Ubiquitous & Pervasive Computing: A Technology-driven Motivation Friedemann Mattern ETH Zürich März 2000 ET ETH Eidgenössische Dagstuhl, August 2002 Technische Hochschule Zürich
Hardware Trends F.Ma. 2
Size Number Computing: A Clear Trend One computer (mainframe) for many people One computer (PC) for everyone Many computers for everyone F.Ma. 3
The Trend… What Next? Number smart dust? Many computers Size for everyone F.Ma. 4
„Yesterday's Computers Filled Rooms...“ IBM Selective Sequence Electronic Calculator (1948) F.Ma. 5
„Yesterday's Computers Filled Rooms - So Will Tomorrow's.“ 00:45 1 2 3 4 5 6 7 8 9 0 IBM Selective Sequence Electronic Calculator (ca. 1948) F.Ma. 6
Internet Trends F.Ma. 7
The Qualitative Growth of the Internet 2002 Mobile Internet WWW Email Research network people to people to people machines Internet time line F.Ma. 8
The Qualitative Growth of the Internet 2002 Mobile Internet WWW Email Research network people to people to people machines Internet time line F.Ma. 9
The Qualitative Growth of the Internet � Networked embedded systems 2002 � machines talking to machines Embedded � Era of ubiquitous and pervasive computing Internet Services � information appliances Mobile � smart things Internet WWW Email Research network people to people to machines to people machines machines Internet time line F.Ma. 10
Ubiquitous Computing � Today, the Internet connects all computers � Tomorrow everyday objects will become smart � embedded processors � ...and they will all be interconnected � wireless communication F.Ma. 11 image source: “Die Zeit”
Everything Smart? F.Ma. 13
Friedemann Mattern Embedded Computing Enables „Cooperating Smart Things“ Real world objects are enriched with information processing capabilities 1. Embedded processors � in everyday objects � small � cheap � lightweight 2. Wireless communication � spontaneous networks 3. Sensors F.Ma. 14
Friedemann Mattern Embedded Computing Enables „Cooperating Smart Things“ Real world objects are enriched with information processing capabilities 1. Embedded processors � in everyday objects � small � cheap � lightweight 2. Wireless communication � spontaneous networks 3. Sensors F.Ma. 15
What If All Things Were Smart? � And communicate with I‘m each other? smart hello! F.Ma. 16
Smart Objects? „A Dancing Toaster“ (Rich Gold, XEROX PARC) F.Ma. 17
Smart Objects! I‘m � Can remember pertinent events smart! � they have a memory � Show context-sensitive behavior � they may have sensors � � location / situation awareness hello! � Are responsive � communicate with their environment � networked with other smart objects F.Ma. 18
Networked with Other Smart Objects? I‘M SORRY, DAVE. ANOTHER BEER, I CAN‘T DO THAT. PLEASE, HAL... THE BATHROOM SCALE AND THE HALL MIRROR ARE REPORTING DISTURBING FLAB ANOMALIES Cartoon by Jeff MacNelly F.Ma. 19
Happy Networking? Water! Hug me! Time for lunch! Lending period expired! Coffee gets cold! F.Ma. 20
Calm Technology? F.Ma. 21
Calm Technology! � The Coming Age of Calm Technology � „As technology becomes more imbedded and invisible, it calms our lives by removing the annoyances while keeping us connected with what is truly important“ � 1988 Notion of Ubiquitous Computing � „In the 21st century the technology revolution will move into the everyday, the small and the invisible…“ � Mark Weiser � The Disappearing Computer � 1952 - 1999 � „The most profound technologies are those that disappear. They weave them- � XEROX PARC selves into the fabric of everyday life until they are indistinguishable from it.“ F.Ma. 22
Disappearing Computer � Information processing moves to the background � human centered: concentrate on the task, not the tool � the notion „computer as a tool“ does no longer hold � New picture of computing as an invisible, ubiquitous background assistance � specialized, invisible computers will become an integral part of the natural human environment � „computing without computers“ image source: Phillips F.Ma. 23
4 Reasons for Ubicomp F.Ma. 24
Four Reasons for Ubicomp F.Ma. 25
First Reason for Ubiquitous Computing: Moore‘s Law (1965) � Processing speed and storage capacity double every 18 months � „cheaper, smaller, faster“ � Exponential increase � will probably go on for the next 10 years at same rate F.Ma. 26
Moore‘s Law Electronics, April 19, 1965 F.Ma. 27
Moore‘s Law F.Ma. 28
Moore‘s Law F.Ma. 29
Moore‘s Law The complexity for minimum 16 15 component costs has increased PER INTEGRATED FUNCTION 14 NUMBER OF COMPONENTS at a rate of roughly a factor of 13 two per year (see graph on next 12 page). Certainly over the short 11 LOG 2 OF THE 10 term this rate can be expected 9 to continue, if not to increase. 8 Over the longer term, the rate is 7 a bit more uncertain, although 6 5 there is no reason to believe it 4 will not remain nearly constant 3 for at least 10 years. That 2 means by 1975, the number of 1 0 components per integrated 1959 1960 1961 1962 1963 1964 1965 1966 1967 1968 1969 1970 1971 1972 1973 1974 1975 circuit for minimum cost will be 65,000. YEAR F.Ma. 30
Transistors Per Die F.Ma. 32
Example: CMOS Performance Increase Relative Device Performance 20 Conventional Bulk CMOS SOI (silicon-on-insulator) 10 High mobility 8 Double-Gate 6 4 2 1 0.8 0.6 0.4 0.2 1988 1992 1996 2000 2004 2008 2012 Source: Year IBM F.Ma. 33
SIA Roadmap 1998 Year 1999 2002 2005 2008 2011 2014 structural size (nm) 180 130 100 70 50 35 cost / transistor 1'735 .580 .255 .110 .049 .022 clock [MHz] 1250 2100 3500 6000 10000 16900 2 ] 340 430 520 620 750 900 chip size [mm F.Ma. 35
Disk Storage Density Hard Disk Storage Density 10000000 Kilobits per sq. inch 1000000 100000 10000 1000 A 1GB disk 100 drive in 10 a compact 1 flash card 1955 1960 1965 1970 1975 1980 1985 1990 1995 2000 format (2001) Year Scientific American, May 2000 F.Ma. 36
Bit Storage Density F.Ma. 37
Friedemann Mattern Generalized Moore‘s Law � Most important technology parameters double every 1 – 3 years: Problems: � computation cycles - increasing cost - energy � memory, magnetic disks � bandwidth � Consequence: scaling down F.Ma. 38
WatchPad1.5 Functions Bluetooth module IrDA port (band joint) 320x240 B/W LCD Piezo speaker bottom Touch panel Accelerometer Battery terminal on board (right side) Microphone Stem switch 3 Buttons Fingerprint sensor Vibrator around here F.Ma. 39
Use as a Universal Remote Controller Home Office Car The Original of Bluetooth � King of Denmark 940-981 � Son of Gorm the Old (King of Denmark) and Thyra Danebod (daughter of King Ethelred of England) Robot F.Ma. 40
Use as a Personal Identificator � Automated check-in at hotel and air counter � Cashless payment at restaurant and station � Medical history and prescription retrieval � Specify your preference Station/Airport Hospital Hotel Restaurant Street F.Ma. 41
Use as a Display for Location Based Services � Provide personalized advertisement and offering information � Timetable and flight schedule at station and airport � Navigate you at stations Station/Airport Restaurant Hotel Schedule Store Today’s special News Seat location Tourist info. F.Ma. 42
Energy Crisis: Not Everything Obeys Moore‘s Law! Generalized Moore‘s Law Generalized Moore‘s Law Battery capacity Battery capacity Source: Source: Jan Rabaey Jan Rabaey F.Ma. 43
Barriers Exponential increase of aircraft speed? Speed [km/h] Concorde Sound barrier 10 3 747 10 2 Wright brothers 10 1900 1920 1940 1960 1980 2000 F.Ma. 44
About Predictions (1955) "Bei diesem besonders schnittigen Modell aus dem Jahr 1975 handelt es sich um einen eleganten Zweisitzer mit Heck-Atomantrieb; die Sitze befinden sich dicht hinter den Vorderrädern..." F.Ma. 45
How Accurate Were Previous Predictions? � Moon colonies � � too expensive? � � Submarine towns � � … + unconvenient? � � Flying cars � � … + too dangerous? � � Paperless office � � too early? F.Ma. 46
How Accurate Were Previous Predictions? � Moon colonies � Supersonic planes � � too expensive? � � what about the Concorde? � � Submarine towns � Home robots � � … + unconvenient? � � too early? � � Flying cars � Magnetic suspension trains � � … + too dangerous? � � too expensive? � � Paperless office � Controled nuclear fusion � � too early? � � too optimistic? F.Ma. 47
Recommend
More recommend