NTraX – Mobile Device Interface Traces

In the linked compressed SQLite database android_device_interfaces.db , we provide overviews for the data generated by mobile user device interfaces. The overall table architecture is:

  • device_{01-30}_{down,up,sum} for the entire device network traffic by direction
  • device_{01-30}_{cell,wifi}_{down,up,sum} for the different network interface traffic by direction

In each table, you find the zero-indexed timestamps in the time column for the measurement (in 2 minute bin intervals) and the corresponding amount of data as size in byte.

A more detailed overview can be found at
http://people.cst.cmich.edu/seeli1p/NTRAXReports/index.html

Please reference the following publication if you use this data:

S. Mead, N. Veeramachaneni, and P. Seeling. An Overview of Mobile Device Network Activities: Characteristics of Heterogeneous Network Interfaces. In Proceedings of  IEEE Consumer Communications and Networking Conference (CCNC), Las Vegas, NV, USA, January 2016.

NTraX – Network Usage Tracking across Applications and Interfaces (IEEE OnlineGreenComm 2013)

In this demonstration, we present the NTraX mobile application and server-side processing framework for profiling of mobile device network utilization. The NTraX mobile application component allows fine grained capturing of network activity on the mobile device without negative battery impacts. Submission of the captured data to the remote server allows for server-side processing and profiling (for individual users and anonymously amongst users), ultimately resulting in recommendations to the user how to optimize their communications strategically while providing a rich anonymous dataset for mobile optimizations.

A PDF outlining our demo is available. The application is available in the Play Store at https://play.google.com/store/apps/details?id=edu.cmich.cps.ntrax

An Inexpensive Testbed for Mobile Device Power Measurement (IEEE OnlineGreenComm 2013)

We outline our mobile device power measurement testbed, which is based on commonly available and inexpensive hardware components. Our setup offers whole mobile system power consumption measurements in a variety of scenarios and can readily modified to suit a broad range of implementation configurations.

A PDF outlining our demo is available.

An archive file with the scripts required to execute the framework is available here: MobilePowerTestbed.