contextualised service delivery in internet of things, smart parking for smart cities
TRANSCRIPT
ContextualisedServiceDeliveryinInternetofThingsSmartParkingforSmartCi<es
AliYavari*,PremPrakashJayaraman†,DimitriosGeorgakopoulos†*RMITUniversity,Melbourne,Australia†SwinburneUniversityofTechnology,Melbourne,[email protected]
RMIT University - July 2015 2
Inthelate1960s,communicaEonbetweentwocomputerswasmadepossiblethroughacomputernetworkIntheearly1980stheTCP/IPstackwasintroduced.CommercialuseoftheInternetstartedinthelate1980WorldWideWeb(WWW)becameavailablein1991InternetofThingstermbyKevinAshton1998(“TheInternetofThingshasthepoten0altochangetheworld,justastheInternetdid.Maybeevenmoreso”)WebofThings(WoT)in2000MITAuto-IDcentrepresentedtheirIoTvisionin2001IoTwasformallyintroducedbyInternaEonalTelecommunicaEonUnion(ITU)in2005More“thingsorobjects”wereconnectedtotheInternetthanpeople.2008-2009[Cisco]
19601980199620002001
CiscoIBSGprojecEons,UNEconomic&SocialAffairsh`p://www.un.org/esa/populaEon/publicaEons/longrange2/WorldPop2300final.pdf
6.307
6.721 6.894 7.347 7.83
0
10
20
30
40
50
2003 2008 2010 2015 2020
Billion
sofD
evices
WorldPopulaEon
50 Billion
SmartObjects RapidadopEonrateofdigitalinfrastructure
5 x faster than electricity & telephony
“Things”perperson
InflecEonPoint
1.1 Billion Data points generated by sensors daily 500 Gigabytes
Data generated by an offshore oil rig weekly
1000 Gigabytes Data generated by an oil refinery daily
10,000 Gigabytes Data generated by a jet engine every 30 minutes
2.5 Billion Gigabytes Data generated worldwide daily
90% of the world’s data Has been created in the last 2 years!
• CiscoIBSGprojecEons,UNEconomic&SocialAffairsh`p://www.un.org/esa/populaEon/publicaEons/longrange2/WorldPop2300final.pdf
Sensors and other Internet-connected devices that are all connected to theinternetandtheyinteractintelligentlytomakethedevelopmentanddeliveryofnewservicesandproducts
Anewparadigmwhichconnectsavarietyofthings-Allthethingsthathavetheabilitytocommunicate
IdenEfy
Communicate
Sense
Control
“Thepriceoflightislessthanthecostofdarkness.”–ArthurCNielsen
Wisdom
Knowledge
Informa<on
Data
Process
ParkingSpaceinaSmartCity
• Directdriverstoemptyparkingspaces– R.E.Barone,T.Giuffrè,S.M.Siniscalchi,M.A.Morgano,andG.Tesoriere,“Architectureforparkingmanagementin
smartciEes,”IETIntell.Transp.Syst.,vol.8,no.5,pp.445–452,2014.– R.Lu,X.Lin,H.Zhu,andX.Shen,“SPARK:anewVANET-basedsmartparkingschemeforlargeparkinglots,”in
INFOCOM2009,IEEE,2009,pp.1413–1421.– R.Grodi,D.B.Rawat,andF.Rios-GuEerrez,“Smartparking:ParkingoccupancymonitoringandvisualizaEonsystem
forsmartciEes,”inSoutheastCon2016,2016,pp.1–5.– Y.Zheng,S.Rajasegarar,andC.Leckie,“ParkingavailabilitypredicEonforsensor-enabledcarparksinsmartciEes,”in
IntelligentSensors,SensorNetworksandInforma0onProcessing(ISSNIP),2015IEEETenthInterna0onalConferenceon,2015,pp.1–6.
– J.Cherian,J.Luo,H.Guo,S.-S.Ho,andR.Wisbrun,“Poster:ParkGauge:GaugingtheCongesEonLevelofParkingGarageswithCrowdsensedParkingCharacterisEcs,”inProceedingsofthe13thACMConferenceonEmbeddedNetworkedSensorSystems,2015,pp.395–396.
– Z.Ji,I.Ganchev,M.O’Droma,L.Zhao,andX.Zhang,“Acloud-basedcarparkingmiddlewareforIoT-basedsmartciEes:designandimplementaEon,”Sensors,vol.14,no.12,pp.22372–22393,2014.
• ProvideanesEmateofaveragewaiEngEmetopark– P.R.deAlmeida,L.S.Oliveira,A.S.Bri`o,E.J.Silva,andA.L.Koerich,“PKLot–Arobustdatasetfor– A.Koster,A.Oliveira,O.Volpato,V.Delvequio,andF.Koch,“RecogniEonandrecommendaEonofparkingplaces,”in
Ibero-AmericanConferenceonAr0ficialIntelligence,2014,pp.675–685.– J.Rico,J.Sancho,B.Cendon,andM.Camus,“ParkingeasierbyusingcontextinformaEonofasmartcity:Enablingfast
searchandmanagementofparkingresources,”inAdvancedInforma0onNetworkingandApplica0onsWorkshops(WAINA),201327thInterna0onalConferenceon,2013,pp.1380–1385.
– E.Akhavan-Rezai,M.F.Shaaban,E.El-Saadany,andF.Karray,“Onlineintelligentdemandmanagementofplug-inelectric
– Y.GengandC.G.Cassandras,“New‘SmartParking’systembasedonresourceallocaEonandreservaEons,”IEEETrans.Intell.Transp.Syst.,vol.14,no.3,pp.1129–1139,2013.
ParkingSpaceinaSmartCity
• Wetakeintoaccounteachdriver'scontextthatmayinclude:– driver’spreferences(e.g.,parkinacoveredparkingspace), – drivingexperience(e.g.,avoidnarrowparkingspaces),– car’slocaEon(e.g.,collectedfromthedriver’ssmartphone),– vehicle’sproperEes(e.g.,vehicletype,length,height,etc.),– otherparkinginformaEonprovided(e.g.,theproperEesofparkingspaces,
suchasshaded,covered,etc.).
Observe
Orient
Decision
AcEon
Internet-connectedDevices
Connec<vity
Internet-scaledData
DataReduc<on
IoTServicesandApplica<on
Presenta<on
Contextualisa<on
Context
ContextorcontextualinformaEonisanyinformaEonaboutanyenEtythatcanbeused to effecEvely reduce the amountof reasoning required (viafiltering,aggregaEon,andinference)fordecisionmakingwithinthescopeofspecificapplicaEons.
ContextualisaEon
• ContextualisaEon excludes irrelevant data fromconsideraEonandhas thepotenEal to reducedatafromseveralaspectsincludingvolume,velocity,andvarietyinIoTapplicaEons
• ContextualisaEonof IoTdatacanhelp improve thevalueofinformaEonextractedfromIoT
• ContextualisaEon improve the data processing andknowledgeextracEoninIoTapplicaEons
ContextCollecEon
ContextualisaEon
DisseminaEonofthecontextualised
data
• AnapproachtorepresentandcontextualiseddataoriginaEngfromIoTdevices.• AmechanismtoefficientlyquerythecontextualisedIoTdata• AnexampleofasmartparkingspacerecommenderapplicaEon• AnexperimentalevaluaEonoftheproposedcontextualisedIoTdataquerying
approachusingsyntheEcdatageneratedfromMelbournecitydatasetsh`ps://data.melbourne.vic.gov.au/
ContextualisedServiceDeliveryintheInternetofThings
A.Yavari,P.P.Jayaraman,D.Georgakopoulos,andS.Nepal,‘‘ContaaS:Anapproachtointernet-scalecontextualisaEonfordevelopingefficientinternetofthingsapplicaEons,’’inHawaiiInternaEonalConferenceonSystemSciencesHICSS
Conclusion
• Scalable and real-Eme contextualisaEonof IoT data has thepotenEal tosignificantly improve data processing for large scale IoT applicaEons inSmartCiEes
• Weproposedanapproach to contextualise andquery Internet scale IoTdata and we exemplify the approach via a smart parking spacerecommenderapplicaEonforSmartCiEes.
• Theexperimental scenario in thispaper illustrates that contextualisaEonof IoTdatareducesqueryEmesfor IoTservices(suchasasmartparkingspacerecommender)bymorethan3EmesincomparisonwithasituaEonwherethequeryiscontextualisaEonagnosEc.
ThankYou!