Book proposal
MOBILE INFORMATION SYSTEMS – INFRASTRUCTURE AND DESIGN FOR
ADAPTIVITY AND FLEXIBILITY
(the MAIS approach)
B. Pernici, Politecnico di Milano (ed.)
Approx 300-350 pages
Target
Industrial research community in multichannel and mobile information systems.
The goal is to provide mobile IS designers and researchers in an industrial setting
contributions to current research questions, offering not an introductory book, but a systemic
vision on innovative aspects, which can be adopted as a basis of single projects or the
development of new frameworks.
Short description of the contents
The current fast evolution in the ICT world suggests that it will be possible to offer access to
information systems, in a ubiquitous way, through various kinds of interaction devices (PCs,
laptops, palmtops, cellular phones, TV sets, and so on).
The MAIS project main research goal is to provide a flexible environment to adapt the
interaction and provided information and services according to ever changing requirements,
execution contexts, and user needs. The focus of the project is adaptivity in general, and its
implications both at design and at infrastructural level.
The MAIS project is a three year project, ending in November 2005, funded by Basic
Research Funds (FIRB Program) of the Italian Department of Education (MIUR).
The project approaches the main research theme from a multi-disciplinary perspective,
coordinated within a general framework defined in the project. The involved research areas
are information systems, database systems, human computer interaction, computer networks
and telecommunication, hardware design, middleware, management engineering.
The goal of the book is to present the reference models, new architectures, and prototypes to
validate the theoretical proposals developed in the project.
The project involves research teams from six Italian Universities, two research centers, and
two industries.
The book will present a general reference model for multichannel adaptive information
systems developed in the project, to allow integration and relating requirements and technical
solutions emerging from the above mentioned areas, in particular for mobile information
systems (MobIS).
The first chapters of the book will discuss the design requirements and illustrates a general
framework for Mobile Information Systems. The book will present Mobile Information
Systems as cooperative systems based on a service-oriented approach, in which the service
oriented paradigm is extended with tools to improve flexibility, such as dynamic invocation,
and an extended UDDI registry with an e-service ontology and quality of service information.
Context description and management based on a reflective architectural paradigm will be
presented.
The second part of the book will contain chapters describing enabling technologies
constituting the technological infrastructure for multichannel, and in particular mobile,
information systems: wireless networks, data management in micro-devices, hardware
solutions.
The third part of the book will discuss design of mobile information systems and present
demonstrators developed in the project in different application domains. The prototype
applications are being developed in the areas of tourism, university and education, and risk
management in archeology.
Project web site: http://www.mais-project.it
Table of Contents
Part I – Core technologies for Mobile Information Systems 3 intro - 110
1. Basic concepts – with MAIS intro 25
2. Reference Framework functional, architectural 20
3. Reference Framework: context (with user profile) and quality 20
4. E-services 30
5. Reflective middleware 30
Part II – Enabling technologies 3pp intro - 75
6. Adaptive networks 30
7. Data management on microdevices 20
8. Low power architectures 25
Part III – Mobile Information Systems design 3pp intro- 110
9. Design Methodologies and tools: front end design 30
10. Design Methodologies and tools: back end design 20
11. Knowledge-based tools for e-service profiling and mining 20
12. Applications 30
Appendices
Quality registry 4
Classification of functions 3
Glossary 4
Bibliography
Description of chapters
PART I - CORE TECHNOLOGIES FOR MOBILE INFORMATION SYSTEMS
1. Basic concepts (B. Pernici) 30 pag.
The chapter presents mobile information systems in general, research and methodological
issues and provide a general introduction to the project.
Basic concepts will be introduced.
2. Reference Framework (editor B. Pernici, coauthors V. De Antonellis et al.) 30 pag.
The reference framework developed in the project is presented: the framework is based on
a Service Oriented Architecture, and it allows defining services, channels, user
characteristics, from the different points of view of the project, from modeling and
evaluating e-services to be used in mobile cooperative environments, down to specific
requirements of hardware components to be adopted in mobile environments. An
introduction to the quality of service model developed in detail in Part III is provided.
The reference framework is represented as a UML schema, providing an integrated view
to all concepts and terms used in the following chapters.
The reference framework provides an innovative approach to service oriented
development and design of mobile information systems, defining all terms and providing
an integrated perspective.
3. E-services (editor B. Pernici, coauthors D. Bianchini, A. Maurino, M. Mecella, P.
Plebani, D. Presenza et al.) 30 pag.
In MAIS, starting from well-accepted standards such as WSDL, BPEL, and UDDI, we
define new modular and interoperable architectures supporting the definition of added
values flexible and adaptive Web Services including new dimensions such as QoS,
behaviour, and semantic description. New languages and models presented in the book
constitute a solid theoretical background supporting our architecture.
Two service execution environments are considered:
Web Services in a service provider context, where web-services are coordinated by
an orchestrator and selected on the basis of their functional properties and quality
of service and on the basis of a variable user’s execution context;
orchestration of services in ad hoc networks and small portable devices, such as
handhelds with constrained execution environments and variable configuration of
the network.
A rigorous methodological approach is also defined to help the designer in the definition
of added value Services, and providers in the publication of their Web Services.
In particular, adaptivity can impact on costs and benefits in several ways. On one hand, it
is likely to increase design costs, since systems should embed additional functionalities to
adapt to varying requirements in real time. On the other hand, it can decrease management
costs by simplifying system operations through the automatic satisfaction of changes in
requirements and related exceptions. Furthermore, the ease with which services can be
composed and replaced can significantly modify the distribution of benefits among
cooperating organizations and lead to different and unconventional pricing mechanisms.
The following issues are specifically addressed:
A model of infrastructural costs is defined to analyze the impact of the
dynamic allocation of computing and communication capacity on
infrastructural costs and corresponding revenues from SLA agreements.
A pricing model is defined based on the economic coalition approach, adapted
to support the principles of a Vickrey combinatory auction.
4. Reflective middleware (editor F. Tisato, Univ. Milano Bicocca, coauthors M. Adorni,
F. Arcelli et al.) 30 pag.
The reflective middleware environment supports the construction of services which
can adapt themselves to the context, where “context” denotes, on one side, the user
requirements and, on the other side, the features and constraints of the technological
platforms.
A significant approach, developed in the project, relies on the concept of
“Computational Reflection” and, in particular, of “Architectural Reflection” in an
Object-oriented context. The key idea is that the system resources (interaction devices,
computing resources, services, components of the communication subsystem) are
modelled by meta-objects which expose, at a suitable level of abstraction, their QoS-
related features. The applications observe and manipulate not only objects modelling
the application domain, which are exploited to fulfill functional requirements, but also
meta-objects modelling the system itself, which are exploited to fulfill non-functional
requirements.
PART II – ENABLING TECHNOLOGIES
5. Adaptive networks (editor P. Giacomazzi, Politecnico di Milano, coauthors) 30 pag.
The bursty nature of packet traffic makes it difficult to plan and manage a multiservice
network capable of providing Quality of Service to end users and be efficient at the same
time.
The modern communication networks must reach a trade-off between the reservation of
resources, which is necessary to provide traffic flows with the required QoS even in case
of congestion, and the statistical sharing of resources among different services and
different users. The network must also be adaptive to cope with heterogeneous services
and mutable requirements.
In MAIS, we are investigating how statistical models can provide the required QoS to
applications with high probability and also achieve high network efficiency. This is made
possible by using suitable Connection Admission Control algorithms, to limit the amount
of traffic in the network and suitable scheduling disciplines, to provide QoS to sources
with different requirements.
The approach adopted has been to maintain the traditional model proposed by the
ISO/OSI protocol stack for the lower protocol layers: physical, data link and network.
For each layer of the protocol stack, some main research topics will be discussed:
Physical layer (PHY): the focus will be on a novel architecture for an adaptive
wireless modem based on the OFDM transmission technique, which is one of the main
candidate technologies envisioned for 4G wireless systems; in particular, we focused
on the dynamic, joint optimization of modulation and channel coding with the
objective of minimizing the transmit power;
Data link layer (DLC/MAC): a novel MAC technique for vehicular ad-hoc networks,
i.e. a very challenging environment for adaptive wireless networks, is discussed;
different logics to enhance the flexibility of today’s main DLC/MAC protocols, IEEE
802.11 (WiFi) and IEEE 802.16 (WiMax) are also proposed and discussed;
Network layer (NET): new methodologies and tools for monitoring, discovering and
inferring inter-domain network configurations and relationships between IPv6 and
IPv4 networks are illustrated; edge-to-edge delay of traffic flows in a Differentiated
Services Network domain estimation and Connection Admission and Control are
discussed.
6. Data management on microdevices (editors F.A.Schreiber, L. Tanca, Politecnico di
Milano, coauthors C. Bolchini, Politecnico di Milano) 30 pag.
The use of handheld devices, such as Smart Cards, Portable Data Assistants (PDA), Palm
PCs and Cell Phones, to store data locally and to issue transactions against both local and
remote data of some Information Systems is discussed in this chapter. Features required
by portable devices in order to manage data are, for some aspects, similar to those found
in Embedded Database systems, and range from very simple file system functions to a full
set of database management capabilities, including some ACID transactions properties.
Databases for very small devices – henceforth called Very Small Data Bases (VSDB) - are
useful in various circumstances, among which personal (micro)information systems, the
so-called citizen’s cards, personal medical record, personal financial database, i.e. the
“stock portfolio”, storing the owner’s financial assets along with the interesting
information about the stock and bond market, the personal travel database, recording all
the travel (e.g. tourist) information considered interesting by the device owner.
The emerging field of Intelligent Sensors Networks adds a new dimension to the problem
of Data Management. Here, the problems of small (flash) secondary memory dimensions
and of Low Power operation by far prevail on the classical transactional requirements and
ask for novel data management techniques both at the physical storage level and at data
collection and transmission times.
The methodology we propose for data design on microdevices shares many issues with
the methodologies for distributed/ federated database design. However, three main
differences w.r.t. the traditional design methodologies are introduced:
the peculiarities of the storage device(s) must be taken into account from the early
steps, thus a logistic phase is added after the usual conceptual and logical phases,
which supports the designer in the physical design task, by taking into account the
logistic aspects of data storage;
since most interesting micro-devices are portable and/or can be disseminated on an a-
priori undefined area, the main mobility issues are to be considered along with data
distribution;
context awareness is included in the data design issues to allow a full exploitation of
context sensitive application functionalities.
The architecture of a prototype VSDB is presented.
7. Low power architectures (editor M.G. Sami, Politecnico di Milano, coauthors) 30
pag.
A number of MAIS requirements drive research in computer architectures:
ever higher performances are requested – yet, present CPU architectures (provided
with various types of Instruction-Level Parallelism) have apparently reached the limit
in performances that can be extracted from a given application, while “conventional”
multiprocessor architectures are hardly suited to environments such as mobile systems
and anyway usually require massive program reorganization in order to achieve high
performances.
power consumption is appearing as the most important constraint posed to the
designer; higher performances today mean higher power consumption, and in fact the
increase in power requirements is not justified by a comparable performance increase.;
mobility and connectivity – the “every time, everywhere” key availability request –
lead to continued execution of communication protocols and to increased security
requirements – that, in turn, involve further requests on performances and lead to
further battery life criticality.
The ideal solution is for new execution paradigms, where multiple levels of parallelism
could be exploited depending on the characteristics of the application; in this way, high
performances could be achieved by identifying and exploiting the “proper” level of
parallelism in a dynamic way (i.e., in relation to different sections of an application) and
without keeping relevant areas of the processing system idle (but still “on” as far as power
consumption is requested). Architectures should be “heterogeneous”, i.e., provide for
application-specific units within the system whenever particular functions grant their
usefulness: cryptography – the basic building block for security solutions – is a typical area
for introduction of such units.
The MAIS environments focuses not (or not only) on the individual processing node, but
rather on the network (a wireless network). From this point of view, the strongest impact on
architectures (on the hardware as well as on the protocol-management segments of the
operating system) derives once more from power consumption demands and constraints:
power consumption is obviously mandatory for battery-powered nodes, and protocols
should be managed so as to reduce power consumption both at the processing node level
and at the processing system level - by optimising protocol management at node and at
network level in a much finer way than the present solutions provide.
The chapter will discuss design choices on these issues, and innovative perspectives for
future mobile information systems based on forthcoming architectures.
PART III – MOBILE INFORMATION SYSTEMS DESIGN
8. Quality of service and user profile (editor C. Batini, Univ. Milano Bicocca,
coauthors) 30 pag.
A major characteristic of the MAIS is approach is the definition of quality characteristics
for service and a user context, to describe requirements and profiles of users accessing the
e-services. To facilitate e-service adaptation and customization, user requirements and
profiles provide information useful to characterize the context constraints.
A quality dimensions ontology has been developed, and it is the basis for the e-service
design methods presented in the following chapter and for characterizing e-service
properties.
User profiling is considering also accessibility characteristics and their implication in
presentation of information. User profiling tools have been developed in the project as a
basis for e-service recommendations.
A detailed description of the quality and user profile models will be provided as a
reference model for MobIS developers.
9. Design and Integration Methodologies and tools (editor R. Torlone, Univ. Roma
Tre, coauthors) 40 pag.
The various devices that can be used as interfaces or terminals for network services
exhibit different features. These devices include not only cellular phones, PDAs, and
terminals for disabled people (which are already in use nowadays), but also new kinds of
devices, possibly embedded into tools such as household appliances, or vehicle
dashboards. The characteristics of the various devices are so different that the issues
related to the delivering of information and services on the Web involve, in addition to
presentation aspects, structural and navigational aspects as well. Moreover, user profiles
can be profitably used to filter data and deliver only relevant information.
The goal of the work on design and integration methodology and tools in MAIS is the
definition of models, methods and techniques for the automatic generation of Web sites,
which are adaptive with respect to context information and user profiles. This activity
consists of the following three main tasks.
- Methodologies and cooperative design tools for multi-channel information systems,
considering both context aware web applications and multimodal user interfaces
- Tools for the automatic generation of interfaces
- Techniques and Tool for usability and accessibility evaluation
10. Applications (editor B. Pernici) 30 pag.
The MAIS approach has been experimented on some demonstrator applications, providing
specific e-services and using elements of the technological platform:
- E-learning (editor L. Sbattella, Politecnico di Milano)
- Tourism (editor D. Presenza, Engineering)
- Risk Management in Archeology (editors M. Brioschi, Cefriel, A. Maurino, Univ.
Milano Bicocca)
Short Resumè of Prof. Barbara Pernici
Barbara Pernici is full professor of Computer Engineering at Politecnico di Milano.
Her research interests include cooperative information systems, workflow management
systems, information systems modeling and design, mobile information systems, temporal
databases, applications of database technology.
She holds a Dr. Eng. Degree from Politecnico di Milano and a Master of Science in Computer
Science from Stanford University.
She has published around 35 papers in international journals, including IEEE and ACM
Transactions, co-edited 10 books, and published about 120 papers at international level. She is
an editor of the Requirements Engineering Journal. She has participated in several
ESPRIT/IST projects (TODOS, Equator, ITHACA, F3, WIDE, Chorochronos). She coedited
project books for TODOS, WIDE, and Chorochronos. She is chief scientist of the Italian
FIRB MAIS (Multichannel Adaptive Information Systems) Project, 2002-2005. She is chair
of Working Group 8.1 on Design and Evaluation of Information Systems of IFIP
(International Federation for Information Processing).
In 2004 she has been program co-chair of the International conference on Business Process
Modeling (Potsdam), of IFIP TC8 International Working Conference on Mobile Information
Systems, Oslo, and of the CaiSE workshop on Data and Information Quality (Riga). She will
be program co-chair of COOPIS in 2005.
Web site: http://www.elet.polimi.it/people/pernici