Introduction to Embedded and Ubiquitous Systemsjupiter.plymouth.edu/~wjt/OpSys/EmbeddedSystems1.pdf · 3 ICS212 WQ05 (Dutt) Introduction 5 Reading Materials zRecommended Books Peter

1

ICS212 WQ05 (Dutt) Introduction 1

Introduction to Embedded and Ubiquitous Systems+

Nikil DuttUC Irvine

ICS 212 Winter 2005

+Copyrighted Material adapted from Peter Marwedel, Rajesh Gupta, Frank Vahid and Tony GivargisTemplates from Prabhat Mishra


Welcome to ICS 212

InstructorNikil Dutt [email protected]

Course Pagehttp://www.ics.uci.edu/~dutt/ics212.htmlVisit regularly for updates and announcements

Class MeetsMon-Wed 2-3:20 PM CS 253

2


Course Requirements

No graduate level courses necessary

Assumes knowledge ofComputer OrganizationData Structures and AlgorithmsProgramming in C ( or C++/VHDL/Verilog)

Basic undergraduate education in CS


Evaluation and Grading

Details will be announced next week

Tentatively:HomeworksES case studiesFinal exam

3


Reading Materials

RecommendedBooks

Peter Marwedel – Embedded Systems Design, Kluwer 2003

Vahid and Givargis – Embedded System Design, Prentice Hall 2002

Both available at Bookstore and on reserve at the Science Library

ReferencesLecture Notes

Embedded Systems CoursesSangiovanni-Vincentelli @ Berkeley, Gupta @ UCSD, …

Newsgroupscomp.arch.embedded, …

Conferences/JournalsACM TECS/TODAES, IEEE TCAD/TVLSI, Kluwer DAES,…

Web Resourceswww.embedded.com, www.eet.com, …


Embedded SystemsAutomobiles

Handheld

Medical

AirplanesMilitary

Entertainment

4


What’s an Embedded System?

Embedded systems =information processing systems embedded into a larger product

Two types of computingDesktop – produced millions/yearEmbedded – billions/year

Non-Embedded SystemsPCs, servers, and notebooks

The future of computing!Automobiles, entertainment,

communication, aviation, handheld devices, military and medical equipments.


An Example Embedded System

Digital Camera Block Diagram

5


Story of the Humble Transistor

1947 – Shockley, Brattain, and Bardeeninvented the transistor at Bell Labs1961 – First commercial IC by Fairchild/TI1963 – CMOS invented1965 – Moore’s law1968 – State of the art: 64 transistor chip1978 – IA 8086: ~10K transistors1986 – IA 386: ~100K transistors1990 – IA 486: ~1 M transistors1998 – IA P2: ~ 10 M transistors2004 – IA P6: ~ 1 B transistors


Moore’s Law

Source: Intel

Driving force behindElectronic systems research and industry

Proliferation of embedded computing systems

Transistor capacity doubles every 18 months

6


1,000,000,000N

um

ber

of

Tra

nsi

sto

rs

100,000,000

10,000,000

1,000,000

100,000

10,000

10001970 1975 1980 1985 1990 1995 2000 2005

4004 8008

8080

8086

286

386

486

Pentium 4

Pentium II

Pentium

Pentium III

NVIDIA NV30 GPU

Sony Graphic SynthesizerNVIDIA NV40

Intel

Intel

Intel

Intel

Intel

Intel

Intel

Intel

IntelIntel

Intel

Sony Graphic Synthesizer

IBM Power4

NVIDIA NV25 GPU

NVIDIA NV20 GPU

AMD Athlon XPIntel Centrino

NVIDIA NV35 GPUATI Radeon X800

Motorola G4

High-end Processors: Intel only?


1,000,000,000

Nu

mb

er o

f T

ran

sist

ors

100,000,000

10,000,000

1,000,000

100,000

10,000

10001970 1975 1980 1985 1990 1995 2000 2005

4004 8008

8080

8086

286

386

486

Pentium 4

Pentium II

Pentium

Pentium III

NVIDIA NV30 GPU

Sony Graphic SynthesizerNVIDIA NV40

Intel

Intel

Intel

Intel

Intel

Intel

Intel

Intel

IntelIntel

Intel

Sony Graphic Synthesizer

IBM Power4

NVIDIA NV25 GPU

NVIDIA NV20 GPU

AMD Athlon XPIntel Centrino

NVIDIA NV35 GPUATI Radeon X800

Motorola G4

High-end: Embedded Processors!

7


Embedded Systems

Embedded Computer Systems

Devices other than desktop PCs, servers, and notebooks

Electricity running throughPerform something intelligent

Hardware/software which form a component of a larger system, but are concealed from user

Computers camouflaged as non-computers

The future of computing!


Profusion of Embedded Systems

GartnerGroup estimates 70 Billion µP used in embedded systems in 2001

Other estimates say 50 to 120 Billion µP

Average embedded system has 4 µP

Of all µP sold, 90% go into “non-computers”, 10% in “computers”

You will most likely end up working with a “non-computer” at some point in your career

8


Components of Embedded Systems

Analog Digital Analog

Memory

Coprocessors

Controllers

Converters

Processor

Interface

Software(Application Programs)

ASIC


Components of Embedded Systems

Analog ComponentsSensors, Actuators, Controllers, …

Digital ComponentsProcessor, CoprocessorsMemoriesControllers, BusesApplication Specific Integrated Circuits (ASIC)

Converters – A2D, D2A, …Software

Application ProgramsException Handlers

9


Computing Elements in ES

ProcessingTransformation of dataUse processors

Storage Retention of dataUse memories

CommunicationTransfer of dataUse buses

Peripherals


Example: BMW 745i

2, 000, 000 LOCWindows CE OS53 8-bit µP11 32-bit µP7 16-bit µPMultiple NetworksBuggy!

10


Characteristics

Application SpecificApplications are known a prioriOptimize for cost, area, power, and performance

Digital Signal ProcessingSignals are represented digitally

ReactiveReacts to changes in the system’s environment

Real-timeCompute certain tasks before deadline

Distributed, Networked, …


Characteristics

ReliabilityProbability of system working correctly provided

that is was working at t=0

MaintainabilityProbability of system working correctly d time

units after error occurred.

SafetyNot harmful for user

SecurityConfidential and authentic communication

11


Traditional Design Challenges

Low cost

Light weight

Reliability

Low power

Portable

Complexity

Ease of use

Mixed digital/analog requirements

Shrinking time-to-market

Short product lifetime

Real-time processing

Inherent concurrency

HW/SW co-design


Recent Design ChallengesDesign Complexity

Ultra low powerHighly adaptiveActive power management (voltage scaling, etc.)Alternative energy source (scavenge, solar, etc.)

Internet awareIncorporate RF technologiesNetworking capabilities

Verification

Security

12


ICS212 – Lecture 2Embedded Applications


OverviewHybrid embedded systems

Aerospace, automobiles, robotics, process control, and sensor nets

Multimedia

Consumer electronicsAppliances, office electronics, and home/office automation

Network componentsBridges, routers, switches, and hubs

Medical instrumentsPatient monitoring, MRI, and artificial hearts

E-BusinessATM, vending machines

Distributed & grid computing

OV

ER

LA

PP

ING

13


Hybrid Embedded SystemsComputation systems whose behavior is tightly integrated with the physical world

E.g., the behavior of an unmanned aerial vehicle (UAV) can be modeled by a combination of differential equations (the aerodynamics and low level feedback controllers) and a finite state automata (high level flight path decisions, such as to ascend or descend).

Behavior is governed by both continuous-state dynamics from the physical world and discrete-state dynamics from the computation

Passage of time during computation affects the state of the physical world

Inherently concurrent (inherent vs. built-in concurrency?)


Aerospace

Flight controlStability: real-time differential feedback loops

Positioning & navigationGPS, INS

InstrumentationData acquisition, display, processing, and archive

RadarCommunication

14


Automobiles

Engine managementFuel, ignition, timing

Emission controlInstrumentation

Data acquisition, display, processing, and archive

Safety & stabilityAirbags, active control

Entertainment & comfortRadio, A/C, …


Robotics

Implies autonomous operationN physical degree of freedomArtificial intelligenceControl heavyMission oriented

Repair, search, rescue, investigate, and perform physically difficult tasks

15


Process Control

Industrial automationPlant monitoring and production controlSimilar to control systems but with emphasis on management


Sensor NetsMany sensor nodes each capable (but limited) of sensing, computation/storage, and communication

Structure safetySearch and rescueMilitary use

Self organizationEnergy EfficientDistributed

Modern Sensor Nodes

UC Berkeley: COTS Dust

UC Berkeley: COTS DustUC Berkeley: Smart Dust

UCLA: WINS Rockwell: WINS JPL: Sensor W ebs

16


MultimediaAn exercise in

Signals, analog to digital conversion, quantization, sampling, processing, and digital to analogue conversionInformation theory, entropy, Huffman codes, compression, lossless compression Images, audio, video

Virtual RealityPresentationQuality of serviceThink lots of data (formats and standards too)!


Consumer ElectronicsHome appliances

Yesterday’s appliances: add computationTomorrow’s appliances: add networking (Internet)

Office electronicsIntegration Electronic paper (filing, printing, sending, and receiving)

Home/office automationCommon fantasy about the automated home or office of the future with lights and appliances that operate by themselves or with minimal effort

17


Network ComponentsStitching LANs

BridgeConnects two parts of the same network

RouterLink networks using different network identities

Extending portsSwitch

Transmit to recipient onlyHub

Transmit to all

Handle large volume of highly structured data with little transformation


Medical InstrumentsPerform diagnosis (screening/evaluation)

Data collectionAppraisal of that dataDeveloping a plan of action

Observation or monitoring

Sensing and instrumentationAccuracy and precision

Other applications:Radiation therapyArtificial hearts, arms, legs, …

18


E-Business

Information processing systems

ATM

Cash registers

ScannersCredit-card readers

Often the interface behind a database

Automation and convenience


Distributed & Grid Computing

Coordinated resource sharing and problem solving

The grid is static, reliable, and has infinite resource (for practical purposes)

Users (the mobile device e.g., PDA) has limited resources

Middleware mitigates the resource sharing and coordination efforts

19


Summary of Application Domains

Looked at a number of application domains

Large amount of overlap between these domains

Often each domain has associated standards, design methodologies, and certification programs

The future appears to suggest a fusion of design differences into a single methodology


Real-Time Systems

A systems where correctness depends on logical results and the time the results are produced

Safety-critical: incorrect operation leads to human lossMission-critical: incorrect operation leads to failed mission

J.A. Stankovic et al. “Strategic Directions in Real-Time and Embedded Systems”

20


Course Outline

ConceptConceptSpecificationSpecification

HW/SWHW/SWPartitioningPartitioning

Hardware ComponentsHardware Components

Software ComponentsSoftware Components

Estimation Estimation --ExplorationExploration

HardwareHardware

SoftwareSoftware

DesignDesign

(Synthesis, Layout,

(Synthesis, Layout, ……))

DesignDesign(Compilation,

(Compilation, ……))

Validation and Evaluation (area, power, performance, Validation and Evaluation (area, power, performance, ……))


Course Outline1/5 W Course overview, Introduction to ES, Typical ES Applications

1/10 M ES Components (Hardware) 1/12 W ES Components (Software)

1/17 M MLK Holiday1/19 W Embedded Design Domains

1/24 M Models, Languages, Tools1/26 W Models, Languages, Tools

1/31 M HW/SW Codesign2/2 W Reconfigurable Computing

2/7 M Timing2/9 W SW4ES: RTOS, Scheduling

2/14 M SW4ES: Compilers, Firmware, Middleware2/16 ES: Validation and Testing

2/21 M DSP2/23 W Control Systems

2/28 M Networked ES: Mobile, wireless3/2 W Networked ES: Ad-hoc networks, Sensornets

3/7 M Ubiquitous Computing (Tentative!)3/9W Embedded systems and HCI/CSCW (Tentative!)

3/14 M Fault Tolerance3/16 W ES Case Studies

Introduction to Embedded and Ubiquitous Systemsjupiter.plymouth.edu/~wjt/OpSys/EmbeddedSystems1.pdf · 3 ICS212 WQ05 (Dutt) Introduction 5 Reading Materials zRecommended Books Peter

Documents