A Survey of Context Awareness for monitoring human daily routine

behavior in the smart environment

Khamla Non-Alinsavath1, Lukito Edi Nugroho2, Agus Bejo3
Department of Electrical Engineering and Information Technology
Faculty of Engineering,
Universitas Gadjah Mada
Yogyakarta 55281, Indonesia
1 [email protected], [email protected], [email protected]

Abstract There are a number of context aware systems have

been proposed. Context awareness system plays an important
role in improving the performance with the technology of
context awareness that it leads us to solve the problem and
build systems especially for the smart home technology. This
paper reviews the literature of context awareness based on
existing works that have been published in journals about
context awareness for smart home monitoring as a control
system that needs to innovate using motion sensor technology,
the system can learn the pattern in its users behavior with
respect location and activities (walking, lying, working with
electronic devices) system along with the temporal data like
time and date in the system to set the time for controlling
human behaviors. The system can be measured with motion
sensor for improving their quality of their daily routine life
and decreasing energy using rate is the outcome of the smart
home technology.
Keywords: context-awareness, monitoring human behavior,
smart home, ontology, motion sensor.

I.
INTRODUCTION
With the moment time we spend talking about
learning human habits and context awareness becomes
significant for the building block of smart application that
constitutes smart space. The system for monitoring human
behavior can be used in many applications like the wireless
sensor applications to process data checking, noticing,
controlling, and analysis of result. In order to reduce
energy use in case the user forgets to turn off the lights, or
if we have to take time to walk to the installation of the
switch control board, this period of time causes long term
energy consumption. The information is used to
communicate the situation for an entities that is considered
relevant between user and application that we called
Context. Context awareness means that one is able to use
context information that can be used to interrupt explicit
operation, to interpret context information, to use context
information and to adapt the functionality of a system to
the current situation[1]. This information can be used to
tag information as collected in the field and to enable
selective response such as triggering alarms or retrieving
information relevant to the task as hand. It is absolutely
one of the highest technology in it today. Compared with
virtualization and cloud computing, context is a

technological capability that will implement to transform

the technology world within this era.
In the late 1980s, there was a period of beginning
activity of context computing, with a lot of interests and
expectations. Very few of context aware computing was
met the interest. However, the activity seems to be
increasing dramatically. Nowadays, to overcome new
challenges and requirement found in context aware
computing. Many researchers have made efforts to design
and implement network, user infrastructure and
middleware which are enough to provide users with
context aware services. Many journals and conferences
have published researches related with context aware
computing and many people are interested in context
aware system. So this a good opportunity for reviewing
analysis, since it has been over a year since many papers
were published[2]. The goal of this paper is to look back
and review the literatures that were published in journals
that there is still has the difficult to compare the articles in
different journals, the objectives of this review are:
- To summarize research papers that related with
context aware computing
- To provide the architecture and modelling of
context aware
- To propose the human behavior monitoring
methodology based on context aware system
- To highlight certain area that needs further
research
- To discuss and give suggestion for context aware
system based on literature review
This review paper is structured as follow details.
The section 2 has introduced the existing work that related
with context aware system. Section 3 reviews various
definitions of Context awareness. The system architecture
for context aware computing is presented by section 4,
section 5 presents Ontology-based context modelling and
reasoning, the discussion and suggestion is contained in
section 6. Finally, this research concludes the paper with
the outcome and result as the section 7.

II. RELATED WORKS (E XISTING WORKS)

Many context aware computing projects in the
past years have studied feature-oriented approaches to
context aware system. In this section introduces some
typical systems for context awareness.
INRIA in France proposed a general infrastructure based on
contextual objects to design adaptive distributed
information systems in order to keep the level of the
delivered service despite environmental variations.
AT&T Laboratories Cambridge in UK [12]
introduced a platform for context aware computing which
enables applications to follow mobile users as they move
around a building. The platform is particularly suitable for
richly equipped, networked environment. Users are required
to carry a small sensor tag, which identifies them to the
system and locates them accurately in three dimensions.
Arizona State University presented reconfigurable
context-sensitive middleware (RCSM) which made use of
the contextual data of a device and its surrounding
environment to initiate and manage ad hoc communication
with other device. Jason Hong and his colleagues proposed
an open-infrastructure approach the encapsulation
underlying technologies into well-established service that
can be used as a foundation for building application. [12]
The European Smart-Its projects proposed a
generic layered architecture for sensor based context
computation, providing a programming abstract that
separate layers for raw sensor data, features extracted from
sensor and abstract contexts derived from sensor
III.
CONTEXT AND CONTEXT AWARENESS
Definition of Context and Context awareness
With the high technology and development of the
social daily life, many researchers in the different fields
and period of time may have different understanding of the
context, so far there is still no clear definition about this.
Now the most authors define through enumerating
examples, the first definition of context is from Schilit
Adams in 1994[3] which introduce the concept of context
is related to the location, nearby person, hosts or objects, as
well as changes of them over the time. Brown, Bovey said
that information such as location, time, season,
temperature and so forth into several aspects of users
context in[4][5]. Context information includes the
capacities of the mobile device, the characteristics of the
network connectivity and user specific information.
Furthermore, emotional state, attention focus. Orientation,
date and time of day, object and people in the users
environment can be viewed as the aspect of context.
However, there is no an agreement the weather to use
users environment, the application environment or other
areas.
Compared to the above definition methods,
enumerating example is applied in practice more easily and
can believe that the three important aspects of context are
where you are, who you are with, and what resources are
nearby. Because of this, the authors divide context into
three parts:
- Computing context: available processors, devices
accessible for user input and display, nearby resources
3.1

such as printers, displays, and workstations, network

capacity, connectivity, costs of computing and
communication, and bandwidth.
- User context: users location, collection of nearby
people, user profile and social situation
- Physical context: lighting, temperature, noise and
humidity level, traffic conditions
In order to achieve a better understanding more about
context, Chen and Kotz add time context such as time of a
day, week, month, season of the year, and time zone. Time
information like a contexts history that adds to this
definition.
As we can see the above definition that we can
conclude context is the information that can be used to
characterize the situation for an entity (persons, places, or
objects) that is considered relevant to the interaction
between user and application [1]. And defines a context
from the view of user center: computing context, user
context, physical context, time context, and social context.
According to the definitions of context that
mentioned above statements. Then the researchers define
the Context awareness, which is the ability of a program or
computing device to sense and act upon the information
about its environment such as location, time, and user
identities. This information can be used to tag information
as collected in the field and to enable selective response
such as triggering alarms or retrieving information relevant
to the task as hand. It is absolutely one of the highest
technology in it today.
Context awareness means that one is able to use
context information that can be used to interrupt explicit
operation, to interpret context information, to use context
information and to adapt the functionality of a system to
the current situation. Context awareness is the ability of a
program or computing device to detect, sense, interpret,
act and respond to an aspect of the environment such as
location, time, temperature, or user identity or from the
perspective of adapting an application to context. [1]
3.2

Context and Context awareness Categorizations

Context categorization can discover the possible
context easily simplify the context manipulation and can
be helpful in providing quality context information. Two
possible broad categorization viewpoints are:
- Conceptual viewpoint it descripts contextual space in
terms of the actors, the actions and the relationships
between them.
- Measurement viewpointit contains continuous context,
enumerative context, state context, and descriptive context,
physical and virtual context.
A Context could be classified as static or dynamic
context. Another similar meaning is continuous and
discrete context. Some researchers do the categorization of
the context of the following points of view: internal and
external context, material context, or social context,
physical or virtual context, real time and unreal time
context, natural context, technology context [6][7].

Although the context information is varied, some

types are more important than others, such as location,
identity, time and activity that follow with these answers of
5Ws: who, what, when, where, and why, which is
minimum necessary set of context information in the
ubiquitous computing environment, but also can be a
source of other context information.
According to the influence of system behavior.
Context awareness categorization can be divided into
active and passive context awareness
- Active context awareness: an application
automatically adapts to discover context through changing
the applications behavior.
- Passive context awareness: an application
presents the new or update context to an interested user or
make the context persistent for the user to retrieve later.
Context awareness also can be broadly classified
into the following areas: self-contained and infrastructure
based context awareness. If considering from information
acquisition tools, it also consists of hard context awareness
which can be obtained by hardware sensors measuring
context awareness, and soft context awareness which
knowledge inference based on the information repository
of the context awareness.
3.3

Context awareness system

There are three main components of the context
awareness system which are concerned. To begin with
system gathers context information available from the user
interface, pre-specified data or sensor and adds it to a
repository, the system uses context and gives a reaction,
and reveals the appreciate context to the user.
A.
Context Acquisition and sensing.
Context acquisition is the most basic level of context
awareness. In general, there are three ways to acquire
context in the following:
- Sensed context: Environment information and
user physical information can be acquired by physical or
software sensors. And the user interaction habit, and
interactive historical record are gained mostly by logical
sensor from the host.
- Derived context: This kind of contextual
information can be computed on the fly.
- Context explicitly provided: For example, users
preferences can be obtained when they are explicitly
communicated to the requesting application.
Though a lot of context information is in the system,
we pay attention to the identity, location, activity and time
than others. There are two different ways of approach to
sense the location information. The first approach is client
based, just like Global Positioning System, which is the
most widely used and accurate methods, Infrared and
Radio Signals, RFID, Wi-Fi, Bluetooth, and the indoor
GPS signal generator. The other one receives the location
information from the network of a certain area. Some
identity information can be given to the context awareness
system explicitly, such as a users gender. Time

information is easy to obtain from the built-in clock of the

computer. In addition, high-level context information such
as the users current activity is a big challenge. One
approach to acquire is machine vision, which is based on
camera technology and image processing. Another possible
approach is to consult the users calendar directly to find
out what the user is supposed to do at certain times. A third
method is to use artificial intelligence techniques to
recognize complex context by combining several simple
low-level sensors [8].
B.

Context representation and application

In order to depict the user surroundings and

situation, the system uses context modeling and
representation in the context awareness computing.
Context modeling must be formulated to provide a
resource for applications to avail of. There is almost not a
problem for all kinds of context modeling independently,
but it is a challenger to handle all kinds of context in a
unified and general context modeling in today's technology
conditions. According to the current situation, the unified
context modeling can be divided into two levels: Different
context uses the same data structure to express mode, such
as Key-Value Models, Markup Scheme Models, etc. It also
can support the semantic unity, such as ontology Based
Models, Graphical Models, etc.
How to effectively use context is an important issue,
applications can be roughly broken down into the
following categories:
- Proximate selection
- Automatic contextual reconfiguration
- Contextual information and commands
- Context-triggered actions
There are two more applications:
- Metadata tagging: Context information, along
with existing pieces of information, gives us more implicit
information.
- Terminal adaptively: Because of small and
cluttered user interface in mobile devices, context
information can help to increase usability presenting the
user with only relevant information [9].
C.
Context storage and retrieval
Raw context data as well as filtered and fused
context information can be stored in some places for
further retrieval. Just as passive context awareness is
defined which an application presents the new or updated
context to an interested user or makes the context
persistent for the user to retrieve later.
Context should be well organized into various
data structures, such as table, object, tree, graphic, etc.
Moreover, context storage is necessary to get historical
records in order to make processed context information
out. The architecture for context storage can be either
centralized or distributed.
The effective and reliable storage of context
information related to many problems, such as
instantaneous, the relevance between context and objects,
context retrieval requiring with range rather than single

conditions, etc. In addition, it associates with the context

modeling and representation.
IV.

THE SYSTEM ARCHITECTURE

FOR CONTEXT

AWARE COMPUTING

A context aware computing architecture consisting of

three logical components: device agents, active maps, and
user agents. Each of these components as based on the
dynamic environment publish-subscribe communication
model.
- Device agents are dynamic environment servers
specialized to monitor and provide information for
workstation, printer, palmtops or other particular devices.
The agent records device-specific information in a device
object, when information about the device change. Device
agents manage device-objects that are published on the
active map or the appropriate user agent if they are owned,
one important attribute contained in the device-object is
located.
In a context aware system, there may be many kinds of
device agents, for example:

how to get there. Rooms are in a building and buildings are

contained in a region. The active map supports functions to
describe the hierarchy, and the comparing of two locations
to see if one is contained within another.
- User agent: In a context-aware system each user
has a collection of personal information that might include,
among other things, their preferences, the characteristics of
devices currently in-use by that user, and the time and
location of the most recent interaction. In our system, each
users information is logically centralized in a user agent.
This provides a well-known place for a users own
applications, as well as applications run by others, to find
out about the user. This section describes the user agent
from a functional viewpoint and proceeds with a
presentation of the agents distributed organization and
protocols that gives applications some degree of
functionality while disconnected from other system
components [3][10].

Name:
The name of the device, e.g., Badge: 38.
Location: The location of device.
Agent:
Connection information for the functional
network interface of the device if one exists.
DE:
Connection information for the Dynamic
Environment network interface for the device
Table 1: Device Agent Global Attribute
Fig 2: User Agent organization

- Active map service (AMS) publishes dynamic

environment object having a location as well as other
geographic, each AMS supports a set of locations called a
region. For a particular region, the active map is the
meeting place where clients can find each other as well as
relevant located objects. The AMS acts as a locationoriented directory service: location serves as the key that
clients use as a first step in determining most of users
context. In addition to provide a dynamic environment of
located-objects, the AMS also manages information about
the relationships that exist between locations.

Fig 3: Context Aware Computing Architecture

V.

ONTOLOGY -BASED CONTEXT MODELLING AND

REASONING

Fig 1: Active Map containment hierarchy and graph of Location.

From above figure we can see that an objects in a

room it is also in the building containing that room and
knowing how far it is from one place to another as well as

In this section, we discuss the use of ontology an

OWL; and describe the ontology based approach for
modeling context
5.1 Context aware home scenario
A context aware home is a smart home environment
which is equipped with various network sensors/actuator
devices such as, x. 10 curtain sensor, etc [10].
For example: Julia wants to have a barbeque dinner
outside the house tonight. She quickly consults her meal
arrangement agent which is able to advise her whether it is
possible. The meal arrangement agent consults the
networked fridge for available food items based on their
food preferences for all family member queries on an

external weather service for the weather condition tonight.

After a while, she realizes that the barbeque dinner is not
possible due to weather condition. After dinner, when Julia
sits on the sofa in the living room and turn on the TV, the
light begins to dim .
At 23:00. Mrs. Park leaves the living room and enters the
bedroom, the air condition and TV in the living room are
turned off. The light in the bedroom is turning on, and just
the brightness to low.
At noon, Mr. Lee enters the living room; the room
temperature is 30 C, the air conditioner will automatically
turn on to decrease the temperature. At the same time, the
TV is turned on and the news report channel is turned[8] .

Rule 4: IF (LocatedIn (presentation, room) AND

Notuse(projector))
THEN (LocatedIn (discussion, room)) [5][7][11]
Start
False

True

Change of subareas
brightness?

Activity
inferring

Retrieved the
brightness
requirement

True

Compare the
requirement and
current brightness

Changed?

End

True

Changed?

False
Deliver relevant data
to allocation model

Fig 4: Smart Space ontology

A. The rule context reasoning approach

Reasoning is used to decide what response should
be made when any change of context occurs. Whenever a
context item is added, changed or deleted, the reasoning
approach will be activated. Because all contexts in the
model are relevant to the activity context except for the
brightness of Subarea, if the detected change is not about
the brightness of sub areas, then activity-inferring process
will start to deduce the living activity, if the activity stays
the same, the reasoning will end, if the detected change is
about the brightness of subareas, the brightness requirement
will be retrieved with the current activity context, and then
the current brightness will be compared with requirement to
decide whether any adjustment of lighting is needed, if it is
needed, the allocation model will be activated with the
current brightness and the requirement as input.
We define some rules based on common sense for
activity inferring. The following samples show a partial set
of these rules. To illustrate these rules more generally, we
just represent them in the form of predicate logic. For
instance, the item LocatedIn (Jim, Room406) indicates Jim
is now located in Room 406.
Rule 1: IF (LocatedIn (person, room) AND LocatedIn
(none, room))
THEN (Participate (person, room) AND LocatedIn
(discussion, room))
Rule 2: IF (LocatedIn (person, discussion) and LocatedIn
(discussion, room)
THEN (participate (person, discussion))
Rule 3: IF ((LocatedIn (discussion, room) AND
((LocatedIn (computer, room) AND Use (computer)) OR
(LocatedIn (projector, room) AND (Use (projector)))),
THEN (LocatedIn (presentation, room))

Fig 5: the flowchart of Reasoning.

To decide whether the current brightness satisfies

the requirement, the brightness of each subarea should be
compared with corresponding item in the requirement
profile. The current to the corresponding required interval.
5.2 The performance of the context aware for smart
environment.
We have implemented the automatic control system to
reduce the use of energy when the system could to turn
on/off the lighting as the smart system by capture human
movement using motion sensor such as PIR (Passive
Infrared) sensor the system can capture human motion and
behavior which according to activities that will send alerts
in the network range which working with electronic
prototype platform to read data from sensors and measure
the entities (people, objects, places). Temperature and
humidity sensors are used to monitor the environment and
make sure that the temperature and humidity is suitable
inside the room

Fig 6: Coding to Capture human movement behavior by PIR

sensor

methodology and real application are needed to solve with

the above issues

Fig 7: the output of capturing by PIR sensor when we moved in / out

of the sensor range read from Aduino Uno platform.

VI.
DISCUSSION AND SUGGESTION
A lot of researchers have been concerned about
context aware system and context awareness and the
systems are still developing, many projects have been
published increasingly. However context aware systems are
not fully implemented in real life. Much research has
focused on the concept and research layer. The scope of
application or service in the most articles is limited to small
regions such as laboratory, school, hospital, smart room and
so on. Furthermore the strategic alternative model for
gaining the revenue by using context aware systems are
very few, technology related to the context aware system
are merely standardized. The architecture, method of
context modeling, inference algorithm, network
implementation and device of users on each project are
different. Moreover middleware, application and service
make use of different level of contexts and adapt the way
they behave according to the current context. Therefore,
according to the level and type of context along with the
goal of context aware system, the context modeling
methodology, inference algorithm, agent structure and
interaction method of agent are changed. Although the
interaction between agents in the same middleware and
cooperation and components of the same architecture are
investigated, a standard for interaction, cooperation and
operation in the different context aware application has not
been studied. Finally, we found that the context sensing,
context managing and context aware service and
application are included in ubiquitous computing
environment. Context aware computing relies on many
technological developments. One of them is multimodal
user interface technology. In order to improve the
performance of user interface naturalness and high
efficiency with accurate and interact with the computer
system by speech recognition, gesture input, feeling
feedback, and some new technology
Context awareness needs to deal with
environmental context just like the way of human does,
which require seriously studies of human cognition method,
process and proposes suitable cognition model for computer
using. Moreover, when the user is in multiple tasks, how to
distinguish the user cognition model at present and switch
timely is also a problem
To quickly provide suitable service to users,
context aware systems should manage variety, diversity and
numerous amount of context. Moreover, previous research
suggests a concept to control these problems, therefore a

VII.
CONCLUSION.
This paper is based on a literature review paper on
context awareness the using keywords index and articles
title search and we reviewed for the concept and application
that related with context aware computing. The review
paper introduced the context aware system concept,
network infrastructure, middleware and application. We
found that the activity that related with context aware
computing seems to be increasing dramatically and can be
classified into designing and implementing network, which
are enough to provide users with a context aware system,
concept and research being the base of context aware
application and services. Context awareness can be used for
application more humanization and more convenient,
especially in the pervasive computing field due to the users
demand and the rapidly changing environment, using
context information is more valuable.
This can be summarized that different
methodologies like psychology, cognitive science, and
human behavior, sensor technology or monitoring system
could implement context aware computing, different
behavior and environment are also having difficulty to
obtain the exact information that can be connected to each
other. So this system is able to detect and resolve context
conflict, different context and not only be dedicated to
people who would like to monitor their daily routine but
also for those who care and attention as a health care tool
that can help checking or monitoring their behavior with
smart system and smart application.
REFERENCES
W. Liu, X. Li, and D. Huang, A survey on context awareness, in
Computer Science and Service System (CSSS), 2011 International
Conference on, 2011, pp. 144147.
[2] J. Hong, E. Suh, and S.-J. Kim, Context-aware systems: A
literature review and classification, Expert Syst. Appl., vol. 36, no.
4, pp. 8509 8522, 2009.
[3] B. Schilit, N. Adams, and R. Want, Context-Aware Computing
Applications, in Mobile Computing Systems and Applications,
1994. WMCSA 1994. First Workshop on, 1994, pp. 8590.
[4] R. KOUSHAEIAN, AN ONTOLOGY AND CONCEPTUAL
GRAPH BASED BEST MATCHING ALGORITHM FOR
CONTEXT-AWARE APPLICATIONS,
MIDDLE
EAST
TECHNICAL UNIVERSITY, 2011.
[5] C. Li, L. Suna, and X. Hua, A context-aware lighting control
system for smart meeting rooms, Syst. Eng. Procedia, vol. 4, no. 0,
pp. 314 323, 2012.
[6] T. Winograd, Architectures for context, Hum.-Comput. Interact.,
vol. 16, no. 2, pp. 401419, 2001.
[7] T. Gu, X. H. Wang, H. K. Pung, and D. Q. Zhang, An ontologybased context model in intelligent environments, in Proceedings of
communication networks and distributed systems modeling and
simulation conference, 2004, vol. 2004, pp. 270275.
[8] W. N. Schilit, A system architecture for context-aware mobile
computing, Columbia University, 1995.
[9] Y. Raja Vara Prasad and P. Rajalakshmi, Context aware building
energy management system with heterogeneous wireless network
architecture, in Wireless and Mobile Networking Conference
(WMNC), 2013 6th Joint IFIP, 2013, pp. 18.
[10] N. D. Rodrguez, M. P. Cullar, J. Lilius, and M. D. Calvo-Flores,
A Survey on Ontologies for Human Behavior Recognition, ACM
Comput Surv, vol. 46, no. 4, pp. 43:143:33, Mar. 2014.
[11] S. Bechhofer, OWL: Web Ontology Language, in Encyclopedia
of Database Systems, L. LIU and M. T. ZSU, Eds. Springer US,
2009, pp. 20082009.
[12] J.-W. Chang and H.-J. Lee, Context-Aware Architecture for
Intelligent Application Services in Ubiquitous Computing, in
Semantic Computing, 2007. ICSC 2007. International Conference
on, 2007, pp. 275281.
[1]