[Dataset] rice/context (v. 2007-05-23)

top

the initial version

@MISC{rice-context-2007-05-23,
  author = {Ahmad Rahmati and Lin Zhong},
  title = {{CRAWDAD} data set rice/context (v. 2007-05-23)}, 
  howpublished = {Downloaded from http://crawdad.cs.dartmouth.edu/rice/context},
  month = may,  
  year = 2007
}
					

We gathered field data about cellular and Wi-Fi networks through participants 
from the Rice community in Houston, Texas, a major US urban area.

In order to check how close we are toward ubiquitous connectivity
in our everyday life, we have gathered network data from a number of 
mobile users on multiple mobile phones through measurement. 
We explore the use of different context information, including time, 
history, cellular network conditions, and device motion, to accurately 
estimate Wi-Fi network conditions without powering up its network interface.

We have used multiple HTC Wizard PDA phones for our data
collection. The HTC Wizard is commercially available under a
variety of brands, including T-Mobile MDA and Cingular 8125. It
is a Windows Mobile 5.0 GSM phone with integrated 802.11b and
is capable of EDGE data connectivity. It has a battery capacity of
1250mAh at 3.7 volts.

We have developed logging software to record various network 
characteristics with minimal intrusion to the normal phone 
operation. We have converted eleven HTC Wizards into
experimental mobile phones by installing our logging software.

14 volunteers from the Rice campus participated in our data
gathering. They carried around our experimental phones for at
least three weeks and could opt to use their own SIM card on the
phone. We requested all participants to carry the phone as they
would carry their own phone. We interviewed each participant
regularly to document any significant diversion from their daily
activities, for example, travels and forgetting carrying the phone.

This dataset is from different time periods than the traces used 
in the mobisys paper [rahmati-context]. We continued our logging after
the paper, and have included those. We also used some logs from older
versions of TowerLogger in the paper that do not exist here.

[Traceset] rice/context/availability (v. 2007-08-01)

top

the initial version

@MISC{rice-context-availability-2007-08-01,
  author = {Ahmad Rahmati and Lin Zhong},
  title = {{CRAWDAD} trace set rice/context/availability (v. 2007-08-01)}, 
  howpublished = {Downloaded from http://crawdad.cs.dartmouth.edu/rice/context/availability},
  month = aug,  
  year = 2007
}
					

This traceset contains logs of network availability, signal levels, and context information,  
measured on Wi-Fi enabled GSM cellular phones.

- Tower Logger: Measuring Network Conditions

Our logging software, called Tower Logger, measures network availability 
and signal levels, and context information.  It records the cell tower ID, 
signal strength, and channel of the currently-associated GSM cell and 
those of up to 6 other visible cells every 30 seconds or 60 seconds. 
It also records the unique Basic Service Set Identifier (BSSID), 
signal strength and the security property of all visible Wi-Fi access points.

Each BSSID / Cell Tower is assigned a unique ID that is consistent
throughout all our traces. Further, participant IDs are randomized 
("01" - "10" in the traces do not correspond to P1 - P10 in our paper).

Wi-Fi:
The BSSID is anonymized except for the first 3 octets. 

Cell towers:
LAC and CI are anonymized for each cell tower (MCC:MNC:LAC:CI).

[Trace] rice/context/availability/GSM (v. 2007-08-01)

top

the initial version

@MISC{rice-context-availability-GSM-2007-08-01,
  author = {Ahmad Rahmati and Lin Zhong},
  title = {{CRAWDAD} trace rice/context/availability/GSM (v. 2007-08-01)}, 
  howpublished = {Downloaded from http://crawdad.cs.dartmouth.edu/rice/context/availability/GSM},
  month = aug,  
  year = 2007
}
					

This trace contains GSM Cellular logs of availability and signal strength, 
measured on 10 participants' Wi-Fi enabled GSM cellular phones.

This trace contains Tower Logger network GSM Cellular logs:
availability and signal strength, measured on 10 participants' 
Wi-Fi enabled GSM cellular phones.

The trace tarball consists of 10 directories "01"-"10", which are 
the logs from 10 participatants' phones ("01" - "10" in the traces 
do not correspond to P1 - P10 in our paper.) Each directory contains
either GSM log or Wi-Fi log.

GSM logs are named 'log2GSM_yyyy-mm-dd.txt'. They are in csv 
(comma-separated value) file format, where each line consists of
the following fields:

        version no.
        connected Orbit Sensor ID (0 = none)
        date
        time
        battery level
        charge status
        no. of visible cell towers
        list of visible cell towers (starting with associated tower):
            Cell ID:
                MCC
                MNC
                LAC (anonymized)
                CI (anonymized)
            signal strength
            channel
            BSIC

[Trace] rice/context/availability/Wi-Fi (v. 2007-08-01)

top

the initial version

@MISC{rice-context-availability-Wi-Fi-2007-08-01,
  author = {Ahmad Rahmati and Lin Zhong},
  title = {{CRAWDAD} trace rice/context/availability/Wi-Fi (v. 2007-08-01)}, 
  howpublished = {Downloaded from http://crawdad.cs.dartmouth.edu/rice/context/availability/Wi-Fi},
  month = aug,  
  year = 2007
}
					

This trace contains Wi-Fi logs of availability and signal strength, 
measured on 10 participants' Wi-Fi enabled GSM cellular phones.

This trace contains Tower Logger network Wi-Fi logs:
availability and signal strength, measured on 10 participants' 
Wi-Fi enabled GSM cellular phones.

The trace tarball consists of 10 directories "01"-"10", which are 
the logs from 10 participatants' phones ("01" - "10" in the traces 
do not correspond to P1 - P10 in our paper.) Each directory contains
either GSM log or Wi-Fi log.

Wi-Fi logs are named 'log2WiFi_yyyy-mm-dd.txt'. They are in csv 
(comma-separated value) file format, where each line consists of
the following fields:
    
        version no.
        connected Orbit Sensor ID (0 = none)
        date
        time
        battery level
        charge status
        list of visible AP info:
            BSSID (anonymized except the first 3 octets)
            signal strength
            encrypted?
            infrastructure mode?

[Author] Ahmad Rahmati

top

[Author] Lin Zhong

top

[Paper] rahmati-context

top

Ubiquitous connectivity on mobile devices will enable numerous new applications 
in healthcare and multimedia. We set out to check how close we are towards 
ubiquitous connectivity in our daily life. The findings from our recent 
field-collected data from an urban university population show that while 
network availability is decent, the energy cost of network interfaces poses a 
great challenge. Based on our findings, we propose to leverage the 
complementary strength of Wi-Fi and cellular networks by choosing wireless 
interfaces for data transfers based on network condition estimation. We show 
that an ideal selection policy can more than double the battery lifetime of a 
commercial mobile phone, and the improvement varies with data transfer patterns 
and Wi-Fi availability. We formulate the selection of wireless interfaces as a 
statistical decision problem. The key to attaining the potential battery 
improvement is to accurately estimate Wi-Fi network conditions without powering 
up its network interface. We explore the use of different context information, 
including time, history, cellular network conditions, and device motion, for 
this purpose. We consequently devise algorithms that can effectively learn from 
context information and estimate the probability distribution of Wi-Fi network 
conditions. Simulations based on field-collected traces show that our 
algorithms can improve the average battery lifetime of a commercial mobile 
phone for a three-channel electrocardiogram (ECG) reporting application by 39%, 
very close to the theoretical upper bound of 42%. Finally, our field validation 
of our most simple algorithm demonstrates a 35% improvement in battery 
lifetime.