Content uploaded by Daniel Schneider
Author content
All content in this area was uploaded by Daniel Schneider on Apr 18, 2017
Content may be subject to copyright.
Towardsanobservatoryformobileparticipatory
sensingapplications
GlauciaMelo
1
,LuizOliveira
1,3
,DanielSchneider
1,2
,JanodeSouza
1
1
PESC/COPPE/UFRJ,GraduateSchoolofEngineering,FederalUniversityofRiodeJaneiro,Brazil
2
DEMAT/ICE/UFRRJ,SchoolofComputerScience,FederalRuralUniversityofRiodeJaneiro,Brazil
3
IFRJ,CampusNiterói,FederalInstituteofRiodeJaneiro,Brazil
{glaucia,lfoliveira,schneider,jano}@cos.ufrj.br
Abstract—This paper presents an observatory for
registering applications that use participatory sensing to
collect data. Cataloging these applications will aid the
scientific community to exchange more information,
facilitating the comparison between different initiatives.
Through an initial research, the applications are categorized
in areas usually considered in the literature. We propose a
survey to validate the platform and also discuss the
taxonomies created as a result of this survey. The main
contributions of this paper include the classification of
crowdsensing applications in different ontological categories,
as well as the proposal of a technology platform that enables
the distributed and collaborative cataloging of crowdsensing
initiatives.
Keywords — crowdsourcing; participatory sensing;
crowdsensing; taxonomy; observatory; survey; literature
review
I.INTRODUCTION
The purpose of this work is to describe an observatory
created specifically for catalogging mobile participatory
sensing applications. In order to do this, a small version of a
survey was carried out, where dimensions that classify these
applicationswerestudied.
The knowledge on the subject has allowed the
identification of important aspects to classify applications
into categories, and in the scope of this work, the categories
are: application domain, sensors, incentive techniques, user
engagement levels, access types and shared data protection
types.Thesefeaturesaredetailedinthispaper,Section3.4.
It will also be possible to address new developments, of
software and hardware, with the goal of benefiting from the
potentialofacrowdofpeoplewhoownamobiledevice.
Some key research questions we aim to answer in this
work include: what is the distribution of the applications in
the investigated dimensions, what information was
unavailable or difficult to obtain and what are the key
academic contributions in the field of participatory sensing.
This paper is not a definitive work: it consists of a previous
study of taxonomies, supported by an observatory built for
thispurpose.
There is evidence of research published in Participatory
Sensing area, and one of the most important papers was
published in 2010 [1]. Since then, there has been significant
progress in the development of mobile applications and
devices, which nowadays have a greater number of sensors,
improved hardware and other aspects such as battery
performance and large scale use of 3G and 4G networks.
This has allowed more people to contribute with data and it
has also created an innovative development scenario.
Therefore, it is necessary to advance research in this area,
dealing with issues that must be considered when performing
the development of an application that has its content
generated by crowds. Moreover, there is a characteristic of
innovation that the development of this technology may
bring to the area, potentially impacting the lives of ordinary
citizens. Creating a taxonomy can help researchers to have
an overview of the area and to have visibility of the many
initiatives that explore the topic. This review will allow
researchers to position themselves and their work within the
field, increasing the synergy between researchers. Thus it is
possible that research challenges are addressed more quickly
through greater interaction and collaboration between groups
as new researchers can achieve their research goals more
rapidly and benefit from the results achieved previously by
otherresearchers.
The rest of the paper is organized as follows. Section II
explores the related work; Section III presents the proposed
observatory and discusses its architecture, Section IV
presents the methodology addressed, Section V discuss
results found of the survey. Finally, Section VI presents some
discussionandconclusions.
II.RELATEDWORK
A.ParticipatorySensingandCrowds
In recent years, with the development of mobile sensing
technologies, the scope of crowd computing systems [54]
using mobile devices has been broadened. Participatory
sensing is described as a sensing paradigm largely based on
the power of devices used by people, including mobile
phones, smart cars, portable devices, and others [2]. There is
also a definition proposed by Burke [3] which encompasses
the range of sensing, as it describes participatory sensing
networks as a mechanism that allows the general public and
professionals to collect, analyze and share local knowledge.
The goal of participatory sensing is to enable ordinary
citizens to share data they know, whether from personal
experiences or the environment in which they find
themselves, through mobile devices, allowing that data can
be analyzed, treated and also distributed according to the
contextforthebenefitofone'sownoracommunity.
B.Taxonomyconstructionapproach
According to Vital [4] NASA (National Aeronautics and
Space Administration), a North American agency, has
defined a classification taxonomy for corporate portals that
arose from the need to integrate experts from the same area,
thus facilitating communication between these experts. A
work conducted by [4] introduced the division of this
practice in two phases, alpha and beta, the former made by a
large group of users and the latter made by experts in the
field. As a result, different levels of relationships between the
proposedtermshavebeendefined.
This creation of taxonomies used in this work has support
in Computer Science [5], and allow multiple users to use the
proposed observatory detailed in Section 3.2 of this work
to create new entries as well as new items for the
taxonomies, which would be the classification alpha. The
first classification made within the scope of this work was
the beta rating since it was done by specialists in the
participatory sensing area. The aim was to provide the
platform for the general public, which in possession of the
guidelines described in Section 3.3 of this paper, will be able
to classify applications areas, thus performing the phase
calledbetarating.
III.ARCHITECTUREOFTHEOBSERVATORY
As the use of spreadsheets was not appropriate to carry
out this work, due to the need of maintaining multiple
hierarchical structures in the inventory that are not supported
by the twodimensional structure, the development of an
observatory or platform that enables the creation of an
application inventory was addressed. This platform is
important because it allows people to collaboratively register
the existing applications while defining new taxonomies and
creating a central repository of data on mobile participatory
sensing applications. The collaboration aspect was also
considered, as it was important that the platform had a fluid
interface for inventory and visualization of data collected. As
the requirements described here were considered to be quite
specific, the authors found no known solution to accomplish
thistask.
The solution chosen by the authors was to develop a
specific platform for cataloging items that could enable the
construction of hierarchical structures on demand. The
platform was developed with the technologies of Wordpress,
MySQLandPHP.
Figure 1 depicts some of the main fields that make up the
registration of an application. As can be seen, the screen is
very intuitive and allows the user (enroller) to see at the
same time all the metadata available to be filled or selected.
The application architecture allows the researcher to create
hierarchical ontologies related to the studied entity, which in
our case is the application. In this way, it is possible to
associate the application with one or more categories, one or
more sensors, levels of engagement, incentive techniques,
amongotherpossibilities.
For each ontology, the researcher can choose one or more
items already available in the list of options, or add a new
option on the same screen where the application is cataloged,
with no need to interrupt the cognitive process of registration
accessing other screens to register domain data. For each of
the ontologies it is possible to consult (during the cataloging
process) the most commonly used options, contributing to
achievegreaterconvergenceofthenomenclaturesused.
Figure1:Interfaceforregisteringanapp
The main screen of the observatory can be visualized in
Figure 2. In this screen it is possible to observe a search
field, a list of applications and facets calculated according to
the defined ontologies. The search field helps the researcher
to find an application by keywords. This feature also allows
the researcher to search for an application before registering
it, thus avoiding duplication of records. In the left part of the
screen, the applications resulting from a selection made by
the user are listed, either through a textual search or by
choosing one of the options defined by the ontologies. On
the right side are displayed facets, which express very rich
information from the observatory. Through the facets it is
possible to filter applications according to a specific type of
one of the ontologies, as well as to visualize the number of
applicationsregisteredinagivenontologicalcategory.
Figure2:Observatory’shomescreen
The platform provides customization facilities and
facilitates the visualization of data collected during the
research. The retrieval of the registered data can be made at
the http://cadb.demoro.net/ page. The application of
registration is made only by authenticated users, which may
be authorized by an administrator. Data inserted on the tool
canbeautomaticallyseenatthetimetheyaresubmitted.
IV.METHODOLOGY
A.Dimensions
This paper proposes a classification for mobile
participatory sensing applications supported by the
aforementioned observatory. The starting point to create this
main classification that describes the purpose of the
applications was inspired by the work of Khan [6]. The terms
peoplecentric and urbansensing for participatory sensing
have been widely used so far, so we found it appropriate to
continue using them. From this starting point it was possible
to build leads as a result of the participation of other
researchers and these leads could be validated through the
application of dozens of frameworks that could be
inventoried and classified according to the proposed
taxonomy.
To perform a survey to validate the proposed platform,
initial terms were used and search strings were formulated
and submitted to the following digital libraries: IEEE,
Journal of Systems and Software, ACM Digital Library and
Scopus. The following terms were used to search the initial
papers on the main subject: ( “crowd sensing” OR
“crowdsensing” OR “participatory sensing”) AND app.
From the list of papers published in journals and
conferences, it was possible to extract a list of applications
that make use of participatory sensing technologies, such as
CenceMe and NoiseSpy [7], UbiFit Garden and PEIR [8],
LiveCompare, MoVi and DietSense [9]. We set at 40 the
number of applications that would be classified, and with this
list in hand, papers describing these apps were then searched
on Google Scholar database. All applications were registered
and classified in the proposed observatory. By using this
tool, it was possible to view information registered about
eachapplication.
B.Tutorialforapplicationregistration
In order to get the best use of the registration platform of
participatory sensing applications, a stepbystep catalog is
describedbelowanditshouldbefollowed.
1) Identify an application that has its content created by
participatorysensing.
2) Checkifitwasnotalreadyregistered.
3) If the application has been identified in a paper, check
thereferencesinthepaperrelatedtotheapplication.
4) Search on Google Scholar database the papers that
describetheapplication,itsproposalandoperation.
Based on the information read in the paper, fill out the
registration platform, Figure 1. In addition to these catalog
information, each application must be classified according to
defined taxonomies, as described in the next section of this
paper.
V.PROPOSEDCLASSIFICATIONANDRESULTS
This section discusses the results achieved with the
registration of 40 selected applications in the developed
platform. Six different dimensions in which the cataloged
applicationswereclassifiedaredescribedbelow.
1) Application working area: The apps were classified in a
hierarchical tree structure, so that they are always associated
with a leaf node of the tree. This leaf node objectively
describes the business domain of the application, ie its most
basic purpose. It should be a description in a few words and
simple enough so that other applications can also fit into this
classification in case they have the same purpose. The
description of the most internal nodes can be seen in Table 1
andtheleafnodesinTable2.
2) Sensors: In the scope of this work, sensors are devices that
can be embedded or connected in mobile devices. When built
in, the device already has the sensors independently of
whether they are used by applications sensors. There are also
derived sensors, which are those that have a particular
purpose, but if adjusted to the context, can be used for other
purposes, such as using the flash and the camera lens
together to detect the heartbeat by lightning the fingertip and
detecting the blood flow movement. The sensors may also be
coupled to the mobile device, such as a drone coupled to the
device or connected via a network with other devices and
lastly, may be human sensors that are actually feelings and
sensations that cannot be mapped by sensors, for example,
mood or the safety feeling in a particular place. The sensors
used by the application are usually explicitly described in the
articleortheapplicationpageonthespecifications.
ParticipatorySensing
Contentcentered
Focusedonaspectsofcontent,suchasnews
andhistoricalevents.
Outdoor
environmentcente
red
Focusedonnaturalaspectsoftheenvironment
suchasplants,animals,airquality,andothers.
Peoplecentered
Focusedonaspectsoftheperson,theones
whichoccuraroundaperson,howtomonitor
yourhealth,yourperformanceinsports,your
interactioninsocialnetworksandothers.
Placeand
servicescentered
Focusedonaspectsofsitesandservices,such
asservicetimes,hearingproductsprices.
Table1:Node’sdescription
3) Incentive techniques: Describes which techniques are
used to encourage users to share information via
participatory sensing applications. They are classified as
financial incentives when participants have the value of
their work paid in cash or coupons [10] and
nonfinancial, the latter having more than one
classification level, which are shown on Figure 3 and
describedbelow.
Classification
Application
Monitoringtimeattendanceto
Demoro[11],[12]
services
Monitornoiseandambience
Noisetube[13]
MoVi[14]
Monitoringofnoisepollution
NoiseSpy[15]
EarPhone[16]
Monitoringatmosphericconditions
WeatherSignal[17]
Enrichmentofsocialmedia
CenceMe[18]
MicroBlog[19]
Calculatingtheimpactassociated
withexposuretoenvironments
PEIR[8]
Encouragephysicalactivity
UbifitGarden[20]
Monitoringgarbage
GarbageWatch[21]
Classifygalaxypictures
GalaxyZoo[22]
Monitoringbirds
eBird[23]
Auditingprices
MobiShop[24]
PetrolWatch[25]
LiveCompare[26]
Monitoringplants
Floracaching[27]
Documentdietarychoices
DietSense[28]
Determinemedicalconditions
HealthSense[29]
Ambiencesensingandadjust
accordingly
SenSay[30]
Monitoringandrecordingsports
experience
BikeNet[31]
Bikeastic[32]
SkiScape[33]
Diabetesmanagement
JogFalls[34]
Monitoringtheairquality
MobAsthma[15]
HazeWatch[35]
PollutionSpy[15]
ThermalColumns
Ikarus[36]
Urbansensingandtracking
MetroTrack[37]
Monitoringtrafficconditionsand
roads
vTrack[38]
MapQuest[39]
Nericell[40]
VirtualTripLines[41]
CarTel[42]
Routefuelefficiency
GreenGPS[43]
Logicallocationviaambience
fingerprinting
SurroundSense[44]
Trackfoodintakeandcaloric
expenditure
Balance[45]
Monitoringsafetyperception
Transafe[46]
Discussionofgreenissues
Fresh[15]
Monitoringheartrate
HeartPhones[47]
TableII:Applicationclassification(taxonomy’sleafnode)
A. Civility: The user contributes in order to make where
youliveabetterplace[10].
B. Collective: The user contributes towards a common
good, so that his information when analyzed together
with the information provided by others will generate
knowledgeforthecollectivegood[7].
C. Fun: The user considers the act of contributing a fun
thingtodo,anditmotivateshimtodoit[7].
D. Gamification: Uses hedonistic components (such as
videogames)inordertomotivateusers[10].
E. Give and take: The user must supply information to the
application in order to receive information from other
users[7].
F. Selfcentered: the user contributes for his own benefit,
to follow a routine, or watch closely his personal
evolution. The data may or may not be shared with
otherusers[7].
G. Social interaction: The user contributes to interact with
otherusersandcreateorincreasehissocialcircle[7].
In regard to the incentive techniques that were chosen to
guarantee the user interaction with the application, the
informationinventoriedisshowninFigure4.
Figure3:Incentivetechniques
4) User engagement levels: The proposed classification
shows how the user data is collected by the application, and
it can be participatory, opportunistic or hybrid. If the user
requires explicit interaction (participatory) or if the data
collection is done automatically when the sensors themselves
realize the collecting opportunistically. Some applications
also use a hybrid form of user engagement [13] [28] [29]
[30] [45]. The hybrid form uses the two levels at different
times, or collect the data opportunistically and even request
further user confirmation. In the case of the DietSense
application [28], the user’s diet is recorded through
photographs in an opportunistic manner by a device worn on
the neck, but there is also the need for the user to inform
participatively feedback on the captured images. In
participatory format there is an active user cooperation, and
in opportunistic format, the user is not involved in
contributing his data to the system, ie, in data sharing [6].
With regard to the degree of user participation, the
opportunistic data collection one that happens without the
user being aware of its occurrence addresses different
technological challenges, since there is no human interaction.
Therefore, the development of tools and the use of sensors
should be directed to this end, seeking ways to detect the best
time in which data collection must begin and end. This active
monitoring of the behavior of people through mobile devices
that they carry with them has the potential to optimize the
society as a whole. Examples would be the possibility to
check the distribution of people in order to conduct urban
planning and the perception of the presence of a person in
shopping centers to generate marketing campaigns [48].
When the collection is explicit [49], ie the user interacts
explicitly, there are other challenges to be addressed, such as
notifications to the user to alert about the possibility to
interact with the application and ways to create a real interest
in the user to share good quality data, ie, true and also real
dataontheirownexperience.
TheinventorieddataonthismatterisshowninFigure4.
Figure4:Userengagementlevels
5) Access methods: This paper describes the following
application access methods: authenticated, anonymous and
hybrid. The user may or may not be logged in to share data,
that means the user must create an account in the application,
fill the username and password. Applications can also work
in hybrid form, for a particular need the user have to be
authenticated, and for others there may occur an anonymous
interaction. The authenticated form creates privacy
challenges for the application, as users’ personal data used in
the registration will be linked to the data that will be shared.
TheinventorieddataonthismatterisshowninFigure5.
Figure5:ApplicationAccessMethods
6) Data protection types: used to perform protection of data
sharedbyusers.Theyaredividedinto:
A. Anonymity: guaranteed by the application, forcing the
usertorelyontheservicetoretainhisinformation.
B. Authorize content sharing: The user explicitly authorizes
eachpieceofinformationhewantstoshare.
C. Exclusion and selective retention: The user chooses to
delete data collected before being shared with the
application server or keep on his local device specific
data [6]. It is different than to authorize the sharing of
content, because this type does not allow the user to
deletetherecordeddatalocallyonthedevices.
D. Data Aggregation: Data aggregation is any process in
which information is gathered and expressed in a
summary form, for purposes such as statistical analysis.
A common aggregation purpose is to get more
information about particular groups based on specific
variablessuchasage,profession,orincome[6].
E. Data Encryption: The data is encrypted by the
applicationbeforeitistransferredtotheserver.
F. Data Perturbation: intentionally perturbs the sensor
samples by adding artificial noise to the data at the
mobilephoneside,suchasGaussian[6]noise.
G. Spatial cloaking: a technique that blurs a user’s exact
location into a spatial region in order to preserve his
location privacy. The blurred spatial region must satisfy
the user’s specified privacy requirement. The most
widely used privacy requirements are kanonymity and
minimumspatial[6].
H. Hiding sensitive data: sensitive data can be selected by
the participants and protected using selective location.
This option is better than the disturbance data and
cloakingbecauseitensuresdataconsistency[6].
I. Keep data: Keep the data stored inside the mobile and
do not transfer it to any server or provide it to any other
person[19][8].
J. Pseudonymity: Pseudonymity is the use of pseudonyms.
Instead of transmitting names in plain text, all
interaction with the application is performed under an
alias[6].
K. Turn off the device: The user can turn off the device that
collects the data whenever he wants the data not to be
collected[20].
L. Not mentioned: When the aspect of privacy is not
mentionedinthepaper.
M. Not yet treated: The privacy aspect is mentioned in the
paper, but as a future work plan or not as a real problem
thatthepaperistryingtosolve.
TheinventorieddataonthismatterisshowninFigure6.
Figure6:Dataprotectiontypes
A.Researchinterpretation
This section will report on what was found as the survey
result, after reading the papers that described the
applications. It was possible to divide the findings into
subjectsandtocorrelatethem.
1) Privacy: Despite being an important issue to consider,
some applications do not mention the privacy of data
collected, as HeartPhones [47], Transafe [46], Balance [45],
HazeWatch [35], SkiScape [33], Jog Falls [34], Sensay [30],
HealthSense [29], FloraCaching [27], MobiShop [24], eBird
[23], Galaxy Zoo [22] and UbiFit Garden [20]. Privacy is an
important aspect since if there is the intention, any
application can perform an opportunistic data collection, that
is,withouttheuserknowingthatthedataisbeingcollected.
2) Development status: Not all applications were in fact
developed; some are only proposals and were not released.
Therefore, some tools were not available for download and
others do not even have a Web page describing them. It is
possible to find information about them in scientific papers,
butyoucannottestthem.
3) Coverage: Some applications have been developed in
order to conduct research at local communities, to
understand and improve the daytoday for small groups,
such as Transafe [46], which was developed for the city of
MelbourneinAustralia.
4) Hybridity of classifications: The same application, such as
eBird [23], may have two classifications for incentive
techniques. In the work of Jaimes [7] the incentive described
for this app is fun, and in the application paper from Wiggins
[23] it is claimed that the incentive techniques is for
collective good, because the app can be used as a knowledge
baseforallspeciesofbirdsaroundtheworld.
5) User participation: There is some debate on how not to
depend only on the user to gather information, but it strongly
depends on the application characteristics and the types of
sensors on which it is based. The opportunistic collection
generates a strong concern for user privacy. The most
modern mobile devices have sensors capable of performing
accurate data collection on sensitive information such as
user’s location at any given time and even over a long
period. The sound that can be collected through the
microphone and the images from the camera have a high
definition quality. This collection combined with the use of
mobile internet can allow user data to be transmitted even in
real time. There is a need to create mechanisms to ensure
users the purpose of the use of their data or warn users that
their data is being collected, in order not to generate a sense
of distrust, or there may be barriers, created by users, to
allowopportunisticcollections.
6) Monetization: No application investigated in the first
selected application group uses financial incentives to
encourage users to contribute. All the investigated tools work
with nonfinancial incentives, by using other techniques that
encourageparticipantstoconsumeandsharedata.
VI.DISCUSSIONANDCONCLUSION
The evolution of hardware and software of mobile devices
and the development of new sensors in these devices not
only allows but also facilitates the sharing of personal
information, since these devices are used massively by
people all over the world. The mobile phone is now an
essential item for communication, and it has become more
accessible and cheaper over time. The collection of
information about people enables a completely customized
experience, and the advancement of machine learning also
contributes to the development of new applications. Thus,
search is no longer adapted to what is offered by the market,
but the market creates solutions that automatically fits to its
users.
New mobile devices are also being created based on these
assumptions, such as the SmartWatches [50], SmartGlasses
[51] and items for smart homes, like the Amazon Echo [52]
and the Google Home [53]. These devices also work with the
concept of engaging applications to their system, as well as
mobile phones, and will rely heavily on participatory sensing
to provide service to its users. Moreover, it is possible to
adapt external sensors to mobile devices and to make
communication between them via bluetooth or wireless, as
thesefunctionsarenowcommontomobiledevices.
This work enables researchers to position themselves in the
present state of the art in mobile participatory sensing design
space, and contributes to the development of taxonomies
related to participatory sensing in mobile devices, which will
guide interested researchers on this field and also application
developers for mobile devices. It also describes a tool created
to be an observatory of applications on participatory sensing,
whichmadethisresearchpossible.
As a future work, we propose to extend the research to
consider more articles, such as [54] and others, as well as
opening the platform for general researchers, so that they can
registertheapplicationsondemand.
REFERENCES
[1] N.D.Lane,E.Miluzzo,H.Lu,D.Peebles,T.Choudhury,and
A.T.Campbell,“Asurveyofmobilephonesensing,”
CommunicationsMagazine,IEEE,vol.48,no.9,pp.140–150,
2010.
[2] B.Guo,Z.Yu,X.Zhou,andD.Zhang,“Fromparticipatory
sensingtomobilecrowdsensing,”inPervasiveComputing
andCommunicationsWorkshops(PERCOMWorkshops),2014
IEEEInternationalConferenceon,IEEE,2014,pp.593–598.
[3] J.A.Burkeetal.,“Participatorysensing,”Centerfor
EmbeddedNetworkSensing,2006.
[4] J.DutraandJ.Busch,“Enablingknowledgediscovery:
TaxonomydevelopmentforNASA,”2003.
[5] L.P.VitalandL.M.A.Café,“Práticasdeelaboraçãode
taxonomias:análiseerecomendações,”ENCONTRO
NACIONAL,2007.
[6] W.Z.Khan,Y.Xiang,M.Y.Aalsalem,andQ.Arshad,
“MobilePhoneSensingSystems:ASurvey,”IEEE
CommunicationsSurveys&Tutorials,vol.15,no.1,pp.
402–427,2013.
[7] L.G.Jaimes,I.J.VergaraLaurens,andA.Raij,“ASurveyof
IncentiveTechniquesforMobileCrowdSensing,”IEEE
InternetofThingsJournal,vol.2,no.5,pp.370–380,Oct.
2015.
[8] M.Munetal.,“PEIR,thepersonalenvironmentalimpact
report,asaplatformforparticipatorysensingsystems
research,”inProceedingsofthe7thinternationalconference
onMobilesystems,applications,andservices,ACM,2009,pp.
55–68.
[9] D.Christin,A.Reinhardt,S.S.Kanhere,andM.Hollick,“A
surveyonprivacyinmobileparticipatorysensing
applications,”J.Syst.Softw.,vol.84,no.11,pp.1928–1946,
Nov.2011.
[10] D.S.Gallo,C.Cardonha,P.Avegliano,andT.C.Carvalho,
“TaxonomyofCitizenSensingforIntelligentUrban
Infrastructures,”IEEESens.J.,vol.14,no.12,pp.4154–4164,
Dec.2014.
[11] L.F.Oliveira,D.Schneider,J.MoreiradeSouza,andS.A.
Rodrigues,“Leveragingthecrowdcollaborationtomonitorthe
waitingtimeofdaytodayservices,”inComputerSupported
CooperativeWorkinDesign(CSCWD),2015IEEE19th
InternationalConference,IEEE,2015,pp.109–114.
[12] L.F.Oliveira,D.Schneider,F.Oliveira,J.M.deSouza,andS.
A.Rodrigues,“Designingformonitoringthewaitingtimeof
daytodayservices,”in2016IEEE20thInternational
ConferenceonComputerSupportedCooperativeWorkin
Design(CSCWD),2016,pp.492–497.
[13] N.Maisonneuve,M.Stevens,M.E.Niessen,andL.Steels,
“NoiseTube:Measuringandmappingnoisepollutionwith
mobilephones,”inInformationTechnologiesinEnvironmental
Engineering,I.N.Athanasiadis,A.E.Rizzoli,P.A.Mitkas,
andJ.M.Gómez,Eds.SpringerBerlinHeidelberg,2009,pp.
215–228.
[14] X.BaoandR.RoyChoudhury,“Movi:mobilephonebased
videohighlightsviacollaborativesensing,”inProceedingsof
the8thinternationalconferenceonMobilesystems,
applications,andservices,2010,pp.357–370.
[15] E.Kanjo,J.Bacon,D.Roberts,andP.Landshoff,“MobSens:
Makingsmartphonessmarter,”PervasiveComputing,IEEE,
vol.8,no.4,pp.50–57,2009.
[16] R.K.Rana,C.T.Chou,S.S.Kanhere,N.Bulusu,andW.Hu,
“Earphone,”inProceedingsofthe9thACM/IEEE
InternationalConferenceonInformationProcessinginSensor
NetworksIPSN,2010.
[17] “WeatherSignal,TheCrowdSourcedWeatherMap,[online]
Available:online.”2016.
[18] E.Miluzzoetal.,“Sensingmeetsmobilesocialnetworks:the
design,implementationandevaluationofthecenceme
application,”inProceedingsofthe6thACMconferenceon
Embeddednetworksensorsystems,ACM,2008,pp.337–350.
[19] S.Gaonkar,J.Li,R.R.Choudhury,L.Cox,andA.Schmidt,
“Microblog:sharingandqueryingcontentthroughmobile
phonesandsocialparticipation,”inProceedingsofthe6th
internationalconferenceonMobilesystems,applications,and
services,ACM,2008,pp.174–186.
[20] S.Consolvoetal.,“Activitysensinginthewild,”in
ProceedingofthetwentysixthannualCHIconferenceon
HumanfactorsincomputingsystemsCHI,2008.
[21] D.EstrinandJ.Burke,“ParticipatoryCampaignsfor
Sustainability(GarbageWatch),CitizenScience(Networked
Naturalist),andActiveLiving(CycleSense),Centerfor
EmbeddedNetworkedSensing,UCLA,”2009.[Online].
Available:
http://research.cens.ucla.edu/urban/2009/urban11.pdf.
[22] K.L.Mastersetal.,“GalaxyZoo:barsindisc
galaxies\ding72,”Mon.Not.R.Astron.Soc.,vol.411,no.3,
pp.2026–2034,Nov.2010.
[23] A.Wiggins,“eBirding:technologyadoptionandthe
transformationofleisureintoscience,”inProceedingsofthe
2011iConference,ACM,2011,pp.798–799.
[24] S.Sehgal,S.S.Kanhere,andC.T.Chou,“Mobishop:Using
mobilephonesforsharingconsumerpricinginformation,”in
DemoSessionoftheIntl.ConferenceonDistributed
ComputinginSensorSystems,vol.13,2008.
[25] Y.F.Dong,S.Kanhere,C.T.Chou,andN.Bulusu,
“Automaticcollectionoffuelpricesfromanetworkofmobile
cameras,”inInternationalConferenceonDistributed
ComputinginSensorSystems,2008,pp.140–156.
[26] L.DengandL.P.Cox,“Livecompare:grocerybargainhunting
throughparticipatorysensing,”inProceedingsofthe10th
workshoponMobileComputingSystemsandApplications,
ACM,2009,p.4.
[27] “Floracaching,[online]Available:
http://research.cens.ucla.edu/
education/highschoolscholars/2011/docs/Floracaching.pdf.”
2001.
[28] S.Reddy,A.Parker,J.Hyman,J.Burke,D.Estrin,andM.
Hansen,“Imagebrowsing,processing,andclusteringfor
participatorysensing:lessonsfromaDietSenseprototype,”in
Proceedingsofthe4thworkshoponEmbeddednetworked
sensors,2007,pp.13–17.
[29] E.P.Stuntebeck,J.S.Davis,G.D.Abowd,andM.Blount,
“HealthSense,”inProceedingsofthe9thworkshoponMobile
computingsystemsandapplicationsHotMobile,2008.
[30] D.P.Siewioreketal.,“SenSay:AContextAwareMobile
Phone.,”inISWC,2003,vol.3,p.248.
[31] S.B.Eisenman,E.Miluzzo,N.D.Lane,R.A.Peterson,G.S.
Ahn,andA.T.Campbell,“BikeNet,”ACMTrans.Sen.Netw.,
vol.6,no.1,pp.1–39,Dec.2009.
[32] K.Shilton,“Fourbillionlittlebrothers?,”Commun.ACM,vol.
52,no.11,p.48,Nov.2009.
[33] S.B.EisenmanandA.T.Campbell,“SkiScapesensing,”in
Proceedingsofthe4thinternationalconferenceonEmbedded
networkedsensorsystemsSenSys,2006.
[34] L.Nachmanetal.,“Jogfalls:apervasivehealthcareplatform
fordiabetesmanagement,”inPervasiveComputing,Springer,
2010,pp.94–111.
[35] Z.Hailian,“HazeWatchproject:Communicationbetween
sensorsandmobilephone.”
[36] M.vonKaenel,P.Sommer,andR.Wattenhofer,“Ikarus,”in
Proceedingsofthe12thWorkshoponMobileComputing
SystemsandApplicationsHotMobile,2011.
[37] G.S.Ahn,M.Musolesi,H.Lu,R.OlfatiSaber,andA.T.
Campbell,“MetroTrack:PredictiveTrackingofMobileEvents
UsingMobilePhones,”inDistributedComputinginSensor
Systems,vol.6131,R.Rajaraman,T.Moscibroda,A.Dunkels,
andA.Scaglione,Eds.SpringerBerlinHeidelberg,2010,pp.
230–243.
[38] A.Thiagarajanetal.,“VTrack:accurate,energyawareroad
trafficdelayestimationusingmobilephones,”inProceedings
ofthe7thACMConferenceonEmbeddedNetworkedSensor
Systems,ACM,2009,pp.85–98.
[39] Z.A.Khan,“UsabilityevaluationofwebbasedGIS
Applications,”BlekingeInstituteofTechnology,2010.
[40] P.Mohan,V.N.Padmanabhan,andR.Ramjee,“Nericell,”in
Proceedingsofthe6thACMconferenceonEmbeddednetwork
sensorsystemsSenSys,2008.
[41] B.Hohetal.,“Virtualtriplinesfordistributed
privacypreservingtrafficmonitoring,”inProceedingofthe
6thinternationalconferenceonMobilesystemsapplications,
andservicesMobiSys,2008.
[42] B.Hulletal.,“CarTel,”inProceedingsofthe4thinternational
conferenceonEmbeddednetworkedsensorsystemsSenSys,
2006.
[43] R.K.Ganti,N.Pham,H.Ahmadi,S.Nangia,andT.F.
Abdelzaher,“GreenGPS,”inProceedingsofthe8th
internationalconferenceonMobilesystemsapplications,and
servicesMobiSys,2010.
[44] M.Azizyan,I.Constandache,andR.R.Choudhury,
“SurroundSense,”inProceedingsofthe15thannual
internationalconferenceonMobilecomputingandnetworking
MobiCom,2009.
[45] T.Denningetal.,“BALANCE:towardsausablepervasive
wellnessapplicationwithaccurateactivityinference,”in
Proceedingsofthe10thworkshoponMobileComputing
SystemsandApplications,ACM,2009,p.5.
[46] M.Hamilton,F.Salim,E.Cheng,andS.L.Choy,“Transafe,”
SIGCASComput.Soc.,vol.41,no.2,pp.32–37,Dec.2011.
[47] M.Z.Poh,K.Kim,A.D.Goessling,N.C.Swenson,andR.
W.Picard,“Heartphones:SensorEarphonesandMobile
ApplicationforNonobtrusiveHealthMonitoring,”in2009
InternationalSymposiumonWearableComputers,2009.
[48] T.Higuchi,H.Yamaguchi,andT.Higashino,“MobileDevices
asanInfrastructure:ASurveyofOpportunisticSensing
Technology,”JournalofInformationProcessing,vol.23,no.
2,pp.94–104,2015.
[49] B.Guoetal.,“MobileCrowdSensingandComputing,”
CSUR,vol.48,no.1,pp.1–31,Aug.2015.
[50] “AppleWatch.Available:http://www.apple.com/watch/.”.
[51] “Google.Googleglasses.Available:
https://www.google.com.br/glass/start.”.
[52] “Amazon.AmazonEcho.Available:
https://alexa.amazon.com.”.
[53] “Google.GoogleHome.Available:https://home.google.com/.”
[54] P.Carreño,F.J.Gutierrez,S.F.Ochoa,andG.Fortino,
“Supportingpersonalsecurityusingparticipatorysensing,”
Concurr.Comput.,vol.27,no.10,pp.2531–2546,Jul.2015.
[54] C.Padoa,D.Schneider,J.M.DeSouza,andS.P.J.Medeiros,
“Investigatingsocialcurationwebsites:Acrowdcomputing
perspective”,In:2015IEEE19thInternationalConferenceon
ComputerSupportedCooperativeWorkinDesign(CSCWD),
2015,pp.253258.
