCP protocol including architecture, data transmission, default congestion control, etc. Investigate how MPTCP is used in companies.

Enhuan Dong [22][23][24][25][26][27][28]
Traffic Data Analysis (assigned to Michael Debono)

Great amount of traffic data are generated everyday from private cars, subway, taxi and buses, etc. Traffic data analysis is of great help to understand the patterns of people mobility, transport planning, urban management and policymaking. And it is also an interesting way to learn some basic knowledge about big data and machine learning.

[Shichang Ding -- shichang.ding@informatik.uni-goettingen.de] [29][30]
Robo advisors and AI

A robo-advisor (robo-adviser) is an online wealth management service that provides automated, algorithm-based portfolio management advice without the use of human financial planners. Robo-advisor is one of new examples which show how AI begin to take place of human beings in high-end service like finance, laws, education and even research. Beside gaining basic knowledge about AI, it is also a good way to understand how AI change our future work markets.

[Shichang Ding -- shichang.ding@informatik.uni-goettingen.de] [31]
Deep Learning and Alphago(Master)

Alphago is one of the best players in board games. One of the important reasons for its great success is deep learning. Deep learning is a class of machine learning algorithms that use a cascade of many layers of nonlinear processing units for feature extraction and transformation. It is now broadly studied and used in following areas: Automatic speech recognition, Image recognition, Natural language processing, Customer relationship management and so on. Alphago (its upgraded version called Master) is one of the most famous and successful applications of deep learning. It is a good way to gain knowledge about this interesting area.

Shichang Ding [32][33]
Large-Scale Mobile Traffic Analysis - A Survey (Assigned to Mian Athar Naqash)

This study is to provide a comprehensive study of large-scale mobile traffic analysis. You should survey the related work about this topic and focus on some classical research work. You also need to give your own opinion on the topic.

Tao Zhao [34]
Understanding and modelling individual human mobility

This study is to provide a comprehensive study of understanding and modelling individual human mobility. You should survey the related work about this topic and focus on some classical research work. You also need to give your own opinion on the topic.

Tao Zhao Take a look at related papers in well known conferences/workshops/journals, e.g., [35]
Recommendations in Location-based Social Networks - A Survey (Assigned to Hussain Nauman)

This study is to provide a comprehensive study of recommendations in Location-based Social Networks. You should survey the related work about this topic and focus on some classical research work. You also need to give your own opinion on the topic.

Tao Zhao [36]
ICN - Information Centric Networking (Assigned to Wazed Ali)

Content Centric Networking (CCN) is a new ambitious proposal to replace the IP protocol. A better and faster content distribution, improved privacy, integrated cryptography and easy P2P communication are among the key elements of this architecture. On the other hand problems like efficiency and scalability of the name-based routing, support of existing application and new ones and the possibility to actually deploy this technology are still open and actively discussed, making CCN one of the most active research field in networking.

By choosing this topic you will gain a general knowledge of the many architecture proposed for ICN and will have to gain insight into one of the problems like routing or security, or solutions (i.e. applications on top of NDN).

  - topics available: Routing in ICN, IoT with ICN, ICN Architectures
- NDN technical report
- ICN Base line scenarios
Sripriya Adhatarao (sripriya-srikant.adhatarao@informatik.uni-goettingen.de) For general introduction:
NFV Frameworks for deployment of Middleboxes and Network Functions in Telco/ISP/Data Center Networks - A Survey

Focus of this topic is to present a comprehensive study of the Industry and academic works targeted towards deployment of NFV in Telecommunications, Data Center and Enterprise networks. Understand and Analyze the key aspects of the predominant NFV frameworks, and characterize them in terms of the adopted standards, resource requirements, deployment factors and constraints, performance metrics, support for service function chaining, etc.

Sameer Kulkarni [37] [38] [39] [40] [41]
Service Plane for Network Functions: Network Service Headers and Other alternatives

Focus of this topic is to understand 'Service Function Chaining of Network Functions', the state-of-the-art proposals like Network Service Headers and related academic works. Reason and justify the need for service plane and then try to propose new mechanisms and design of the data plane to support network services, and the control plane functions necessary to manage these data plane functions.

Sameer Kulkarni [42] [43] [44]
NFV state-of-the-art and Future trends - A survey

Study and Understand Network Function Virtualisation (NFV), the real world use cases and deployment trends of NFV in the Datacenter, telecommunication, private networks. Survey on the reports by standardisation committees and open workgroups like IEFT/ETSI/OPNFV, primarily the specification and requirements for the NFV, and the NFV deployment models. Compare with the available open-source/commercial products if any in the market and make the study of NFV characteristics, the Key Performance Index(KPIs) for NFV and identify the open issues and challenges towards adopting to NFV. Student can choose to carry out either breadth or in-depth on particular aspect of NFV.

Sameer Kulkarni [45] [46] [47] [48] [49]
Towards SDN and NFV Fault Management and High Availability

Network Function Virtualisation (NFV), is gaining rapid momentum, but are they reliable? can they conform with the Telecom operators latency and availability requirements of Fine Nines or Six Nines? The focus of this work is to first study and understand the concerns with NFV in terms of their failures, what amount of availability can they support. Second, study the state-of-the-art in terms of techniques that have been provided in the Cloud and Data Center networks for the traditional Virtual Machine based approaches and make the clear distinction of what aspects can and cannot be adapted? and what are the characteristics of NFV that make them differ from traditional VM based solutions? and aspects and solutions that can be adapted to achieve scalability, efficiency, and reliability in the NFV environments.

Sameer Kulkarni [50] [51] [52] [53]
Learning from Imbalanced Data (assigned to Christoph Rauterberg)

When building and training classifiers for classification problems, one commonly encountered problem is that of imbalanced data. For instance, in the case of a binary classifier, this means that one class is hugely overrepresented in the data available. Training classifiers for this kind of datasets has been a problem for some time. In this work, your task is to i) precisely introduce the imbalanced data problem, ii) discuss the state of the art of approaches for mitigating this problem (both from the perspective of learning algorithms and data manipulation techniques) and iii) find out what issues still remain open until today. Note that this topic requires a background in data science, and in particular in classification algorithms. Also, this topic requires a comparatively high reading effort.

David Koll [54]
Deep Learning and its (possible) flaws (assigned to Sven Voigt)

One recent trend in machine learning is 'deep learning', where neural networks are employed for solving a wide range of problems. One prominent example of such problems is image classification. While neural networks are in fact delivering sometimes great results, they may also have some weak spots. In this work, your task is to i) make yourself familiar with neural networks, ii) discuss the state-of-the-art in image classification, and iii) to investigate some possible flaws in neural networks. Note that for this topic a background in data science, and in particular in classification algorithms, is strongly recommended. Also, this topic requires a comparatively high reading effort.

David Koll [55]
How do self-driving cars work?

The topic title is pretty self-explanatory :) (however, you need to understand the math behind it to some extent).

David Koll [56]
Multicast Video Streaming

In network communication, the transmission of information to multiple recipients can greatly benefits from multicast technology in terms of bandwidth efficiency. In particular, video streaming and downloads are beginning to take a larger share of bandwidth and will probably grow to more than 80% of all consumer Internet traffic by 2020. Although multicast transmission can easily resolve the bandwidth limitations, several issues persist, like flow and congestion control. The study involves analyzing and comparing the different solutions proposed in both research and industry. NOTE:This topic could be a good entry for master project and thesis.

Jacopo De Benedetto [57] [58] [59]
Adaptive Video Streaming (assigned to Nikolaj Kopp)

Today's Internet is a heterogeneous networking environment. In such an environment, resources available to multimedia applications vary. To adapt to the changes in network conditions, both networking techniques and application layer techniques have been proposed. The study must give an overview of the different techniques proposed and some real use-case scenarios (ever heard about a company named Netflix??)

Jacopo De Benedetto [60] [61] [62]
D2D Proximity Services

Sometimes referred as "digital sixth sense", Device-to-device (D2D) proximity discovery enables spectral reuse via D2D communications as well as a range of innovative proximity services, such as enhanced social networking and location services, thus helping in the offload of local data transmission. The study involves analyzing the actual and experimental technological solutions that enables the proximity services and the underlying communication protocols. NOTE:This topic could be a good entry for master project and thesis.

Jacopo De Benedetto [63] [64] [65]
Microsoft Natick

Natick is Microsoft research project to manufacture and operate an underwater datacenter. The goal of this topic is study the impact of underwater datacenters on environment and performance compared to modular datacenters.

Abhinandan S Prasad [66]
Big Data Optimization Algorithms

Big data is a current buzz word in both industry and academia. The goal of this topic is to study atleast two convex optimization based big data optimizations like firts-order, randomization, etc.

Abhinandan S Prasad [67]
Prediction Markets

Prediction markets are exchange-traded markets created for the purpose of trading the outcome of events. The market prices indicate the probability of an event. The goal is to study and understand how prediction markets work.

Abhinandan S Prasad [68][69][70]
Privacy-Aware Platform for Personal Data (Assigned to Alireza)

Take a look at the open IoT platforms and provde a summary of the differences, similarities and vision.

Mayutan Arumaithurai Take a look at Fiware, Sophia, Crystal and other open platforms
Edge Computing for Internet of Things (Assigned to Andrea Melina)

A study of the various edge computing solutions that exist for IoT

Mayutan Arumaithurai Take a look at Amazon Lambda, Amazon IoT, Amazon greengrass and solutions by other companies
Brief Introduction to High Performance Computing Acceleration (assigned to DOAN Ho Anh Triet)

The aim of this work is to take a look at HPC and study their similarities, differences, pros and cons

Mayutan Arumaithurai Take a look at recent papers in well known conferences/workshops.