View - Academy Publisher

86 JOURNAL OF SOFTWARE, VOL. 2, NO. 6, DECEMBER 2007 

In and out of Reality: Janus-Faced Location 

Awareness in Ubiquitous Games 

Alexander Höhfeld 

University of Trier/Department of Computer Science, Trier, Germany 

hoehfeld@syssoft.uni-trier.de 

Abstract — Many future ubiquitous devices will be able to 

determine their precise physical position using different 

types of localization techniques (e.g. GPS antennas for outdoor-, 

RFID tags, infrared beacons or triangulation etc. in 

indoor settings). Especially for games in a mobile environment 

usage of the devices, respectively player’s location will 

capacitate new game functionality and content. Location 

Awareness in this context is twofold though, corresponding 

to the games’ focus: Either the player’s actions can be influenced 

by providing additional information according to the 

devices physical location (augmented or mixed reality) or location 

awareness directly or indirectly alters content and 

state of the game (location immersion). This work focuses on 

the presentation of two ubiquitous strategy games, which 

are based on the same uniform middleware for MANETs, 

implementing the above mentioned different forms of location 

awareness. 

Index Terms — Location Awareness, Mobile Learning, Multi-Hop 

Ad Hoc Networks, Ubiquitous and Pervasive Strategy 

Games, Network Centric Operations. 

I. INTRODUCTION 

As GPS-antennas are getting smaller and more reliable 

each day and the emerging of a broader range of indoor 

localization techniques, especially mobile devices will 

benefit from this development. The devices themselves 

possess increasing storage capacities and processing power, 

providing the user not only with additional performance 

for his or her everyday work, but as well with superior 

ubiquitous gaming possibilities. Combined with location 

and context awareness, these games will feature 

new types of immersion experiences and new unique content. 

Depending on game design and type different approaches 

to use the devices and players physical location 

within the applications have evolved. From a perspective 

of reality these range from augmentation, in which reality 

is enriched with virtual or simulated information, to immersion, 

where the position impacts the actual game 

state. This work presents two different ubiquitous games 

which are taking advantage of location awareness: A 

Treasure Island scenario for the ncoML framework [1], 

using augmented reality techniques to influence the users 

behavior and thus implicitly changing the game content 

and UbiQuest [2], a science-fiction strategy game, using 

location immersion to directly modify certain 

characteristics within the game according to the players 

© 2007 ACADEMY PUBLISHER 

physical location. The following figure illustrates the disposition 

of the two ubiquitous games in the context of 

reality: 

Location Immersion Augmented Reality 

UbiQuest 

Figure 1. Disposition in the context of reality 

ncoML 

Chapter II will introduce the different game designs of 

the presented ubiquitous strategy games and their technical 

realization. The follow-on chapters deal with the localization 

strategies used within the presented games and 

the experiences made during their development and testing. 

The middleware environment JANE, both applications 

are based on, is discussed in chapter V. Prior to presenting 

a short summary and conclusion, chapter VI features 

related work in the area of ubiquitous games and location 

awareness. 

II. GAME DESIGN AND TECHNICAL REALIZATION 

A. UbiQuest 

UbiQuest is a real-time science fiction strategy and 

conflict simulation, similar to well known games like Master 

of Orion, Civilization or The Settlers. The basic idea 

of these games is ported to a multi-hop ad hoc environment. 

In UbiQuest the player controls a giant starship (represented 

by a location aware mobile device of PDA size) 

with a remainder of the earth’s population after a global 

crisis on its way to find new hospitable planets. 

Starting with only a minimum of offensive, defensive 

and economical capabilities, the ship can be upgraded by 

gathering resources from bypassed asteroid fields or planets, 

building new on-board facilities, researching advanced 

technology and increasing the ships population. 

Nearly all actions within the game are interconnected. As 

an example, population growth is only possible if enough 

habitats for them are available. These on the other hand 

can only be constructed if sufficient energy- and building 

resources are stored, while the speed of the construction 

in turn is depending again on the population size, their 

training level, etc.

JOURNAL OF SOFTWARE, VOL. 2, NO. 6, DECEMBER 2007 87 

On its waay 

through thhe 

simulated universe (accom 

plished by pphysical 

moveement 

of the ddevice 

hostingg 

the 

UbiQuest appplication), 

a sstarship 

can enncounter 

and iinter 

act with otheer 

starships (ddevices 

of otheer 

players) as well 

as with speccial 

entities, sso 

called star bases. Thesee 

star 

bases are staationary, 

non mmobile 

devicees 

within the MA- 

NET which aare 

equipped with higher pprocessing 

andd 

sto- 

rage power. TThis 

could bee 

e.g. a libraryy 

information com 

puter or a priivately 

ownedd 

desktop PC that features awireless card annd 

is taking part 

in the gammes 

network. CCom 

munication wwith 

other deevices 

or starr 

bases is accom 

plished ad hhoc. 

No centtral 

backbonee 

infrastructurre 

or 

access pointss 

are necessaryy. 

Accordingg 

to the games 

ad hoc enviironment, 

it distin 

guishes betwween 

three ddifferent 

leveels 

of interaaction 

[Fig.2]. Locaal 

interaction includes the mmicromanageement 

of the ships resources, faccilities 

and reesearch. 

Tradee, 

di- 

rect communnication 

(singgle 

hop) and ( (military) connflicts 

are situated on the tacticaal 

interaction level. Finallyy 

the 

strategic level 

encompassses 

longer terrm 

diplomacyy 

and 

espionage, aas 

well as ageent 

based commmunication 

with 

starships or bases not inn 

direct commmunication 

rrange 

(multi hop). 

Figure 2. 

Strategicc 

Interaction 

- Diplomacy 

Tactical Interaction 

- Trade 

- (Military) conflicts 

Local Innteraction 

- Micromanagement 

- Facility Managemennt 

- Research 

Interaction Levells 

Trade (on the tactical innteraction 

leveel) 

is performmed 

in 

two different 

ways. If a suitable tradee 

partner is wwithin 

communicatiion 

range (sinngle 

hop), a staar 

ships owneer 

can 

initiate a tradde 

request, ressulting 

in a poositive 

or neggative 

feedback from 

the trade ppartner. 

Takinng 

into accounnt 

the 

short time peeriod 

two mooving 

device mmay 

have conntact, 

another posssibility 

is auttomated 

tradee. 

Here the sship’s 

owner sets thresholds foor 

selling annd 

buying ceertain 

goods. If annother 

ship oor 

star base iin 

communiccation 

range dispersses 

matching offers, trade iis 

performed wwith 

out direct useer 

input. 

UbiQuest has been devveloped 

in JAAVA 

within thhe 

si- 

mulation envvironment 

JANNE, 

which is described in more 

detail in thee 

follow-on chhapter. 

The ffirst 

prototypee 

has 

been tested on HP-iPaq H5550 Pockket 

PCs, equiipped 

with a Holuux 

GM-270 CCompactFlash-GPS 

adapterr 

and 

integrated WWLAN 

capabillities. 

For JAVVA 

support on 

the 

mobile devicces 

we used thhe 

IBM Virtuaal 

Machine j9. 

The ad hooc 

communicaation 

within tthe 

game usees 

the 

routing frammework 

of the aforementionned 

simulationn 

en- 


viron nment JANE. . Multi-hop ccommunication 

n is based onn 

the DSR D [3] routiing 

protocol. AAnother 

impo ortant commu- 

nica ation service iis 

an upgradeed 

beaconing functionality, , 

prov viding not only 

information 

about the ships s locationn 

and presence, buut 

as well dispersing 

the current tradee 

offer rs etc. This iss 

accomplisheed 

by using th he PeriodiCastt 

beac coning protocol 

adapted to the trade off fer/request in- 

form mation [4]. Coommunicationn 

in a cluster red MANET, , 

with hout routes to certain nodess, 

is implemen nted by usingg 

softw ware agents wwith 

certain ttasks. 

These try t to reach a 

spec cified device oor 

region, perrform 

their ta ask and returnn 

to a predefined hoome 

zone or llocation 

of the eir originator. . 

Typical 

tasks for aagents 

are dipplomacy 

and espionage. e 

Figuree 

3. UbiQuest ddeployed 

on PDAs 

B. A Treasure Islland 

Setting foor 

ncoML 

The 

m-learningg 

framework ncoML was originally o de- 

sign ned to support the training aand 

preparatio on of humani- 

taria an, catastrophhe 

aid and mmilitary 

missi ions that aree 

base ed on the tenets 

of Network 

Centric Op perations [5]. . 

Sinc ce it is a veryy 

flexible frammework, 

it ca an be used ass 

well l to depict sceenarios 

not dirrectly 

related to its originall 

inten nt. In the conntext 

of this year’s 

children n’s university, , 

whe ere kids have tthe 

opportunitty 

to get an ins sight into uni- 

versity 

life, it is planned 

as a veehicle 

to playf fully test theirr 

com mmunication annd 

cooperationn 

abilities. Fo or this, a Trea- 

sure e Island-like ggame 

scenarioo 

has been developed d andd 

impl lemented baseed 

on the ncoMML 

framewor rk. 

The 

system is ddesigned 

to be 

highly flexi ible and to al- 

low many differeent 

mission tyypes 

within th he underlyingg 

fram mework. One typical scenarrio 

from its original o appli- 

catio on domain (e.g. 

in a UN oobserver 

mission) 

could bee 

the discovery d of ccommon 

graves 

or weapon deposits, e.g. . 

in a region that haas 

been or is bbeing 

harassed 

by civil un- 

rest or war. Witthin 

the Treaasure 

Island scenario, thee 

child dren’s objectiive 

is to find a hidden treasure, 

by wor- 

king g together andd 

by using inddividually 

ava ailableresour- ces, within a limitted 

timescale. 

Every E player is 

endowed wwith 

a GPS-eq quipped tablett 

PC providing p meddium-range 

coommunication 

n capabilities, , 

such h as WLAN. A digital mapp 

of the game area is storedd 

on the t device. Fuurthermore 

evvery 

player features 

certainn 

(sim mulated) abilities 

and posssesses 

prede efined assets. . 

Whi ile moving thhrough 

the ggame 

area ph hysically, thee


player can innteract 

with thhe 

simulated wworld 

by usinng 

his 

or her abilitiies 

or by perrforming 

certaain 

actions, wwhich 

will result inn 

environmental 

or objectivve 

changes. AAssu 

ming the devvice 

will inform 

the playerr 

about a grouup 

of 

native populaace, 

which haave 

gathered iin 

the area shee 

just 

entered, by uusing 

her dipplomatic 

or laanguage 

skillss 

she 

can try to gaather 

information 

about the games objecttives. 

In the Treasuure 

Island sceenario, 

it is noot 

possible too 

suc- 

ceed in accomplishing 

thee 

mission alonne. 

The task oof 

the 

players is to learn that coooperation 

will 

lead to winnning 

the game fasster 

and with lless 

required rresources 

thann 

in a 

competing ennvironment. 

Figuure 

4. Treasure Island Scenario ffor 

ncoML 

The basic design of thee 

learning frammework 

is divvided 

into two diffferent 

parts: The “framewwork” 

part andd 

the 

“scenario” paart. 

Both partss 

are completeely 

separated from 

each other, thhough 

withouut 

a scenario, tthe 

frameworkk 

will 

not provide any challengge 

to the plaayer, 

and witthout 

framework aany 

scenario iss 

purely theorretical. 

Scenarrio 

as 

well as mapp 

information is provided to the systemm 

via 

XML configguration 

files. 

They descrribe 

the simuulated 

“world” the pplayers 

can innteract 

with, ass 

well as the aavail 

able resourcees 

(abilities & assets). 

The frameework 

providdes 

the user interface andd 

the 

game map. FFurthermore 

iit 

calculates thhe 

success orr 

fail- 

ure of decissions 

and upd dates the scennario 

accordingly. 

Most importtantly 

the frammework 

provvides 

the tools 

the 

users will neeed 

to communnicate 

and exccel 

in fulfillinng 

the 

mission as a team (chat, wwhiteboard, 

etcc.). 

The frameework 

allowss 

the analysiss 

of technicall 

and 

group dynammical 

effects in a mobile eenvironment, 

with 

players phyysically 

situatted 

in an auugmented 

reality. 

According too 

scenario or ssupervisor 

preeference, 

the ccolla 

borative leveel 

can be adjuusted 

from alllowing 

no usser 

to 

share and access 

any inforrmation, 

up too 

the possibiliity 

to 

display otherr 

player’s actiions, 

featuress 

and status onn 

the 

map without hiding any innformation. 

Onne 

important qques 

tion whose aanswer 

can be derived fromm 

the logged sstatis 

tic data is e.gg. 

the consequuence, 

an in- or decrease oof 

the 

collaborationn 

level will have on gaame 

success and 

efficiency. 

NCOml iss 

implementedd 

in JAVA as well and doees 

not 

require a ccentral 

commmunication 

innfrastructure, 

like 


WLA AN access pooints, 

but is bbased 

on mult ti-hop ad hocc 

com mmunication, pprovided 

by thhe 

simulation n environmentt 

JAN NE, which willl 

be describedd 

in more detail 

in a follow- 

on paragraph. p 

Central C part off 

the learningg 

framework is i a graphicall 

user r interface, whhich 

allows thee 

player to int teract with thee 

simu ulated world while physiccally 

passing g through thee 

gam me area. In thee 

first prototyype, 

the middleware 

JANEE 

allow wed to simulate 

the ad hooc 

network within w a LAN. . 

Here e physical moovement 

was simulated by y allowing too 

mov ve a virtual chharacter 

thoughh 

the game ar rea by using a 

remo ote-control-likke 

add-on too 

the user in nterface. Thee 

current 

version oof 

the framewwork 

runs on n Tablet-PCss 

usin ng the JANE pplatform 

modee. 

All A rules, objeccts, 

assets andd 

abilities with hin the systemm 

are provided by an external XML configuration 

file, , 

whic ch is uploaded 

to the playyer’s 

device on 

system andd 

gam me start. Gamee 

objectives annd 

detailed in nformation aree 

henc ceforth availaable 

to the pllayer. 

Objects 

include thee 

diffe erent team-meembers 

and thheir 

attributes and abilities, , 

as well w as simulaated 

real-worrld 

objects, th he player cann 

inter ract with. 

Actions A include 

using or intteracting 

with h objects, mo- 

difying 

them, pickking 

them up or dropping them, etc. In- 

terac cting with thee 

simulated rreal 

world usu ually requiress 

rand domized abilitty 

checks. All abilities are distributed d onn 

a pe ercentage scalee; 

where 0% mmeans 

no kno owledge/profi- 

cien ncy about the ttopic 

and 100% % ability mastery. 

Externall 

influ uences like (siimulated) 

weaather- 

or terrai in-conditions, , 

utiliz zation of diffeerent 

assets, eetc. 

can modify fy certain basee 

abili ity scores. E. .g. player “CCpt. 

Jacks” “observation” 

“ ” 

abili ity (50%) cann 

be modifiedd 

by a “spygl lass (observa- 

tion x 1.5)” to 775%, 

since thhe 

spyglass modifies m any-- 

body y’s observatioon 

ability by tthe 

factor 1.5 5. Deciding too 

use the spyglass tto 

search for his lost shove el, but rollingg 

an 85%, 8 will result 

in not finnding 

it durin ng the currentt 

roun nd. Upon enteering 

a certainn 

part of the game g area, ex- 

istin ng terrain andd 

environmentt 

factors are automaticallyy 

affec cting defined abilities and aassets 

(like fog 

affects visi- 

bility y, trails or waater 

courses mmobility, 

etc.). 

Every E interaction 

between pplayers 

will co ommence in a 

certa ain amount off 

time (round) , – with either r a positive orr 

a ne egative result. Again, this tiime 

is random mized and mo- 

difie ed according tto 

the skill levvel 

of the user. 

During thiss 

time e, the player ccan 

not start a new intera action withoutt 

abor rting the runnning 

processs. 

The game e will finishh 

succ cessfully if thhe 

player willl 

fulfill certain 

predefinedd 

crite eria (objectivees). 

If these tasks are no ot performedd 

with hin a predefineed 

timeframe, , or if certain negative con- 

ditio ons occur, the player group will lose the game. g 

All A player inpuut, 

start and ennd-states, 

as well w as all in- 

term mediate resultss 

of any actioon 

or ability are a stored in a 

log file, f which is aautomaticallyy 

uploaded to the t supervisorr 

syste em after eithher 

failure or success of the t game forr 

furth her analysis. 

III. 

LOCATION IMMMERSION 

VS. REALITY AUG GMENTATION 

A. Location L Immeersion 

in UbiQQuest 

Using U the device’s 

availablle 

localization n services re- 

sults s in a close innterdigitation 

of reality an nd simulation, ,


since the environment, the player respectively the device 

is situated in, directly influences the games state and statistics. 

This is typical for a location immersion technique. 

The basic principle in UbiQuest is: The more positive the 

real environment is for the player, the more the in-game 

development benefits from. This technique is supposed to 

motivate the player to situate him or herself in a positive 

environment. E.g. as a basic examples on a university 

campus physically moving in the park-like environment 

between the lecture halls grants a boost to the birth rate of 

the ships population, while being located in less hospitable 

surroundings decreases it. 

On every device an XML-based map of the game area 

is stored, containing information about the different types 

of “space” the star-ship passes through. Every area-type 

encompasses several modifiers to in-game characteristics. 

If the ship enters, respectively leaves an area, these modifiers 

are applied or removed from the list of current modifiers. 

Areas can overlap as well, e.g. a parking lot near a 

big building could be represented by an asteroid field in 

the vicinity of a large planet. In every area additionally 

different kind of random encounters can occur with a predefined 

probability, like the discovery of alien artifacts or 

space pirates. 

B. Reality Augmentation in ncoML 

In ncoML as well a digital map of the player’s surroundings 

is available. Here the university campus (for 

the Treasure Island scenario) or a military training area 

etc. is depicted within the device, augmented by virtual 

information relevant for the scenario (e.g. the player can 

meet a group of aborigines at a certain location, which 

does not exist in reality). 

The ncoML framework keeps track of the player’s 

physical position and maps it to the position within the 

game, while supplying him or her with corresponding (simulation) 

information. This localization technique focuses 

on the enrichment of the reality with additional data 

and does not directly influence in-game characteristics. 

This is done implicitly by the users, whose actions are governed 

by the information he perceives or is shown by the 

framework. 

IV. EXPERIENCESS 

A. UbiQuest 

During the tests within the simulation environment 

using scripted building strategies and moving according 

to a mobility source adapted from the random waypoint 

model, no abnormal system behavior was detected. After 

porting and testing the software on the PDA devices the 

first real but expected problems were encountered. In the 

near “perfect” virtual environment, we had constant communication 

ranges of the devices. In the “real” world, 

every infrastructural change in the environment influenced 

the communication capabilities of the devices as 

well. Due to multi-user interference, shadowing in the 


network etc., close-by neighboring devices were sometimes 

not even discovered by the beaconing protocol. 

The reception of GPS signals posed another challenge. 

Location awareness was integrated using GPS receivers 

without the more modern SIRFstar III chipset which resulted 

in deviations of the devices from their original location, 

slow updates (more than 30m range and latency 

values of more than several seconds) and black spots, 

caused e.g. by shadowing. Since location awareness was 

implemented using GPS receivers, only outdoor usage 

was possible. For indoor use, WLAN signal strength triangulation 

techniques have been conceptualized but were 

not implemented. 

Finally, due to resource restrictions only five PDAs 

were available for the field tests. Fortunately, the feedback 

from the testers were positive. The integration of the 

physical location into the game was a feature appreciated. 

Though “increased population growth” did not motivate 

the players to really position themselves in the mentioned 

“positive” areas, the general idea of location immersion 

in ubiquitous games and applications in general was 

highly approved. 

B. ncoML 

NcoML has been tested currently only with military 

personnel. Using it with a game scenario and a totally different 

group of users in the context of the children’s university 

will be an interesting experiment. The results so 

far show, that simulating movement in an office environment 

and physical movement through an exercise area 

with augmented information really does make a difference. 

Though in the simulated world the movement 

speed, as well as the special behavior of the individual 

user have been adapted, all test results in a real environment 

show a significant (positive) decrease in mission 

time [Fig. 5]. The trainees do not have to make any necessary 

abstractions, but can concentrate directly on the 

mission success ahead of them. 

Other results from the simulation with a smaller scenario 

show that mission success increased drastically by utilizing 

higher collaborative levels: Alone the ability to 

communicate in a bidirectional way decreased mission 

time by nearly 33%; introduction of additional collaborative 

tools like white board functionality, etc. again nearly 

halved the mission time. [Fig. 6]. 

Technical results were less promising though. Using a 

level training area without any obstacles yielded the best 

communication results. As expected, buildings, trees and 

other obstacles influenced the WLAN connectivity between 

the different devices drastically, resulting in smaller 

communication radii and connection losses due to interferences 

and shadowing. Field tests have been performed 

with FujitsuSiemens Tablet PCs with integrated WLAN 

and Bluetooth Holux GPSlim 236 receivers. Since the 

original ncoML framework has been developed for outdoor 

usage, no indoor localization methods have been 

examined.


00:36:00 

00:28:48 

00:21:36 

00:14:24 

00:07:12 

00:00:00 

Simulation vs. Reality 

0 1 2 3 4 5 6 

Group (1‐5) 

Figure 5. Comparison of simulated an real field test results 

Figure 6. Relationship collaboration level – mission time 

V. MIDDLEWARE CONCEPT 

Simulation 

Reality 

Collaboration Level vs. Mission Time 

00:20:10 

00:17:17 

00:14:24 

00:11:31 

00:08:38 

00:05:46 

00:02:53 

0 1 2 3 4 

Collaboration Level 

JANE is a simulation environment for multi-hop ad 

hoc networks (MANETs) [6, 7], comparable to Ns/2 [8] 

or GloMoSim/QualNet [9, 10] which has been developed 

at the University of Trier. It provides network developers 

with a middleware, which can be used to easily implement 

protocols and applications. The key ingredients of 

JANE are component based design and asynchronous 

communication between these components. This approach 

has been chosen because the traditional view of 

the network stack and synchronous communications does 

not fit well in a highly dynamic ad hoc network. 

Within JANE every component is a service. This is valid 

for applications and basic middleware services, as 

well as for routing algorithms and even hardware components 

(e.g. a GPS receiver). Services cannot be accessed 

directly though - this is accomplished by an event-based 

communication pattern, supplied by an abstracted operating 

system, which provides service management and – 

interaction, event execution management and timeout management. 

A network simulated in JANE is implemented as a set 

of services and event handling objects. The network itself 

is represented on the link layer OSI-level. It provides 


basic asynchronous message oriented unicast- and broadcast 

communication with direct neighbors in an ad hoc 

environment. Besides a collision free and a shared network, 

the detailed implementation of an IEEE 802.11 

MAC protocol is available too. 

Atop of the chosen network, JANE provides a generic 

routing framework, allowing the network developer to 

implement, combine and test existing as well as new routing 

services according to his or hers research interests, 

e.g. by combining DSR with geographic greedy routing 

[11]. 

Development of mobile applications using JANE as simulation 

environment or middleware component follows 

a 3-tier development process. The first is the implementation 

within the “pure” simulation environment. Since the 

main purpose of JANE is to support developers in implementing 

applications in an ad hoc environment, it allows 

to connect real hardware in a hybrid-mode [Fig. 5] as 

well. Here real world devices, like PDAs or Tablet-PCs 

that are physically connected to the simulation server can 

be transparently merged into the simulated environment, 

while the integrated middleware e.g. emulates WLAN 

802.11 within a local area network. The only new software 

modules that have to be developed within this second 

development step are the necessary user interfaces 

for direct interaction with the system. 

To use the hybrid mode, every physical device has to 

be mapped to a simulated device. Messages destined to 

the virtual device will be transparently forwarded to the 

real world device. Technically, these connections are 

realized using JAVA RMI (Remote Method Invocation), 

allowing to connect “external” devices on any physically 

attached network. 

(virtual) 

WLAN 

LAN 

Mobile Device 

JANE Simulation 

Environment 

simulated devices 

Figure 7. The JANE HybridMode 

JANE 

MW 

Mobile Device 

JANE 

MW 

WLAN 

Figure 8. The JANE Platform Mode 

physical devices 

physical devices 

Mobile Device 

JANE 

MW


For complete integration of real devices and as the last 

step in a development process, JANE offers a so-called 

platform mode, which can act as a middleware for multihop 

ad hoc networks as shown in Fig. 6. Since a graphical 

user interface has already been developed for the hybrid 

mode, porting applications to real devices is an easy and 

fast endeavor. 

As mentioned above, all applications (UbiQuest and 

ncoML) have been developed within the simulation environment 

and subsequently have been ported to use 

JANEs hybrid-mode. Additionally UbiQuest has been implemented 

and field-tested on GPS-and WLAN equipped 

PDAs using the JANE platform mode. 

VI. RELATED WORK 

In the domain of location-aware ubiquitous games, several 

projects with different objectives are currently researched. 

They differ in the type of location awareness 

used, ranging from augmented- and mixed reality to location 

immersion techniques. Additionally they can be classified 

by the technical implementation of location awareness. 

Many projects are using GPS receivers for positioning, 

but RFID, IR-barks and triangulation technologies 

are also common – especially for indoor applications. A 

few systems are combining different types of technologies. 

The communication capabilities are heterogeneous 

as well, ranging from close range ad hoc Bluetooth-, over 

WLAN- to the integration of hybrid and larger infrastructural 

networks, via GPRS, UMTS, etc. The most often 

used platforms for pervasive games are subdivided 

between handheld/PDA-sized devices down to mobile 

phone which gain more and more functionality each day. 

The other end is characterized by Tablet-PCs, sub-notebooks, 

UMPCs and portable gaming platforms (e.g. Playstation 

Portable). 

Prominent projects regarding ubiquitous games are 

“Uncle Roy all around you” and its predecessor “Can you 

see me now” [12, 13, 14]. Both games are played outdoors 

and are using GPS devices respectively manual positioning 

and are a mixture between reality- and online 

game, dealing with a portation of the well-known children’s 

game “Catch me if you can”. Communication 

means are WLAN respectively GRPS-able PDAs. 

Catchbob! [15, 16] is a collaborative game, in which 

groups of three persons have to locate a virtual object on 

the Lausanne University campus. It runs on WLAN 

enabled Pocket- and Tablet PCs and uses SOAP to 

communicate game data with a central university server. 

An early pervasive adaption of an existing “normal” 

strategy game is “Pirates!” which is played indoors on a 

PDA and uses special close-range transmitters for location 

awareness (allowing only relative localization) and 

WLAN communication [17, 18]. 

A project based on even higher technical prerequisites 

in the realm of augmented reality is “Human Pacman”, 

where the player wears a head-mounted display, providing 

him with information about virtual cookies he has to 

gather and ghosts he has to avoid, while physically walking 

or running though a virtual environment [19]. Communication 

means here are Bluetooth and WLAN. In 


New York a less technically demanding version of this 

idea was life tested in 2004 [20]. Information about position 

and ghosts/cookies was here communicated “primitively” 

via mobile phones. 

GeoCaching [21] is the hunt for a hidden “treasure” for 

which only real world coordinates are available, but not 

the exact location, using only a GPS receiver. Though it 

does not necessarily require a ubiquitous computer device 

(and thus not necessarily fitting the description of a ubiquitous 

computer game) it nonetheless can be 

implemented perfectly on mobile devices combining 

map- and GPS functionality. 

Mogi [22] is a typical example for ubiquitous game 

hype in Japan for several years, though it is just a “simple” 

collection and communication game. It uses GPS as 

well as cell-based localization services and runs on mobile 

phones. 

VII. CONCLUSION 

This paper presented two location aware ubiquitous 

games for multi-hop ad hoc networks. Though both are 

using location awareness as a new feature within the 

game, both are using it in different ways. For application 

development, the scope and planned deployment setting 

governs how to integrate location awareness into them. 

This always is a specific decision and not every application 

can take advantage of these available new features. 

Technically it is possible to integrate localization services 

of the different kinds into ubiquitous games, though 

some aspects still require additional in-depth research. 

Most applications only concentrate on one specific technology, 

though especially the transition from in- to outdoor 

scenarios and vice-versa is an interesting and challenging 

task. The same is valid for the networking capabilities. 

Different physical situations might provide hybrid 

networks, not only relying on ad hoc WLAN connections 

that could be used for in-game communications 

as well. An interesting approach for that is e.g. the pervasive 

game UbiSettlers, developed by the University of 

Luxembourg [23] which intelligently combines ad hoc 

and infrastructural communication means. 

In general using location based services within ubiquitous 

games is not only a feature which is technically possible, 

but which provides new game content, interaction 

possibilities, physical movement and last but not least fun 

– the most important ingredient in a mobile game. 

ACKNOWLEDGMENT 

The author wishes to thank Prof. Rothkugel (Computer 

Science Department of the University of Luxembourg) 

for providing additional hardware resources. This work 

was supported in part by the German Airforce Office 

(ncoML). 

REFERENCES 

[1] A. Höhfeld, “A ubiquitous m-learning framework for 

Network Centric Operations” in Proceedings of the 5 th 

International Conference on Pervasive Computing and 

Communications, New York, 2007, pp. 171-176.


[2] C. Hutter, „Interaktionsformen in Selbstorganisierenden 

Anwendungen aus dem Bereich Ubiquitous Computing“ 

Diploma Thesis, University of Trier, Trier, 2002. 

[3] D.B. Johnson, D.A. Maltz, “Dynamic Source Routing in ad 

hoc wireless networks”, in Mobile Computing, Kluwer 

Academic Publishers, 1996. 

[4] H. Fray, J.K. Lehnert, D. Görgen, P. Sturm, “A Generic 

Background Dissemination Service for Mobile Ad Hoc 

Networks”, Technical Report, University of Trier, Trier, 

2004. 

[5] Network Centric Operations & Warfare. 

http://ncow.nps.edu/wiki. 

[6] D. Görgen, H. Frey, C Hiedels, “JANE – A Simulation 

Platform for Ad Hoc Network Applications” in Demos of 

the 9 th International Symposium on Modeling, Analysis and 

Simulation of Wireless and Mobile Systems, Torremolinos, 

2006 

[7] J.K. Lehnert, D. Görgen, H. Frey, P. Sturm, “A Scalable 

Workbench for Implementing and Evaluating Distributed 

Applications in Mobile Ad Hoc Networks” in Proceedings 

of the Western Simulation MultiConference, San Diego, 

2004 

[8] The Network Simulator ns-2. http://www.isi.edu/nsnam/ns 

[9] L. Bajaj, M. Takai, R. Ahuja, K. Tang, R. Bagrodia, M. 

Gerla, “Glomosim: A scalable network simulation environment”, 

Technical Report, UCLA CS Department, Los 

Angeles, 1999 

[10] Scalable Network Technologies: Qualnet family of 

products. http://www.qualnet.com 

[11] H. Frey, “Geographical Cluster Based Routing with Guaranteed 

Delivery” in Proceedings of the International 

Conference on Mobile Ad Hoc and Sensor Systems 

MASS’05, Washington DC, 2005 

[12] S. Benford, R. Anastasi, M. Flintham, A. Drozt, A. Crabtree, 

C. Greenhalgh, N. Tandavanitj, M. Adams, J. Row- 

Farr, “Coping with uncertainty in a location-based game” 

in IEEE Pervasive Computing 2003, Vol. 2, Issue 3, pp. 

34-41, 2003 

[13] S. Benford, M. Flintham, A. Drozt, R. Anastasi, D. Rowland, 

N. Tandavanitj, M. Adams, J. Row-Farr, A. Oldroyd, 

J. Sutton, “Uncle Roy All Around You: Implicating the 

City in a Location-Based Performance” in Proceedings of 

the International Conference on Advances in Computer 

Entertainment ACE’04, Singapore, 2004 

[14] S. Benford, C. Magerkurth, P. Ljungstrand, “Bridging the 

Physical and Digital in Pervasive Gaming” in Communications 

of the ACM, Vol. 48, Issue 3, pp. 54-57, 2005 

[15] N. Nova, F. Girardin, P. Dillenbourg, “The Underwhelming 

Effects of Automatic Location-Awareness on Collaboration 

in a Pervasive Game” in Proceedings of the International 

Conference on the Design of Cooperative Systems 

COOP’06, Carry-le-Rouet, 2006 

[16] N. Nova, F. Girardin, P. Dillenbourg, “Location is not 

enough! An Empirical Study of Location-Awareness in 


Mobile Collaboration” in Proceedings of the International 

Workshop on Wireless and Mobile Technologies in 

Education WMTE’05, Tokushima, 2005 

[17] S. Björk, J. Falk, R. Hansson, P. Ljungstrand, “Pirates! 

Using the Physical World as a Game Board” in Proceedings 

of the International Conference on Human and Computer 

Interaction InterAct ‘01, Tokyo, 2001 

[18] J. Falk, P. Ljungstrand, S. Björk, R. Hansson, “Proximity- 

Triggered Interaction in a Multi-Player Game” in Extended 

Abstracts on Human Factors in Computing Systems CHI 

’01, pp. 119-120, 2001 

[19] A.D. Cheok, K.H. Goh, W. Liu, F. Farbiz, S.W. Fong, S.L. 

Teo, Y. Li, X. Yang, “Human Pacman: A Mobile, wide 

area entertainment system based on physical, social and 

ubiquitous computing” in Personal and Ubiquitous Computing, 

Vol. 8, Issue 2, pp. 71-81, 2004 

[20] M. Delio, “Manhattan Gets Pac-Man Fever” in Wired 

News: 

http://www.wired.com/gaming/gamingreviews/2004/05/63 

400, 2004 

[21] GeoCaching – The Official Global GPS Cache Hunt Site. 

http://www.geocaching.com 

[22] B. Joffe, “Mogi – Location and Presence in a Pervasive 

Community Game” in Proceedings of the International 

Conference on Ubiquitous Computing UbiComp ’05, 

Tokyo, 2005 

[23] C. Hoff, U. Wehling, C. Hiedels, S. Rothkugel, “UbiSettlers 

– A dynamically adapting multi-player game for hybrid 

networks” in Proceedings of the 5 th International Conference 

on Pervasive Computing and Communications, 

New York, 2007 

Alexander Höhfeld was born 1975 in Erlangen/Germany 

and graduated in computer science at the University of the Federal 

Armed Forces in Munich/Germany, focusing on operations 

research and self-organization, receiving his diploma in 

1999. 

He is a former German Air Force Officer who has been working 

in Germany, the Baltic States and the US in the development 

of military software for Germany and the NATO. After 

having finished his military career in 2006 he began writing his 

Ph.D. thesis at the University of Trier, where he is currently 

working in the computer science department/system software 

and distributed systems. His professional interests include selforganized 

and distributed systems, e/m-learning, MANETs, 

ubiquitous computing and the 3D-Internet. 

Mr. Höhfeld is member of the German computer society GI 

(“Gesellschaft für Informatik”).

View - Academy Publisher

Create successful ePaper yourself

Delete template?

Save as template?