AUTOMATIC CONTEXT DETECTION OF A MOBILE USER

Uta Christoph, Karl-Heinz Krempels, Janno von St

ulpnagel and Christoph Terwelp

Informatik 4, Intelligent Distributed Systems Group, RWTH Aachen University, Aachen, Germany

Keywords:

Mobile user, User context, Mobile device, Conﬁguration space, Context detection.

Abstract:

Mobile devices have obtained a signiﬁcant role in our life providing a large variety of useful functionalities

and features. It is desirable to have an automated adaptation of the behavior of a mobile device depending on

a change of user context to fulﬁll expectations towards practical usefulness. To enable mobile devices to adapt

their behavior automatically there is a need to determine the mobile user’s context.

In this paper we introduce an integrated approach for the automatic detection of a user’s context. Therefore, we

summarize and discuss existing approaches and technologies and describe a service architecture that takes into

account information from the interaction of the mobile device with communication networks and positioning

systems, from integrated sensors, and planned behavior of the user from e.g his calendar or activity list.

Additionally it considers the social network of the user to derive further information about his context and

ﬁnally it takes into account his customs through a behavior model.

1 INTRODUCTION

The omnipresence of mobile devices requires the abil-

ity to adapt the device’s capabilities. Simple imple-

mentations of this feature are already in place in most

common mobile devices. So the user can limit the

usage of the device’s interaction features for example

muting the ring tone, deactivating the network inter-

face radio, etc. To simplify the conﬁguration the set-

tings are often grouped into conﬁguration proﬁles, so

the user only has to select a deﬁned proﬁle and gets

a suitable setup for a situation. In real environments

the context continuously changes and therefore it is

desirable to support automatic detection of the user’s

current context. This would increase the usage com-

fort due to improved adaptability of the device. As

an additional feature the current context of the user in

combination with his current position can be used to

improve the location based services by providing this

information to the service.

In this paper we discuss existing and new ap-

proaches to determine a mobile user’s context. We

analyze their prospects, possible drawbacks and the

technical requirements. The paper is organized as fol-

lows, in Section 2 we illustrate a simple example sce-

nario from the domain of discourse to motivate our

research. In Section 3 we give an overview of existing

work in this area. Section 4 discusses the conﬁgura-

tion space of a mobile device which should be adapted

to individual context. In Section 5 we focus on exist-

ing approaches and introduce some new ideas to de-

termine a mobile user’s context and in Section 6 we

describe the architecture of our idea of an integrated

approach for context detection. Finally in Section 7

we summarize the results and describe our vision for

future development.

2 SCENARIO

The following example scenario motivates the auto-

matic detection of a user context and provides the

reader with an overview of the arising problems.

Imagine a researcher who travels to a distant city

(e.g. Milan, Italy) for a conference. Therefore, the

researcher has to proceed along a chain of activities.

So, he prepares for the trip at Sunday evening, goes

to sleep, gets up at Monday 8:00 am, takes a shower

and has breakfast with coffee, takes the bus at 9:12

am to the train station, arrives there at 9:34 am, waits

until 9:46 am, takes the train to the airport, arrives at

10:34 am, checks in for his ﬂight, gets a coffee, waits

until 12:45 pm, boards the airplane, gets another cof-

fee and lunch, and prepares the slides for his confer-

ence presentation. The plane touches down at 2:30

pm, he leaves the plane 2:39 pm, picks up his lug-

gage at the baggage claim, leaves the airport, takes

the metro, walks from the metro station to the hotel,

checks in at the hotel, takes a shower, takes dinner

189

Christoph U., Krempels K., von Stülpnagel J. and Terwelp C. (2010).

AUTOMATIC CONTEXT DETECTION OF A MOBILE USER.

In Proceedings of the International Conference on Wireless Information Networks and Systems, pages 189-194

DOI: 10.5220/0003030701890194

 SciTePress

at the hotel restaurant, drinks an espresso and joins

to the initial meeting with the other researchers in a

conference room. At this point we leave our fellow

researcher and state that his context changed at least

20 times until this point.

The common researcher is very busy thinking

about his research work, so it is very inconvenient for

him to always keep the context setting of his mobile

device up to date. But it would be preferable that the

mobile device itself adapts automatically according to

the situation. For example, disabling all radio systems

of the mobile device during the ﬂight, switching the

device to silent mode and enabling visual signals (a

ﬂashing light) for incoming calls and messages dur-

ing the meeting, but deactivating vibration alarm be-

cause the device is on the table and not in the pocket,

switching to a louder ring tone and vibrating alert in

the metro, changing the input mode from pen to ﬁn-

ger touch and the output mode to speech synthesis and

symbols while walking or driving. But also travel as-

sistance like information about the exit station just be-

fore arriving there or information about the menu of

the hotel restaurant at dinner time could be provided.

3 STATE OF THE ART

The state of the art for context detection is to use one

kind of specialized sensor either attached to a person

or located in several different places across so called

smart rooms. For example (Karantonis et al., 2006)

uses only one accelerometer to detect movement of

the user. (Kern et al., 2003) use multiple identical

accelerometers to detect activity context information.

In most cases where different sensor types are used,

each type of sensor detects a certain kind of context

for a certain kind of purpose. Only very recent re-

search works use more than one sensor type to rec-

ognize a single context, e.g. (Berchtold and Beigl,

2009), where an accelerometer and a microphone are

used to recognize ”knocking on the table in apprecia-

tion”. Currently most research approaches, as (Ran-

ganathan et al., 2004) and (Brdiczka et al., 2007),

use pattern recognition instead of programmed rules

to derive contexts from sensor information.

One of the current theoretic challenges is to ﬁnd

a standardized description and deﬁnition of context.

(Dey, 2001), for example, gives an operational def-

inition of context whereas (Henricksen and Indul-

ska, 2005) deﬁnes conceptual models for context-

aware applications. A survey of context modeling

approaches is given in (Strang and Linnhoff-Popien,

2004), but often researchers use their own deﬁnitions.

In the last years the uncertainty aspect of context

detection got more and more attention because with

pattern recognition methods the detection is never en-

tirely certain and applications can use a context de-

tection service better if they get real time information

about the quality of the detection. Examples of work

in this research area are (Berchtold and Beigl, 2009)

and (da Silva et al., 2006).

4 DEVICE CONFIGURATION

The device conﬁguration space consists of all possible

conﬁgurations of a given device. To effectively man-

age this conﬁguration space of a device proﬁles are

used which are preset conﬁgurations combined with a

suitable name.

Some of the settings from a device conﬁguration

space are listed here to help the reader get an impres-

sion of conﬁguration possibilities of a mobile device.

For example, as output modes the user can choose

among icon or text based representation of a content,

or even voice synthesis. Furthermore, for the textual

and graphical output the text size and screen direc-

tion (portrait or landscape) can be conﬁgured, and for

voice synthesis the volume and perhaps a few addi-

tional characteristic parameters can be adjusted, e.g.

speaking speed, the pitch and timbre of the voice or

the gender of the speaker. As input modes the user can

choose for example among a microphone with voice

recognition, a touch screen (ﬁnger or pen mode), or

the traditional key-pad. There are also many addi-

tional settings, e.g. microphone and touch screen sen-

sitivity, etc. As general behavior settings the user can

conﬁgure the ring volume, ring tone, screen bright-

ness, earphones volume, and so on.

5 CONTEXT DETECTION

APPROACHES

Today the predominant way to adjust a mobile device

to a changing context is that the user conﬁgures a set

of different proﬁles for his device in advance and later

selects the suitable proﬁle for his current context. If

the context changes very often it is inconvenient to

make such adjustments manually. Thus, it is desirable

to automatically detect the user’s context and adapt

the conﬁguration of the device accordingly. In the

following we brieﬂy discuss several approaches to de-

termine the context of a user.

WINSYS 2010 - International Conference on Wireless Information Networks and Systems

190

5.1 Planning

A very simple approach to context detection is to de-

duce it from the digital calendar, that contains all the

events and trips the user has planned. Each event may

consist of several activities. Since the classical cal-

endar functionality is provided by many mobile de-

vices, especially smartphones, the scheduled events

and trips are already available for further processing.

Therefore, the actual time is used to determine the

corresponding event from the user’s digital calendar.

The actual context of the user can be deduced from

the description of the event, e.g. meeting, ﬂight, etc.

5.2 User Behavior Model

This approach is based on the assumption that user

behavior is to some degree consistent. So, for ex-

ample the user goes to bed at almost the same time

every day and he has lunch at a speciﬁc time. From

this we derive that user behavior can be deﬁned as

the correlation of time, space, activity and context.

Consequently, the user behavior can be deduced from

his past, by observing former behavior tuples and

matching them to the current situation (Sama et al.,

2008) (da Silva et al., 2006).

5.3 Radio Signals

Mobile devices are able to interact with several

communication networks and there exist several ap-

proaches to derive the precise position of a mobile

device from the interaction with these communication

networks, like GSM (Kunczier and Anegg, 2004),

UMTS/3G (Kos et al., 2006), and WLAN (Jan and

Lee, 2003) (Krempels and Krebs, 2008); all of them

providing localization estimations with acceptable ac-

curacy.

The context of a mobile user can be derived from

his position. For that purpose the geographical posi-

tion must be processed by a GIS (Geographical Infor-

mation System) to obtain additional information about

the user context, e.g. street name, building name, or

name of a point of interest. From the type of build-

ing, e.g. train station, the user context can often be

derived directly. If the localization accuracy is high

the user context can also be determined in a more de-

tailed manner, e.g. if the user is in a restaurant or in

the business lounge inside the train station. This ad-

ditional information can be used to detect time shifts

or schedule deviations in conjunction with plan based

context detection, as discussed in section 5.1.

5.4 Sensors

The majority of todays mobile devices have a number

of different sensors at their disposal. Most of them

are only built in to serve for one or two special appli-

cations but they also can help to determine the user’s

context.

Accelerometer. The accelerometer not only mea-

sures acceleration of the device but also how the de-

vice is tilted. It is possible to match the movement

pattern of the device to certain situations, such as the

device is in the user’s hands, laying on a desk or held

to the user’s ear. While in the user’s pocket it is also

possible to decide for example if he is walking, run-

ning or sitting down (Karantonis et al., 2006).

Camera. The camera’s use for detecting user con-

text is limited because most of the time when auto-

mated context detection is needed the mobile device is

in the user’s pocket, but sometimes it can be used for

image recognition (Luley et al., 2005). For example

it can help to determine wether the mobile device is

in a pocket or not. It also helps to recognize the light-

ing conditions of the surrounding even if you use this

information only to adapt the brightness of the dis-

play. It is possible to use 2D barcodes and the camera

to get information about a location that is equipped

with them. The public transportation system in the

city of Berlinfor example is using them on bus stops.

If people scan the bar code they get a hyperlink to a

timetable with real-time bus schedules.

Microphone. The microphone has many possibili-

ties to detect the user context. It is easy to measure the

volume of the background noise and adapt the volume

of the audio alarm or the speakers accordingly. It is

also possible to identify places based on the character-

istic background noise. For example (Ma et al., 2003)

were able to recognize with background noise only if

they are located in an ofﬁce, at football match or at the

beach. They were able to differ between ten location

with an overall accuracy of 91.5%. It should also be

possible to use voice recognition to detect how many

people are in the surroundings and who they are.

Compass. Some mobile devices have a built in elec-

tronic compass, because GPS can only give a position

but no direction information. The electronic compass

gives information about the movement of the user

and can be used to compliment the data from the ac-

celerometer and GPS.

AUTOMATIC CONTEXT DETECTION OF A MOBILE USER

191

5.5 Near Field Community Interaction

We see a promising area for further development in

the sharing of context information with other users.

Since applications for instant messaging and social

networking became available for many mobile de-

vices a user can share all the information with respect

to its position and context with the people connected

to him in his social network or through his instant

messaging contact list.

To derive the user’s actual mobile context from the

context of the connected people in his social network

we assume that he has either a similar or an identical

context as his colleagues and friends.

We suggest that the mobile context of the user

could also be derived with the help of a majority vote

from the actual contexts of his near ﬁeld community.

A straight forward way to implement a near ﬁeld com-

munity service is to advertise the user if one of his

colleagues or friends is in the same cell of the mobile

network, the same WLAN access point range, or even

in the same train, building, plane, etc. After the adver-

tising the mobile device can accept the new context as

its current context or can use it to determine a simi-

lar context by combining the advertised context with

other relevant knowledge.

6 INTEGRATED CONTEXT

DETECTION ARCHITECTURE

As mentioned in Section 3 existing approaches only

consider a certain aspect or technology to discover a

user’s context. Thus, we propose that a combination

of several approaches is desirable to offer an overall

adaptation to a mobile user’s context. We also pro-

pose that detection of a mobile users’s context should

be provided by the mobile device as a service, so that

all applications can have access to the context infor-

mation and can adapt their behavior accordingly. In

Figure 1 we show an architecture of a Context Detec-

tion Service which comprises all aspects of context

detection mentioned above and computes a suitable

context that is provided as a context description for

other applications for further processing.

The integration approach of the Context Detection

Service is based on a three layer architecture. The

data or signal source layer consists of the available

sensors, radio network interfaces, the built in clock,

and or even the connection interface to the user’s com-

munity. The source layer provides information in

form of data or signals which comprise the informa-

tion layer. This information needs to be transformed

to deﬁned knowledge so it can be processed by the

Context Detection Service. The transformation of in-

formation to knowledge is done by additional Infor-

mation Processing Services which match the informa-

tion events from the source layer to knowledge tags

with the help of suitable patterns deﬁned in the Pat-

tern Repository.

The most signiﬁcant criteria to determine one’s

mobile context seem to be his geographical location,

the current time, and his planned activities. On one

hand all these criteria have a direct impact on ones

context and on the other hand the corresponding tech-

nologies, like calendars, location based directories,

and time dependend schedulers are already part of

prevalent mobile applications.

For the detection of the geographical location,

technologies such as GPS or UMTS, GSM, and WiFi

networks are available to easily determine geograph-

ical locations and thus it should be considered as one

of the main parameters for the context detection ser-

vice. But the coordinates of a geographic location

alone might not be sufﬁcient to determine the context,

since it does not provide any descriptive information

regarding the corresponding place. So the detected

location must at least be enriched with more details

from a map or yellow-pages service to become more

useful in terms of context detection, e. g. by adding

descriptive knowledge tags like cinema, restaurant,

airport, etc. to the detected location.

An identiﬁed description tag of the surroundings,

e.g. airport or even an ofﬁce inside the airport, has

a static character with respect to the temporal dimen-

sion, since the context of a user can change over time

even if he stays at the same place. In an ofﬁce in-

side the airport for example the context might be of-

ﬁce work in the morning, tutorial in the afternoon, and

in between several coffee breaks. One possibility to

draw more precise conclusions about the actual user

context is to make a correlation of a description tag

with the planned acitivities of the user. Therefore, we

take into account appointments from the user’s cal-

endar and planned activities from his TODO lists in

combination with the current date and time to detect

his activity context for the given location description

tag. Another possibility to reﬁne the actual user con-

text can be achieved by using noise patterns recorded

by the microphone to distinguish among e.g. quiet

ofﬁce work, a coffee break, or a presentation, etc.

An additional way for context detection are mov-

ing patterns of the user (e.g. riding a bus, a train or a

bycicle, walking, running, not moving etc.) that can

be detected with the help of sensors available in the

device, e.g. accelerometer, compass, or even a micro-

phone. Those patterns can be used alone or even in

combination with other identiﬁed context description

WINSYS 2010 - International Conference on Wireless Information Networks and Systems

192

Information

Processing

Service

Information

Processing

Service

Radio Signals

(GPS,UMTS,WLAN)

Accelerometer

Camera

Compass

Microphone

Context Detection

Service

User

Location

Brightness

Movement

Pattern

Code

Clock

Activity

Tag

Calendar

Activity

List

Near Field

Community

Context

Noise

Pattern

Noise

Tag

Context

Repository

Pattern

Repository

Information

Layer

Knowledge

Layer

Information

Processing

Service

Context

Tag

Conﬁguration

Service

Conﬁguration

Repository

Device

Conﬁguration

Context

CDS

API

Date / Time

Data / Signal

Layer

Movement

Tag

User

Behavior

Recognition

Brightness

Tag

Location

Tag

Light

Sensor

Figure 1: Context Detection Service (CDS) Architecture.

tags to derive a certain device conﬁguration.

The Context Repository containts context deﬁn-

tions consisting of a context name and a list of re-

quired context description tags provided by informa-

tion processing services. Everytime when all the re-

quired description tags of a deﬁned context arrive in

a well deﬁned time interval at the Context Detection

Service, the corresponding context is matched and

provided as output for the Conﬁguration Service or

for context sensitive applications on the Context De-

tection Service Application Programming Interface

(CDS API). The context descriptions provided by the

Context Repository consist of a list of description tags

< T

>, . . . , < T

> derived from the signals of the

input sensors of the source layer describing a well de-

ﬁned user context Context

. A list of context detec-

tion rules from the Context Repository could be:

(< T

1,1

>, . . . , < T

1,m

>) ⇒ Context

(< T

2,1

>, . . . , < T

2,m

>) ⇒ Context

. . . . . .

(< T

n,1

>, . . . , < T

n,m

>) ⇒ Context

The Conﬁguration Service selects a suitable de-

vice conﬁguration description from the Conﬁgura-

tion Repository for a detected context and applies

it to the mobile device to adapt its behavior to the

corresponding context of the mobile user. For ex-

ample for a given Context

a suitable conﬁgura-

tion will be Con f iguration

, for Context

it will be

Con f iguration

, and so on.

Context

> ⇒ Con figuration

Context

> ⇒ Con figuration

. . . . . .

Context

> ⇒ Con figuration

Finally, the Context Detection Service can reﬁne

a context description rule by comparing the larger set

of the identiﬁed context description tags with the de-

scription tags of a certain rule that deduced the actu-

ally detected context. However, these tags character-

ize also a user’s behavior providing the possiblity to

update or renew the deﬁntion of the context rules in

the Context Repository.

7 SUMMARY & OUTLOOK

In this paper we discussed different approaches to au-

tomatic context detection and proposed an integrated

service architecture which can combine information

from the different approaches. We think this is nec-

essary to gain a precise view on the user’s context

which is the main preposition to developing context-

aware mobile applications. As a fundamental for an

integrated service the context information from the

data or signal source layer is transformed into knowl-

edge which can be interpreted by the context detec-

tion service and then be provided to mobile applica-

tions running on the device or as basis for the device

conﬁguration. The selection criteria for the most suit-

able context detection approaches should be on the

AUTOMATIC CONTEXT DETECTION OF A MOBILE USER

193

one hand pervasiveness of the underlying technology,

e.g. UMTS and WLAN networks of the radio sig-

nal approach, and on the other hand the obtainable

accuracy for the derived user context. Todays mo-

bile environments are characterized by highly avail-

able, pervasive mobile communication networks, mo-

bile calendar based job itineraries, and mobile devices

with high computational power. Under these precon-

ditions it is recommendable to combine it with radio

based approaches in UMTS/GSM networks for the

detection of the user’s current geographical region.

A rough context of the mobile user can be deduced

therefrom in combination with the event and activ-

ity list from his calendar. This rough context can be

reﬁned with the help of the radio based approach in

WLAN infrastructures that the user will enter, cross,

or leave. To determine the context of a mobile user

in an automated way it is necessery to process knowl-

edge from several sources and services. Thus, there

is need for common ontologies for the description of

the context concept, knowledge description tags char-

acterizing a deﬁned context, and even the events pro-

vided by the signal layer. The next implementation

steps for the proposed context detection architecture

are the deﬁnition of a suitable context ontology and

the interaction design of the discussed components of

a Context Detections Service.

REFERENCES

Berchtold, M. and Beigl, M. (2009). Increased robustness

in context detection and reasoning using uncertainty

measures - concept and application. In Proceedings

of the European Conference on Ambient Intelligence

(AmI’09), Salzburg, Austria. Springer.

Brdiczka, O., Crowley, J. L., and Reignier, P. (2007). Learn-

ing situation models for providing context-aware ser-

vices. In HCI (6), pages 23–32.

da Silva, B. C., Basso, E. W., Perotto, F. S., C. Bazzan,

A. L., and Engel, P. M. (2006). Improving reinforce-

ment learning with context detection. In AAMAS ’06:

Proceedings of the ﬁfth international joint conference

on Autonomous agents and multiagent systems, pages

810–812, New York, NY, USA. ACM.

Dey, A. K. (2001). Understanding and using context. Per-

sonal Ubiquitous Comput., 5(1):4–7.

Henricksen, K. and Indulska, J. (2005). Develop-

ing context-aware pervasive computing applications:

Models and approach. Pervasive and Mobile Comput-

ing, In, 2:2005.

Jan, R.-H. and Lee, Y. R. (6-9 Oct. 2003). An Indoor Ge-

olocation System for Wireless LANs. Parallel Pro-

cessing Workshops, 2003. Proceedings. 2003 Interna-

tional Conference on, pages 29–34.

Karantonis, D. M., Narayanan, M. R., Mathie, M., Lovell,

N. H., and Celler, B. G. (2006). Implementation of a

real-time human movement classiﬁer using a triaxial

accelerometer for ambulatory monitoring. Informa-

tion Technology in Biomedicine, IEEE Transactions

on, 10(1):156–167.

Kern, N., Schiele, B., and Schmidt, A. (2003). Multi-sensor

activity context detection for wearable computing. In

In Proc. EUSAI, LNCS, pages 220–232.

Kos, T., Grgic, M., and Sisul, G. (June 2006). Mobile

User Positioning in GSM/UMTS Cellular Networks.

Multimedia Signal Processing and Communications,

48th International Symposium ELMAR-2006 focused

on, pages 185–188.

Krempels, K.-H. and Krebs, M. (2008). Directory-less

WLAN Indoor Positioning. In Proceedings of the

IEEE International Symposium on Consumer Elec-

tronics 2008, Vilamoura Portugal.

Kunczier, H. and Anegg, H. (5-8 Jan. 2004). Enhanced Cell

ID Based Terminal Location for Urban Area Loca-

tion Based Applications. Consumer Communications

and Networking Conference, 2004. CCNC 2004. First

IEEE, pages 595–599.

Luley, P. M., Paletta, L., and Almer, A. (2005). Visual ob-

ject detection from mobile phone imagery for context

awareness. In MobileHCI ’05: Proceedings of the 7th

international conference on Human computer interac-

tion with mobile devices & services, pages 385–386,

New York, NY, USA. ACM.

Ma, L., Smith, D., and Milner, B. (2003). Environmen-

tal noise classiﬁcation for context-aware applications.

In Mar

ık, V., Retschitzegger, W., and Step

ankov

a, O.,

editors, DEXA, volume 2736 of Lecture Notes in Com-

puter Science, pages 360–370. Springer.

Ranganathan, A., Al-Muhtadi, J., and Campbell, R. H.

(2004). Reasoning about uncertain contexts in perva-

sive computing environments. IEEE Pervasive Com-

puting, 3(2):62–70.

Sama, M., Rosenblum, D. S., Wang, Z., and Elbaum, S.

(2008). Model-based fault detection in context-aware

adaptive applications. In SIGSOFT ’08/FSE-16: Pro-

ceedings of the 16th ACM SIGSOFT International

Symposium on Foundations of software engineering,

pages 261–271, New York, NY, USA. ACM.

Strang, T. and Linnhoff-Popien, C. (2004). A context mod-

eling survey. In In: Workshop on Advanced Context

Modelling, Reasoning and Management, UbiComp

2004 - The Sixth International Conference on Ubiq-

uitous Computing, Nottingham/England.

WINSYS 2010 - International Conference on Wireless Information Networks and Systems

194