1. Feb 12, 20071 ECET 581/CPET/ECET 499 Mobile Computing Technologies & Apps Context Aware Computing 3 of 3 Lecture 12 Paul I-Hai Lin, Professor Electrical and Computer Engineering Technology Indiana University-Purdue University Fort Wayne

2. Feb 12, 20072 Context-Aware Computing Introduction Introduction Ubiquitous or Pervasive Computing Ubiquitous or Pervasive Computing Definitions and Types of Context Definitions and Types of Context Context-Aware Computing and Applications Context-Aware Computing and Applications Middleware Support Middleware Support Ubiquitous Knowledge Discovery Ubiquitous Knowledge Discovery Summary Summary

3. Feb 12, 20073 Middleware Support Challenges - Context-aware applications Challenges - Context-aware applications 1.Contextual info – from various heterogeneous and distributed sources 2.Same type of contextual info may have to be obtained from different sources at different times 3.The low-level contextual info must be abstracted to be useful 4.Context-awareness is most relevant when the environment is highly dynamic (mobile)

4. Feb 12, 20074 Middleware Support Contextual info – from various heterogeneous and distributed sources Contextual info – from various heterogeneous and distributed sources Hardware and software sensors: motion detectors, noise, temperature sensors, location systemsHardware and software sensors: motion detectors, noise, temperature sensors, location systems System recorded input – user-system interaction historySystem recorded input – user-system interaction history Other applicationsOther applications User’s personal computing space: User’s personal computing space: Schedules, calendars, address books, contact lists, and to-do listsSchedules, calendars, address books, contact lists, and to-do lists Distributed computing environment Distributed computing environment Obtained from applications running in the vicinity of these devices: shopping malls, freeway, etcObtained from applications running in the vicinity of these devices: shopping malls, freeway, etc

5. Feb 12, 20075 Middleware Support 2.Same type of contextual info may have to be obtained from different sources at different times GPS receiver – outdoor positioning systemGPS receiver – outdoor positioning system Indoor positioning systemIndoor positioning system 3.The low-level contextual info must be abstracted to be useful GPS position info (latitudes and longitudes)GPS position info (latitudes and longitudes) → Tour guide, location info→ Tour guide, location info 4.Context-awareness is most relevant when the environment is highly dynamic (mobile) Real-time detection, trigger commands, auto reconfigurationReal-time detection, trigger commands, auto reconfiguration

6. Feb 12, 20076 Middleware Support – Contextual Services Middleware Infrastructure that can provide the following services (acquire contextual input from sensors) Middleware Infrastructure that can provide the following services (acquire contextual input from sensors) 1.Contextual subscription and delivery service 2.Context query service 3.Context transformation service 4.Context synthesis service 5.Discovery and management service  Actuator Services  Perform a context-dependent output function

7. Feb 12, 20077 Middleware Support – An Example The Context Toolkit - Publications, http://www3.cc.gatech.edu/fce/contexttoolkit/ The Context Toolkit - Publications, http://www3.cc.gatech.edu/fce/contexttoolkit/ http://www3.cc.gatech.edu/fce/contexttoolkit/ Software components for context acquisition Software components for context acquisition Context widgetsContext widgets An interface between sensors and applications An interface between sensors and applications Provides an abstraction layer Provides an abstraction layer State: a set of attributes State: a set of attributes Behavior: call back function Behavior: call back function Persistent entities that can be shared by multiple applications Persistent entities that can be shared by multiple applications Context interpreters (info translators)Context interpreters (info translators) Context aggregatorsContext aggregators Entity: person, room, software systems, hardware devices Entity: person, room, software systems, hardware devices DiscovererDiscoverer Discovery and management services Discovery and management services

8. Feb 12, 20078 Ubiquitous Knowledge Discovery Following excerpts are from the Proceedings of the Workshop on Ubiquitous Knowledge Discovery for Users, Sept. 18-22, 2006, Berlin, Germany, http://vasarely.wiwi.hu-berlin.de/UKDU06/#Proceedings Following excerpts are from the Proceedings of the Workshop on Ubiquitous Knowledge Discovery for Users, Sept. 18-22, 2006, Berlin, Germany, http://vasarely.wiwi.hu-berlin.de/UKDU06/#Proceedings http://vasarely.wiwi.hu-berlin.de/UKDU06/#Proceedings A key research area for the coming years A key research area for the coming years It encompasses the whole process of turning data into knowledge: It encompasses the whole process of turning data into knowledge: AlgorithmsAlgorithms Business/Application UnderstandingBusiness/Application Understanding Data UnderstandingData Understanding Data PreparationData Preparation ModelingModeling Evaluation, andEvaluation, and DeploymentDeployment

9. Feb 12, 20079 Ubiquitous Knowledge Discovery Users play a pivotal role in this process Users play a pivotal role in this process They create dataThey create data Documents and/or references Documents and/or references Links between documents Links between documents Data are related to themData are related to them Explicit reactions to questions such as the input of registration data, and the behavior that leave traces, biometric measurements, etc. in log files Explicit reactions to questions such as the input of registration data, and the behavior that leave traces, biometric measurements, etc. in log files They are the ultimate beneficiaries of the discovered knowledgeThey are the ultimate beneficiaries of the discovered knowledge The user models – foundation for general-audience or personalized improvement of services and devices The user models – foundation for general-audience or personalized improvement of services and devices Activities which directly benefit the end user or another human stakeholder (Web site owner, government agency, etc) Activities which directly benefit the end user or another human stakeholder (Web site owner, government agency, etc)

10. Feb 12, 200710 Ubiquitous Knowledge Discovery Dimensions of Ubiquity Dimensions of Ubiquity Ubiquity of devices and dataUbiquity of devices and data Ubiquity (distributedness) of processingUbiquity (distributedness) of processing Ubiquity of people and contextsUbiquity of people and contexts Information ubiquityInformation ubiquity Standards for UbiquityStandards for Ubiquity

11. Feb 12, 200711 Summary Google Location Dependent Search Engine: http://local.google.com/lochp Google Location Dependent Search Engine: http://local.google.com/lochphttp://local.google.com/lochp Find businessFind business Get directionsGet directions Serendipity – MIT Media lab project: a wireless, proximity-based dating service Serendipity – MIT Media lab project: a wireless, proximity-based dating service can accidentally discovers something fortunatecan accidentally discovers something fortunate http://www.media.mit.eduhttp://www.media.mit.eduhttp://www.media.mit.edu BlueJAQ- Pull on an unsuspecting stranger in close proximity who has a Bluetooth- enabled phone: http://www.bluejackq.com/ BlueJAQ- Pull on an unsuspecting stranger in close proximity who has a Bluetooth- enabled phone: http://www.bluejackq.com/ http://www.bluejackq.com/