CRAWDAD upb/mobility2011

This is the data from an Android Bluetooth tracing experiment performed for a period of 35 days in an academic environment (University Politehnica of Bucharest) between November 18 and December 22 2011. 

date/time of measurement start: 2011-11-18

date/time of measurement end: 2011-12-22

collection environment: The trace was collected in an academic environment, at University Politehnica of Bucharest. The participants were Bachelor and Master students and research assistants. The data was collected only inside the grounds of the faculty between 8 AM and 8 PM during week-days. There were a total of 22 participants, chosen to be as varied as possible in terms of year, in order to obtain a better approximation of mobility in a real academic environment. Thus, there were twelve Bachelor students (one in the first year, nine in the third and two in the fourth), seven Master students (four in the first year and three in the second) and three research assistants. The participating members were asked to start the application whenever they arrived at the faculty and to turn it off when they left, because we were only interested in the mobility patterns and social interaction in the academic environment.

network configuration: Data was collected using an Android application (http://code.google.com/p/social-tracer/) that registers contacts between mobile devices with Bluetooth. The participants were organized into an opportunistic network, since the tracing was performed using Android phones.

data collection methodology: Data was collected using an Android application (http://code.google.com/p/social-tracer/) that registers contacts between mobile devices with Bluetooth. The application separates between internal encounters (i.e. encounters with other participants in the experiment) and external encounters (i.e. with other devices that have their Bluetooth on). The tracing was performed during weekdays between 8 AM and 8 PM.

sanitization: Data was anonymized by assigning a unique ID for each user from 1 to 22. No MAC addresses or IMEI codes were collected and/or used.

limitation: Since Bluetooth is not a very reliable protocol, it may be possible that some encounters are missed. Also, the polling interval for each device could be controlled by the user (between 5 and 30 minutes), so the timers on two devices would not be necessarily synchronized. For example, device A might see device B at one point, but device B might miss the encounter altogether (or may see just a part of it).

note: Some of the participants have not attended courses or have not uploaded their data, so they should probably be eliminated when using the trace.

This is the data from an Android Bluetooth tracing experiment performed for a period of 35 days in an academic environment (University Politehnica of Bucharest) between November 18 and December 22 2011.

Traceset

upb/mobility2011/bluetooth

Bluetooth encounter trace collected from Android phones in an academic environment

• files: interactions.dat, social.dat
• description: This is the data from an Android Bluetooth tracing experiment performed for a period of 35 days in an academic environment (University Politehnica of Bucharest) between November 18 and December 22 2011.
• measurement purpose: Educational Use, Social Network Analysis, Human Behavior Modeling, Opportunistic Connectivity
• methodology: The trace was collected in an academic environment, at University Politehnica of Bucharest. The participants were Bachelor and Master students and research assistants. The data was collected only inside the grounds of the faculty between 8 AM and 8 PM during week-days. Data was collected using an Android application (http://code.google.com/p/social-tracer/) that registers contacts between mobile devices with Bluetooth. The application separates between internal encounters (i.e. encounters with other participants in the experiment) and external encounters (i.e. with other devices that have their Bluetooth on). The tracing was performed during weekdays between 8 AM and 8 PM. 

upb/mobility2011/bluetooth Traces

• android:
• configuration: Data was collected using an Android application (http://code.google.com/p/social-tracer/) that registers contacts between mobile devices with Bluetooth.

  The participating members were asked to start the application whenever they arrived at the faculty and to turn it off when they left, because we were only interested in the mobility patterns and social interaction in the academic environment.

• format: There are two files in this trace. The first one (interactions.dat) contains all the interactions between devices participating in the experiment. Each line contains the following entries:

Observer ID --- Observed ID --- Encounter start --- Encounter end --- Number of

encounters between these two devices --- Duration until the next encounter

between these two devices

The IDs start from 1 and go to 22 for internal devices, and higher than 22 for

the external ones. The encounter start and end times are in seconds since

January 1 1970, while the duration until the next encounter between the same two

devices is in milliseconds.

The file is sorted first by observer ID, and encounters with the same observer

are sorted by encounter start time. It contains both internal encounters

(observed ID is lower than or equal to 22) and external encounters (with the ID

higher than 22).

The second file (social.dat) contains the social network between the

participating nodes. It is a comma-separated matrix with each line corresponding

to a device (line 1 is device with ID 1, line 2 is device with ID 2, etc.). Each

column represents a node in the experiment (first column is node 1, etc.), and

it can be either 1 if the nodes have a social relationship (i.e. if the device

owners are friends on Facebook) and 0 otherwise.