[Dataset] novay/cosphere (v. 2009-05-01)

top

the initial version

@MISC{novay-cosphere-2009-05-01,
  author = {Arjan Peddemors},
  title = {{CRAWDAD} data set novay/cosphere (v. 2009-05-01)}, 
  howpublished = {Downloaded from http://crawdad.cs.dartmouth.edu/novay/cosphere},
  month = may,  
  year = 2009
}
					

The CoSphere (Communication Context for Adaptive Mobile Applications) trial gathered 
network traces on the personal mobile devices of 12 trial participants over a period 
of approximately one month in the February/March 2007 time frame.

(adapted from Rice data license @ crawdad)

1. We grant you a nonexclusive, nontransferable license to use the 
data and/or code for commercial, educational, and/or research purposes 
only. You agree to not redistributed the data/code without our previous 
express written approval.

2. The traces we provide are anonymized. To respect the privacy of 
those human subjects whose activity is captured by the data, you will
not attempt to reverse the anonymization process. This may include but
is not limited to identifying specific cell ids, access points, node 
addresses, the actual users, or their location.

3. You agree to acknowledge the source of the data, i.e., the 
MobilityModels'08 paper or one of our more recent papers describing the 
data.

THIS DATA IS PROVIDED ON AN "AS IS" BASIS, WITHOUT ANY WARRANTY OR IMPLIED 
FITNESS FOR A PARTICULAR PURPOSE. IN NO EVENT SHALL WE BE LIABLE FOR ANY 
DIRECT, INDIRECT, INCIDENTAL, SPECIAL, EXEMPLARY, OR CONSEQUENTIAL DAMAGES.

The objective of the CoSphere Trial is to collect network availability and 
network status traces as observed on multi-homed devices of real users. 
We ran this experiment with 12 users in a six week period in February/March 2007, 
logging cellular, 802.11 wireless lan, Bluetooth and (indirectly) fixed USB events 
on Windows CE devices.

We want to use the traces of this experiment to investigate the possibilities 
for 'experience-based network resource usage on mobile hosts'. The basic idea 
is that knowledge on the mobility patterns of the user helps to optimize the 
usage of network resources by applications on mobile hosts, especially for 
those applications that have a degree of delay-tolerance. We have focused on 
the prediction in time of a mobility event of interest, such as 'getting in 
range of the home 802.11 network'. 

We propose a method using preceding events as predictors for the event of 
interest, as discussed in the papers Density Estimation for Out-of-Range 
Events on Personal Mobile Devices and Predicting mobility events on personal 
devices.

The CoSphere trial gathered network traces on the personal mobile devices
of 12 trial participants over a period of approximately one month in the
February/March 2007 time frame. The dataset contains information from three 
different wireless network interfaces - cellular, 802.11 and Bluetooth - 
and provides insight into the richness and dynamics of the visibility of 
wireless networks from a user oriented perspective.

The traces were collected using the Network Abstraction Layer (NAL) software
available at http://cosphere.novay.nl/nal/ .

One participant experienced several clock resets to the extent that it was 
not possible to restore the chronological order of the logged events. The
traces of this participant are not included in the dataset, reducing the
total number of participants to 11. 
To safeguard the privacy of the participants, the data is anonymized in the 
following way. All references to cellular operator names, operator numbers,
location area codes, cell ids, SSIDs, BSSIDs, BD_ADDRs are replaced with
randomized identifiers. The mapping of the real identifiers to the 
anonymized identifiers is consistently over all participants, so that, for 
example, it is possible to determine whether two participants have been
attached to the same 802.11 access point.

[Traceset] novay/cosphere/visibility (v. 2009-05-01)

top

the initial version.

@MISC{novay-cosphere-visibility-2009-05-01,
  author = {Arjan Peddemors},
  title = {{CRAWDAD} trace set novay/cosphere/visibility (v. 2009-05-01)}, 
  howpublished = {Downloaded from http://crawdad.cs.dartmouth.edu/novay/cosphere/visibility},
  month = may,  
  year = 2009
}
					

The CoSphere (Communication Context for Adaptive Mobile Applications) trial gathered 
network traces on the personal mobile devices of 12 trial participants over a period 
of approximately one month in the February/March 2007 time frame.

The dataset contains the following visibility information:
- the cellular operator name, operator number, location area code (LAC), 
  and cell id (CID) of the associated base station (monitored continuously, 
  storing a new base station when the association changes)
- the cellular operator name and operator number of all in-range cellular 
  networks (scanned once every 5 minutes)
- the service set identifier (SSID), the basic service set identifier 
  (BSSID) and the association state of all in range 802.11 access points 
  (monitored during 1 minute, every 10 minutes)
- the Bluetooth device address (BD_ADDR) and device name of all in range 
  Bluetooth nodes (scanned once every 5 minutes)
- the remaining battery power percentage
- device attachment to AC power

[Trace] novay/cosphere/visibility/arff (v. 2009-05-01)

top

the initial version

@MISC{novay-cosphere-visibility-arff-2009-05-01,
  author = {Arjan Peddemors},
  title = {{CRAWDAD} trace novay/cosphere/visibility/arff (v. 2009-05-01)}, 
  howpublished = {Downloaded from http://crawdad.cs.dartmouth.edu/novay/cosphere/visibility/arff},
  month = may,  
  year = 2009
}
					

The CoSphere (Communication Context for Adaptive Mobile Applications) trial gathered 
network traces on the personal mobile devices of 12 trial participants over a period 
of approximately one month in the February/March 2007 time frame.

After a scan on a network interface, the visibility state is taken to
be constant until the next scan, which means that for 802.11 a change may 
occur every 10 minutes (more often when the user interacts with the device)
and for Bluetooth a change may occur every 5 minutes.

The traces are stored in files of the Attribute-Relation File Format (ARFF,
see http://www.cs.waikato.ac.nz/~ml/weka/arff.html ), in sparse form. Every
line describes the visibility of cellular, 802.11 and Bluetooth networks at
a specific time. Lines are not printed at regular intervals, but only when
the current visibility state differs from the state described in the previous
line. 

The sparse .arff file starts with a section describing the 'attributes', 
followed by a 'data' section describing the value of these attributes. The
sparse notation assigns a default value of 0 (zero) to attributes not
specified on a data line. We use binary values to indicate whether a base
station, access point or node is in range. 

Example .arff file:
@RELATION visibility
@ATTRIBUTE timestamp date                                      % attr 0
@ATTRIBUTE after-reset {0, 1}                                  % attr 1
@ATTRIBUTE batterylevel NUMERIC                                % attr 2
@ATTRIBUTE on-ac {0, 1}                                        % attr 3
@ATTRIBUTE "cid-copname01-copnumber01-lac1398-cid82300" {0, 1} % attr 4
@ATTRIBUTE "cid-copname01-copnumber01-lac1398-cid23897" {0, 1} % attr 5
@ATTRIBUTE "wlap-inrange-ssid58034-bssid27992" {0, 1}          % attr 6
@DATA
{0 2007-03-10T16:08:58, 1 1, 2 81, 4 1, 6 1}
{0 2007-03-10T16:10:46, 2 80, 4 1}
{0 2007-03-10T16:12:06, 2 80, 5 1}

The timestamp, afterreset, batterylevel and on-ac attributes (numbers 0-3) 
are in every .arff file. The rest of the attributes refer to the visibility 
of network resources, using the following naming scheme:
- cop-copname<nr>-copnumber<nr>
  Cellular operator
- cid-copname<nr>-copnumber<nr>-lac<nr>-cid<nr>
  Currently associated cell
- wlap-inrange-ssid<nr>-bssid<nr>
  In range 802.11 access point
- wlap-assoc-ssid<nr>-bssid<nr>
  Currently associated 802.11 access point
- btnw-bdaddr<nr>
  In range Bluetooth node
These attributes vary per participant. When the afterreset attribute is set
the device was restarted in between the time of the last line and the time
of the current line. 
For convenience, we have included a Python script to transform the .arff 
files into .txt files in which the full names of visible network entities
are used (tested under Python version 2.5.1). Running
> python arfftotxt.py */*.arff
from the directory of this README file will generate the .txt files.

[Author] Arjan Peddemors

top