0RELOH$JHQW6ROXWLRQVIRU$FFRXQWLQJ0DQDJHPHQWLQ0RELOH&RPSXWLQJ
Paolo Bellavista, Antonio Corradi, Silvia Vecchi
'LSDUWLPHQWRGL(OHWWURQLFD,QIRUPDWLFD6LVWHPLVWLFD8QLYHUVLW\RI%RORJQD
^SEHOODYLVWDDFRUUDGLVYHFFKL`#GHLVXQLERLW
$EVWUDFW
7KHFRQYHUJHQFHRIPRELOHWHOHFRPPXQLFDWLRQVDQGWKH
,QWHUQHWJOREDOV\VWHPIRUFHVWRUHFRQVLGHUWUDGLWLRQDOFOL
HQWVHUYHUVROXWLRQVIRUQHWZRUNDQGV\VWHPVPDQDJHPHQW
7KHSDSHUFODLPVWKDWDFFRXQWLQJLQWKHPRELOLW\HQDEOHG
,QWHUQHW UHTXLUHV VXSSRUW LQIUDVWUXFWXUHV KRVWHG LQ WKH
IL[HGQHWZRUN7KHVHLQIUDVWUXFWXUHVVKRXOGPRQLWRUFRQ
WURODQGUHJLVWHUUHVRXUFHFRQVXPSWLRQORFDOO\ZLWKLQWKH
GRPDLQV ZKHUH XVHUVWHUPLQDOV G\QDPLFDOO\ PRYH WR
ZLWKRXWUHTXLULQJFRQWLQXRXVFRQQHFWLYLW\ZLWKUHPRWHDQG
FHQWUDOL]HG DFFRXQWLQJ KRPH PDQDJHUV ,Q DGGLWLRQ WKH
SDSHU VKRZV WKDW WKH 0RELOH $JHQW 0$ WHFKQRORJ\ LV
VXLWDEOH WR RYHUFRPH WKH OLPLWV RI WUDGLWLRQDO DFFRXQWLQJ
VROXWLRQV LQ VHYHUDO PRELOLW\HQDEOHG XVDJH VFHQDULRV
0$VFDQPD[LPL]HORFDOLW\LQDFFHVVLQJPRQLWRULQJGDWD
FDQHQDEOHDFFRXQWLQJHYHQLQFDVHRIWHPSRUDU\GLVFRQ
QHFWLRQ FDQ LQVWDOO QHZ PRQLWRULQJFRQWURO EHKDYLRU G\
QDPLFDOO\ DQG FDQ VXSSRUW VHVVLRQGHSHQGHQW VROXWLRQV
7KHSDSHUILQDOO\SUHVHQWVWKHGHVLJQDQGLPSOHPHQWDWLRQ
RI WKH 0$EDVHG 0LGGOHZDUH IRU 0RELOLW\ $FFRXQWLQJ
0DQDJHPHQW 0$0 WRJHWKHU ZLWK VRPH XVH FDVHV
VKRZLQJWKHDGYDQWDJHVRIWKH0$DGRSWLRQ
,QWURGXFWLRQ
Advances in mobile telecommunications and portable
device miniaturization propose new service scenarios
where users, access terminals and even service compo-
nents can geographically move during provisioning. This
forces to face several technical challenges at different lev-
els of abstraction, from network connectivity to location
tracking, from device-dependent service adaptation to lo-
cation-dependent service provisioning. In addition, mobile
computing and the Internet are converging and likely to
merge, with the aim of providing a unique support infra-
structure, capable of hosting network elements, service
components, and fixed/mobile access terminals [1].
This mobility-enabled Internet also changes the per-
spective of network, systems and service management.
Traditional management solutions are tailored to fixed in-
frastructures of network elements, and management re-
search activities have proposed architectures to deal with
geographic distribution and heterogeneity of resources and
service components. IETF and OSI have proposed man-
agement models based on Client/Server (C/S) interaction
[2, 3] and on several variations of it. If one can organize
hierarchies of C/S components to achieve decentralization
and scalability, the interaction among management entities
is usually statically determined for clients and servers in
exchanging remote information. These solutions do not fit
well global systems in rapid evolution where the enter-
ing/exiting of unknown heterogeneous mobile components
is usual. These cases ask for the possibility of dynamically
installing new management behavior without any service
suspension. Some significant management research efforts
investigate the support of code mobility in a secure and
interoperable way. They have been tagged with different
names, such as Management by Delegation [4], Active
Networks [5], Programmable Networks [6], Mobile
Agent-based management [7, 8].
Among the different activities belonging to the man-
agement domain, accounting can be defined as the process
of monitoring, controlling and registering the amount of
administered resources that an authorized user (or a user
group) exploits. Accounting becomes particularly relevant
and challenging in mobile computing. Accounting solu-
tions require tracking user locations in a global environ-
ment and coordinating remote resources, possibly located
in different networks with heterogeneous mechanisms for
monitoring and access control. Accounting systems should
face temporary disconnections and network partitioning.
In addition, they should organize control strategies on the
base of previous user actions in remote network localities,
e.g., to permit the access to one service component if the
client has not used up the time quota reserved for that
service.
The current hardware/software limits of several catego-
ries of mobile devices, from personal digital assistants to
programmable cell phones, impose an infrastructure over
the fixed Internet to support specific issues for their net-
work connectivity and service access [1]. In the same way,
a support is necessary to provide accounted services in
any network locality willing to open its resources to mo-
bile accessibility. The paper claims that mobility-enabled
accounting management requires a dynamic and extensi-
ble support infrastructure, capable of evolving during
service provisioning depending on client mobility. The
infrastructure should be in charge of monitoring, control-
ling and registering resource consumption locally where
mobile users/terminals move to, without requiring con-
Proceedings of the Seventh International Symposium on Computers and Communications (ISCC’02)
1530-1346/02 $17.00 © 2002 IEEE