Transcript 설명: 설명: 설명 - Ubiquitous Computing Lab

유비쿼터스 컴퓨팅 소개
Fall 2009
경희대학교 컴퓨터 공학과
이승룡 교수
Ubiquitous Computing Laboratory
Kyung Hee University
Contents
1
Introduction
•
•
•
•
2
The Vision
Ubicomp Scenarios
Major Trends in Computing
What is Ubicomp?
Essential Technologies
• Wireless Sensor Networks
• Context-aware Middleware
3
Ubicomp Applications
• Smart Home, Telematics, U-Healthcare, Military,
Agriculture, Robot, Metrology & u-Government
4
Future Trends & Summary
• Social Mega-trends, IT-Technology Trends,
R&D Strategy in Major Countries
• Summary
-2-
Contents
1
Introduction
•
•
•
•
2
The Vision
Ubicomp Scenarios
Major Trends in Computing
What is Ubicomp?
Essential Technologies
• Wireless Sensor Networks
• Context-aware Middleware
3
Ubicomp Applications
• Smart Home, Telematics, U-Healthcare, Military,
Agriculture, Robot, Metrology & u-Government
4
Future Trends & Summary
• Social Mega-trends, IT-Technology Trends,
R&D Strategy in Major Countries
• Summary
-3-
The Major Trends in Computing
-4-
Picture from Prof. Mattern, Porquerolles
Yesterday's Computers Filled Rooms
-5-
… So Will Tomorrow’s
-6-
The Vision
“In the 21st century the technology revolution will
move into the everyday, the small and the
invisible…“
Mark Weiser, 1988
“The most profound technologies are those that
Mark Weiser
disappear. They weave themselves into the fabric
(1952 – 1999)
of everyday life until they are indistinguishable
XEROX PARC
“Palo Alto Research Center” from it”
http://www.parc.xerox.com Mark Weiser, 1991
-7-
What is Ubicomp
• computing everywhere
• many embedded, wearable,
handheld devices
communicate transparently
to provide different services
to the users
• devices mostly have low
power and short-range
wireless communication
capabilities
• devices utilize multiple onboard sensors to gather
information about
surrounding environments
-8-
Contents
1
Introduction
•
•
•
•
2
The Vision
Ubicomp Scenarios
Major Trends in Computing
What is Ubicomp?
Essential Technologies
• Wireless Sensor Networks
• Context-aware Middleware
3
Ubicomp Applications
• Smart Home, Telematics, U-Healthcare, Military,
Agriculture, Robot, Metrology & u-Government
4
Future Trends & Summary
• Social Mega-trends, IT-Technology Trends,
R&D Strategy in Major Countries
• Summary
-9-
Essential Technologies
 Hardware technologies






Processors, memories, …
Today’s focus
(Wireless) networking
Sensors, actuators
Power
Packing and integration
Potentially: entirely new technologies (optoelectronics,
biomaterials)
 Software technologies





Operating environments
Networking
Context-aware Middleware
Platform technologies
User interfaces
- 10 -
Wireless Sensor
Networks
- 11 -
Wireless Sensor Networks (WSNs)
 Wireless Sensor Network Infrastructure
- 12 -
Wireless Sensor Networks (WSNs)
A network of small, battery-powered, wireless
computing devices
Capabilities:
 Processing
 Sensing
 Communication
- 13 -
WSN Constraints
Limited processing
Limited memory
Storage
Limited battery
Limited radio range
Limited power
A miniature computing device
- 14 -
Sensors
Processor
Radio
P
o
w
e
r
WSN Sensing Capabilities
Temperature, light sensors
Radiation sensors
Acoustic sensors
Motion sensors
Cameras, microphones
- 15 -
A temperature
sensor
WSN Applications
Monitoring
 For scientific applications in remote areas
Surveillance and tracking
“Smart” environments
 Smart homes
 Smart office
- 16 -
WSN Challenges
Tradeoff between performance and network
lifetime
Resource friendly protocols
 For example: routing
Scalability
 Nodes can become
arbitrarily large
Different devices in the same
network (Heterogeneity)
Security and privacy
- 17 -
Context-aware
Middleware
- 18 -
Context
- 19 -
What is context?
Context is ANY INFORMATION that can be
used to CHARACTERIZE the SITUATION of an
ENTITY (5W 1H).
For example, the engine of a car is running or
stopped.
• ENTITY – Car’s engine,
• CONTEXT- Stopped or Running
- 20 -
Low and High level contexts
Low level Contexts: Time, noise level,
temperature, nearby object, network bandwidth,
LOCATION etc.
 Or SENSED CONTEXTS
Defined Contexts: user profile, schedule,…
High level Contexts: User’s current activity,
emotions, rationality, intelligence etc…
 Or DEDUCED/INFERRED CONTEXTS
- 21 -
Context-Aware Middleware
• Steps in processing context
Perception: symbolic observables
Sensing: numerical extracted features
Context History
Context and Situation Identification
Discovery/Recovery
Context Usage
Privacy/Security/Trust
– Sensing technologies, gathering sensed context data,
sensor data fusion
– Modeling & storing context data,
– Reasoning/inferring & learning, and
– Delivering/disseminating the inferred contexts to applications
Levels of abstraction for a general-purpose infrastructure for context-aware computing
- 22 -
Context-aware Middleware
Designating those steps to Middleware will:
 Free the application developer from underlying tasks:
Let him focus on implementing application logic
 Reusability: built once – used by everyone
 Separation of concerns: Context-aware Middleware
decouples application layer with lower layers => more
efficient to develop
- 23 -
Context-Aware Middleware
- 24 -
Our proposed Architecture
Context-Aware
Middleware for
Ubicomp
Systems
More details could be found in our research papers and technical reports
Ubiquitous Computing LAB
-(25/100)
25 -
Context-Aware Middleware
App
Feature Ontology
Location
Sensor Data
Sensed values
Feature Tuples
Feature Context
markups
Context data
App
App
App
lookup
CDelivery Manager
register
CA
CA
CA
REManager
RE
RE
RE
Context Object
(low and high level contexts)
Context markups
CDManager
Context Object
(low level contexts)
Context repository
FCMDataMngr
FCMapping
Internal DB
FTSManager
Tuple
Space
Feature tuple
Mgr.
Hardware Abstraction Layer
FXA
System Admin
Drivers
Configuration
data
Raw data
from sensor
- 26 -
Sample Dataflow
 FEATURE : Unified format of data from sensors
 <sensorTypeID, featureTypeID, sensorID, quantizedLevel,
value, probability, timestamp>
 Ex : <6,1,1,null,105,1,20041210120000>  RFIDTagID = 105, from sensor #1,
probability=1, timestamp=20041210120000
 The feature is mapped into context markup
<RFIDTag>
<value>105</value>
<probability>1</probability>
<timestamp>20041210120000</timestamp>
<RFIDTag>
 Combining many kinds of features with the data about users, sensors, places, using
some reasoning mechanism, the system can infer the location of user
<Person resource=“http://hs.ac.kr/doctors#doctorA ”>
<locatedIn resource=“http://hs.ac.kr/rooms#room125”/>
</Person>
 The location info will be saved in the context repository, and can be delivered to the
applications/services by Location Context Aggregator
 The system can also use reasoning to infer higher-level contexts, such as user’s
activities, based on location, user’s intention and the environment features, then
provides some smart services.
Ubiquitous Computing LAB
- 27 -
Key characteristics of CAMUS
• Support for heterogeneous and distributed sensing agents
– Unified Sensing Framework: data structure & access interface
– Make it easy to incrementally deploy new sensors and context-aware
services in the environment
– Autonomic Sensing Agents are being developed
• Provide different kinds of context classification mechanisms
– Pluggable engines
– Different mechanisms have different power, expressiveness and decidability
properties
– Rules written in different types of logic (first order logic, description logic,
temporal/spatial logic, fuzzy logic, etc.)
– Probabilistic-based reasoning mechanisms, e.g Bayesian nets
– Data mining & User Preference Learning are being investigated
(28/100)
Ubiquitous Computing LAB
Key characteristics of CAMUS (cont)
• Follow a formal context model using ontology
– Using OWL
– To enable syntactic and semantic interoperability, and knowledge sharing
between different domains
• Dealing with uncertainty to enhance the quality of context
– A really challenge for system usability
– We deal with uncertainty in all levels: sensor data (probability attribute
associated with feature value), context data (probabilistic reasoners, &
modeling uncertainty being developed)
(29/100)
Ubiquitous Computing LAB
Key characteristics of CAMUS (cont)
• Service Discovery & Delivery
– Semantic Matchmaking & Policy based Autonomic Access Control
• Privacy policy and security mechanism
– Trust-based Security Infrastructure
• Facilitate for applications to specify different behaviors in
different contexts easily
– Graphical development tool to ease developers in writing context-aware app.
– Will be developed along with the application scenarios (smart spaces)
(30/100)
Ubiquitous Computing LAB
Context-Aware Applications
What is “Context-aware” Application?
 An application's ability to detect and react to
environment variables
Some examples





Telephone forwarding
Turn off cell phone in theaters, meeting…
Automatically adjust brightness / volume
Multimedia / Bandwidth adaptation
Smart Homes
- 31 -
Contents
1
Introduction
•
•
•
•
2
The Vision
Ubicomp Scenarios
Major Trends in Computing
What is Ubicomp?
Essential Technologies
• Wireless Sensor Networks
• Context-aware Middleware
3
Ubicomp Applications
• Smart Home, Telematics, U-Healthcare, Military,
Agriculture, Robot, Metrology & u-Government
4
Future Trends & Summary
• Social Mega-trends, IT-Technology Trends,
R&D Strategy in Major Countries
• Summary
- 32 -
Smart Home
Smart Home Perspective ....
a new kind of service delivery environment providing value to
the consumer
19th Cenrury
Elementary tasks based
on Human Capability &
Flexibility
20th Century
Functional System
Integration makes life
easier
- 33 -
21st Century
Integration of
comprehensive
Living Environment
Smart Home Scenario
Motes gather
information about
activities of daily living
for monitoring by a
caregiver. This
increases client
independence and
permits reduction of
number of full-time
caregiving staff.
(source: Ross, 2004, IEEE
Spectrum online)
- 34 -
Telematics
 Telematics is a combination of telemetry
(receiving
measurements
from
Global
Positioning Satellites as data, communicated
via the GSM network) and informatics
(interpretation and distribution of information).
 Telematics incorporates the latest advances
in Internet, geo-positioning, software and
cellular technologies to assist drivers and
vehicle owners.
 Telematics creates the ability to manage
vehicles from the Internet or cell phone linked
to backup emergency services.
- 35 -
Telematics in Korea
Je-Ju Island Pilot Project Overview
Je-Ju
119
Emergency call
Vehicle control
Emergent
going out order
Travel/Traffic
(Shortcut
information)
Travel/Traffic
(Traffic
information)
Travel/Traffic
Travel/Traffic
(Sightseeing)
(Accommodation)
Vehicle and emergency call control
Je-Ju
Already made
contents
Cyber Je-Ju
(http://cyber.jeju.go.kr)
E Je-Ju Mall
(http://www.ejeju.net)
Cyber Samdaguan
Emergency rescue
(linking with 119)
Trave/Traffic
Je-Ju culture
Emergency rescue
(vehicle location
information)
Leisure
Telematics Service Center
Leisure
A 코스
(Climbing)
Leisure
(Weather)
C 코스
Weather
Climbing, Golf, Fishing,
Karaoke, Game, MP3,
Stock trading, Weather
V-Shop
(Local
specialty)
GPS satellinte
V-Shop
Contents provider
Cultural
event
B 코스
Leisure
(Golf)
Entertainment
Entertainment
(Game)
$
CDMA 1x (EvDO)
Je-Ju ITS Center
Telematics service (text etc.)
CDMA station
Telematics terminal
Entertainment
(Stock trading)
$
Entertainment
(Karaoke)
$
Wireless LAN
WLAN
AP
36
Telematics service (multimedia contents etc.)
$ : charged
- 36 Stacy
Source:
Kang, “The Current Status of Telematics in Korea”, KOTBA, 6 Jun 2005.
Ubiquitous Healthcare
Expansion of time and space
Expansion of supplier and consumer
Reduction of cost
Daily healthcare
Various services
- 37 -
Military
 WSN in Military
 Asset monitoring and
management
 Surveillance and battlespace monitoring:
Vibration and magnetic
sensors can report
vehicle and personnel
movement, permitting
close surveillance of
opposing forces
 Urban warfare:
monitoring movement of
enemy in buildings for
snipers
 Protection for sensitive
objects: detect and
warning intruders.
- 38 -
Agriculture & Environmental Monitoring
 Precision Agriculture
 Monitoring soil, crop and
climate in a field
 Generalizing the results to a
decision support system
 Taking real-time action:
fertilizer, lime, pesticide,
tillage…
 Environmental Monitoring
 Monitoring of freshwater
quality
 Zebranet: tracking the
movement of zebras in
Africa (The Zebranet project
at Princeton)
 Disaster detection: early
detect forest fire and floods
- 39 -
Bio-Metrology
SCO(Smart
Cooperative Object)
모바일 이동단말(MHD)
1.센서로부터 전달받은 데이터를 기반으
로 사용자에게 실시간 정보 제공
2.중앙기상서버로 데이터를 전달하여, 기
상정보의 가공 및 다양한 어플리케이션
으로 응용
MHD
(Mobile Hand-held Device)
환경 센서(SCS)
1.사용자 주변의 환경 정보를 습득하여
모바일 단말기로 전달
2.다양한 환경정보(기온, 습도, UV등)의
습득이 가능
SCS
MHD
사용자
SCS(Smart
Cooperative Sensor)
사용자
IMT2000
3G/4G
Cellular
PCS
Wireless
PAN
Wired
Comm.
Wireless
LAN
BcN
Back-end Server
(IP Based Core Network)
FTTH+x
DSL
DMB
Satellite
WiBro
HSDPA
중앙 기상 서버
1.사용자로부터 기상 정보 습득
2.모바일 단말기로 기상정보 전달
3.인터넷을 통해 기상정보 제공
사용자
인터넷을 통해 기상정보 제공
1.사용자가 온라인으로 기상정보를 제공받음
2.현재 기상에 관한 상세한 정보를 비롯 예보 정보 습득
- 40 -
u-Robot
2006
2010
2015
홈서비스용
로봇
인간과 공존 – 정리정돈/심부름 로봇
청소, 교육용 로봇
실버▪라이프
케어로봇
헬스케어 로봇
노약자지원 – 실버메이트/신체지원 로봇
사회안전▪
국방 로봇
재난극복▪인명구조 로봇
응용분야 확대 – 실내외경비/소형투척형 로봇
초정밀▪초소형
제조 로봇
협업/지능화 – 초정밀제조용/초정밀바이오 로봇
자동차제조용 로봇
- 41 -
u-Government (u-Seoul)
정부
조직
SVC
Infra
U-복지
U-교육
Broadband
기업
조직
시민
사회
U-환경
U-안전
Mobile
GIS
- 42 -
U-교통
세계
체제
U-레저
RFID
U-문화
IPv6
Contents
1
Introduction
•
•
•
•
2
The Vision
Ubicomp Scenarios
Major Trends in Computing
What is Ubicomp?
Essential Technologies
• Wireless Sensor Networks
• Context-aware Middleware
3
Ubicomp Applications
• Smart Home, Telematics, U-Healthcare, Military,
Agriculture, Robot, Metrology & u-Government
4
Future Trends & Summary
• Social Mega-trends, IT-Technology Trends,
R&D Strategy in Major Countries
• Summary
- 43 -
Social Mega-Trends
인구 구조
 저출산/고령화: 사회/의료/복지 비용증가, 지식 산업의 고도화, 이민자 증가, 정체성 혼란
 핵 가족 재분화: Family Value 증가, 가족단위 사회구조 증가
 신 모계사회: 정치/경제/문화 영역에서 여성 파워 증가, 여성의 사회진출 확대 및 경제력 향상
사회 가치
 유비티즌의 보편화: Web 2.0, 참여/개방/공유 문화 확산, Social Networks
 개인주의 만연: 개인의 개성/창의력 중시, 개인지향 사회 법제도 및 시스템 확대
라이프스타일
 삶의 질 추구: 삶의 질 추구 가치관 확산, 행복의 조건이 안전하며/즐겁고/윤택하며/건강한 삶
 경험 감성 중시: 문화의 중요성 확산, Fun 추구, 감성/디자인/엔터테인먼트 산업 활성화
글로벌 환경
 Glocalization: WTO와 FTA의 공존, 국가간 글로벌 역할 분담 및 공존
 환경 자원의 지배: BRIC 등장으로 지구적 자원고갈, 온난화/산업화로 인한 환경 재앙
기술 혁명
 유비쿼터스 기술: 디지털 르네상스의 촉매제, 새로운 기술/산업/사회/문명 창출 촉진
 융합 기술: 이기종 기술 및 산업간의 융합화 가속
- 44 -
IT-Technology Trends
- 45 -
u-Convergence
- 46 -
R&D Strategy in Major Countries
- 47 -
Summary
• 창조적 지식혁신 인재 (과학인)
• 글로벌 마인드를 지닌 인재 (국제인)
• 문화창조 능력을 보유한 인재 (문화인)
비전
( 삶의 지향)
인간중심/
수요자중심
세계화
Social
Network
Ubiquitous 르네상스 <-> 21세기 문명
종교/인종
갈등
환경오염
가속화
기술독점/
지적재산권
분쟁
고령화
인구감소
에너지
위기
BRICs
등장
양극화
심화
21세기 선진 국가
인재
강국
지식정보화 사회
개방, 공유, 참여사회
기술
강국
문화
강국
U-Korea
유비쿼터스사회
환경
강국
외부환경: 위기
와 도전
산업
강국
국가 경쟁력
강화 목표
신 성장동력 산업
인간중심사회
- 48 -
21세기
사회 파라다임