A MINIMALLY INTRUSIVE WIRELESS SOLUTION
FOR CONTEXT- AND SERVICE AWARENESS ENABLEMENT
IN MOBILE COMMUNICATIONS
Máirtín O’Droma
Telecommunications Research Centre, University of Limerick, Limerick, Ireland
Ivan Ganchev
Telecommunications Research Centre, University of Limerick, Limerick, Ireland;
Department of Computer Systems, Plovdiv University, Plovdiv, Bulgaria (on leave)
Nikola.Nikolov
Department of Computer Sciense and Information Systems,University of Limerick, Limerick, Ireland;
Telecommunications Research Centre, University of Limerick, Limerick, Ireland
Paul Flynn
Corvil, Dublin, Ireland;
Telecommunications Research Centre, University of Limerick, Limerick, Ireland
Keywords: Ubiquitous Consumer Wireless World (UCWW); Next Generation Networks (NGN); Wireless Billboard
Channel (WBC); Advertisement, Discovery and Association (ADA); Abstract Syntax Notation v.1 (ASN.1);
Wireless Services; Context Awareness; Always Best Connected and best Served (ABC&S); Personalised
Information Retrieval (PIR).
Abstract: A wireless solution for context- and service-awareness in mobile communications is the theme of
this paper. Respecting mobile users’ desire for minimal intrusion of unsolicited advertisements, here we
show how the novel push-advertisement technology and medium of ‘wireless billboard channels’ (WBCs)
could be employed by service providers to broadcast advertisements of their wireless services to mobile
terminals. While the word ‘billboard’ here seems to hint at in-your-face intrusiveness, in fact the service
functions in the background in a near-transparent un-intrusive manner to the user. When combined with
user-driven, dynamic, smart user-profile functionality, inclusive of smart optimisation of current user’s
‘always best connected and best served’ (ABC&S) policies, the system has the potential to provide an
effective, pro-active, wireless-based, context-aware and service-aware infrastructure.
The exposition here includes detailed descriptions of the WBC concept, its associated
advertisement, discovery and association (ADA) functionality, full technical details of the WBC
advertisement service description techniques, formats and attributes, and operational aspects, such as the
WBC bit stream structure. It also includes discussion of algorithmic approaches towards optimising smart
user profile functionality on mobile terminal which will drive the ABC&S decision-making in ways
matched especially to the user needs, in particular those schemes utilising Personalised Information
Retrieval (PIR) systems.
118
O’Droma M., Ganchev I., Nikolov N., Nikolov N. and Flynn P.
A MINIMALLY INTRUSIVE WIRELESS SOLUTION FOR CONTEXT- AND SERVICE AWARENESS ENABLEMENT IN MOBILE COMMUNICATIONS.
DOI: 10.5220/0005414901180128
In Proceedings of the First International Conference on Telecommunications and Remote Sensing (ICTRS 2012), pages 118-128
ISBN: 978-989-8565-28-0
Copyright
c
 2012 by SCITEPRESS – Science and Technology Publications, Lda. All rights reserved

1 INTRODUCTION
Wireless billboard channels (WBCs) originated
as a foundational infrastructural component in the
recently proposed Ubiquitous Consumer Wireless
World (UCWW), (O’Droma, 2007) & (O’Droma,
2010), with the main role of direct-to-user push-
based advertisement, discovery and association
(ADA) service for wireless access network
providers (ANPs) and application service providers
(ASPs) promoting use of their networks and
services, (Flynn, 2007). Through the WBC-ADA
service, mobile users, as consumers, may discover
the ‘best’ services and associate with them
following a user-driven ‘always best connected and
best served’ (ABC&S) paradigm, (O’Droma, 2006).
The physical radio characteristics of WBCs include
them being narrowband, unidirectional, point-to-
multipoint (P2MP) broadcast channels. They could
be created in a number of different stand-alone
infrastructures, but for various reasons, economic as
well as technological, it would be better that they
are integrated into existing wireless broadcast
infrastructures, such as a digital television or digital
radio broadcast platform.
The WBC service advertisement concept has the
attractive advantage of being non-intrusive service
advertisement vehicle and infrastructure. This
contrasts with the current approach of cellular
operators and service providers seeking to attract
roaming travellers to use their networks and
services by intrusive unsolicited SMS text message
advertisements. It is a phenomenon, for example,
experienced by many on turning on their mobile
phone on arrival in the airport of a new country.
While it is of the nature of being a consumer to
want to get best value and thus to want to know
about competing prices and services, and to have
the freedom to choose among them, usually this
SMS promotional bombardment does not serve this
need in a reasonably informative way. Another
growth area of these intrusive unsolicited SMS
advertisements is location-based services and other
such like services on offer in the area of the mobile
user’s present location.
The WBC-ADA service described in this paper
is well fitted for this advertisement need and
service, and to a much wider extent. Moreover, as it
functions in the background in a near-transparent
un-intrusive manner, it is a comprehensive solution
to this growing consumer problem of ‘spam’ SMS.
Following an exposition of the WBC concept
and underpinning technology, how these channels
may be made to serve as a wireless solution for
context- and service awareness in mobile
communications is presented in this paper. The
approach, which facilitates both the minimally-
intrusiveness criterion and background near-
transparent functionality, is presented as well.
2 WIRELESS BILLBOARD
CHANNELS (WBC)
Taking into consideration the potentially huge range
of wireless services already available to mobile
users, the demand for an efficient and easy
mechanism for services’ ADA adapted to the
terminal’s capabilities, user preferences and user
location, is clearly foreseen. The approach we are
proposing to facilitate ADA is to use WBCs that
will provide mobile terminals with information
about the services available to them. Each terminal
can compare this information with its capabilities,
the current user profile’s preferences, the user
location, and other context information such as the
time of day, to select the best services to use to
achieve a particular goal. As well as advertising the
service, the WBC will also provide information to
help with the process of discovering and associating
with that service.
The WBC should have the following
characteristics (Flynn, 2007):
 Simplex and broadcast: Simplex here applies
not just to the unidirectional physical nature
of the channel but also to the unidirectional
nature of the WBC service. This attribute has
the additional benefit of easing the WBC
physical deployment and operation. If the
channel is duplex, then in a way it simply
becomes another wireless access network
infrastructure for two-way wireless
communications, and thus in the end no
different in its general attributes from any
other wireless access network infrastructure.
And if this, then bandwidth-spectrum
allocation becomes a much more significant
issue, as it has been for instance for existing
cellular spectrum allocations.
 Limited bandwidth: Given the proposed usage
– P2MP unidirectional service of
advertisements – bandwidth requirements
will be relatively narrow. This has the added
advantage of enhancing the WBC likely
success, e.g. of global agreements on
spectrum allocations for them. With limited
spectrum available, this would improve the
A Minimally Intrusive Wireless Solution for Context and Service Awareness Enablement in Mobile Communications
119

WBC’s chances of becoming a worldwide
standard.
 Maximum coverage area: These channels
should ideally be available anywhere and at
any time. No matter where it is (indoor or
outdoor), a mobile terminal should have the
ability to discover what services are available
to it in the current location provided by local,
regional, national or international service
providers. Terminal mobility should not
affect the ability to receive information on the
channel.
 Different versions for different areas: The
number and types of WBCs could correspond
to the local/regional/national/international
interests of advertisers and users. In practice
there would be growth; perhaps a start-up
situation would be one national WBC
channel, advertising all the services that are
relevant on a national level, which could
include advertisements of local, regional or
interregional significance. And then separate
regional WBCs channels, advertising the
services available in that region.
 Operated by non-ANP service providers: In
consideration of the need for fair competition
in the ANP marketplace and equity of access
to WBC advertisement space, i.e. equally
open to all ANPs, it is preferable for WBCs
service providers (WBC-SPs) to be fully
independent and physically separate from
ANPs and their networks. This requirement
may need the support of some regulation.
3 WBC-ADA MODEL
The main purpose of the WBC from the consumer’s
perspective is to allow wireless services to be
discovered by mobile terminals. Service discovery
is a networking concept whereby a client can
automatically discover services that are available to
it in a network. There are a number of protocols in
existence that deal with service discovery. Three of
the well-established protocols, namely Jini (Oaks,
2000), Service Location Protocol (SLP), (Guttman,
1999), and Salutation (Pascoe, 1999), were used as
a basis for the development of the WBC-ADA
model. Though very different, these protocols use
the same basic model that relies on a central registry
of service descriptions (SDs), which is accessible in
a request-reply manner. A client makes a request for
a service, based on the service type and attributes it
would like, and receives a reply from the registry
containing information about services that match
the request. A WBC, however, is a simplex
broadcast channel, which does not allow for
requests to be made. The solution then is to
broadcast periodically all SDs over the WBC
(making allowance for flexibility, e.g., that some
SDs be broadcast more frequently than others).
The flow diagram for the WBC-ADA model is
shown in Figure 1 and described below:
A. Service providers register their SDs with a
WBC-SP’s central registry using some external
method, e.g. via a web portal.
B. WBC-SP broadcasts all registered SDs,
repeatedly, on a WBC (service advertisement) in
compliance with the specific agreements with each
advertising service provider client (e.g., frequency
of re-transmissions).
C. Mobile terminals tune to WBC and listen,
selectively, to broadcast to receive desired SDs
(service discovery).
D. Each terminal freely associates with the
chosen service provider (server) to use its service
(association).
Figure 1: The WBC-ADA model.
3.1 WBC SD Format
The WBC SD format is based around some of the
fields used in the well-established service discovery
protocols mentioned above.
3.1.1 Service Type
Every wireless service is assigned a service type,
which is a field that indicates the basic function
performed by the service. The purpose of the
Service Type field is to allow similar services to be
grouped together in a WBC broadcast as this makes
it possible to apply power saving techniques.
First International Conference on Telecommunications and Remote Sensing
120

A hierarchical structure for the Service Type
field is considered as this allows the order of SDs
on a WBC broadcast to be more structured. It is a
five-octet field, composed of four subfields.
Service-Type ::= SEQUENCE
{
division OCTET STRING(SIZE(1)),
category OCTET STRING(SIZE(1)),
type OCTET STRING(SIZE(2)),
version OCTET STRING(SIZE(1))
}
The division subfield specifies whether it is, for
example, an access network’s communication
service or an application service. The category
subfield describes what category of services within
its division a service falls into. The type subfield
specifies the actual type of service, within its
category. The version subfield allows new versions
of templates for service types to be published. This
could be used to update service types or to create
subtypes.
3.1.2 Scope List
The Scope List field identifies which scopes the
service belongs to. Scopes are a way to
administratively group services together. A service
may have as many scopes as necessary. Mobile
terminals are also configured with scopes. A
terminal will only pay attention to SDs that belong
to one of the same scopes as itself. Services may
also belong to a default scope, meaning that all
terminals will pay attention. If a terminal receives a
SD that has neither one of the same scopes as itself
nor the default scope, then it will ignore that SD.
An example of uses for scopes would be to
advertise services that are only available to a certain
group of people, e.g. government officials,
emergency service employees, employees of large
corporations, etc.
A scope list is a list of one or more scopes.
ScopeList ::= CHOICE
{
defaultScope SEQUENCE OF Scope
OPTIONAL,
notDefaultScope SEQUENCE OF Scope
}
Any advertiser that wishes should be assigned a
globally unique scope. For this reason, the Scope
List field should be large enough so that scopes do
not run out; or a variable length scheme is
incorporated. For a fixed length, a 32-bit field
seems more than sufficiently large to cater for
future expansion.
Scope ::= OCTET STRING(SIZE(4))
3.1.3 Length
The Length field is an integer that specifies the
length of the Attributes field in bytes. As 65,535
bytes is more than enough to accommodate any
possible list of attributes, it could be defined as:
Length ::= INTEGER(0..65535)
This means that the Length field has a fixed size
of two bytes.
3.1.4 Attributes
The Attributes field is the main one in a SD. It
carries the information for advertisement, discovery
and association of a service. Advertisement
attributes are used by a terminal to decide if its user
wants to use the service (based on information
stored in the user profile). Discovery and
association attributes are used to enable a
user/terminal to use the service. The format of the
Attributes field depends on the service type. The
size of the field is not fixed, meaning that the total
size of the SD is not fixed. Perhaps service
providers would pay more for advertisement of a
larger SD than for a smaller one.
3.2 WBC SD Encoding
One of the desired properties for the WBC is that it
should use as little bandwidth as possible. In Jini,
SDs are Java objects, and passing the descriptions
involves serialising the objects. Serialised Java
objects are a very inefficient way of encoding data.
In SLP, the SDs are encoded as text which, while
more efficient than Java objects, is still not efficient
enough. The Salutation SD, which uses the Abstract
Syntax Notation (ASN.1), (ITU-T, 2008a), is the
most efficient approach in this case. ASN.1 data can
be encoded in a number of different ways,
depending on which encoding rules are followed.
The Packed Encoding Rules (PER), (ITU-T,
2008b), give the most efficient encoding (close to
optimal). For this reason we chose for the SDs to be
ASN.1 structures and encode them using the PER.
The following is the SD format specified in
ASN.1.
ServiceDescription ::= SEQUENCE
{
service-Type Service-Type,
scopeList ScopeList,
length SDLength,
attributes Attributes
}
The encoded SD ASN.1 structures are broadcast
on a WBC. The Service Type and Scope List have
A Minimally Intrusive Wireless Solution for Context and Service Awareness Enablement in Mobile Communications
121

the same format in every SD so a terminal that
receives a SD knows what is the type of service and
to which scopes it belongs. For the Attributes field,
however, the format depends on the service type.
This field makes no sense unless the terminal
knows what its format is. Therefore, for each
service type an attribute template should exist, as
there is in SLP, which lays out the format of the
attributes for a particular service type. This way,
advertisers, WBC-SPs and terminal manufacturers
know what format to follow. Attribute templates
would form additions to the basic WBC standard.
One of the benefits of using attribute templates
is that terminal software only has to be programmed
to understand the meaning of service types that are
relevant to that terminal. For instance, a terminal
with no video capabilities has no interest in SDs for
video services. Another benefit is that as new
service types come into existence, new templates
can be published, without having to make changes
to the main WBC standard. However, this could
leave some terminals unable to understand the new
formats, while other smarter terminals will be able
to update their WBC software appropriately. It is
necessary then, for a terminal to be able to ignore
Attributes fields whose format it does not
understand. This is why the Length field is needed,
to tell terminals how many bytes they must ignore
before the next SD begins.
4 SERVICE TYPES AND
TEMPLATES
While WBCs would be used for ADA of access
networks’ communication services –for many its
primary goal–, the focus of this paper is on the
application services.
The vision for next generation networks (NGN)
is for all-IP networks where terminals will receive
application services from service providers over IP.
There are many different service types currently
available to terminals and in NGN that number will
be greater. A service type should include all
services that perform a similar function. For
example, a voice-call service should be one service
type rather than having many different service types
depending on the Voice over IP (VoIP) protocol
being used. These, instead, should be included as
attributes of the voice-call service.
In addition service types can themselves be put
into broader categories. For example,
communication services, information services,
location-based services, etc. This allows similar
service types to be grouped together on the WBC
broadcast. The ITU-T Focus Group on Next
Generation Networks’ (FGNGN), (ITU-T, 2004)
Working Group 1 is concerned with NGN services.
Drawing on its NGN Release 1 Service
Classification (Carugi, 2005) as a guide to what
services will be available in NGN, possible service
categories and types advertised to users/terminals
over the WBC would include, but by no means are
limited to, the following:
 Communication Services – Voice call; Video
call; Push-to-talk; Conference call;
Multimedia interactive communication
service; Incoming call; Voicemail; Call
diversion.
 Messaging Services – SMS; MMS; Email;
Fax; Instant messaging; Chat rooms; Bulletin
boards.
 Information Services – News; Weather;
Sport; Finance; Yellow pages directory.
 Entertainment Services – Games (download /
play on-line / on-line multiplayer); Music
(download / streaming); Video clips
(download / streaming); Ringtones / Logos /
Icons / Screensavers / etc (download);
Television (streaming); Radio (streaming);
Competitions; Gambling; Dating; Voting /
Surveys (maybe related to TV programmes,
e.g. reality TV).
 Educational Services – mLearning (lectures,
tutorials, labs, tests, examinations).
 mCommerce Services – mPay facilities;
mShopping portals; mBanking and share
trading; Bookings and ticketing; M-market
type services (eBay).
 Location-Based Services – Nearest restaurant,
cinema etc; Local weather; City guide;
Walking/driving directions; Car parking;
Traffic information; Public transport services
/ timetables; Location-based advertising;
Location-based dating.
 Emergency Services – Police, Fire brigade,
First medical aid, Rescue service, Alerts for
natural disasters (in the area), etc.
 Terminal reconfiguration services – upgrades
of mobile software, operating systems and
applications.
 Other Services – On-line data storage; Cloud
computing; Organisational services –
calendar / address book etc; Virtual Private
Network (VPN); Social networking.
First International Conference on Telecommunications and Remote Sensing
122

Some of these listed service types may not be
suitable for advertising on the WBC, and other
(suitable) service types may have been omitted.
Also, some of the service types may belong to more
than one service category.
Each service type needs a service template to
specify the attributes that are included in a SD for
advertisement, discovery, and association. Making
decisions between different services is automatic,
based on preferences specified by the user in his/her
profile and by taking into account the cost, quality
and supported features of the available services.
After selecting a service by the user, the
terminal (operating in a background mode, i.e.
without disturbing the user) will choose the ‘best’
available instance of the desired service and will
associate with it by using information obtained from
the WBC. The first thing needed is the client-side
software to be installed on the terminal (if not
already installed/pre-loaded). For this an attribute is
needed, which tells the terminal how to download
this software. There should then be an attribute
specifying the software itself and its version so that
the terminal can see if it needs to be updated. Rather
than having an attribute that says where to get the
software (or software update), it could be better to
have one software-download service (which would
also be advertised as any other service on the
WBC), which allows for downloading of all
additional software needed to use services
advertised on the WBC. Having installed the
software, there are some attributes specific to a
particular service (e.g. IP address and port numbers)
that also need to be known.
Each service type needs a service template,
which is an ASN.1 specification of the attributes
used to define a service of that type. Considering
the voice-call service as an example, a list of
attributes for its advertisement and association is
presented in Table 1.
Table 1. Attributes for the voice-call service type.
Attribute Meaning Use
Cost
Contains information about the cost of using the
voice-call service (e.g. cost per minute to various
recipients in different regions, countries etc).
Used by a terminal to compare the cost of the
service from one SP with the maximum cost for
the call that the user wishes to pay, and with that
for the same service offerings from other SPs.
Quality of
Service
(QoS)
Contains third-party measurements of the quality
of this service (e.g. calls dropping rate, sound
quality, etc).
A terminal selects the service (provider) that is
expected to perform ‘best’.
Protocols
Specifies which communication protocols are
used by this service.
A terminal may not support the use of certain
protocols. Reconfiguration requirements and
decisions.
For advertisement
Features
A list of features apart from the basic that might
differentiate one voice-call service from another
(e.g. ability to send text during a call, music
playing in the background, etc).
Used by a terminal to compare additional features
desired by the user in order to choose the ‘best’
service.
Software
An identifier that tells a terminal what software is
needed to access this service.
A terminal can see if this software is already
installed. If not, it can be downloaded from a
software-download service provider, etc.
Supported
versions
The version numbers of the software that are
supported by this service.
A terminal can see if the version of the currently
installed; software is suitable/up to date. If not,
then if a new version may be downloaded, based
on WBC advertisement information.
For association
Software
parameters
Information about this particular service needed
by the terminal software (e.g. IP address and port
numbers). This may have different fields for each
value of the software attribute.
A terminal software uses this information to
associate with the service.
A Minimally Intrusive Wireless Solution for Context and Service Awareness Enablement in Mobile Communications
123

Specified in ASN.1, the coded voice-call service
definition itself would have the following form:
VoiceCallService ::= SEQUENCE
{
cost VoiceCallCost,
performance
VoiceCallPerformance,
protocols VoiceCallProtocols,
features VoiceCallFeatures,
software SoftwareIdentifier,
versions SoftwareVersion,
softParams
VoiceCallSoftwareParameters
}
It is worth noting that the ASN.1 service
specifications are designed in such a way as to
accommodate also future service types and updates.
5 WBC BIT-STREAM
STRUCTURE
5.1 Data Scheduling
A SD contains the values of all the attributes that are
specified in the service template. These are encoded
into binary data using the PER of ASN.1. Depending
on the service type and the values of certain
attributes, SDs are not all going to be the same size.
The WBC broadcasts SDs continuously so that
terminals listening to the channel can receive the
service descriptions they need. The simplest
approach is to broadcast all the SDs, one after the
other and when the end is reached, start at the
beginning again (called a flat broadcast). This,
however, might not be the most efficient way, or the
way desired either by the advertisers or by the
WBC-SP seeking ways to make their business more
profitable. Also from the consumer’s perspective
‘flat broadcast’ can result in inefficiencies and
delays. If a certain SD is demanded by terminals
more frequently than another, then it makes sense to
broadcast it more frequently. The priority should be
to minimise the average access time of the entire
system. Access time is that time from the moment
when a mobile terminal first starts accessing a data
item (SD) on the WBC, until the moment it receives
full data item (SD). Minimising that time, averaged
across all required SDs by all terminals, is a
common-sense goal.
There are quite a few data scheduling algorithms;
however, most of these are pull-based, meaning that
they rely on the clients making requests on a back
channel for certain data items to be broadcast. These
algorithms are not suitable for the WBCs, which are
simplex broadcast channels. Hence only push-based
algorithms, such as the Broadcast Disks and Priority
Index Policies, were considered. A modified and
improved version of the Broadcast Disks algorithm
was elaborated and described in (Ji, 2008) for data
scheduling on WBCs.
5.2 Data Indexing
Efficient use by mobile terminals of their limited
battery energy is another factor. Any techniques to
reduce power consumption thereby extending their
battery charge-life are attractive especially if they
are cost-neutral. Power is saved by minimising the
time spent by terminals for listening to the WBC
channel to receive the required SD (referred to as
tuning time). Tuning time could be reduced by the
use of indexing because without it, a terminal would
have to tune into the WBC and listen to the
broadcast continuously until the required SD is
transmitted. By adding indexing data to the
broadcast, terminals can tune in, find out when the
required SD will be transmitted, then tune out and
wait until that time to tune back in again. By adding
redundant data to the broadcast, however, the
average access time is going to be increased.
There are several indexing schemes that aim to
provide a good trade-off between having a low
tuning time and a low access time. All these
schemes have several features in common: the
attribute that the data items (records) are indexed by
is called the key (in the case of the WBC, the key is
the service type); the order of the records in the
broadcast sequence is arranged by the value of their
keys; the broadcast sequence is divided into fixed
size parts called buckets; the indexing tells the client
in which bucket to look to find the first record with
the required key (as buckets are of a fixed size, the
client will know the exact time that a particular
bucket will be broadcast).
Short description and comparison of the main
indexing schemes are provided in (Flynn, 2010). It
may be that flat-broadcasting schemes could be
adapted to accommodate non-flat broadcasting as
well. Based on these, a new ‘Simple Indexing
Scheme’ was elaborated for use in WBCs as
described in (Flynn, 2010).
First International Conference on Telecommunications and Remote Sensing
124

6 SMART USER-PROFILE
FUNCTIONALITY
In future networks, the ultimate goal of service
providers will be to provide services to mobile users
in ABC&S mode, i.e. at the ‘best’ QoS and cost
levels, and by applying personalisation and context
awareness. As described in (Stavroulaki, 2009),
personalisation supports different user types and
roles, whereas context awareness considers the state
of the user (user identity and roles), mobile terminal
(capabilities) and environment (location and time
zone). To achieve this, proper management of user
preferences and terminal capabilities is required by
means of accurate description and representation of
information along with the configuration and update
of user profiles.
The user profiles could be structured with two
groups of parameters as proposed in (Stavroulaki,
2012):
 Observable parameters – these include the
services that are currently running on the user
terminal, corresponding QoS parameters, and
context parameters such as user location, time
zone, service rank, etc.
 Output parameters – the value of these is
dynamically changed over time depending on
the value of observable parameters. The main
output parameter is the so called utility
volume, which provides a ranking of service-
QoS pairs by order of user preferences. The
latter, of course, may change over time and
may vary depending on the user context. An
approach for dynamically learning of user
preferences w.r.t. the perceived QoS level per
service with the use of Bayesian statistics
concept is presented in (Stavroulaki, 2009).
A generic intelligent iWBC client, which sits on
the mobile terminal and operates in the background
in support of the mobile user becoming fully aware
of all relevant wireless services available (in the
current location) and advertised according to the
structured WBC format, has been designed and
successfully implemented (Ji, 2011). Information
about available services is proactively pushed to the
terminal which, with the help of this application, can
thus discover the services and all the necessary
details about their offerings, sufficient to make
informed ABC&S decisions about using them,
including knowing how to associate with the access
networks to obtain these services in the best possible
way through the ADA procedure. Hence this novel
iWBC application amounts to a significant advance
in consumer-driven ABC&S capabilities and
services.
Through the WBC system, including the iWBC
client application, all wireless service
advertisements, become a background activity. That
is, the data is captured, analysed and filtered
according to individual user’s ABC&S policies and
presented to the user as and when s/he likes. It is
only when the user seeks to use a service that s/he
might become aware of the service offerings
garnered from the WBC advertisements by the
iWBC application. The iWBC would then present
these over a friendly user interface allowing the user
to assess options and attributes in an information-
structured and comparative manner, and make an
ABC&S choice decision on which service to use.
In fact it can all be even more transparent and
un-intrusive to the user, if the ABC&S decisions are
also made in the background following default (e.g.,
lowest price) or preset ABC&S policies and profile
settings. This ‘full transparent ABC&S’ mode will
mean a minimum disruption to the user, and yet it
will still be user-driven ABC&S.
(Ji, 2011) demonstrates a practical use case
example when a new ‘best’ wireless service instance
is becoming available in the current area of the
mobile user. In this case, the iWBC client
application automatically discovers this new service
offering by a background extraction of this
information from the WBC’s service descriptions,
logs this information in the terminal’s database,
examines it in accordance with the consumer-driven
ABC&S policies and makes decisions on its usage,
informing the user of the new service offerings, and
so on.
7 PERSONALISED
INFORMATION RETRIEVAL
The UCWW environment presents a shift from
the currently dominating subscriber-based wireless
access towards a consumer-centric one (O’Droma,
2007) & (O’Droma, 2010). Allowing the user to
choose a provider for a particular service from a list
of alternatives opens the opportunity for stronger
competition between providers and a better service
for the user as a result. However, giving the user the
freedom to choose should not become an
overwhelming burden when the number of options,
as well as factors that choice depends on, is too
large.
A possible solution is to introduce an extra
feature to the WBC-ADA system that facilitates the
A Minimally Intrusive Wireless Solution for Context and Service Awareness Enablement in Mobile Communications
125

user’s choice by suggesting a ranked list of
providers for the desired service as per the past
experience of other users in the same physical
context. The retrieved list of service providers could
then be adjusted and personalised for the user and
their physical context by utilising the context
parameters in the user profile.
Systems which retrieve information that is both
relevant to the submitted queries and personalised
for the user are known as personalised information
retrieval (PIR) systems. They have been the subject
of extensive research in the last couple of decades
with a major application in Web search as well as in
other areas such as eLearning and news
dissemination (Ghorab, 2012). For example, a PIR
framework for a wearable computer in ubiquitous
computing environment was proposed by (Hong,
2005). This framework is particularly oriented
towards retrieving personalised information from
objects in physical proximity to the user in a
typically cluttered with data environment. Although
the iWBC client application would also work on
wearable devices in a ubiquitous computing
environment, it will need to retrieve information
from a relatively less heterogeneous data source (the
stream of SDs broadcast on WBC) than the complete
variety of objects in physical proximity to the user.
Thus, the design principles for a typical PIR system
(for example, a PIR system for Web search) may be
followed, taking into account also the physical
context of the user.
In order to provide personalised results, an
iWBC PIR client must gather and store data about
the information needs of the user. There are two
approaches to representing and storing the gathered
information. It can be either stored in an
individualised user model (Zhang, 2007) &
(Speretta, 2005) or in an aggregated one (Agichtein,
2006) & (Smyth, 2006). A recent survey proposes a
classification of PIR systems with respect to the
scope of personalisation into three categories:
individualised personalisation, community-based
personalisation, and aggregate-level personalisation
(Ghorab et al. 2012). The individualised
personalisation is limited to the information gathered
about the user who queries the system. Similar to it,
the community-based personalisation also employs a
user model for providing personalised information
retrieval. However, it also allows sharing of
information between users. The aggregate-level
personalisation, on the other hand, makes no use of
an explicit user model and decisions are based on
aggregate usage data.
A community-based PIR (Teevan, 2009) &
(Sugiyama, 2004) would seem to suit WBC needs
best as it offers personalised results to the user based
not only on their previous behaviour and experience
but also on the experience of other users who have
used the service in a similar context. There are three
aspects of a PIR system which need to be addressed
in the design: information gathering, information
representation, and implementation and execution of
personalisation (Gauch, 2007). An immediate next
step in the WBC research is to decide what
information about users should be tracked, how this
information will be gathered and stored in a user
model and then how user models will be used for
retrieving personalised results. The research
literature offers a wide spectrum of solutions for
each of the stages of PIR (Ghorab, 2012). The
majority of them have been developed for the needs
of Web information retrieval. However, they should
be easily adaptable for the needs of selecting
providers of services in NGN.
From another point of view, user modelling and
PIR can be a feature that can make the concept of
WBC-ADA attractive to the current wireless service
providers. Instead of having third-party
organisations providing the PIR service for all users,
PIR can be distributed between SPs. That is, PIR can
be a service that a user will request from a particular
provider they are subscribed to or have used as a
consumer. For example, consider the following
scenario. User A is receives PIR services provided
by SP X. User A needs to make an international
phone call and tunes on WBC to receive information
of a provider that can allow them make the call for a
fixed price under 1€. The iWBC client finds a
number of providers who satisfy this user request
and returns a personalised ranked list. The top
provider in the list is Y and the user selects it for the
phone call. That is, the user has used the PIR
provided by X to find out that Y is the provider that
best suits their requirements for a phone call at their
particular physical context. In this case, even though
X will not provide the phone call, X can still be
motivated in providing the best PIR for that user.
Thus, the competition between providers can stay
healthy and they can preserve their user base.
8 CONCLUSIONS
Advertisement and discovery are part of an
important R&D challenge to find solutions for the
automated enabling of the entire process of
advertisement, discovery and association (ADA) of
wireless services and for the evolution of wireless
communications. It will contribute significantly to
continuity of connection to wireless services
provided to mobile terminals. To facilitate this, a
wireless equivalent of the shop window and
First International Conference on Telecommunications and Remote Sensing
126

billboard advertising, called a wireless billboard
channel (WBC), was presented in this paper. This
would provide access network providers (ANPs) and
application service providers (ASPs), with a very
active dynamic means to advertise (i.e. proactively
‘push’) their presence and services through WBCs
operated by non-ANP service providers. The pro-
active push advertisement nature of the WBC would
correspond well to ANP and ASP competitive
desires to use all means to reach the mobile users
from their existing ‘loyal’ subscribers through to the
impulse buying consumers. Advertisement and
discovery of wireless communication services and
application services deployed and accessible in a
particular area/location, and procedures for terminal
association with these, are defining characteristics of
contemporary networks.
WBCs would also benefit the users due to
automated discovering functionality of terminals
scanning the WBCs and updating service offerings
and availability information, matching these against
user profiles and proposing, and enabling casual or
persistent consumer-type association links for
different user-desired services. The result of this
process will be more up-to-date information for
user-driven always best connected and best served
(ABC&S) decisions. Ideas on smart user profile
functionality and user-driven ABC&S decision-
making based on personalised information retrieval
(PIR) systems adapted to this UCWW context have
been set out and discussed.
Implicit in this vision is the global
standardisation of these proposed WBCs as a vehicle
and a wireless infrastructural support for ADA
procedure. Practically it makes sense to have several
WBCs directed at different geographic extensions.
The utilisation of the Digital Video Broadcast -
Handheld (DVB-H) standard – a standard for
broadcasting digital television to handheld devices
(ETSI, 2004) – as a candidate carrier technology for
WBC was extensively studied, e.g. in (Ji, 2010a).
DVB-H extends the DVB-Terrestrial standard with
time-slicing and other additions to greatly reduce
receiver’s power consumption. Designed for
handheld devices, it allows for high terminal
mobility and thus is ideal for WBC use. A ‘WBC
over DVB-H’ prototype system was designed,
implemented, evaluated and successfully tested, e.g.
(Ji, 2010b) & (Ji, 2009).
ACKNOWLEDGEMENTS
The authors wish to acknowledge the support of the
Telecommunications Research Centre (TRC), UL,
Ireland and the NPD of Plovdiv University under
Grant No. NI11-FMI-004.
REFERENCES
Agichtein, E., Brill, E., Dumais, S., Ragno, R., 2006.
Learning user interaction models for predicting Web
search result preferences. In SIGIR 2006, 29
th
Annual
International ACM SIGIR Conference on Research
and Development in Information Retrieval. ACM.
Carugi, M., Hirschman, B., Narita, A., 2005. Introduction
to the ITU-T NGN Focus Group Release 1: Target
Environment, Services, and Capabilities. In IEEE
Communications Magazine, 43(10).
ETSI, 2004. EN 302 304 V1.1.1. Digital Video
Broadcasting (DVB); Transmission System for
Handheld Terminals (DVB-H). http://www.etsi.org/
deliver/etsi_en/302300_302399/302304/01.01.01_60/e
n_302304v010101p.pdf
Flynn, P., Ganchev, I., O’Droma, M., 2007. Wireless
Billboard Channels: Vehicle and Infrastructural
Support for Advertisement, Discovery, and
Association of UCWW Services. In Annual Review of
Wireless Communications, vol. 2. International
Engineering Consortium, Chicago, Ill, USA.
Flynn, P., 2010. Wireless Billboard Channels for the
Advertisement, Discovery and Association of Mobile
Services in a Ubiquitous Consumer Wireless World.
Master’s Thesis. University of Limerick, Ireland.
Gauch, S., Speretta, M., Chandramouli, A., Micarelli, A.,
2007. User profiles for personalized information
access. In The Adaptive Web, Springer. Berlin, 1
st
ed.
Ghorab, M. R., Zhou, D., O’Connor, A. and Wade, V.,
2012. Personalised Information Retrieval: survey and
classification. In User Modeling and User-Adapted
Interaction, Springer. Netherlands.
Guttman, E., Perkins, C., Veizades, J., Day, M., 1999.
Service Location Protocol, version 2. IETF RFC 2608.
http://www.ietf.org/rfc/rfc2608.txt
Hong, D., Park, Y.-K., Lee, J., Shin, V., Woo, W., 2005.
Personalized Information Retrieval Framework. In 1
st
International Workshop on Personalized Context
Modeling and Management for UbiComp
Applications.
ITU-T, 2004. Focus Group on Next Generation Networks
(FGNGN). http://www.itu.int/ITU-T/ngn/fgngn/
ITU-T, 2008. Information technology - Abstract Syntax
Notation One (ASN.1): Specification of basic notation
(Recommendation X.680). http://www.itu.int/rec/T-
REC-X.680-200811-I/en
ITU-T, 2008. Information technology – ASN.1 encoding
rules: Specification of Packed Encoding Rules
(Recommendation X.691). http://www.itu.int/rec/T-
REC-X.691-200811-I/en
A Minimally Intrusive Wireless Solution for Context and Service Awareness Enablement in Mobile Communications
127

Ji, Zh., Ganchev, I., O’Droma, M., 2008. Efficient
Collecting, Clustering, Scheduling, and Indexing
Schemes for Advertisement of Services over Wireless
Billboard Channels. In ICT’2008, 15th International
Conference on Telecommunications.
Ji, Zh., Ganchev, I., O’Droma, M., 2009. Performance
Evaluation of 'WBC over DVB-H' System. In IEEE
Transactions on Consumer Electronics, 55(2).
Ji, Zh., Ganchev, I., O’Droma, M., 2010. Performance
Analysis of 'WBC over DVB-H' Link Layer. In
EURASIP Journal on Wireless Communications and
Networking, vol. 2010.
Ji, Zh., Ganchev, I., O’Droma, M., 2010. `WBC over
DVB-H` Testbed Design, Development and Results.
In EURASIP Journal on Wireless Communications
and Networking, vol. 2010.
Ji, Zh., Ganchev, I., O’Droma, M., 2011. “An iWBC
Consumer Application for 'Always Best Connected
and Best Served': Design and Implementation”. In
IEEE Transactions on Consumer Electronics, 57(2).
Oaks, S., Wong, H., 2000. Jini in a Nutshell. O'Reilly
Media. 1
st
edition.
O’Droma, M., Ganchev, I., Chaouchi, H., Aghvami, H.,
Friderikos, V., 2006. `Always Best Connected and
Served` Vision for a Future Wireless World. In
Journal of Information Technologies and Control, IV
(3-4).
O’Droma, M., Ganchev, I., 2007. Toward a Ubiquitous
Consumer Wireless World. In IEEE Wireless
Communications, 14(1).
O’Droma, M., Ganchev, I., 2010. The Creation of a
Ubiquitous Consumer Wireless World through
Strategic ITU-T Standardization. In IEEE
Communications Magazine, 48(10).
Pascoe, B., 1999. Salutation Architectures and the newly
defined service discovery protocols from Microsoft
and Sun (white paper). Salutation Consortium.
http://salutation.org/wp-content/uploads/2012/05/Jini-
UPnP.pdf
Smyth, B., Balfe, E., 2006. Anonymous personalization in
collaborative Web search. In Inf. Retr. 9.
Speretta, M., Gauch, S., 2005. Personalized search based
on user search histories. In WI 2005, International
Conference on Web Intelligence. IEEE/WIC/ACM.
Stavroulaki, V., Kritikou, Y., Darra, E., 2009. Acquiring
and Learning User Information in the Context of
Cognitive Device Management. In IEEE ICC’2009,
IEEE International Conference on Communications.
Stavroulaki V. et al., 2012. Knowledge Management
Toolbox: Machine Learning for Cognitive Radio
Networks. In IEEE Vehicular Technology, 7(2).
Sugiyama, K., Hatano, K., Yoshikawa, M., 2004.
Adaptive Web search based on user profile
constructed without any effort from users. In WWW
2004, 13
th
International Conference on World Wide
Web. ACM.
Teevan, J., Morris, M.R., Bush, S., 2009. Discovering and
using groups to improve personalized search. In
WSDM 2009, 2
nd
ACM International Conference on
Web Search and Data Mining. ACM.
Zhang, H., Song, Y., Song, H.-T., 2007. Construction of
ontology-based user model for Web personalization.
In UM 2007, 11
th
International Conference on User
Modeling.
First International Conference on Telecommunications and Remote Sensing
128