=Paper=
{{Paper
|id=Vol-1138/poster1
|storemode=property
|title=Requirements for Context-Aware Systems
|pdfUrl=https://ceur-ws.org/Vol-1138/poster1.pdf
|volume=Vol-1138
|dblpUrl=https://dblp.org/rec/conf/refsq/EvansBA14
}}
==Requirements for Context-Aware Systems==
https://ceur-ws.org/Vol-1138/poster1.pdf
Requirements for Context-Aware Systems
Carl Evans, Lindsey Brodie, Juan Carlos Augusto
Research Group on the Development of Intelligent Environments
Department of Computer Science
Middlesex University
London, United Kingdom
C.Evans@mdx.ac.uk
Abstract. Originating from ubiquitous computing and augmented by the rapid onset
of mobile technology, the field of intelligent environments (IE) is maturing to a level
at which a range of sophisticated applications are emerging. By its very nature, an IE
typically comprises a fusion of sensors, networks, intelligent software and end-user
interfaces with intrinsically complex interoperability [1]. Such systems aim to be
context-aware, especially being adaptable to dynamic circumstances, and possibly
unpredictable circumstances. Deploying a reliable IE system can be critical. For in-
stance, in the case of a pervasive system designed to facilitate ambient assisted living
(AAL), in which the end-user could be vulnerable to some degree, a system failure
could have disastrous consequences. Recently, it has been acknowledged that numer-
ous IE systems have been developed without adopting best practices in software engi-
neering. The reality is that the development of such a promising area is so far guided
by ad hoc and disconnected initiatives and there is an increasing acceptance that IE
researchers and system developers share a responsibility to design holistically safer,
more efficient, and more user-customised systems. However, engagement with core
software engineering principles has, to date, been quite limited within the IE commu-
nity [2]. Our work focuses on the requirements engineering stage and presents a
framework for IE systems, in which an intrinsic component is context-awareness. Our
study of previous work in this area has highlighted a number of prominent themes
which provide focus for our research. In particular, we are currently working as part
of the POSEIDON1 consortium towards a distributed technological infrastructure to
foster the development of services (based on both static and mobile smart environ-
ments) which can support people with Down’s Syndrome [3]. Aspects of the
POSEIDON project have some affinity with the work by Sutcliffe et al. [4]. Each
incorporates a higher-order goal of ‘empowerment’ of an individual with disability,
recognises the benefit of scenario-based elicitation techniques, and identifies a de-
marcation between individual user requirements which demand a degree of personali-
sation, or customisation, and distinct user groups that support an individual who also
present distinct sets of requirements. In particular, we nominate five requirements
stakeholder categories as part of our framework. Another salient feature of our work
is the introduction of a core ethical model. Previous work has addressed the issues of
1
http://www.poseidon-project.org
social context (as an aspect of geographical context), cultural issues and ethnicity. We
have enhanced this idea with the adoption of a set of ethical principles within the
overall framework that also considers the important issue of privacy. More specifical-
ly, within our research group we have created an ethical framework based on services
which collectively are consistent with a number of higher level ethical principles
which are designed to protect users from informal and rushed system development.
The framework for context-aware systems that we propose can be described as a form
of workflow model, whereby we have identified a series of core requirements gather-
ing categories (establish high-level objectives and scope, identify tasks and perfor-
mance qualities, determine stakeholder profiles for context users) against which dedi-
cated requirements elicitation activities proceed, which are complemented by two
further activities that play a vital supporting role (operational support and harmonisa-
tion). Central to the process is a determination of stakeholder profiles for those users
who require context data capture (not all stakeholders will require context profiles).
This phase comprises a number of sub-activities which include the use of psychologi-
cal questionnaires, scenario-based techniques and the application of the ethical
framework that is mentioned above. Whilst our requirements engineering framework
targets context-aware systems in general and our intention is to apply it to a range of
projects, the first application of the model has commenced with the POSEIDON
project. The POSEIDON project is in its early stages. In terms of our framework,
high-order objectives and scope have been established having been considered at
early stages through the definition of the project and subsequent interaction with the
European Commission (who are funding the project). The POSEIDON consortium
has designed online questionnaires which have been filled in by more than 300 users.
Approximately 30 families have been interviewed, and a user-focused workshop
event has taken place. It is envisaged that this real-world application of the framework
will endorse its conception and facilitate refinement of the model.
References
1. Augusto JC, Callaghan V, Cook D, Kameas A, Satoh I (2013) Intelligent Environments: a
manifesto. Human-centric Computing and Information Sciences, 3:12. doi: 10.1186/2192-
1962-3-12.
2. Preuveneers D, Novais P (2012) A survey of software engineering best practices for the de-
velopment of smart applications in Ambient Intelligence. Journal of Ambient Intelligence
and Smart Environments, 4:3 pp 149-162. doi: 10.3233/AIS-2012-0150.
3. Augusto JC, Grimstad T, Wichert R, Schulze E, Braun A, Rodevand GM, Ridley V (2013)
Personalized Smart Environments to Increase Inclusion of People with Down’s Syndrome.
In proceedings of the 4th International Joint Conference on Ambient Intelligence, pp 223-
228.
4. Sutcliffe A, Fickas S, Sohlberg M.M, (2006) PC-RE: a method for personal and contextual
requirements engineering with some experience. Requirements Engineering 11 (3) pp 157-
173. doi: 10.1007/s00766-006-0030-0.