A COMPUTING MODEL TO BRIDGE A GAP BETWEEN REAL

SPACE AND WEB SPACE

Kenji Sugawara, Shigeru Fujita

Fuculty of Information and Network Science, Chiba Institute of Technology, Japan

Norio Shiratori, Takuo Suganuma

Research Institute of Electrical Communication, Tohoku University, Japan

Tetsuo Kinoshita

Informatrion Synergy Center, Tohoku University, Japan

Keywords:

Post-Ubiquitous Computing, Symbiotic computing, e-Gap, awareness, watch over system.

Abstract:

In this paper, we propose a computing model named ”symbiotic computing” to bridge an e-Gap between the

real space (RS) and the web space (WS). The symbiotic computing is a post ubiquitous computing model based

on an agent-based model to bring in social heuristics and awareness of the RS into the WS. The symbiotic

computing is deﬁned as a space of computing model spanned by three axes of the Ubiquitous computing, the

Web computing and P/S computing, which consists of Perceptualware and Socialware. The last section of this

paper describes a prototypical system of a symbiotic computing-based application of uEyes, which is a gentle

watch over system for elderly people in an ubiquitous computing environment.

1 INTRODUCTION

Various kinds of computing models were proposed

to develop more efﬁcient and useful software sys-

tem easily, such as the object-oriented computing, the

Web computing and the ubiquitous computing (Lyyti-

nen and Yoo, 2002). The technologies leaded the

traditional society to the information network society

where people can exchange information easily via the

WWW efﬁciently. On the other hand, emergent prob-

lems also occurred with the Internet society, such as

the digital divide, security, network-based crimes, and

so on. These problems have been caused due to so-

cial and human difﬁculty rather than due to the com-

puter and network technology. However, the technol-

ogy should tackle these difﬁcult problems by bringing

in the sociality and humanity into computing models.

In this paper, we propose a computing model

named ”symbiotic computing” which is formalized to

bridge an e-Gap (eGAP-EU, 2006) between the real

space (RS) we are living and the web space (WS)

which is a cyber space constructed using the com-

puter and network technology. We considered that the

e-Gap which causes the problems was made because

of the lack of mutual awareness between the RS and

the WS. That is, people cannot receive its advanced

services without IT skills, and the WS cannot provide

a service suitable with a person depending on his/her

situation. The WS also cannot provide a safe and se-

cure service without heuristics on persons’ activities

in a society, such as custom, law, expertise and so on.

The symbiotic computing we proposed is deﬁned

as a space of computing model spanned by three axes

of the Ubiquitous computing, the Web computing and

the P/S computing. P/S computing is a methodology

to acquire and process awareness of the RS and the

WS using the Ubiquitous computing and to store so-

cial knowledge, persons’ heuristics to live in a region

and maps and facilities of the region, and to use the

knowledge for people in the region.

The last section of this paper describes a prototyp-

ical system of a symbiotic-computing-based applica-

tion of uEyes, which is a gentle watch over system for

children and elderly people in the ubiquitous comput-

ing environment, using awareness and social knowl-

edge.

107

Sugawara K., Fujita S., Shiratori N., Suganuma T. and Kinoshita T. (2007).

A COMPUTING MODEL TO BRIDGE A GAP BETWEEN REAL SPACE ANDWEB SPACE.

In Proceedings of the Third International Conference on Web Information Systems and Technologies - Society, e-Business and e-Government /

e-Learning, pages 107-112

DOI: 10.5220/0001270601070112

 SciTePress

2 E-GAP PROBLEMS

Although a huge amount of social capital has been

invested in the IT infrastructures and application sys-

tems, many of us in the Internet era still have not been

able to reap satisfactory beneﬁts from the services we

have received in our daily lives. For example, the dig-

ital divide problem has blocked out unskilful people

who has expected the useful helps form IT services.

Fig.1 explains cause of the phenomena in terms of

an e-Gap which isolates the RS from the WS where

various kinds of IT services and resources are in-

stalled. The RS has invested a huge amount of social

capital and human resources to develop better services

of the WS expecting that people can enjoy the satis-

factory services in their daily lives and work places.

However, due to the e-Gap, the people feel disap-

pointed because they cannot receive satisfactory and

suitable services they required. Therefore, we deﬁne

the e-Gap by the differences between the expectation

and the disappointments of the RS.

Real Space

Web Space

investment

limited

services

e-Gap

expectation

disappointment

Figure 1: e-Gap Problem.

3 CONCEPT OF SYMBIOTIC

COMPUTING

In order to bridge the e-Gap, we propose a comput-

ing model to construct an e-Bridge over the e-Gap

which enable users to receive rich services of the WS.

A basic concept of an e-Bridge is based on a concept

of mutual awareness between the RS and the WS as

shown in Fig.2(Pentland, 2005). The e-Bridge helps

services of the WS recognize users’ requirements and

needs of help of people living and working in the RS.

Then the e-Bridge can provide the adequate services

automatically based on the recognition. The e-Bridge

also helps each person ﬁnd adequate services and re-

sources in the WS as if he/she is familiar with them.

RS awareness in Fig.2 is a notion on recognizing

the RS, using functions provided by e-Bridge, to cap-

ture users’ requirements in real time, to recognize sit-

uation of the users, to acquire data of a region where

users live, such as weather, trafﬁc, position of each

person, etc..

WS awareness is a notion on intuitive interface

and visualization of WS to help users ﬁnd adequate

Real Space

Web Space

e-Bridge

RS awareness

WS awareness

Real Space

Web Space

e-Bridge

RS awareness

WS awareness

Figure 2: Mutual Awareness through e-Bridge.

resources in WS.

The concept of the symbiosis implies a biological

terms of ”the relationship between two different living

creatures that live close together and depend on each

other in particular ways, each getting particular bene-

ﬁts from the other” (Oxford, 2000). We are applying

this concept to a relation between the RS and the WS,

in order to bridge the e-Gap so that users and soci-

ety easily get maximum beneﬁts from the WS. The

WS also gets principal investment from the RS when

it recognizes that the investments brings richness and

comfort to people living in it. The symbiotic comput-

ing provides models and technology to develop the

e-Bridge which promotes the above symbiotic rela-

tionships.

The symbiotic computing we proposed is de-

ﬁned as a space of computing model spanned

by three axes of the Ubiquitous computing the

Web computing and the P/S computing as shown

in Fig.3(Shiratori, 2005). (Symbiotic Computing,

http://symbiotic.agent-town.com),

In this deﬁnition, the ubiquitous computing pro-

vides an interface between RS and e-Bridge, which

is composed of embedded sensors, mobile devices,

wireless LAN and so on, and the Web computing pro-

vides WWW services for the e-Bridge to compose

services which users request. The P/S computing

in Fig.3 is a model to develop functions to perceive

the situations of users and communities in the RS,

to acquire and save social knowledge and personal

heuristics to live in a region and to provide context-

adaptive(Coutaz et al., 2005) and social-aspect ser-

vices(Hattori et al., 1999) for users using the above

functions and services by WWW.

Ubiquitous computing

P/S computing

Symbiotic

Computing

Web computing

Ubiquitous computing

P/S computing

Symbiotic

Computing

Web computing

Figure 3: Symbiotic computing and P/S computing.

WEBIST 2007 - International Conference on Web Information Systems and Technologies

108

4 A MODEL OF E-BRIDGE

The symbiotic computing is a methodology to de-

velop an e-Bridge between a community in a region

and WS to make the community safe and comfortable,

using the ubiquitous technology and Web technology.

An e-Bridge is composed of Perceptualware and So-

cialware, as shown in Fig.4.

The Perceptualware is a collection of software and

data for e-Bridge to perceive RS and WS through

ubiquitous infrastructure and the web service inter-

face(Billsus et al., 2002). The Socialware is a collec-

tion of software and data for e-Bridge to understand

RS and WS, to save knowledge on RS and WS and

to make decisions on how to support persons in RS

properly and safely.

Functions of the Perceptualware is categorized

into the followings, as shown in Fig.4,

(1) RS Perception

Functions of RS Perception are to transform signals

and data from sensors and embedded devices into

some data forms deﬁned to perceive persons’ activ-

ities, surrounding situation, climate, etc.. RS per-

ception is categorized into (a)Location-based percep-

tion, (b)Event-based perception, (c)Vision-based per-

ception, (d)Audio-based perception.

(2) RS Operation

Functions to operate mobile terminals, actuators and

embedded devices in order to enhance person-system

interactions and person-person communication in

ubiquitous environment.

(3) WS Perception

Functions to perceive structure of Web Space and to

transform web pages into data form which the mod-

ule of “Recognition & Action” in Perceptualware can

recognize. The model of perceiving WS is based on

the model of Web Service(IBM, Webservices, 2006).

(4) WS Operation

Functions to maintain web pages which the e-Bridge

provides for to WWW for persons in the region.

(5) Recognition and Action

Functions to recognize situation of RS and to make

actions to solve problems collaborating with func-

tions in the Socialware. This module recives data of

lower level of perception from RS Perception ans WS

Perception, and transfer them into higher level of per-

ceptional representation on RS ans WS. This module

also make lower level of decisions on reactive actions

of e-Bridge.

Functions of the Socialware are categorized into

the followings,

(1) Situation Understanding

Functions to understand activities and, surrounding

situations of each person, collaboration among per-

Ubiquitous

Infrastructure

(region)

Recognition & Action

Broadband Infrastructure

Perception

e-Bridge

Perception

retrieve

mining

sensing

watching

Operation

Socialware

Perceptualware

display

operation

maintain

Figure 4: Architecture of e-Bridge.

sons, etc. using the perceived data form provided by

the RS Perception functions.

(2) WS Understanding

Functions to understand information and services

which are provided for in WS and create new infor-

mation from them to meet person’s request.

(3) Social Knowledgebase

Functions to store knowledge and common social

heuristics in order for persons to use/reuse.

(4)Region-Directed Knowledgebase

Functions to store the region-directed knowledge and

information which are acquired by the function of the

Situation Understanding and the WS Understanding.

location

by GPS

daily

school

route

walking

route

recognition

deviation

surrounding

situation

recognition

unsafety

alarm

policy

send

alarm-mail

WWW

mining

Perceptualware

Socialware

WWW

notice

Figure 5: Simple Example of e-Bridge of Watch-over-

support for children.

An example of an e-Bridge that we are develop-

ing in order to watch over children in a region is illus-

trated in Fig. 5. Th e-Bridge watches over children’s

walking routes using location data acquired by GPS,

referring to map information of the region, which is

stored in a Region-directed knowledgebase. When a

module of ”detection of deviation” detects a devia-

tion from the child’s daily school route stored in the

Region-directed knowledgebase, a message of the de-

tection is sent to a module of ”detection of unsafety”

in ”Recognition and Action”. If the module decides

that the child may be in a dangerous situation, using

the surrounding situation provided by the Socialware,

it activates ”send alarm-mail” module in RS Opera-

tion and ”WWW notice” module in WS Operation.

These modules are installed using an agent frame-

work described in the next section.

A COMPUTING MODEL TO BRIDGE A GAP BETWEEN REAL SPACE AND WEB SPACE

109

5 AN AGENT FRAMEWORK

We are developing the functions in Fig.4 using an

agent framework of ADIPS along with the technolo-

gies of the Ubiquitous computing and the Web com-

puting. The ADIPS framework was proposed and im-

plemented to develop functions of ﬂexible networks

(Shiratori et al., 1996).

We have developed network middleware and dis-

tributed applications (Suganuma et al., 2003; Sug-

awara, 2005) using ADIPS framework. Features

of ADIPS framework are described as follows (Ki-

noshita and Sugawara, 1998);

(1) Agentﬁcation of resources in WS and devices in

An ADIPS agent is an autonomous module consisting

of programs and devices, which is generated based on

the wrapping approach as shown in Fig.6. An agent

consists of a wrapper module and a resource consist-

ing programs, devices, web pages, databases, knowl-

edge bases, actuators, and so on. The wrapper in

Fig.3 enables an agent communicatewith other agents

through Agent Communication Language (ACL). The

wrapper also controls a resource based on algorithms,

a production model, BDI model , which was im-

plemented using Java based on the ADIPS frame-

work(Kinoshita and Sugawara, 1998).

resource

wrapper

operation/

sensing

operation/

sensing

ACL

agent

resource

wrapper

agent

resource

wrapper

operation/

sensing

operation/

sensing

ACL

agent

resource

wrapper

agent

Figure 6: Wrapper-based Agentiﬁcation of Resources.

(2)Agent Virtual Machine (AVM)

AVM supports agents to be generated, to run, to com-

municate, to be uniquely named and to be persistent.

R-agent

Repository

Agent Virtual Machine (AVM)

organization

reorganization

change of

environment

adaptation

symbiotic

agent

R-agent

Repository

Agent Virtual Machine (AVM)

organization

reorganization

change of

environment

adaptation

symbiotic

agent

R-agent

Repository

Agent Virtual Machine (AVM)

organization

reorganization

change of

environment

adaptation

symbiotic

agent

Figure 7: Repository-based Programming and self-

organization.

(3) Repository-based programming

System developers programs R-agents and save them

into a repository as class agents in Fig.7. An R-agent

is transformed into an instance agent onto an AVM

according to a request from the AVM.

(4)Repository-based Self-Organization

A multi-agent system working on an AVM or on dis-

tributed AVMs has a property to adapt its functions

to change of the social/system’s requirements eas-

ily and rapidly (Fujita et al., 1998). This property

gives the WS a cost-effective evolutional mechanism

to serve the RS progressively. Fig.7 shows a frame-

work of developing functions having the above prop-

erties. This repository-based framework generate a

multi-agent system from a repository consisting of R-

agents which produce agents in a socialware(Hattori

et al., 1999; Sugiyama et al., 1999), perceptualware

and networkware.

6 SYMBIOTIC APPLICATION

6.1 An Overview of eEyes

Building a gentle watch over system for children or

elderly people in the ubiquitous computing environ-

ment has led to a big challenge(Helal et al., 2005).

Due to the lack of context awareness of devices, soft-

ware and networks around the user, they cannot pro-

vide appropriate watching services to the user. Fur-

thermore users are losing sense of ease and con-

veniences in the watching because of lack of user-

oriented aspects in operation and quality of service

(QoS) of the system. In this chapter, we explain a

prototypical system to watch over elderly people de-

veloped based on the symbiotic computing-based ap-

proach. In this system, multiple contexts including

user’s presence, location, detailed requirements for

watching, device status, available bandwidth of net-

work etc, in both watched site and watching site, are

comprehensively considered for quality control of the

live video streaming, based on the concept of Sym-

biotic Computing. We build an application of watch-

ing over system to provide support for elderly people

in home, and show the effectiveness of the idea with

some initial experiments with the prototype system.

Fig.8 shows an overview of uEyes and the struc-

ture of e-Bridge in uEyes. Socialware is a set of pro-

grams, heuristics, data regarding watching over el-

derly people, which are developed and acquired based

on the socialware (Hattori et al., 1999).

Perceptualware is a set of programs and data

to process RS awareness and WS awareness based

WEBIST 2007 - International Conference on Web Information Systems and Technologies

110

watched

person

Watching

person

Display

Camera

RFID receivers

Tag

Wireless

access

Sonic receivers

Camera

Networkware

Perceptualware

Socialware

Web

e-Bridge

watched

person

Watching

person

Display

Camera

RFID receivers

Tag

Wireless

access

Sonic receivers

Camera

Networkware

Perceptualware

Socialware

Web

e-Bridge

Figure 8: An Overview of uEyes and Components of e-

Bridge.

on the Perceptualware using ubiquitous devices de-

scribed in the next section.

Networkware is a set of programs to support com-

munications among agents and resources in WS based

on the ﬂexible network model.

6.2 Ubiquitous Devices Used in uEyes

Fig.9 shows a snapshot of example of user terminal

for the prototypical system of uEyes. The user termi-

nal is a device which watching persons always carry

with them. Although this terminal can receive video

streaming in low quality, users can check roughly sit-

uation of their family anytime and anyplace. This ter-

minal also becomes a sensor to acquire the data of

their location and becomes an anytime-anyplace in-

terface with agents in the socialware.

Data communication card (PHS)

Video receiver

application

GUI of user requirement

Data communication card (PHS)

Video receiver

application

GUI of user requirement

Figure 9: User Terminal.

In this prototypical system, we use a handheld

PC Sony VAIO type-U (Windows XP, Celeron M

900MHz, 256MB memory) for the device. It is con-

nected to the network with a CF type data commu-

nication card (PHS) that provides a link of 128 kbps

bandwidth.

We use the following three kinds of location sen-

sors to obtain location information of uses in the

room, (1) ultrasonic-based sensor system in Fig.11

which measures location, moving velocity and direc-

tion, direction of sight line, (2) passive type RFID

systems, and (3) an active type RFID system using

315MHz radio frequency., which can recognize lo-

cation of the tag that is located within 2 meter from

receiver, in minimum setting.

Table

PC3

PC2

PC4

Living room

100Mbps Ethernet

802.11b

Wireless access

PC1

Table

PC3

PC2

PC4

Living room

100Mbps Ethernet

802.11b

Wireless access

PC1

Figure 10: Sensing Devices in Watched Side.

PHS

(128kbps)

Ultra sonic

receives

Digital

appliance

(emulation)

Digital TV

(emulation)

PHS

(128kbps)

Ultra sonic

receives

Digital

appliance

(emulation)

Digital TV

(emulation)

Figure 11: Sensing Devices in Watched Side.

6.3 Agent-based Architecture of

e-Bridge in uEyes

An e-Bridge in the prototypical uEyes in Fig.8

is being implemented using the ubiquitous-directed

ADIPS agent framework named AMUSE. The e-

Bridge was designed as a distributed agent system

working on the distributed PCs as shown in Fig.10

and 11. The networkware in Fig.8 is an agent-based

middleware to control QoS of multimedia communi-

cation between PCs which controls multimedia de-

vices such as cameras, microphones, displays. El-

derly people never want to be watched over always by

camera if they are in good health. However, they may

want to be watched over with high quality of multime-

dia information captured by camera and microphones

if they are in badly illness. They may choose certain

level of the QoS of the multimedia information if they

feel vague anxiety or loneliness in order to tell the

feeling with their family implicitly. The networkware

connects data channel among multimedia devices in

the uEyes system, controls QoS of every communica-

tion channels, and keep the important channel even if

A COMPUTING MODEL TO BRIDGE A GAP BETWEEN REAL SPACE AND WEB SPACE

111

the troubles happens in the network system used by

uEyes.

The perceptualware of uEyes is a multi-agent

system to perceive situation and health condition

of watched persons and to control the networkware

based on a decision instructed by a designated agent

in the socialware.

For example, an agent in the perceptualware of

which roll is to count a pulse-beat and check blood

pressure notiﬁes an agent in the socialware, of which

roll is to manage the health information of the per-

son, of abnormal state in the persons health. If the

health information management agent in the social-

waere know that it is very dangerous situation for the

person according to a medical instruction by a per-

sonal doctor,the agent in perceptualwarenotiﬁes mul-

timedia QoS agents of setting multimedia channels to

his/her family and doctors in the highest QoS.

The socialware of uEyes consists of agents to store

the social knowledge and heuristics for taking care of

elderly people, to store information of families such

as jobs, daily routine, roles in family, to know map of

the region and facilities in the region. A socialware

agent designated to the elderly person inform persons

who take care the watched person of prescribed alam

and data based on the contract between the agent and

the persons.

7 CONCLUSION

In order to bridge the e-Gap, we propose a computing

model named symbiotic computing to construct an e-

Bridge over the e-Gap which enables users to receive

rich services of the WS. An e-Bridge is composed of

Perceptualware and Socialware which are working in

web servers.

The Perceptualware is a collection of software and

data for e-Bridge to perceive RS and WS through

ubiquitous infrastructure and the web service inter-

face. The Socialware is a collection of software and

data for e-Bridge to understand RS and WS, to save

knowledge on RS and WS and to make decisions on

how to support persons in RS properly and safely.

Based on the above concept, we prototyped an ex-

perimental system of uEyes which watch over elderly

people. An advantage of uEyes to the conventional

approach of the ubiquitous computing is that it can

watch over more gently because the system tries to

keep both of their privacy and safety based on the so-

cial knowledge to take a warm watch.

REFERENCES

Billsus, D., Brunk, C. A., Evans, C., Gladish, B., and Paz-

zani, M. (2002). Adaptive interfaces for ubiquitous

web access. Communications of ACM, pages 34 – 38.

Borrie, G., Chalemers, M., LaMarca, A., and Nixon, P.

(2005). Delivering real-world ubiquitous location sys-

tems. Communications of ACM, pages 36 – 41.

Coutaz, J., Crowley, J. L., Dobson, S., and Garlan, D.

(2005). Context is key. Communications of ACM,

pages 49 – 53.

eGAP-EU (2006). http://www.egap-eu.com/.

Fujita, S., Hara, H., Sugawara, K., Kinoshita, T., and Shira-

tori, N. (1998). Agent-based design model of adaptive

distributed systems. Applied Intelligence, 9:57–70.

Hattori, F., Ohguro, T., Yokoo, M., Matsubara, S., and

Yoshida, S. (1999). Socialware: multiagent sys-

tems for supporting network communities. CACM,

42(3):55–61.

Helal, A., Mann, W., Elzabadani, H., King, J., Kaddourah,

Y., and Jansen, E. (2005). Gator tech smart house:

A programmable pervasive space. IEEE Computer,

pages 64–74.

IBM, Webservices (2006). http://www-

128.ibm.com/developerworks/webservices.

Kinoshita, T. and Sugawara, K. (1998). Adips frame-

work for ﬂexible distributed systems. Lecture Notes

in Artiﬁcial Intelligence 1599,Multiagent Platform,

Springer, pages 18–32.

Lyytinen, K. and Yoo, Y. (2002). Issue and challenges in

ubiquitous computing. CACM, 45(12):63–65.

Oxford (2000). Oxford advanced learners dictionary.

Pentland, A. (2005). Socially aware computation and com-

munication. IEEE Computer, pages 33–40.

Shiratori, N. (2005). Toward future development of ubiq-

uitous information society -symbiotic computing-.

Technical report, Key Note Address of 2005 NEC

Technology Forum ( Symbiotic Computing for Ubiq-

uitous Information Society).

Shiratori, N., Suganuma, T., Sugiura, S., Chakraborty, G.,

Sugawara, K., Kinoshita, T., and Lee, E. S. (1996).

Framework of a ﬂexible computer communication

network. Computer Communications, 19:1268–1275.

Suganuma, T., Imai, S., Kinoshita, T., Sugawara, K., and

Shiratori, N. (2003). A ﬂexible videoconference sys-

tem based on multi-agent framework. IEEE Trans. on

Systems, Man, and Cybernetics - Part A: Systems and

Humans, 33(5):633–641.

Sugawara, K. (2005). Agent-based support system for

project teaming for teleworkers. Lecture Notes in Ar-

tiﬁcial Intelligence 3371,Intelligent Agents and Multi-

Agent Systems, Springer, pages 279–290.

Sugiyama, T., Kinoshita, T., and Shiratori, N. (1999). On a

software architecture for supporting virtual workspace

an agent-based architecture of socialware . Proceed-

ings of 13th International Conference on Information

Networking, pages 11B–2.1–6.

WEBIST 2007 - International Conference on Web Information Systems and Technologies

112