[Dataset] upmc/rollernet (v. 2009-02-02)

top

the initial version

@MISC{upmc-rollernet-2009-02-02,
  author = {Jérémie 
Leguay and Farid Benbadis},
  title = {{CRAWDAD} data set upmc/rollernet (v. 2009-02-02)}, 
  howpublished = {Downloaded from http://crawdad.cs.dartmouth.edu/upmc/rollernet},
  month = feb,  
  year = 2009
}
					

This data includes traces of any opportunistic sighting of Bluetooth devices 
by groups of rollerbladers carrying iMotes in the roller tour in Paris, France.

- You are welcome to use these traces for your research. 
We ask you in return to accept the following terms:

 1- To acknowledge the use of the data in all resulting publications.

Please reference the following paper:

@InProceedings{RollerNet-INFOCOM09,
    author = {Pierre Ugo Tournoux and J\'er\'emie Leguay and Farid Benbadis and Vania Conan and Marcelo Dias de Amorim and John Whitbeck},
    title = {The Accordion Phenomenon: Analysis, Characterization, and Impact on DTN Routing},
    booktitle = {Proc. {IEEE INFOCOM}},
    year = {2009}
}

 2- Not to redistribute the data set to anyone without our permission.

 3- Not to use the data for any other purpose than education and research.

 ====
 NB: In order to gain access to this file and the data we provide, you
     should have already confirmed these terms to us by e-mail. If this is
     not already the case, please send us just an e-mail to confirm.

Every Friday evening and every Sunday afternoon in Paris, weather permitting, 
groups of between 5,000 and 15,000 people go rollerblading. Over the course 
of three hours, the rollerbladers typically cover 30 km, crossing a large portion 
of the city. They are guided by staff members and assisted by public safety forces. 

In order to analyze the mobility of participants, we perform experiments in which 
we deploy sensors, called iMotes, on approximately a hundred volunteers, who could 
be organisers' friends, members of rollerblading associations or members of staff. 

The data set has been collected on August 20, 2006. According to organizers
and police information, about 2,500 people participated to the rollerblading
tour (few rain showers just before the tour resulted in a number of participants
below the average). The total duration of the tour was about three hours,
composed of two sessions of 80 minutes, interspersed with a break of 20 minutes.

The data that we collect allows us to measure and characterize the interactions 
between people over the duration of the roller tours. Such information is helpful 
in the design of new forms of applications in the domains of emergency response, 
location services, and content delivery. These applications will be available 
on mobile phones and will take advantage of the ability of phones to communicate 
directly to other phones in their vicinity, without passing through the traditional 
cell phone network.

In our experiment, 62 iMotes were distributed to a group of people to collect
any opportunistic sighting of other Bluetooth devices (including the other
iMotes distributed). Our deployment plan was as follows:

During an outing, staff members are organized into groups. Over the duration of 
an outing these groups have an almost static position relative to each other: 
there are four groups on each side of the crowd, one group at the front, 
another one at the end, and two groups within the crowd itself. We entrusted 
iMotes to each group. Deploying in this manner should distribute coverage across 
the crowd, though we expect that the wireless network will not be fully connected 
all of the time (indeed, we are interested in studying the disruptions that 
emerge from changing network conditions for devices of limited range). The remaining 
iMotes were entrusted to others within the crowd, many of whom did not have assigned 
positions. In addition to the iMote sensor deployment, we asked other people 
to activate Bluetooth on their mobile phones.

The iMotes use Bluetooth technology and log at a high frequency the devices 
(other iMotes or people's mobile phones) they meet.

Volunteers holding iMotes were made aware of the privacy issues related 
to the experiment: 
the data that is collected includes the unique identifier of each radio device, 
and the times when it comes into radio contact with other devices. This data 
might be associated with descriptive information concerning the individual 
who bears the device (whether she or he is a staff member or a regular participant, 
for instance), however the trace data is stored separately from the names of 
the participants, and RollerMotes forswears any attempt to associate a trace 
with a person's name. Regarding regular participants who are asked to turn 
on their Bluetooth devices, we ensure that data are anonymised after the experiment.

[Traceset] upmc/rollernet/imote (v. 2009-02-02)

top

the initial version.

@MISC{upmc-rollernet-imote-2009-02-02,
  author = {Jérémie 
Leguay and Farid Benbadis},
  title = {{CRAWDAD} trace set upmc/rollernet/imote (v. 2009-02-02)}, 
  howpublished = {Downloaded from http://crawdad.cs.dartmouth.edu/upmc/rollernet/imote},
  month = feb,  
  year = 2009
}
					

This traceset includes traces of any opportunistic sighting of Bluetooth devices 
by groups of rollerbladers carrying iMotes in the roller tour in Paris, France.

I. Data collection and pre-processing:

We tried to keep the processing of data before public release to a
minimum, to allow any flexibility for possible research use. Some
choices had to be made to reduce power consumption, memory use, and
because of specific capabilities of the iMote prototype.
Before using these data for your research, it may be important to
check that it does not impact any of your findings.

1- periodic desynchronized scanning.

In our experiment, iMotes were distributed to a group of people to collect
any opportunistic sighting of other Bluetooth devices (including the other
iMotes distributed). Each iMote scans on a periodic basis for devices,
asking them to respond with their MAC address, via the paging function.

It takes approximately 5 to 10s to perform the complete scanning. After
initial tests, we observe that most of the contacts were recorded with a
5s scanning time, and this value was used in the experiment.

The time granularity between two scannings is 15s. It is important to avoid
synchronization of two iMotes around the same cycle clock, as each of them
cannot respond to any request when it is actively scanning. Therefore, we
implemented a random dephasing on [-5s;+5s] to handle this case.

2- skip-length sequence.

A contact "A sees B" is defined as a period of time where all
successive scanning by A receive a positive answer by B. Ideally an
information should be kept at the end of each contact period.

After preliminary test it became quite clear that a very large number of
contact periods were only separated by one interval. We decided, to avoid
memory overflow, to implement a skip sequence of "one", meaning that a
contact period will only be stopped after two successive failure of a
scanning response. As a consequence, no inter-contact time of less than
two intervals could have been observed.

3- Manual Time synchronization.

Time between iMotes is not synchronized by a central entity, and traces
belonging to different devices bear times which are relative to the
starting time of each device. We recorded the time at which each iMote was
first powered up, which corresponds to time 0 at that iMote. After
collecting the data, we then converted all times into Unix timestamps
(seconds elapsed since 00:00:00 UTC, Jan 1, 1970).

II. iMotes deployment

In the experiment we performed, we were interested in tracking contacts
between different mobile users.

The data set has been collected on August 20, 2006. According to organizers
and police information, about 2,500 people participated to the rollerblading
tour (few rain showers just before the tour resulted in a number of participants
below the average). The total duration of the tour was about three hours,
composed of two sessions of 80 minutes, interspersed with a break of 20 minutes.

During the tour the iMotes has been deployed in three main group of skaters 
divided as following:
    --Staff members which are themselves organized into six groups:
        -Front left and front right
        -Rear left and rear right
        -Front and rear.
        25 iMotes were entrusted among these six groups. These positions are 
		relative and may have not been always respected by the skaters. 
		There are two iMotes that always stayed at their assignated positions, 
		one at the front and one at the back of the tour.
    --Skating associations, which receveid 26 iMotes. This is a group of 
	skilled skaters which were expected to be highly mobile
    --A set of friends which received 11 iMotes. 

The belonging of each iMote ID to one of these groups is described in the trace 
configuration section.

As in the Haggle experiments, we observed that a number of MAC addresses
recorded were different from a known one only by one or two digit. They
were most of the time recorded once for a single time slot. It is clear
that at least a part of them comes for a corrupted signal received on the
link level by our devices. We filtered the data set retaining only MAC adresses
of device that have been seen at least twice.

To protect participants privacy, we choose not to release the MAC address,
neither from the iMotes nor from other external devices recorded. Every
device is given a unique identifier, usually called ID number in this
document. Depending on which number, it might be an iMote or another MAC
address that were recorded from other active Bluetooth devices around.

[Trace] upmc/rollernet/imote/contacts (v. 2009-02-02)

top

the initial version

@MISC{upmc-rollernet-imote-contacts-2009-02-02,
  author = {Jérémie 
Leguay and Farid Benbadis},
  title = {{CRAWDAD} trace upmc/rollernet/imote/contacts (v. 2009-02-02)}, 
  howpublished = {Downloaded from http://crawdad.cs.dartmouth.edu/upmc/rollernet/imote/contacts},
  month = feb,  
  year = 2009
}
					

This trace includes traces of Bluetooth sightings by groups of 
rollerbladers carrying iMotes in the roller tour in Paris, France.

To ease the understanding of data while keeping a sufficent privacy level,
we provide here the group belonging of iMotes ID:

    - Skaters associations ( skilled skaters ):
                    [1  - 26]
    - Staff:
                    [27 - 51]
        3 Front: 27, 33, 41
        2 Front left: 29, 42
        4 Front right: 32, 47, 48, 49
        5 Rear left: 31, 35, 37, 43, 51
        3 Rear right: 34, 44, 45
        6 Rear: 28, 36, 38, 40, 46, 50
        * Nodes 27 and 38 were respectively known to be always at the head and 
		the tail of the roller tour.
        * Nodes 30 and 39 were rescuer, not affected to any particular place.
    - Set of friends:
                    [52 - 62]

The 1050 external devices ( cell phones, PDAs ...) have IDs from 63 to 1112.

The experiment started on Sunday, 20 Aug 2006 14:24:06 (GMT),
and stopped on Sunday, 20 Aug 2006 17:14:00 (GMT).

The dataset containts two data files as follows:

1. "MAC3Btable.dat"
is a file that contains the three first bytes of the MAC address,
associated with each ID. It could be useful to identify the manufacturer
of each external device.

2. "contacts.dat"
is a file which describes the contact that were recorded by all
devices we distributed during this experiment.
A contact between two devices A and B is reported only once and last
the time that A sees B or B sees A.

========================
Examples taken from table.Exp1.dat (two first columns and first rows)
========================
51      377     1156089135      1156089164      7       498
51      377     1156089399      1156089399      8       235
51      377     1156089428      1156089428      9       29
51      377     1156089569      1156089585      10      141
51      377     1156090078      1156090078      11      493
51      377     1156090532      1156090532      12      454
51      377     1156090833      1156090833      13      301
51      377     1156090914      1156090914      14      81
51      377     1156090946      1156090966      15      32
51      377     1156093426      1156093426      16      2460
51      377     1156093798      1156093798      17      372
51      381     1156088465      1156088465      1       0
51      389     1156087797      1156087797      1       0
51      396     1156085474      1156085474      1       0
51      396     1156085603      1156085603      2       129
51      396     1156087039      1156087051      3       1436
========================
                    
- The first and second columns gives the IDs of the devices of
which the contact is reported.
        
- The third and fourth column describe, respectively, the first and
last time when:
    -the address of ID2 were recorded by ID1
    OR  
    -the adress of ID1 were recorded by ID2 for this contact.
        
- The fifth and sixth column are here for reading convenience. The
fifth enumerate contacts with same ID1 and ID2, as 1,2,... . The last
column describes the time difference between the beginning of this
contact and the end of the previous contact with same ID1 and ID2. It
is by convention set to 0 if this is the first contact for this ID1 
and ID2.

- Times are unix timestamps which correspond to the number of seconds
since midnight January 1, 1970 UTC (referred to as the Epoch).

[Author] Jérémie Leguay

top

[Author] Farid Benbadis

top

[Paper] tournoux-rollernet

top