This post is the reading notes of the following paper:

Stan Kurkovsky, Pervasive computing: Past, present, and future, The 5-th International Conference on Information and Communication Technology (ICICT 2007), 2007: 65-71.

## Abstract

The paradigm of pervasive computing describes ubiquitous computing environments that provide anytime and anywhere access to information services while making the presence of the system invisible to the user. Pervasive computing envisioned by Mark Weiser emerged at the conjunction of research and development in a number of areas which include embedded and devices and systems, wireless communications, and distributed, mobile and context-aware computing.

## I. INTRODUCTION

The fundamental properties of a system comprised of tabs, pads and boards described by Weiser include wireless communications, embedded and mobile devices, distributed computing, and context awareness.

As in any emerging research area, there are many challenging problems in pervasive computing. One of the most important and open questions is how to ensure that a computing system is seamlessly and invisibly embedded in the environment and how to minimize the possible impact of its intrusiveness on a user’s perception.

Generally, there are two mutually complementing approaches to solving the problem of unobtrusiveness of a pervasive system: by miniaturization of devices and embedding of the system’s logic into wearable, handheld, and mobile devices, as well as into the environment, and by achieving a level of intelligence of the system that will be able to anticipate the actions of the user in the context of the factors in the environment.

As a result, such a pervasive system will “fade into the background” and the users will only need to interact with such a system at their leisure without giving it much thought [14,23,28,30,32].

## II. COMPONENTS OF PERVASIVE COMPUTING SYSTEMS

In addition to making the system physically invisible, the concept of pervasive computing provides for making systems logically invisible.

Such a logical invisibility can be achieved if the system can adjust itself to the behavioral patterns exhibited by its user and to the factors of the surrounding environment, which include time, location, other users, objects and systems. Such capabilities can be achieved through context-awareness, which is a necessary component in providing the system’s logical invisibility to complement its physical invisibility ensured by device embedding [28].

Pervasive computing systems comprise of a large number of sensors, devices and mobile systems; Software architectures that may be better suited to such an environment requires that the application logic be decomposed into atomic functions, which are distributed across ubiquitous processing, storage and input/output resources and that it be capable of reassembling in varying combinations as may be needed by a particular task.

Distributed systems generally include components playing one of three possible roles: resources, resource managers, and clients [3].

Every resource within the distributed system has a unique identity, independent of the resource location.

A distributed system is said to provide transparency of migration if it supports mobility of its resources.

Context-awareness is a vital factor enabling the invisibility of pervasive computing systems [3,10,28,31]. Context-aware systems can adjust their properties and behavior according to the information about the current state of the user, such as physiological state, behavioral patterns, orientation and position, geographical location, or properties of the surrounding environment, such as time of day, nearby users, objects and other systems [6,15,26,28]

For example, based on locality and availability, a context-aware service with limited resources may be able to off-load processor intensive operations or storage onto other systems located nearby that have more substantial resources. A vehicle navigation system that is aware of the remaining level of gasoline in the tank and the brand and type of fuel preferred by the user, may alert the driver when he or she is within certain proximity of a gas station selling that type of fuel at an acceptable price.

Invisibility in a pervasive computing system depends on how successfully it implements a certain level of intelligence that would enable it to allow the user to focus on a particular task at hand rather on interacting with the system itself [12]. Context-aware systems utilizing this type of intelligence often calculate the intent of the user based on a number of contexts. As a result, context-awareness helps eliminate the bottleneck of the human attention by anticipating the user’s reactions to relevant events [9].

1. Pervasive computing systems need to anticipate the actions of the user and adjust themselves while maintaining a careful a balance between their proactivity and invisibility of the system.

2. The ability of the system to adapt to changing factors in its environment is another important consideration of preserving the system’s invisibility.

3. Finally, pervasive computing infrastructure could dynamically notify the user about the changes in resource availability and offer a choice of corrective actions.

For example, suppose a user is viewing a video stream on a mobile wireless device [20].

## III. MOBILE PHONES AND PERVASIVE COMPUTING

Personalization is typically achieved by implementing different levels of context-awareness, which vary from rudimentary setting of user preferences [27] to intelligent profile matching [34].

## IV. SUMMARY

As Weiser wrote [32], “tabs, pads, and boards are just the beginning [of pervasive computing systems]. The real power of the concept comes not from any one of these devices; it emerges from the interaction of all of them.”

## BibTeX of the Paper

@inproceedings{kurkovsky2007pervasive,
title = {Pervasive computing: Past, present, and future},
author = {Stan Kurkovsky},
year = {2007},
month = {December},
booktitle = {The 5-th International Conference on Information and Communication Technology (ICICT 2007)},
pages = {65--71},
}