Code & Data Highlight

"Writing code is not production, it's not always craftsmanship though it can be, it is design." -- Joel Spolsky

System implementations and data are key part of my research. Besides delivering contributions via scientific publications, CPI Lab always release, if permitted, our work as open-source software and datasets. Here I list software implementations and datasets that are completed through CPI projects. The source codes and datasets can be found in corresponding repositories.

Quick link to Source Codes | Research Datasets

System Implementations

SoftOfflod

SoftOffload is a joint work with Helsinki, Cambridge and Columbia through EIT-SDN project.

SoftOfflod Homepage

SoftOffload for Mobile Data Offloading

SoftOfflod Team

Aaron Ding - TU Delft / Helsinki
Yanhe Liu - Sina / Helsinki
Hannu Flinck - Nokia Bell Labs
Sasu Tarkoma - Helsinki
Henning Schulzrinne - Columbia
Jon Crowcroft - Cambridge

SoftOfflod Code

GitHub repository

MADNet

Metropolitan Advanced Delivery Network (MADNet) is a collaborative WiFi-based mobile data offloading architecture that targets to improve energy efficiency for smartphones. MADNet achieves collaboration by intelligently aggregating the power of cellular operators, WiFi service providers and end-users. It utilizes an energy-aware algorithm to assist the offloading decision for smartphones, which enables them to select the most energy efficient WiFi APs to offload mobile traffic. MADNet project is in collaboration with Deutsche Telekom T-Labs, AT&T and University of Maryland College Park.

MADNet Publications

"Energy Awareness in Mobile Traffic Offloading"
Aaron Ding, Jon Crowcroft
Cambridge University Press: Smartphone Energy Consumption: Modelling and Optimization, 2014.
"Enabling Energy-Aware Collaborative Mobile Data Offloading for Smartphones"
Aaron Ding, Bo Han, Yu Xiao, Pan Hui, Aravind Srinivasan, Markku Kojo, Sasu Tarkoma
IEEE SECON, New Orleans, USA, June 2013.
"Collaborative Communication and Sensing for Mobile Systems"
Aaron Ding, Jon Crowcroft, Sasu Tarkoma
ACM SenSys Doctoral Colloquium, Rome, Italy, November 2013.
"Enable Energy-Aware Mobile Data Offloading for Smartphones through Vertical Collaboration"
Aaron Ding, Pan Hui, Markku Kojo, Sasu Tarkoma
ACM CoNEXT PhD Student Workshop, Nice, France, December 2012. (32 accepted out of 62 submissions)
"Enabling Energy-aware and Collaborative Management of Data Traffic for Mobile Systems"
Aaron Ding, Markku Kojo, Sasu Tarkoma
ACM EuroSys Doctoral Workshop, Bern, Switzerland, April 2012.

MADNet Team

Aaron Ding - TU Delft / Helsinki
Yu Xiao - Aalto
Bo Han - AT&T Research / UMD
Pan Hui - Deutsche Telekom Labs
Sasu Tarkoma - Helsinki

MADNet Code

GitHub repository

NAO

Network-Assisted Offloading (NAO) is a framework that utilizes network assistance to achieve efficient IP traffic offloading. It enables mobile network operators to actively trigger and guide the traffic offloading through dedicated protocols. NAO supports two offloading modes, namely host-driven mode with DHCPv6 and network-driven mode with Neighbor Discovery. NAO project is in collaboration with Nokia and Broadcom.

NAO Publications

"Toward Network Controlled IP Traffic Offloading"
Jouni Korhonen, Teemu Savolainen, Aaron Ding, Markku Kojo
IEEE Communications Magazine (IEEE COMMAG), Volume 51, Issue 3, p.96 - 102, 2013.
"Controlling Traffic Offloading Using Neighbor Discovery Protocol"
Jouni Korhonen, Teemu Savolainen, Aaron Ding
IETF Internet Draft, 2012.
"NAO: A Framework to Enable Efficient Mobile Offloading"
Aaron Ding, Jouni Korhonen, Pan Hui, Teemu Savolainen, Sasu Tarkoma, Markku Kojo
ACM Middleware PDT Workshop, Lisbon, Portugal, December 2011.
Aaron Ding, Jouni Korhonen, Teemu Savolainen, "Controlling Traffic Offloading Using Neighbor Discovery Protocol", In Proceedings of IETF-83, Internet Area, France, 2012.
» Presentation at IETF, Paris.
Aaron Ding, Jouni Korhonen, Teemu Savolainen, "Controlling Traffic Offloading Using Neighbor Discovery Protocol", In Proceedings of IETF-82, Internet Area, 2011.
» Presentation at IETF, Taipei.