Mobile Computing

Document Sample
Mobile Computing Powered By Docstoc
					第1章 移动计算简介
 §1.1 基本概念
 §1.2 相关概念分析
 §1.3 移动应用
The Era of “Mobile Computing”

             Mobile Computing
          Mobile/Wireless Network

Mobile Computing Architecture

Mobile Computing Is
 "taking a computer and all necessary files and software
  out into the field."
 " being able to use a computing device even when
  being mobile and therefore changing location.
  Portability is one aspect of mobile computing."
 " the ability to use computing capability without a pre-
  defined location and/or connection to a network to
  publish and/or subscribe to information . "
 " a variety of wireless devices that has the mobility to
  allow people to connect to the internet, providing
  wireless transmission to access data and information
  from where ever location they may be. "

Mobile Computing
 Mobile computing refers to computing in a distributed system
  in which some processes or processors can move.
       Moving processes  logical mobility, realized by agents.
       Moving processors  physical mobility, realized by moving devices.
 Mobile computing extends a distributed computing
  environment with a new dimension of mobility.
       Most existing mobile computing systems are based on client-server
       computing systems.
       Recent mobile computing solutions consider general distributed
       computing, namely, peer-to-peer computing environments.
       Mobility generalizes the network reconfiguration problem.
 Many mobile computing techniques have their root in
  distributed systems.
Mobile Computing
 Motivations
      Availability of wireless network at reducing communication cost.
      Widespread use of handheld devices with increasing computing
      power and lower cost.
      Integration of communication and computing devices (Tablet
      computer, SmartPhones).
      Detection of the physical or logical location of a device.
      Making use of wireless network to access information and perform
      data processing, resource sharing at high capacity servers.
      Extend the ability and usefulness of handheld devices.
 Goal
      Access to information anywhere, anytime, anyway.
Mobile Computing vs. Others
 Related concepts
      Wireless computing
         computing without a wire.
         the most primitive form, e.g. using 802.11b, 802.11g or bluetooth.
      Nomadic computing
         computing on the move.
         clients or users are moving.
         physically mobile computing, perhaps prolonged disconnection.
      Ubiquitous computing
         a term coined by Mark Weiser in 1991.
         invisible computing or subconscious computing.
         computers and their functionality available to users without being noticed.
      Pervasive computing
         immersive computing.
         use of wearable computers.
         a current name for ubiquitous computing.
Mobile Computing vs. Others
 built upon distributed computing discipline
 closely related to
     networking and wireless communications,
     internet computing and cloud computing,
     pervasive computing.

Mobile Computing vs. Others
 Networking is the base layer, with communication primitives.
 Wireless networking and communications extends networking
  protocols to cover lack of link and movement of devices
       GPRS, 3G (WCDMA), IPv6, WLAN.
 Distributed computing covers fundamental issues to utilize
  computers over a network.
       coordination mechanism to make the solutions work
       general problem solving approaches
 Internet computing is more on the practical issues of
  developing programs and applications that work over internet.
       using Java and related technology.
       can be considered a case of distributed computing.
Mobile Computing vs. Others
 Mobile computing relies on wireless networking and
  communications and extends distributed computing
     new issues of client movement, location management
     and context-awareness, not to mention the limitations of
 Pervasive computing extends again mobile
  computing, making use of small equipment and
     The number of devices and volume of data can increase
     several orders of magnitude
     new issues like data streaming need to be handled.
Pervasive Computing
 Industry revolution vs information revolution
Pervasive Computing
 Devices
Pervasive Computing
 Principles
        based on distributed systems.
        with information synchronization.
        devices and applications are embedded into wireless
        networking infrastructure.
        should target on different users with different needs.
        must be able to manage the diversity.
        boundless connectivity.
        convenient and intuitive to use.
        good and mature human computer interface.
Pervasive Computing
 The MIT wearable computer project
      PolyU ITC is doing something similar now.

                                                  MicroOptical display
                                                  IR tag reader
                                                  CerfBoard (developin
                                                  Internet appliance)
                                                  Body bus junction

                                                  802.11 bridge
                                                  BSEV core (single
                                                  Board computer)
                                                  Body network hub
Recap on Distributed Systems
 A distributed system is a system supporting distributed
 There is no universal definition.
 It is generally a collection of autonomous computers,
  connected together by some communication media, usually a
 They are capable of sharing resources and capable of
  providing service for one another.
 The processors cooperate to achieve an overall goal.
 Users may view the resulting environment as a single
  computing facility.
Recap on Distributed Systems
 Distributed hardware
      Processors and memory are physically separated, connected only
      via communication media.
 Distributed control
      Each processor is autonomous, running its own program.
      They cooperate together in a voluntary manner.
 Distributed data
      Data for a program may scatter around and need to be exchanged to
      carry out a task.
 Resources sharing
      Expensive resources can be shared among different users.
Recap on Distributed Systems
 Openness
      System is visible to user, and can be modified and extended when
      needs arise.
      For example, Unix is an open system and IBM PC has an open
 Concurrency
      Processors are autonomous and running concurrently.
      Several jobs may be executed simultaneously.
 Scalability
      It is relatively easy to build a large system using similar approach to
      build a small system.
      For example, Internet is built from smaller networks.
Recap on Distributed Systems
 Fault tolerance
      Processors more available (high availability), independent failure
      mode and preferably graceful degradation.
      System can still function when some parts are down.
 Transparency
      User should not perceive the system as being composed of different
      parts and the lower level of functionality must be concealed.
      For example, send/receive using a socket should not be concerned
      with data link and network connections.
      Forms of transparency include access, location, concurrency,
      replication, failure, migration, performance, scaling.
New Features in Mobile Computing
   Limited asymmetric bandwidth and high latency
        Wireless channels are of limited bandwidth and asymmetric.
        Perhaps of speed of grade of modems.
   Weak connectivity and low reliability
        Disconnection to channels can easily occur (planned or unplanned) and
        can be prolonged.
        Messages may be lost with much higher probability.
   Low processing power and energy
        Battery is expensive and limited.
        Screen or display is small.
   Mobility of devices
        Need to track location of devices and to provide proper services.
        Invalidation to data upon change of location.
   Low physical security
        Mobile devices are easily lost, broken or stolen.
   High degree of heterogeneity
        Devices are very different in nature, e.g., Palm and Pocket PC are
        programmed differently, worse for embedded systems and devices.
Objectives of Mobile Computing
 To run existing distributed applications
 To provide distributed services
 To share data and resources
 To maintain efficiency
 To keep communication and operational cost down
 To conserve energy
 To achieve consistency among views of different
 To be scalable in supporting numerous users
New Solutions to Old Problems
 Solve problems in distributed systems
      Hardware and computer architecture
      Communication architecture and protocols
      Distributed operating systems
      Synchronization need
      File systems
      Databases systems
      Real-time processing
      Multimedia support with QoS
      Dynamic system reconfiguration
      Application requirement and design
      User interfaces design (on mobile devices)
Mobile Computing Architecture
 Fixed network can be dedicated proprietary network
  for the mobile system.
 Fixed network can also be the Internet or Intranet.
 Infrastructured Systems
      MSS (mobile support station, or base station) +
      MU (Mobile unit, mobile station, mobile client)
      WLAN, Cellular Net., etc.
 Ad hoc systems
      Mobile nodes + mobile nodes
      MANET, WSN, VANET, etc.
  Mobile Computing Architecture


(mobile unit)                              Fixed network
                (mobile support station)

       Wireless radio
       cell (19.2Kbps)                               WLAN      23
Mobile Computing Architecture

 Personal applications
      Web and email access.
      Personal information management.                       personal
      Personalized use of computer resources.
      Weather and traffic report.                     logistics/tourism
      Network games.                                          personal
 Commercial applications                                      finance
      Stock price quotation.
      Inventory control for traveling salesperson.  e-commerce/finance
      Banking activities.                                       finance
      Bus-stop information kiosk at major bus terminals.       tourism
      Push-based advertisement on vehicle.          e-commerce/tourism
      Generic M-commerce applications.                     e-commerce
Location Tracking
 Active badge/RFID card with sensor technology
      An early application of sensor technology in location tracking,
      developed by AT&T Cambridge Lab, for tracking and locating of
      individuals within a building.
      Each building office is equipped with one or more networked
      sensors to detect these transmission signals.

 Biometrics cards with
  bluetooth and RFID deployed
  at US Homeland Security
  Office May 2005.
      Requirement: strong resistance
      to terrorist exploitation and
      rapid electronic authentication.
      Could Hong Kong smart ID do
      Problem: signal could be
 Location and mobility are the two major factors to be
  capitalized in a mobile application.
 Applications can be based on the current location of a user.
 Example
       Emergency vehicle routing to caller of 999.
       Deliver me a pizza as I am hungry!
       Deliver value-added information based on current position.
          Targeted advertisement to customer: E-coupon.
       Logistics arrangement can be made based on location.
          Send the goods to an alternative location if traffic jam is out-of-control
          for receiving the goods.
 A context-aware application is an application that adapts its
  behavior to a changing environment.
 It needs to gather contextual information about its
  environment, usually through the use of sensors.
 User does not need to give all information and the
  application should try to figure out the rest.
 The follow-me type of applications:
       Application knows the location of a user:
          Smart map (Telenor Research) that always maintain the current user
          position at center.
          Nearest restroom/toilet depends on gender.
       When in a room, system can look up user profile for preferred room
       setting, or mode of operation for a speaker.

           Context vs. Location?

   Example of three stages???
Intelligent Home
 Intelligent home networking system with general technology
      Use mobile phone technology to instruct a home control
      computer for home automation equipment.
      WiMAX (802.16), 802.11, Bluetooth etc.
      Transmit signal and control electronic devices via power line.
      Smart TV
Killer App. of Mobile Computing
   Instant messaging?
        10 million iMode users in Japan since 18 months of introduction, also used by
        Greece Olympics.
        15.6 billion SMS messages in China January 2004.
   Mobile games?
        $9 billion sold in 2004.
   Mobile web access?
        WAP, iMode, WWW
        How to resolve the bandwidth problem? 3G?4G?
   Location-dependent services?
        Sensor network infrastructure cost?
   Context-aware services?
        Locating a best restaurant closest by sensing e-cash in e-wallet and dining
   How can the communication infrastructure match with the application
    functionality and bandwidth / mobility requirement?
        High bandwidth can be resulted, using pico-cells, but the client movement speed
        must be limited.
Bandwidth Requirement

                   wireless    current
              3G              Ethernet
 Concept of mobile computing
 Mobile computing vs. related concepts
      Pervasive computing
 Mobile applications


Shared By: