๐ Lecture 1 โ Mobile & Pervasive Computing
๐ Overview: This lecture introduces the foundational concepts of pervasive and ubiquitous computing, tracing its origins from Mark Weiserโs vision at Xerox PARC to modern implementations. It explains how computing can be seamlessly integrated into everyday objects and activities, and surveys key historical projects and modern directions that have shaped this field.
๐๏ธ Topics Covered
The lecture covers the definition and philosophy of pervasive/ubiquitous computing, Mark Weiserโs vision and the original Xerox PARC prototypes (ParcTab, ParcPad, Liveboard), the evolution through IBMโs commercial initiatives and wearable computing, augmented reality, and several landmark projects including Classroom 2000, Aware Home, GUIDE, SenseCam, RADAR, EasyLiving, and Place Lab.
๐ Lecture Summary
Pervasive Computing
Pervasive Computing is defined as โcomputing that is omnipresent and is, or appears to be, everywhere all the time; may involve many different computing devices that are embedded in various devices or appliances and operate in the background.โ It is also described as โthe age of calm technology, when technology recedes into the background of our lives.โ The goal is to make computers so common and accessible that users are not even aware of their physical presence. Ubiquitous Computing (ubicomp) is a post-desktop model of human-computer interaction where information processing is thoroughly integrated into everyday objects and activities, including furniture, clothing, white goods, and even toys and paints.
๐ Definition โ Ubiquitous Computing (ubicomp): A post-desktop model of human-computer interaction in which information processing has been thoroughly integrated into everyday objects and activities.
๐ก Why this matters: The key idea is that technology should become invisible and fade into the background, enhancing our lives without demanding our attention.
Evolution of Pervasive Computing
Mark Weiser coined the original term ubiquitous computing in 1988 at Xerox PARC. The essence of Weiserโs vision is that mobile and embedded processors can communicate with each other and the surrounding infrastructure, seamlessly coordinating their operation to support a wide variety of everyday work practices. Weiser believed that in a ubicomp world, computation could be integrated with common objects rather than being a separate activity. He sometimes referred to this as invisible computing.
The first prototypes from Xerox PARC included:
- ParcTab (or Tab): An inch-scale computer representing a pocket book or wallet. Tabs communicated wirelessly with a ceiling-mounted base station using 10 kbps diffuse infrared signaling.
- ParcPad (or Pad): A foot-scale device serving as a pen-based notebook or e-book reader. ParcPads used a low-bandwidth X-protocol across a radio link, communicating with a base station through a proprietary short-range near-field radio.
- Liveboard: Provides the functionality of a whiteboard. Liveboards were designed around standard computer workstations with much larger pen-based displays and pen-based input.
In the mid-1990s, IBM began a research direction called pervasive computing and created a business unit dedicated to the task. One of the first commercial deployments was a collaboration between IBM Zurich and Swissair in 1999 (IBM Swissair), enabling passengers to check-in using Web-enabled (WAP) cell phones. The phone also served as a boarding pass, showing gate, seat, and flight departure information.
Wearable computing takes another approach by putting the emphasis on a portable computer that can be unobtrusively integrated with a personโs clothing while being comfortable. An important related topic is augmented reality, where a computer can overlay information on top of what a user sees to improve their ability to carry out a task.
๐ Definition โ Invisible Computing: Weiserโs term for computation that is integrated so well into everyday objects that users may not even notice any computers were involved in their work.
๐ Definition โ Augmented Reality: A technology in which a computer overlays information on top of what a user sees in order to improve ability to carry out a task.
Pervasive Computing Projects
Classroom 2000 began in July 1995 with a vision of applying ubiquitous computing to education. It investigated capturing entire lessons in a form useful as a reference. The key challenge was to create automatically generated index points that enabled students to skip to exact points in time that might answer a question. Using an electronic board (the Xerox Liveboard), all teacher annotations and slide transitions were timestamped and used to index the audio and visual record. The combined media, timeline, and indices created a powerful summary available to students when class finishes.
The Aware Home project, founded in 1999, explored how computation and embedded technologies could support everyday activities in a home. A complete residential building was designed as a Living Laboratory, with features for embedded computation, sensing, wiring conduits, and a control center. Systems included cameras and RFID tags to identify and track occupant location, as well as a smart floor with pressure sensors that could identify individuals by their characteristic ambulatory gait.
The GUIDE project was the first mobile electronic guidebook designed and optimized from concept to implementation for use by tourists. It captured the imagination of researchers interested in location-based services in the wild.
๐ Definition โ Living Laboratory: A complete residential building designed from scratch with all expected modern home features plus additions to support embedded computation and sensing research.
Modern Directions
SenseCam is a small wearable computer that periodically captures images of the world as a user moves around. In collaboration with the MyLifeBits project, it provides contextual data about the wearer, augmented by a database of documents and electronic media the individual has accessed. The result is a prosthetic memory aid that enables more detailed recall than unaided means.
RADAR was the first example of a wireless system allowing mobile computers to locate themselves in a building (an indoor GPS). It was designed around the first WiFi (802.11b) radios and used reception maps in a Microsoft building with multiple access points (APs). By comparing the received signal strength indication (RSSI) for each AP at a point of interest with RSSI signal maps on record, the system could determine a mobile computerโs most likely location to an accuracy of 2โ3 meters.
EasyLiving, established in 1999, was Microsoftโs closest project to the original ubicomp vision. It centered on a smart room supporting work and recreational activities. Key features included using image processing with multiple cameras to recognize activities and track occupants, the ability to migrate computing sessions from screen to screen as people moved, and automatic control of lighting and music to suit all occupants.
Place Lab was the best-known project from Intel Research Seattle (IRS), exploring a system that could determine the location of a mobile device by cataloguing and mapping WiFi access points throughout a city. It was later extended to GSM towers and demonstrated effective location-based services for cell phone applications. It had a distinct advantage over GPS: it worked well indoors and did not require additional GPS equipment for outdoor operation.
๐ Definition โ RSSI (Received Signal Strength Indication): A measurement of the power present in a received radio signal, used by the RADAR system to compare against signal maps and determine location.
๐ Definition โ Prosthetic Memory Aid: A device (like SenseCam) that helps an individual recall information about their life by capturing and storing contextual data and electronic media.
โญ Key Takeaways
Students must remember that pervasive/ubiquitous computing is fundamentally about making technology invisible and seamlessly integrated into everyday life, as envisioned by Mark Weiser at Xerox PARC. The evolution moved from early prototypes (ParcTab, ParcPad, Liveboard) through IBMโs commercial initiatives to major research projects like Classroom 2000 and Aware Home. Modern directions include wearable computers (SenseCam), indoor location systems (RADAR using WiFi RSSI), smart environments (EasyLiving), and community-based location services (Place Lab). The defining characteristic of ubicomp is that computation recedes into the background, and key enabling technologies include wireless communication, sensors, location tracking (both indoor and outdoor), and augmented reality.
๐ง Quick Revision Questions
- Who coined the term โubiquitous computingโ and at what institution?
- What were the three main device prototypes developed at Xerox PARC, and what were their scale and primary purpose?
- What was the innovation of the IBM Swissair project in 1999?
- How did the RADAR system determine a mobile computerโs location indoors?
- What advantage did Place Lab have over GPS for location-based services?
๐ Lecture 2 โ Wearable Computing?
๐ Overview: This lecture introduces the concepts of wearable and pervasive computing, illustrating how technology can seamlessly integrate into daily life through proactive and self-tuning systems. It distinguishes pervasive computing from its predecessorsโdistributed systems and mobile computingโand outlines the foundational knowledge from distributed systems that supports pervasive computing.
๐๏ธ Topics Covered
The lecture covers the definition and vision of wearable computing, illustrated through two detailed Aura scenarios. It then defines pervasive computing's key ideas of proactivity and self-tuning. The composition of pervasive computing is traced back through distributed systems and mobile computing, and the foundational technical areas from distributed systems are reviewed, including remote communication, fault tolerance, high availability, remote information access, and security.
๐ Lecture Summary
Wearable Computing?
The lecture opens with a quote from Mark Weiser: โThe most profound technologies are those that disappear. They weave themselves into the fabric of everyday life until they are indistinguishable from it.โ This vision is characterized by omnipresence, background execution, accessibility, and embed-ability. Two example scenarios from a system called "Aura" are provided to illustrate this vision.
๐ Definition โ Wearable Computing: Technology that is integrated into everyday life, often with omnipresence, background execution, and accessibility, to the point where it becomes indistinguishable from the user's environment.
Example Scenarios โ Aura
- Scenario 1 (Airport): XYZ is at Gate 23 in the Lahore airport, trying to email large documents over a poor wireless connection. Aura observes the low bandwidth, consults the airport's network weather and flight schedule services, and suggests she move to Gate 15, where bandwidth is excellent and there are no arrivals/departures for 30 minutes. Aura asks her to prioritize her emails, and she walks to Gate 15. Aura informs her when it's almost done, and the last message is transmitted while she walks back, allowing her to board on time.
- Scenario 2 (Meeting): Fred is preparing for a presentation a 10-minute walk away. He is not ready. Aura transfers his work state from his desktop to his handheld computer, allowing him to make final edits via voice commands while walking. Aura infers his destination from his calendar and the campus location service, downloads the presentation to the projection computer, and warms up the projector. As the presentation proceeds, Aura detects unfamiliar faces in the room using face recognition and warns Fred before he displays sensitive budget information, allowing him to skip that slide and end on a high note.
Pervasive Computing
These scenarios embody two key ideas in pervasive computing:
- Proactivity: Combining knowledge from different layers of the system to anticipate user needs (e.g., consulting network and flight schedule services to suggest a better gate).
- Self-Tuning: Automatically adjusting behavior to fit circumstances and effortlessly moving execution state across diverse platforms (e.g., transmitting the last message as XYZ walks, or transferring Fred's work to his handheld).
๐ Definition โ Proactivity: A system's ability to anticipate user needs by combining and reasoning about information from multiple, diverse sources (e.g., network, calendar, location).
๐ Definition โ Self-Tuning: A system's ability to automatically adapt its behavior to the current context and seamlessly migrate computational tasks across different devices and environments.
Composition of Pervasive Computing
Pervasive computing is a major evolutionary step in a line of work dating back to the mid-1970s, with two distinct earlier steps:
- Distributed Systems
- Mobile Computing
Some technical problems in pervasive computing correspond to those studied earlier, where existing solutions may apply directly or require modification. Pervasive computing also introduces entirely new problems with no obvious mapping to previous work.
Distributed Systems
The field of distributed systems arose at the intersection of personal computers and local area networks. Its conceptual framework and algorithmic base are foundational to pervasive computing. The knowledge base spans several critical areas:
- Remote Communication: This includes concepts like protocol layering, remote procedure call (RPC), timeouts, and end-to-end arguments.
- Protocol layering is a technique to simplify networking by dividing it into functional layers, each with its own protocols.
- The end-to-end principle states that communication protocol operations should be defined to occur at the end-points of a system, or as close as possible to the resource being controlled.
- Fault Tolerance: This includes atomic transactions, distributed and nested transactions, and two-phase commit.
- An atomic transaction is a series of database operations that either all occur or none occur.
- A nested transaction occurs when a new transaction is started by an instruction already inside an existing transaction.
- Two-phase commit is a protocol that coordinates all processes in a distributed atomic transaction to decide whether to commit or abort (roll back) the transaction.
- High Availability: This includes optimistic and pessimistic replica control, mirrored execution, and optimistic recovery.
- Pessimistic replication systems try to guarantee from the beginning that all replicas are identical to each other.
- In optimistic replication, replicas are guaranteed to converge under certain conditions.
- Optimistic Recovery is a technique for transparent recovery from processor failures in distributed systems.
- Remote Information Access: This includes caching, function shipping, distributed file systems, and distributed databases. This makes it possible for multiple users on multiple machines to share files and storage resources over a network.
- Security: This includes encryption-based mutual authentication and privacy.
- Mutual authentication is a security feature where a client must prove its identity to a server, and the server must prove its identity to the client, before any application traffic is sent.
๐ Definition โ Distributed Systems: A field of computing that deals with multiple computers connected by a network, providing a conceptual framework for remote communication, fault tolerance, and resource sharing, which is foundational to pervasive computing.
โญ Key Takeaways
The lectureโs core vision is that pervasive computing makes technology "disappear" into the fabric of daily life through proactivity (anticipating needs by combining knowledge) and self-tuning (automatically adapting to context and moving state across devices). The Aura scenarios are crucial examples of these principles in action. Pervasive computing is an evolutionary step from distributed systems and mobile computing, and it inherits foundational problems and solutions from distributed systems. Key foundational areas from distributed systems that a student must understand include remote communication (e.g., protocol layering, end-to-end arguments), fault tolerance (e.g., atomic transactions, two-phase commit), high availability (e.g., optimistic vs. pessimistic replication), remote information access, and security (e.g., mutual authentication).
๐ง Quick Revision Questions
- What are the four characteristics of a "disappearing" technology, according to the definition of wearable computing?
- Describe two specific examples of self-tuning from the Aura airport and meeting scenarios.
- What is the primary difference between proactivity and self-tuning in the context of pervasive computing?
- What are the two distinct earlier steps in the evolution that led to pervasive computing?
- Explain the end-to-end principle as it applies to remote communication in distributed systems.
๐ Lecture 3 โ Mobile Computing
๐ Overview: This lecture introduces the fundamental constraints and research areas of mobile computing, focusing on how traditional distributed system principles are adapted to handle mobility. It covers key challenges like network variability, resource limitations, and energy conservation, along with enabling technologies for mobile networking, information access, and location-aware systems.
๐๏ธ Topics Covered
This lecture covers the emergence of mobile computing with laptops and wireless LANs in the 1990s, the four key constraints of mobility (unpredictable network quality, lowered trust, resource limitations, battery power), and six active research areas: mobile networking (Mobile IP, IPv6, ad hoc protocols), mobile information access (disconnected operation, bandwidth-adaptive file access, data consistency), adaptive applications (transcoding by proxies), energy-saving techniques (energy-aware adaptation, variable-speed processor scheduling), and location sensitivity (location sensing, location-aware behavior).
๐ Lecture Summary
Mobile Computing
The appearance of full-function laptop computers and wireless LANs in the early 1990s marked the beginning of mobile computing, which is essentially a distributed system with mobile clients. Although many basic principles of distributed system design continued to apply, four key constraints of mobility forced the development of specialized techniques:
- unpredictable variation in network quality
- lowered trust and robustness of mobile elements
- limitations on local resources imposed by weight and size constraints
- concern for battery power consumption
Mobile computing is a very active and evolving field of research. Key areas include:
-
Mobile networking, including Mobile IP, IPv6, ad hoc protocols, and techniques for improving TCP performance in wireless networks. The IP address of a node consists of two portions: a network identifier (network ID) and a host identifier (host ID). The network ID specifies which network a host is on, and the host ID uniquely specifies hosts within a network.
-
Mobile information access, including disconnected operation, bandwidth-adaptive file access, and selective control of data consistency. Data consistency summarizes the validity, accuracy, usability and integrity of related data between applications and across the IT enterprise.
-
Support for adaptive applications, including transcoding by proxies and adaptive resource management. Application adaptation is a means by which applications can respond to a changing operating environment. Transcoding proxies are used as intermediaries between the generic World Wide Web and a variety of client services in order to adapt to greatly varying bandwidths of different client communication links and handle the heterogeneity of possible small screen devices.
-
System-level energy saving techniques, such as energy-aware adaptation, variable-speed processor scheduling, and energy-sensitive memory management. This involves how applications can dynamically modify their behavior to conserve energy.
-
Location sensitivity, including location sensing and location-aware system behavior.
๐ Definition โ Mobile Computing: A distributed system with mobile clients where specialized techniques address constraints of unpredictable network quality, lowered trust, limited local resources, and battery power consumption.
๐ Definition โ Data Consistency: The validity, accuracy, usability, and integrity of related data between applications and across the IT enterprise.
๐ Definition โ Application Adaptation: A means by which applications can respond to a changing operating environment.
๐ Definition โ Transcoding Proxy: An intermediary between the generic World Wide Web and client services that adapts content for different bandwidths and small screen devices.
๐ Concept: IP Address Structure โ An IP address consists of a network identifier (network ID) specifying which network a host is on, and a host identifier (host ID) uniquely specifying hosts within that network.
โญ Key Takeaways
The four key constraints of mobilityโunpredictable network quality, lowered trust, resource limitations, and battery powerโrequire specialized techniques beyond standard distributed system design. Mobile networking relies on Mobile IP, IPv6, and ad hoc protocols, with IP addresses split into network and host identifiers. Data consistency is crucial for mobile information access, ensuring validity across applications. Adaptive applications use transcoding proxies to handle varying bandwidths and device heterogeneity, while energy-aware adaptation and location sensitivity are critical for system performance and context-aware behavior.
๐ก Why this matters: These constraints and techniques form the foundation for all modern mobile and pervasive computing systems, from smartphones to IoT devices.
๐ง Quick Revision Questions
- What are the four key constraints of mobility that forced the development of specialized techniques in mobile computing?
- How is an IP address structured, and what is the role of the network identifier vs. the host identifier?
- What is data consistency and why is it important in mobile information access?
- What is the purpose of a transcoding proxy in adaptive applications?
- Name two system-level energy saving techniques mentioned in the lecture.
๐ Lecture 4 โ Pervasive Computing
๐ Overview: This lecture defines pervasive computing as technology that seamlessly integrates into daily life to the point of becoming invisible. It explores five key research challenges and examines how pervasive computing subsumes mobile computing while introducing unique problems that require new solutions.
๐๏ธ Topics Covered
The lecture covers the definition of pervasive computing and its relationship to mobile computing, five fundamental research challenges: Effective Use of Smart Spaces, Invisibility, Localized Scalability, and Masking Uneven Conditions. Finally, it discusses the composition of pervasive computing problems, including those carried forward from earlier computing paradigms and entirely new challenges.
๐ Lecture Summary
Pervasive Computing
Pervasive computing is described as an environment "saturated with computing and communication capability, yet so gracefully integrated with users that it becomes a 'technology that disappears.'" The agenda of pervasive computing subsumes that of mobile computing, but goes much further in its scope and ambition.
The lecture identifies five major research challenges for pervasive computing:
Effective Use of Smart Spaces: A smart space brings together two worlds that have been disjoint until now: the computing infrastructure and the building infrastructure. This integration allows the physical environment to respond intelligently to users.
Invisibility: The ideal is the complete disappearance of pervasive computing technology from a userโs consciousness. In practice, a reasonable approximation is minimal user distraction. If a pervasive computing environment continuously meets user expectations and rarely presents surprises, it allows interaction at almost a subconscious level. ๐ก Why this matters: Invisibility is the ultimate goal โ technology should serve without demanding attention.
Localized Scalability: As smart spaces grow in sophistication, the intensity of interactions between a userโs personal computing space and surroundings increases. This has severe bandwidth, energy, and distraction implications for a wireless mobile user. The presence of multiple users further complicates this problem. Scalability, in the broadest sense, is thus a critical problem. Like the inverse square laws of nature, good system design must achieve scalability by severely reducing interactions between distant entities.
Masking Uneven Conditions: There exist huge differences in the "smartness" of different environments. The large dynamic range of "smartness" can be jarring to a user, detracting from the goal of making pervasive computing technology invisible. ๐ก Why this matters: Users move between highly smart and minimally smart spaces; the system must mask these transitions to maintain seamless experience.
Composition of Pervasive Computing
Some technical problems in pervasive computing correspond to problems already identified and studied earlier in the evolution of computing. In some of those cases, existing solutions apply directly; in other cases, the demands of pervasive computing are sufficiently different that new solutions must be sought. There are also new problems introduced by pervasive computing that have no obvious mapping to problems studied earlier.
๐ Definition โ Pervasive Computing: An environment saturated with computing and communication capability, yet so gracefully integrated with users that it becomes a "technology that disappears."
๐ Definition โ Smart Space: A space that brings together computing infrastructure and building infrastructure into an integrated, responsive environment.
๐ Definition โ Localized Scalability: The design principle that good system design must achieve scalability by severely reducing interactions between distant entities, similar to inverse square laws in nature.
โญ Key Takeaways
The most critical concept is that pervasive computing goes beyond mobile computing by aiming for complete invisibility โ technology that disappears from user consciousness. Students must understand the five research challenges: Effective Use of Smart Spaces integrates computing and building infrastructures; Invisibility requires minimal user distraction; Localized Scalability demands reducing distant interactions to manage bandwidth and energy; and Masking Uneven Conditions hides differences in environment smartness from users. The composition of pervasive computing problems includes three categories: problems with direct existing solutions, problems requiring modified solutions due to unique demands, and entirely new problems without precedent.
๐ง Quick Revision Questions
- What is the fundamental difference between the agenda of pervasive computing and that of mobile computing?
- Define "invisibility" in the context of pervasive computing and explain its practical approximation.
- Why is "localized scalability" a critical problem in pervasive computing, and what natural principle does it resemble?
- What are the three categories of technical problems in pervasive computing based on their relationship to problems studied earlier?
- What two infrastructures does a "smart space" bring together?
๐ Lecture 5 โ Composition of Pervasive Computing
๐ Overview: This lecture explores the technical composition of pervasive computing by examining problems inherited from earlier computing paradigms and new challenges unique to pervasive environments. It provides a structured overview of six key domainsโsensing, operating systems, computer architecture, software engineering, security, and human-computer interactionโand explains how mobility and context-awareness drive pervasive application development.
๐๏ธ Topics Covered
The lecture covers six main technical areas of pervasive computing: sensing and actuating (including location sensing, activity inference, robotics, and ad-hoc networks), operating systems (for small devices, power management, application-aware adaptation, and disconnected operation), computer architecture (wearable computers, low-power hardware, and rapid prototyping), software engineering (self-healing systems, dynamic reconfiguration, and economic models), security and privacy (location privacy, low-power encryption, and biometric authentication), and human-computer interaction (context awareness, seamlessness, and multimodal I/O). It also discusses the two fundamental characteristics of pervasive applications: mobility and context-awareness.
๐ Lecture Summary
Composition of Pervasive Computing
The technical problems in pervasive computing fall into three categories: problems with existing solutions that apply directly, problems where pervasive computing demands new solutions, and entirely new problems unique to pervasive computing. The lecture organizes these challenges into six key areas.
Sensing and Actuating โ This area covers how pervasive systems perceive and interact with the physical world. Key sub-problems include location sensing (determining where devices or users are), activity inference (deducing what users are doing), robotics (autonomous physical agents), and ad-hoc networks (self-organizing wireless networks without fixed infrastructure).
Operating Systems โ Pervasive computing requires specialized operating systems for resource-constrained devices. Key challenges include operating systems of small things (OS designed for tiny embedded devices), power management (conserving battery life), application-aware adaptation (applications adjusting behavior based on resource availability), and disconnected operation (functioning without network connectivity).
Computer Architecture โ Hardware design must adapt to pervasive computing requirements. This includes wearable computers (computing devices worn on the body), low-power hardware (energy-efficient components), and rapid prototyping (quickly building and testing new hardware designs).
Software Engineering โ Building reliable pervasive software requires new approaches. Key areas are self-healing systems (software that automatically recovers from failures), dynamic reconfiguration (capability to change system behavior without restarting), and economic models (market-based approaches to manage resource allocation).
๐ Definition โ Self-healing systems: Software systems that can automatically detect, diagnose, and recover from failures without human intervention.
Security and Privacy โ Pervasive environments raise unique security concerns. Key challenges include location privacy (protecting information about where users are), low-power encryption (security algorithms efficient enough for constrained devices), and biometric authentication (identifying users using physical characteristics like fingerprints or facial recognition).
๐ Definition โ Location privacy: The ability to prevent unauthorized parties from knowing a user's current or past physical location.
Human Computer Interaction โ Designing how users interact with pervasive systems. This covers context awareness (systems that understand and respond to user context), seamlessness (smooth transitions between devices and environments), and multimodal I/O (multiple input/output methods like voice, touch, and gesture).
Pervasive Application Development
The vision of pervasive computing leads to two fundamental characteristics of pervasive applications: mobility and context-awareness. Both characteristics result from the extremely dynamic nature of pervasive computing environments.
Mobility has three implications. First, applications must run on a wide variety of devices, including devices embedded in environments and devices carried by users. Second, because devices may be transported to locations where a high-bandwidth network connection is not available, applications must work (perhaps in a degraded mode) with low-bandwidth network connections or in the absence of any network connection. Third, applications that make use of a user's location must account for the possibility that the location will change.
๐ Definition โ Context-awareness: The ability of a system to sense, interpret, and respond to aspects of the user's environment, such as location, time, activity, and nearby people or devices.
๐ก Why this matters: The mobility and context-awareness characteristics explain why pervasive applications cannot simply be ported from traditional desktop environmentsโthey require fundamentally different design approaches that account for device diversity, variable connectivity, and changing user context.
โญ Key Takeaways
For the exam, you must understand that pervasive computing technical problems map to six domains: sensing/actuating, operating systems, computer architecture, software engineering, security/privacy, and HCI. Each domain has specific sub-problemsโfor example, location sensing, power management, wearable computers, self-healing systems, location privacy, and context awareness. You must remember that pervasive applications have two fundamental characteristics: mobility (requiring multi-device support, disconnected operation, and location-change handling) and context-awareness (systems responding to dynamic environments). The lecture emphasizes that some problems have existing solutions, some need new pervasive-specific solutions, and others are entirely new. Finally, mobility's three implicationsโdevice variety, degraded-mode operation, and location-change handlingโare critical for application development.
๐ง Quick Revision Questions
-
List the six major technical areas of pervasive computing covered in this lecture.
-
What are the three categories of technical problems in pervasive computing based on how they relate to earlier computing paradigms?
-
What are the two fundamental characteristics of pervasive applications, and what causes them?
-
Name the three implications of mobility for pervasive application development.
-
What is the difference between a problem that has "existing solutions apply directly" versus one where "demands are sufficiently different that new solutions have to be sought"?
๐ Lecture 6 โ Pervasive Application Development
๐ Overview: This lecture defines and explores the key characteristics and development challenges of pervasive applications, including context-awareness, multi-device support, and multimodal interfaces. It establishes the fundamental definitions and distinctions needed to understand how applications adapt to different devices, users, and environmental contexts in pervasive computing.
๐๏ธ Topics Covered
The lecture covers definitions of context-aware applications and how they sense and adapt to their environment, followed by explanations of multi-device and multimodal applications. It distinguishes between thin-client and thick-client architectures, introduces the MVC (Model-View-Controller) application structure, explains disconnectable applications, and describes application context attributes including user location, destination, and task characteristics.
๐ Lecture Summary
Pervasive Application Development โ Core Concepts
A context-aware application is one that is sensitive to the environment in which it is being used, such as the location or particular user of the application. The application can use this information to customize itself to the particular location or user. This implies three technological requirements: identifying and binding to data sources that provide the right information, composing the information from these sources to create information that is useful for an application, and using that information in meaningful ways within the application itself.
๐ Definition โ Context-aware application: An application that is sensitive to the environment in which it is being used and can customize itself to the particular location or user.
๐ Example: A navigation app that detects a user is in a car and automatically switches to a simplified, large-button interface for driving.
๐ก Why this matters: Context-awareness is what distinguishes pervasive applications from conventional ones โ it allows the application to proactively adapt rather than requiring manual user configuration.
Some Definitions
A multi-device application is one that is able to execute on devices with different capabilities. A multimodal application is one that supports multiple user interface modalities such as GUI, voice, and a combination of the two. The lecture considers only networked applications because they represent the bulk of interesting and useful pervasive applications. A device platform is the distributed software platform to which a pervasive application is targeted. A thin-client application is a networked application in which the user interface rendering component is executing on the userโs device, whereas the rest of the application is executing on a networked computer. A thick-client application, on the other hand, has significant application components executing on the userโs device.
๐ Definition โ Thin-client application: A networked application where only the user interface rendering runs on the user's device, with the rest executing on a networked computer. ๐ Definition โ Thick-client application: A networked application where significant application components execute on the user's device.
MVC Application Structure
The well-known MVC application structure separates application components into Model, View, and Controller. The view represents the presentation, and the controller represents the application flow, including the navigation, validation, error handling, and event handling. The view and the controller together deal with the user interaction of the application. The model component includes the application logic as well as the data underlying the application logic.
๐ Definition โ MVC (Model-View-Controller): An application structure where the Model handles application logic and data, the View handles presentation, and the Controller handles application flow including navigation, validation, error handling, and event handling.
Disconnectable Applications and Context
A disconnectable application is one that is able to continue to execute when there are different levels of connectivity between the different components of the application. The attributes of the environment of an application are referred to as the context of the application. The context of an application includes some of the userโs significant attributes, such as location, destination, the identities of other people in the vicinity, and the attributes of the task being performed, such as the objective and the artifacts necessary for the task. A context-aware application is one that is able to sense some aspects of the environment in which it is executing and adapt its behavior to the sensed environment.
๐ Definition โ Disconnectable application: An application that can continue to execute under different levels of connectivity between its components. ๐ Definition โ Context: The attributes of the environment of an application, including location, destination, identities of nearby people, and task attributes. ๐ Definition โ Context-aware application: An application that can sense aspects of its execution environment and adapt its behavior accordingly.
Why Pervasive Application Development Is More Difficult
There are fundamental reasons why pervasive application development is more difficult than conventional application development. End user devices, such as smart phones and PDAs, come in many varieties and have widely varying capabilities, both hardware (form factor, user interface hardware, processor, memory, and network bandwidth) and software (operating system, user interface software, services, and applications).
๐ก Why this matters: This heterogeneity of device capabilities is the core challenge in pervasive computing โ developers cannot assume a uniform platform and must design for varying constraints simultaneously.
โญ Key Takeaways
The lecture establishes that pervasive applications must be context-aware, meaning they sense and adapt to user location, task, and environment. Applications can be classified as thin-client (UI only on device) or thick-client (significant processing on device), and the MVC structure separates presentation/flow from logic/data. Disconnectable applications must function despite varying network connectivity. The fundamental difficulty in pervasive computing stems from the extreme heterogeneity of end-user devices in terms of both hardware and software capabilities.
๐ง Quick Revision Questions
- What are the three technological requirements for building a context-aware application?
- What is the difference between a thin-client application and a thick-client application?
- In the MVC structure, which components handle user interaction and which handle application logic and data?
- What attributes of a user's environment are considered part of the "context" of an application?
- Why is pervasive application development more difficult than conventional application development?