THE MOTIVE CONCEPT

Enabling Mobile Terminals to Act as Sensors

Michael Masikos, Konstantinos Demestichas, Evgenia Adamopoulou

School of Electrical Engineering and Computer Science, National Technical University of Athens, Heroon Polytechneiou 9,

Zographou, 15773, Greece

Christos Desiniotis

Vodafone-Panafon S.A., Tzavela 1-3, Chalandri, Athens, 15231 Greece

Keywords: B3G environment, End User Experience, Positioning Techniques, Ubiquitous Monitoring, Wireless Sensor

Networks.

Abstract: In a mobile telecommunications environment, the mobile terminal collects a variety of network performance

related data that are subsequently used in order to perform basic networking functions, such as cell

selection, handover or power control. Terminals’ processing capabilities and the potentials derived through

their integration with sensors, remain today mostly untapped. Taking advantage of these evolving sensing

capabilities, the MOTIVE (MObile Terminal Information Value addEd Functionality) project intends to

demonstrate the potential stemming from the exploitation of such information. The MOTIVE system

comprises a terminal monitoring module (which operates in a user transparent mode) and a network module

that collects the appropriate set of data and performs the proper processing. The paper examines the

application of this concept in three key areas, related to integrated end-to-end user experience monitoring,

ubiquitous terminal assisted positioning and anonymous mobile community services, taking into account the

terminal functionality evolution as well as the network evolution towards a multi-access composite IP based

network.

1 INTRODUCTION

Recent evolutions in mobile communications

include both the widespread use of mobile terminals

and the development of new technologies. Indeed,

more than 2 billion people around the globe are

mobile communications subscribers, and growth

rates are considerably promising for the future. In

addition, modern mobile terminals have become

technologically sophisticated, encompassing

increased functionality and a variety of extra

features, such as cameras, multiple antennas,

temperature sensors, and short range connectivity

(e.g., Infrared and Bluetooth).

In parallel, the scientific community has recently

shown a tremendous interest in the area of wireless

sensor networks. Significant progress has been

made, regarding sensors’ hardware, operating

systems, embedded software, and networking

enabling technologies. As a result, modern wireless

sensor networks have provided the means for

developing innovative applications, targeted for

environmental monitoring, motion monitoring as a

form of condition-based maintenance (Culler et al,

2004), patient monitoring and assisting (Jovanov et

al, 2001), inventory management, product quality

monitoring and disaster areas monitoring. Hence, the

use of wireless sensor networks is valuable for a

variety of applications and is expected to grow

further in the future.

In this paper, it is argued that the largest wireless

sensor network existing today is the community of

mobile terminals, which still remains unexploited.

Indeed, mobile phones and sensors present a lot of

similarities:

(a) relatively limited processing speed and

storage capacity;

(b) substantial processing capability in the

aggregate, but not individually; and

347

Masikos M., Demestichas K., Adamopoulou E. and Desiniotis C. (2006).

THE MOTIVE CONCEPT - Enabling Mobile Terminals to Act as Sensors.

In Proceedings of the International Conference on Wireless Information Networks and Systems, pages 347-354

DOI: 10.5220/0002094003470354

 SciTePress

What is more, mobile terminals present a number

of useful characteristics. Firstly, in contrast to sensor

nodes, mobile phones have a global identification

(ID), the Mobile Station International ISDN Number

(MSISDN). Secondly, there are already a large

number of devices scattered all over the world.

Thirdly, the location of these devices can be

estimated via a number of indoor and outdoor

localisation methods (Muthukrishnan et al, 2005).

Consequently, a “sensor network” consisting of

mobile terminals can provide applications with

geospatial data at a higher granularity and with

greater coverage than previously possible.

The MOTIVE (MObile Terminal Information

Value addEd Functionality) project introduces the

aforementioned innovative idea, which has not yet

been exploited. Mobile phones collect by default

several measurements, concerning, for example, the

bit error rate and the signal-to-noise ratio, which are

subsequently used in order to perform basic

networking functions, such as handover. These

measurements are currently untapped, in the sense

that they are not fully exploited. Moreover, mobile

phones could be outfitted with several inexpensive

sensors that provide enhanced capabilities.

Based on this idea, MOTIVE’s vision is to build

a powerful wireless sensor network. MOTIVE

enabled terminals are expected to transparently

store, pre-process and upload monitored data to the

network, without, however, obstructing the user

from utilising his device. In more detail, the

MOTIVE project proposes the application of this

innovative “sensor network” in three key areas,

related to: (a) integrated end-to-end user experience

monitoring; (b) ubiquitous terminal assisted

positioning; and (c) anonymous mobile community

services.

This paper describes the architecture of the

MOTIVE concept and the proposed applications in

the three key areas. Section 2 presents the proposed

architecture regarding the terminal and the network

side. Sections 3, 4 and 5 refer, in detail, to the three

proposed applications, as well as the methods in

which the capabilities of the underlying network can

be exploited. Finally, Section 6 concludes the paper

and identifies future research directions.

2 THE MOTIVE

ARCHITECTURE

2.1 Network Architecture

A high level presentation of MOTIVE’s proposed

architecture is depicted in Figure 1.

As can be observed, two main entities comprise

the overall system: (a) the data capturing devices;

and (b) the network server. Communication and data

transferring between these entities is carried out by

exploiting a composite radio network infrastructure,

consisting of diverse, heterogeneous networks.

External

sensor

GSM, GPRS

UMTS

WLAN

WiMax

MOTIVE

Server

Service

Providers

Integrated

sensor

Figure 1: MOTIVE high-level architecture.

The data capturing devices fall into three

categories: (a) external sensors communicating with

the mobile terminal and transferring the collected

data to it through a short-range technology (e.g.,

Bluetooth); (b) sensors integrated to the mobile

terminal; and (c) the mobile terminals, acting as

sensors themselves.

Regarding (a), external sensors are devices

equipped with simple networking functionality and

able to sense different environmental parameters.

Their size or power consumption, however, does not

permit their integration to the mobile terminal. A

typical example of this type of sensors is the gas

sensors (Schmidt et al, 2001). These sensors require

a significant amount of time for heating up,

accompanied by excessive energy consumption

(often around 1W for 1 minute).

Pertaining to (b), there are sensors that their

energy demand, processing requirements and size

render their integration to the mobile terminal

feasible. In fact, a number of mobile devices already

implement such sensing abilities. For example,

microphones can provide valuable information, even

WINSYS 2006 - INTERNATIONAL CONFERENCE ON WIRELESS INFORMATION NETWORKS AND SYSTEMS

348

when using minimal processing power.

Calculations on a microcontroller with less than 200

bytes of RAM proves to contribute useful

information, such as the noise level and the type of

input (noisy, music, speaking).

Concerning (c), mobile terminals can act as

sensors without any extra enhancements. In other

words, the mobile terminal can record and upload

measurements of parameters related to its operation,

as, for instance, the perceived signal strength. This

information can later be used in a plethora of

applications, like location based services.

In order for these captured data to be efficiently

exploited, their further processing is required.

Initially, the data collected are pre-processed at the

mobile terminal. The pre-processed data are then

uploaded to the MOTIVE application server through

the radio access interface used by the terminal. The

scheduling of the transmissions and the cost of

transferring the data are matters that need to be

carefully examined before deploying the

corresponding services. The main volume of the

required processing is conducted by the MOTIVE

server. Appropriate functionality for retrieving,

indexing and storing the collected information is in

place in this entity. Service providers should be able

to retrieve the necessary information through

suitable Application Programming Interfaces.

2.2 Definition of Functional Layers

In what follows, a layered architecture is presented,

serving as the basis for the development of a simple,

yet efficient, system aiming at the provision of

enhanced, context-aware applications. The four

layers comprising the proposed architecture are

depicted in Figure 2. For further comprehension of

the system’s functionality, the role of every layer

will be discussed, and a representative application

will be investigated.

The lower level consists of the data capturing

devices, namely the three abovementioned types of

sensors. These components are responsible for

collecting the required measurements from their

surroundings and, in the case of the external sensors,

transmitting them to the mobile terminals.

The next level involves the formation of cues. By

using the term cue (Schmidt et al, 2001), we refer to

information that has been submitted to a pre-

processing procedure. The concept of cues has

proven to be very useful, since it provides a level of

abstraction from the hardware components and

renders their modification transparent to the higher

layers. In this way, if new sensing devices are

included in the system, only changes to the

corresponding cues must be adapted. Cues can be

realized by using appropriate statistical functions.

Thus, the collected data are either summarized or

processed at a basic level. Typical examples of cues

that are usually investigated are the average value,

the standard derivation and the first derivative. It is

Data Capturing Level

Sensor #1

Cue

1.1

Cue

1.2

Cue

1.3

Sensor #1

Cue

1.1

Cue

1.2

Cue

1.3

Sensor #2

Cue

2.1

Cue

2.2

Cue

2.3

Sensor #2

Cue

2.1

Cue

2.2

Cue

2.3

Sensor #3

Cue

3.1

Cue

3.2

Cue

3.3

Sensor #3

Cue

3.1

Cue

3.2

Cue

3.3

Cue Level

Content Formation

Level

Application

Developer

Application

Developer

Application

Developer

Applications Level

Content

Figure 2: Functional layers of the MOTIVE architecture.

THE MOTIVE CONCEPT - Enabling Mobile Terminals to Act as Sensors

349

noteworthy that each cue is dependent on one single

sensor, but, by using the data of one sensor, multiple

cues can be calculated.

The next level involves the formation of content,

that is the main processing and indexing of the

collected information into organised structures. The

result of this level’s processing leads to data forms

that can be directly exploited by the application

providers of the higher level.

Finally, at the top of the proposed architecture

lies the application level. All possible context-aware

applications utilizing the collected information can

be included in this level.

Taking into account the processing capabilities

of the mobile terminals, the power consumption

limitations raised, as well as the network load

provoked, the following mapping of the functional

entities described above to the network entities of

Figure 1 can be deduced: Data retrieval from sensing

devices and cue development can be integrated in

the terminal-side. More precisely, the MOTIVE

terminals will gather the desired information and

proceed in its statistical pre-processing. This

information will afterwards be uploaded to the

MOTIVE server, where it will be submitted to the

content formation procedures. The content created

by these procedures will then become available to

application developers through the appropriate

interfaces.

In what follows, a simple, yet enlightening,

application will be presented, through which the

functionality of the involved entities will be

clarified. This application is launched by a service

provider who desires to deploy a service providing

temperature information over a large city. Firstly, at

the lowest level, mobile terminals, integrating the

temperature measurement functionality, will record

temperature values, perceived at their current

positions. These measurements will then be pre-

processed by the mobile terminal, in order to form

the corresponding cues. Useful and exploitable

parameters would be, for example, the average

temperature and the corresponding changing rate.

The time window during which the measurements

should be taken is a matter of further study. The

results are then uploaded to the MOTIVE server. At

this point, they are combined with measurements

originating from other terminals, and the main

volume of processing takes place. A useful piece of

information, resulting from this level, would be, for

example, the categorization of the perceived

temperatures, according to the specific time and

place they were measured. Finally, at the higher

level, the application provider would exploit this

formatted content, in order to create maps of the area

under consideration, where measured temperatures

would be depicted in a user-friendly way (e.g.,

different colours indicating different climate

conditions).

3 END-TO-END USER

EXPERIENCE MONITORING

As has already been noted, in a mobile

telecommunications environment, the mobile

terminal collects a variety of network performance

data that are subsequently used in order to perform

basic networking functions, such as cell selection,

handover or power control. Terminal collected data

constitute a valuable, but untapped yet, source of

information. By properly exploiting them, mobile

terminals can turn into ubiquitous super-sensors,

providing information about end-to-end user

experience.

In more detail, the integration of a monitoring

agent will enable mobile terminals to monitor

several Key Performance Indicators (KPIs), related

to: (a) the efficiency of the user interface; (b) the

efficiency of the protocol stack during

communication; and (c) the network’s performance.

Regarding (a), the time it takes to find and

trigger an application through the user interface, as

well as how often an application is used, can be

assessed. Pertaining to (b), several parameters,

including the receiver quality, the voice quality and

the IP protocol performance, can be utilized. Both of

the aforementioned issues are currently addressed

via terminal testing, which lacks in scalability.

MOTIVE’s monitoring agents can complement to

the task of terminal evaluation, helping terminal

manufacturers to improve their products.

Concerning (c), the contribution of network

performance to user experience can also be assessed

at the level of the mobile terminal, by measuring

primarily air-interface parameters related to

coverage quality and air-interface performance,

including the bit error rate and the signal-to-noise

ratio. By acting, in this way, as sensors and relaying

the aforementioned measurements back to the

network side, possibly after some filtering or pre-

processing (e.g., calculation of the average value),

mobile terminals can provide significant feedback to

mobile network operators, in issues such as radio

planning and customer care. Currently, to resolve

these issues, mobile network operators rely on

network verification surveys, KPIs derived from the

WINSYS 2006 - INTERNATIONAL CONFERENCE ON WIRELESS INFORMATION NETWORKS AND SYSTEMS

350

Network Management Systems (e.g., blocked or

dropped calls), or measurements from special

terminal implementations, such as Ericsson TEMS

and Nokia Traffica. The MOTIVE approach is able

to outperform the state-of-the-art, as it is far more

scalable and collects measurements from the actual

end users.

In addition to the benefits for terminal manufacturers

and mobile network operators, the MOTIVE enabled

terminals can prove to be valuable for context radio

systems. Cognitive radio systems can be defined as

systems capable of sensing the RF environment and

easily adjusting to current conditions, by making

intelligent decisions about how to best utilize

spectrum. In this context, mobile terminals can act

as sensors, in order to properly “sniff” the RF

environment and detect spectrum holes, i.e. RF

bands where the interference level is below a certain

threshold, rendering them usable for communication.

The collection of measurements for this purpose and

the use of a feedback channel, so as to transfer them

to the network side, can facilitate the latter to select

the most suitable configuration, at any given time.

4 UBIQUITOUS TERMINAL

ASSISTED POSITIONING

Various research activities are ongoing in the field

of Location Based Services (LBS), a type of services

that is expected to give a boost to mobile

communications. The provision of such services

assumes that the position of the client is known.

Hence, localisation proves to be of paramount

importance for their development.

Several techniques have been proposed for

addressing the localisation problem. These

techniques can be classified according to several

characteristics (Muthukrishnan et al, 2005):

• Outdoor vs. Indoor, depending on the

position area.

• Network based vs. Terminal based,

depending on the side of the implementation

(terminal or network).

• Terrestrial-radio-based vs. Satellite-based vs.

Standalone depending on the use of a

terrestrial radio signal, a satellite signal or no

use of signal for positioning.

• According to the location estimation

technique used, namely Global Positioning

System (GPS), Assisted GPS (A-GPS),

Observed Time Difference (OTD), Time of

Arrival (TOA), Time Difference of Arrival

(TDOA), Angle of Arrival (AOA), Multipath

Fingerprinting, Timing Advance (TA),

Enhanced Forward Link Triangulation (E-

FLT) and Received Signal Strength (RSS).

All of the aforementioned positioning methods

rely on signal measurements. In some cases, the

measurements are collected by a mobile device

acting as a sensor, while in other cases by a

specialised sensor network that may or may not

cooperate with a mobile device. Regarding the

specialised sensor networks, several techniques have

been proposed so far, based on Bluetooth (Forno et

Figure 3: The UTAP architecture.

THE MOTIVE CONCEPT - Enabling Mobile Terminals to Act as Sensors

351

al, 2005), Ultra Wide Band (UWB) (Bocquet et al,

2005), Infrared (Lee et al, 2004), Ultrasonic (Holm

et al, 2005), GPS (Zhao, 2002), or force measuring

sensors (Orr at al, 2000).

All of these techniques can easily be integrated

in the Ubiquitous Terminal Assisted Positioning

(UTAP) concept, which is illustrated in Figure 3.

The main characteristics of the UTAP concept

are:

• It is functional in a B3G environment, where

a variety of network technologies (GSM,

GPRS, UMTS, WLANs, short range radio

interfaces, etc.) are present.

• It can use any of the existing positioning

methods and techniques, either as a stand-

alone or as a hybrid scheme.

• The use of A-GPS is proposed for positioning

in RF-shadowed environments. The basic

idea of assisted GPS (Zhao, 2002) is to

establish a GPS reference network whose

receivers have clear views of the sky and can

operate continuously. The most important

contribution of a reference network to

handset GPS receivers is the reduction of the

frequency uncertainty of satellite signals.

• The Statistical Terminal Assisted Mobile

Positioning (STAMP) (Markoulidakis et al,

2006) approach can utilize the terminal’s

historical positioning data, in order to

improve the terminal positioning accuracy.

STAMP exploits the measurements that the

mobile terminal periodically collects while in

idle mode. Firstly, standard positioning

techniques are applied on the collected

measurements to provide estimations of the

terminal’s positions in successive time

moments. Then, statistical filtering (e.g.,

Kalman filter) is applied on the latter

estimations, in order to accurately infer the

terminal’s current position.

• It allows the concurrent estimation of the

mobile terminal’s speed and direction,

enabling the timely provision of advanced

LBS applications.

• It requires only an additional software

module on the terminal side, while the impact

on the terminal operation is minimal (e.g.,

through the exploitation of parameters which

terminals anyhow measure as part of their

standard operation).

Based on these characteristics, the UTAP

concept intends to provide location estimation under

any circumstances and with high precision. Further

research and real experiments for evaluation

purposes are ongoing under the MOTIVE project.

5 ANONYMOUS MOBILE

COMMUNITY SERVICES

The MOTIVE concept allows for the definition of a

new, innovative type of services, the Anonymous

Mobile Community (AMC) services, which exploit

the capability of the terminals to act as sensors, i.e.

collect information, and present it to the network

whenever required.

Figure 4 depicts the concept of MOTIVE AMC

services. According to this notion, the service

requestor may be either a third-party application

server or a mobile terminal. The MOTIVE AMC

server receives and validates the request, and

produces a response. Depending on the type of

service, there are two alternatives for generating the

appropriate response: (a) the real-time approach; and

(b) the non real-time approach.

Anonymous Mobile

Community

Motive

Terminal

Module

Motive

Terminal

Module

Motive

Terminal

Module

B3G Heterogeneous

Access Network

MOTIVE Anonymous

Mobile Community

(AMC) Server

B3G Composite Core

Network

Mobile

Terminal

Third-party

Application

Server

Figure 4: The concept of MOTIVE Anonymous Mobile

Community Services.

In (a), information is collected from the

community of terminals immediately after validating

the service request. Then, this information is

processed, and the appropriate result is sent back to

the requestor. In (b), the MOTIVE AMC server

generates a response on the basis of the data that

have already been collected from the community of

terminals. In all cases, user anonymity and privacy

should be respected. Privacy considerations are of

high importance, and the relevant requirements, as

WINSYS 2006 - INTERNATIONAL CONFERENCE ON WIRELESS INFORMATION NETWORKS AND SYSTEMS

352

specified in (3GPP, 2002) and (3GPP, 2005), must

be satisfied.

Viewing an anonymous community of mobile

terminals as a grid of active sensors offers the ability

to construct a group of useful, innovative services on

top of it. Representative examples of interesting

services are depicted in the following.

(a) Environmental services. In this case, each

terminal belonging to the anonymous community

will provide information about the current

temperature and the weather in its position.

Requestors (either third-party application servers or

terminals that are members of the community) can

be informed about the average temperature and

climate conditions in the area of their interest.

(b) End-user experience monitoring. As

described in Section 3, the anonymous community

of terminals can be used to sense KPIs, such as the

bit error rate and the signal-to-noise ratio. Mobile

network operators act in this case as requestors.

similar way to (b), terminals belonging to an

anonymous community may be utilized to measure

the value of the electromagnetic field in the used

frequency bands. Special non-profit organizations

will act as requestors, and generate reports about

whether safety levels are met or not.

(d) Traffic conditions monitoring. The retrieval

of road traffic conditions from fellow drivers

(determined by the terminal velocity in that area),

who are subscribers to the corresponding

anonymous community, can prove to be extensively

valuable in modern cities. Third-party application

servers or terminals that are members of the

corresponding community can act as requestors.

In general, provided that privacy requirements are

satisfied in an uncompromised manner, this type of

services is expected to become more and more

popular in the future.

6 CONCLUSIONS

In this paper, the concept of enabling mobile

terminals to act as sensors, by collecting data from

their RF and physical environment, was evolved.

The exploitation of such information in three key

areas, related to integrated end-to-end user

experience monitoring, ubiquitous terminal assisted

positioning and anonymous mobile community

services, was established.

Further research activities include the detailed

specifications and development of the corresponding

applications, as well as the conduction of trials for

the evaluation of the system’s functionality. The

trials will be performed using commercially

available mobile terminals in two different sites,

Athens and Paris, employing several diverse radio

access networks.

ACKNOWLEDGEMENTS

This paper introduces concepts and technologies

deployed within the framework of the project

MOTIVE (FP6-IST-27659), which is co-funded by

the European Commission in the 6th framework of

the IST program.

REFERENCES

3GPP TR 23.871 V5.0.0, 2002. Enhanced support for User

Privacy in location services. Release 5.

3GPP TS 23.271 V7.3.0, 2005. Functional stage 2

description of Location Services (LCS). Release 7.

Bocquet, M., Loyez, C., Benlarbi-Delaϊ A. (2005). Using

Enhanced-TDOA Measurement for Indoor

Positioning. In IEEE Microwaves and Wireless

Components Letters, Vol. 15, Issue 10, 612 - 614.

Culler, D., Estrin, D., Srivastava, M. (2004). Overview of

Sensor Networks. In IEEE Computer’s Society

Magazine, Vol. 37, Issue 8, 41 - 49.

Forno, F., Malnati, G., Portelli G. (2005). Design and

implementation of a Bluetooth ad hoc network for

indoor positioning. In Software Engineering, IEE

Proceedings, Vol. 152, Issue 5, 223 - 228.

FP6-IST4 27659 MOTIVE Technical Annex.

Holm, S., Hovind, O., Rostad S., Holm, R. (2005). Indoor

Data Communications using Airborne Ultrasound. In

IEEE ICASSP.

Jovanov, E., Raskovic, D., Price, J., Chapman, J., Moore,

A., Krishnamurthy, A. (2001). Patient Monitoring

Using Personal Area Networks of Wireless Intelligent

Sensors. In Biomedical Sciences Instrumentation.

Lee, C., Chang, Y., Park, G., Ryu, J., Jeong, S., Park, S.,

Park, J., Lee, H., Hong, K., Lee, M. (2004). Indoor

Positioning System Based on Incident Angles of

Infrared Emitters. In the 30th Annual Conference of

the IEEE Industrial Electronics Society, Vol. 3, 2218 -

2222.

Markoulidakis, J., Desiniotis, C. (2006). Statistical

Terminal Assisted Mobile Positioning in a Beyond 3G

Environment. In the 15

IST Mobile & Wireless

Communications Summit.

Muthukrishnan, K., Lijding, M., Havinga, P. (2005).

Towards Smart Surroundings: Enabling Techniques

and Technologies for Localization. In LoCA

Proceedings. Vol. 3479. Springer-Verlag.

THE MOTIVE CONCEPT - Enabling Mobile Terminals to Act as Sensors

353

Orr, R., Abowd, G. (2000). The smart floor: a mechanism

for natural user identification and tracking. In

Conference on Human factors in computing systems.

Schmidt, A., Karlsruhe, U., Laerhoven, K.-V. (2001).

How to build smart appliances. In IEEE Personal

Communications, Vol. 8, Issue 4, 66 - 71.

Zhao, Y. (2002). Standardization of Mobile Phone

Positioning for 3G Systems. In IEEE Communications

Magazine Vol. 40, Issue 7, 108 - 116.

WINSYS 2006 - INTERNATIONAL CONFERENCE ON WIRELESS INFORMATION NETWORKS AND SYSTEMS

354