O slideshow foi denunciado.
Utilizamos seu perfil e dados de atividades no LinkedIn para personalizar e exibir anúncios mais relevantes. Altere suas preferências de anúncios quando desejar.

Module1 Mobile Computing Architecture

2.895 visualizações

Publicada em

Three Tier Architecture, Design Consideration for Mobile Computing, Architecture of Mobile Computing

Publicada em: Engenharia
  • Seja o primeiro a comentar

Module1 Mobile Computing Architecture

  1. 1. Client layers of the optical layer Module 1 1
  2. 2. Architecture for mobile computing  The first layer is User Interface or Presentation Tier.  Deals with user facing device handling and rendering  Includes a user system interface where user services reside  Second tier is the Process Management or Application Tier.  Here business logic and rules are executed.  Capable of accommodating hundreds of users  Controls transactions  Asynchronous queuing to ensure reliable completion of transactions.  Third tier is Database Management or Data Tier  For database access and design  Provides increased performance, flexibility, maintainability, reusability and scalability, while hiding the complexity of distributed processing from the user 2
  3. 3. Three-tier architecture for mobile computing 3
  4. 4. Three-tier architecture (in detail)  To have universal access, the server is connected to a ubiquitous network like internet.  To have access with network, an application called web browser is required such as Internet explorer, Mozilla etc  Three layers  Presentation Tier (Tier 1)  Application Tier (Tier 2)  Data Tier (Tier 3) 4
  5. 5. The mobile computing architecture 5
  6. 6. Presentation Tier (Tier 1)  First tier  Layer of agents applications and systems.  Applications run on the client device and offer all the user interfaces.  Responsible for presenting the information to the end user.  Humans use audio and video to receive information from the machine  Information is given to the system from keyboard(computer, laptop, cell phones ), pen (tablet, palmtop) etc  It includes web browsers 6
  7. 7. Application Tier (Tier 2)  Considered as engine of the application  Performs the business logic of the processing input, obtaining data and making decisions.  Includes technologies like CGIs, Java, JSP, .NET services, PHP etc. deployed in products such as Apache, Websphere etc.  Tier is presentation and database independent.  Operations such as network management, security are performed using a middleware software  Middleware is defined as a layer of software, which sits in between the operating system and user facing software.  Can also be defined as a software gateway connecting two independent open objects. 7
  8. 8. Middleware categories  Message-oriented Middleware (MOM)  Transaction processing (TP) Middleware  Database Middleware  Communication Middleware  Distributed Object and Components  Transcoding Middleware  Internet Content Adaption protocol (ICAP)  Web services 8
  9. 9. Message-oriented Middleware (MOM)  It connects different applications through asynchronous exchange of messages.  Works over a networked environment without knowing what platform or processor the other application is resident on.  The message can contain formatted data, requests for actions or unsolicited response.  Provides a message queue between any two interoperating applications.  If the destination process is busy , then message is held in temporary storage until it is processed. 9
  10. 10.  It is asynchronous, peer to peer, works in Publish/Subscribe fashion.  In this mode, one or more objects subscribe to an event.  As event occurs it will be published by the loosely coupled asynchronous object.  MOM will notify about this event , Is appropriate for event driven applications  Eg: Message Queue from IBM known as MQ series 10
  11. 11. Transaction Processing (TP) middleware  It provides tools and environment for developing transaction based distributed applications.  Inputs the data into a system at point of information source and output the data at information sink.  Used in data management, network access, security systems, delivery order processing, airline reservations, customer service etc  Capable of providing services to thousands of clients in a distributed client/ server environment. 11
  12. 12.  Includes features such as restarting failed process, dynamic load balancing and ensuring consistency of data.  Independent of database architecture.  It optimises the use of resources by multiplexing many client functions onto much smaller set of application service routines.  Eg: MVC 12
  13. 13. Transaction processing middleware 13
  14. 14. Model view controller MVC  Java uses MVC architecture , which is an example of TP middleware  It splits application into separate layers i.e. presentation, domain logic and data access.  Model  domain specific representation of the information on which the application operates.  Domain logic manipulates and adds the meaning to the raw data  View  Responsible for rendering the model into a form suitable for interaction and understood by the user  Controller  Manages the process and responds to events, typically user actions, may invoke changes in the model 14
  15. 15. Communication middleware  Similar to connecting one application to another with the help of telnet  Used in telecommunication world  User interface is through telnet  A mediation server automates the telnet protocol to communicate with these nodes in the network.  Eg:TN5250, TN3270 15
  16. 16. Distributed Object and Components  Eg: CORBA (Common Object Request Broker Architecture)  Open distributed computing infrastructure being standardized by object management group.  Is vendor – independent infrastructure.  CORBA based program from any vendor on almost any computer, operating system, programming language and network can interoperate with a CORBA based program from same or another vendor  Handles large number of clients at high rates with high reliability. 16
  17. 17. Transcoding Middleware  Used to transcode one format of data to another to suit the need of the client.  To access web pages on phone, the HTML pages are transcoded into WML pages so that mobile phone can access it.  It is used for content adaptation to fit the need of the device.  Content adaptation is also required to meet the network bandwidth.  Content adaptation is done through proprietary protocols 17
  18. 18. Internet Content Adaptation Protocol (ICAP)  Contents present in the internet is accessed from different agents and devices.  Is a protocol aimed at providing simple object based content vectoring for HTTP services.  Lightweight protocol to do transcoding on HTTP messages.  Allows ICAP clients to pass HTTP messages to ICAP servers for some sort of transformation.  The server sends back the transformed messages back to its clients.  Before document is displayed for the agent, it is checked virus. 18
  19. 19. Typical data flow in ICAP environment 19
  20. 20. Steps performed  The user agent makes a request to an ICAP client for an object on an object server.  The client sends the request to the ICAP server  The ICAP server executes the ICAP resource service on the request and sends the possibly modified request or a response to the request ICAP client  The client sends the request, possibly different from the original client’s request to the original server.  The origin server responds to the request  The ICAP client sends the reply to the client. 20
  21. 21. Services provided by ICAP servers  Virus checking for the requested content  Content filtering based on the sensor rating like PG( parental guidance) , R (Restricted)  Local real time advertisement insertion like television  Wireless protocol translation  Image magnification for elderly  Peer to peer compression  Encryption of data and many more. 21
  22. 22. Web services  It provides a standard means of communication and information exchange among different software applications, running on a variety of platforms or frameworks.  Is a software system identified by URL whose public interfaces and bindings are defined using XML  The basic architecture includes web service technologies capable of:  Exchanging messages  Describing web services  Publishing and discovering web service descriptions.  Web services architecture defines the standard for exchange of messages between the service requester and service provider. 22
  23. 23.  Service providers are responsible for publishing a description of the services they provide  Requesters must be able to find and discover descriptions of the services.  Software agents in the basic architecture can take on one or all of the following:  Service requester: requests the execution of the web service  Service provider: processes a web service request  Discovery agency : agency through which a web service description is published and made discoverable. 23
  24. 24. Data (Tier 3)  Used to store data needed by the application and acts as repository for both temporary and permanent data.  The data can be stored in any form of datastore or database.  Data can be stored in XML format for interoperability with other systems and data sources. 24
  25. 25. Database Middleware  Database independence helps in maintenance of the system better.  Allows the business logic to be independent and transparent of the database technology and the database vendor.  Runs between the application program and the database.  Also called as database connectors as well.  Eg: ODBC, JDBC  With the help of these middleware's, the application will be able to access data from any data source.  Data sources can be text files, flat files, spreadsheets or network, relational , indexed, hierarchical, XML database etc. 25
  26. 26. SyncML  Is a protocol is an emerging standard for synchronization of data access form different nodes.  New industry initiative to develop and promote a single common data synchronization protocol that can be used industry wide.  Supports the naming and identification of records and common protocol commands to synchronize local and network data.  Supports identification and resolution of synchronization conflicts.  It works over all networks used by mobile devices both wireless and wireline. 26
  27. 27. Design consideration for mobile computing  The term context means , all the information that helps determine the state of an object or actor  It can be a person, device , a place , a physical or computational object.  Content with context awareness  Build each application with context awareness  There are different services for different client context  Bank providing services on internet, for mobile phones and PDA  Have different URL for each type of device  Content switch on context  Provides adaptation of content within service  Transparent to the client  Here service is the same for internet , PDA and WAP  All access the bank site from same URL 27
  28. 28.  Content transcoding on context  Performs the adaptation of content based on the context and behaviour of the device  Transparent for the client and the application  Middleware is used to perform all the operation  Middleware is intelligent to convert from HTML to XML or vice versa as need occurs. 28
  29. 29. Client Context Manager 29
  30. 30.  Identity information  The device will be in a position to communicate its identity without any ambiguity  Spatial information  Information related to surrounding space i.e. location, orientation, speed , elevation and acceleration  Temporal information  Information related to time such as time of day, date, time zone, season of the year  Environmental information  Related to environment surroundings such as temperature, air quality, moisture etc  Social information  Related to social environment such as who is with you, people nearby, whether the user is in the meeting or party. 30
  31. 31.  Resource that are nearby  Relates to the resources accessible nearby such as devices, hosts or information sinks.  Availability of resources  Information about the device in use such as battery power , processing power, display etc  Physiological measurements  Relates to physiological state of the user such as blood pressure, heart rate, respiration rate , tone of voice  Activity  Relates to the activity of the user such as talking, reading, walking and running.  Schedules and agendas  Relates to the schedules and agendas of the user. 31
  32. 32.  A system is context aware of it can extract , interpret and use context related information to adapt its functionality to the current context.  Three aspects of the context management  Context sensing  The way in which the information is obtained  Context representation  The way in which the context information is stored and transcoded.  Context interpretation  The way in which the meaning is obtained from the context representation.  W3C proposed a standard for context information, i.e. Composite Capabilities/ Preference Profiles (CP/PP) 32
  33. 33. Composite capabilities/ Preference Profiles (CC/PP)  Proposed by W3C  Model is based on RDF( Resource Description Framework) and can be serialized using XML.  Contains a number of attribute names and associated values that are used by an application to determine the appropriate form a resource to deliver to a client.  Help a client/ proxy to describe their capabilities to an origin server or other sender of resource data.  Fir different application to interoperate, some common vocabulary is needed. 33
  34. 34. Steps are  Device sends serialized profile model with request for content  Origin server receives serialized RDF profile and converts it into an in-memory model  The profile for the requested document is retrieved and in- memory model is created.  The device profile model is matched against the document profile model  A suitable representation of the document is chosen.  Document is returned to device and presented. 34
  35. 35. The simplest use of CC/PP 35
  36. 36. Policy Manager  Responsible for controlling policies related to mobility  A policy is set of rules, these rules need to be followed under different conditions.  When mobility is considered, the information will be visible from outside the four walls of the enterprise.  Organizations will have policies regarding the disclosure of the information.  Eg: certain documents can be viewed only at the office of CEO  These kinds of policies must be transferable to mobile computing policy manager.  It will be able to define policy for the documents and assign roles to users.  Each role will have permission , prohibition, obligations associated with it.  Policy manager will have rights to read, write and execute 36
  37. 37. Semantic Web  How to make machines understand policies and make them behave in the expected fashion?  Semantic web addresses this need  Data is defined in the web and linked in such a way that it can be used by the machine not for just display, but for the automation. Security, filtering, integration and reuse of data across various applications.  It is generally built on syntaxes which use URLs to represent data, usually tuple based structures.  These syntax is called Resource Description Framework (RDF) syntaxes. 37
  38. 38. Security manager  Provides a connection between the client device and the origin server.  Security manager will ensure security with respect to  Confidentiality:  The message transacted needs to be confidential  Nobody should be able to see it.  Managed by encryption eg: AES  Integrity  The message transacted needs to be tamper-resistant  No body should be able to change any part of the message  Managed using various hashing algorithms 38
  39. 39.  Availability  The system will be available  Nobody will be able to stop the service  Non-repudiation  Users of the system can be identified  Nobody after using will claim otherwise  Managed by digital signatures  Trust  There are complex issues of knowing what resources, services or agents to trust.  The system will be trusted. 39
  40. 40. Platform for privacy preference project (P3P)  Is an emerging standard defined by W3C  Enables web sites to express their privacy practices in standardized format so that they can be retrieved and interpreted as user agents.  With P3P , users need not read the privacy policies they visit  Instead key information about the content of the web is conveyed to the user  The goal of P3P is to increase the user trust and confidence in the web.  This informs the privacy policies about the site, hence makes the user to decide whether to release personal information or not. 40
  41. 41. Adaptability Manager  Responsible for adapting content, behaviour and other aspects acc. to the context and policy  May take any number of actions depending on the information passed to the context manager  Most common action performed is transcoding of the content so that it may be viewed on a particular device  Appending location-specific information to the document. 41
  42. 42. Content adaption and transcoding  Content adaptation can be done at either content level in server end or the agent level in client device  Can be done in intermediate level in a middleware framework.  Content adaptation needs to consider the following attributes:  Physical capabilities of the device  Screen size, width and height in pixels, color  Logical capabilities of the device  Display image, audio, video  Effective network bandwidth  Payload  The total amount of bits can be delivered to the agent for static parts. 42
  43. 43. Types of transcoding  Spatial  Transcoding in space or dimension  A standard frame is downscaled and reduced.  The frame is changed from one size to different size to suit the target device  Temporal  Reduction of number of frames on time scale  Downscales the no. of transferred frames to suit the target device and network bandwidth  Code  Used to change the coding standard from one to another  Eg: transcoding BMP code into WBMP for wireless devices 43
  44. 44. Types of transcoding  Color  Requested for monochrome clients  Uses less pixels and reduces bandwidth consumption  Modifies the perception of the image  Object or semantic  Comprises various techniques based on the computer vision techniques  Extracts the valuable object from the scene, transfer them with lower amount of compression in order to maintain the both details and speed. 44
  45. 45.  Server side adaptation , resource requirements are  Static content size in bits  Display size such as height, width and area  Streaming bit-rate  Color requirements  Compression format  Hardware requirement such as display for images, support for audio and video 45
  46. 46. Transcoding done through middleware  Benefit of the middleware approach is it is totally transparent to the device and content  Drawback is  Content providers have no control over how their content will appear to different clients  There may be legal issues like copyright  Html tags provides formatting information rather than semantic information  Transcoding of videos and audio is not easy  Developing a general purpose transcoding engine is very difficult. 46
  47. 47. Content Rating and Filtering  W3C has proposed a standard called PICS (Platform for Internet Content Selection) for rating of the web content  Filtering of the content takes place based on this rating  It is set of technical specifications for labels that help software and rating to work together  Rating is allowed to the service provider , since it involves some amount of subjective judgement  Rating can be through self labelling or third part labelling of the content  PICS was concerned about children accessing the inappropriate materials , i.e meta data system, meaning that labels provide any kind of descriptive information about internet material  Controlling was done using : 47
  48. 48.  Digital Signature:  Coupling the ability to make assertions with cryptographic signature block that ensure integrity and authenticity  Intellectual Property Right Management  Using the meta-data system to label Web resources with respect to authors, owners, rights management information  Privacy (P3)  Using Meta-Data system to allow sites to make assertions about their privacy practices and for users to express their preferences for the type of interaction they want to have with those sites.  Personalization  Based on some policy, the content can be personalized to suit the need of user and the service 48
  49. 49. Content Aggregation  Four parties  End User (EU)  Content Provider (CP)  Content Aggregator (CA)  Internet Service Provider (ISP)/ Wireless Network Operator (NO)  NO will have routers, gateways, cache etc to offer service  Any node can be responder and requester  There will be different parameters to determine the content  They can be static or dynamic parameters  Static parameter is received before the service begins  Dynamic parameter are those which are required with every request. 49
  50. 50. Types of service (Content Aggregator ‘s perspective)  Single Service Request  Works at user level  Works for only one user  Group service request  Works for group of users  Initiated at CA level or the NO level 50
  51. 51. Seamless Communication  A system should be available and accessible from anywhere and anytime and through any network or device  The user should be able to access the system when travelling.  Patient travelling from house to hospital via ambulance or helicopter, here information has to exchange without any interruptions.  Seamless Communication will combine seamless handoffs and seamless roaming 51
  52. 52.  Handoff is the process which the connection to the network is moved from one base station to another base station within the same network  Roaming is a process in which connection to the network is moved from one base station of one network to another base station of another base station.  Offers freedom to roam across different wireless networks  Roaming works within homogeneous networks like GSM to GSM or CDMA2000 to CDMA2000 52
  53. 53.  Following aspects needs to be maintained in seamless communication  Authentication across network boundaries  Authorization across network boundaries  Billing and charging the data collection  End to end data security across roaming  Handoff between wireless access points  Roaming between the networks  Session migration  IP mobility 53
  54. 54. Autonomous Computing  The purpose of autonomous system is to free users and system administrators from the details of the system operation and maintenance complexity.  System will run 24 X 7  Combination of the following functions  Self Configurable  Will configure automatically in accordance with high level policies  Self optimizing  Continuously look for ways to improve its operation with respect to cost, resource and performance  Self Healing  Will detect, diagnose and repair localized problems resulting from bugs or failures. 54
  55. 55.  Self Protecting  It will be well defined from external attacks  Will not cascade failures to other parts of the system  Self Upgradeable  Will be able to grow and upgrade itself within the control of the above properties. 55
  56. 56. Context aware Systems  Context manager is to maintain information pertaining to location, mobile devices, networks, users , the environment around each mobile device and other context information deemed relevant  Information such as  Location information  Helps to identify the location of the user/device  Can be achieved using device or through network  From device- GPS, can offer location precision of 10 feet radius  From network- timing advanced technology- precision of 100 feet radius  Device information  The know about the characteristics of the device  To determine the resource capability and the user interface capability  User-Agent parameter of HTTP we can obtain the device information 56
  57. 57.  Network information  Mobile computing environment, device moves from one network to another network  Is required to identify the capability of the network  Includes information such as security infrastructure, services offered by the networks  User information  This information is required to identify the user correctly  The system needs to ensure that the user is genuine and who he claims to be  Information can be validated through authentication independent of device or network  Environment information  Tells about the surrounding environment  Know about the temperature, elevation, moisture etc which is mainly required for sensor based networks 57
  58. 58. GPS  Global Positioning System , is a system that gives exact position on this earth  Funded and controlled by US Department of Defence  GPS satellites are orbiting around the earth, which transmits the signals that can be detected by anyone with a GPS receiver  Has three parts  Space segment  User segment  Control segment  Space Segment  Consists 24 satellites, each in its own orbit  Each GPS satellite takes 12 hours to orbit the earth  Each satellite is equipped with a clock to broadcast signals with respect to time  User segment  Has receivers, which can be users hand, embedded in a mobile device or mounted in a vehicle.  Receives the signal from satellite at the speed of light 58

×